DOCUMENT RESUME ED 270 509 UD 024 855 AUTHOR ... - ERIC

DOCUMENT RESUMEED 270 509 UD 024 855AUTHORHendricks, Glenn L., Ed.; And OthersTITLEThe Hmong in Transition.INSTITUTION Center for Migration Studies, Inc., Staten Island,N.Y.; Minnesota Univ., Minneapolis. Southeast AsianRefugee Studies.REPORT NO ISBN-0-913256-94-3PUB DATE 86NOTE460p.; Papers originally presented at the HmongResearch Conference (2nd, Minneapolis, MN, November17-19, 1983).AVAILABLE FROM Center for Migration Studies, 209 Flagg Place, StatenIsland, NY 10304 ($17.50 cloth, $12.95 paper; 10 ormore 20% discount).PUB TYPE Speeches/Conference Papers (150) ReportsResearch /Technical (143) -- Books (010)EDRS PRICEDESCRIPTORSIDENTIFIERSMF01 Plus Postage. PC Not Available from EDRS.*Acculturation; Asian Americans; *Cultural Traits;Economic Factors; Family Attitudes; *Health; HealthServices; Immigrants; *Lao; *Laotians; SocialAttitudes; Social Integration; Social Services*Hmong; *Hmong People; Southeast AsiansABSTRACTPapers presented at a 1984 conference on research onthe Hmong in the United States are collected in this book. The 24papers are organized around four central themes (among culture andchange, adaptation to a new society, language and literacy, andhealth care issues). Titles and authors are: (1) "Stages of HmongCultural Adaptation" (William A. Smalley); (2) "The Hmong of Laos:Economic Factors in the Refugee Exodus and Return" (Robert Cooper);(3) "Processes of Identity Maintenance in Hmong Society" (TimothyDunnigan); (4) "Culture and Adaptation: Hmong Refugees in Australia"(Gary Yia Lee); (5) "The Miao in Cnntemporary China" (Louisa Schein);(6) "Geomancy as an Aspect of Upland-Lowland Relationships" (NicholasTapp); (7) "Training Hmong Women: For Marginal Work or Entry into theMainstream" (Sarah R. Mason); (8) "Changes in Hmong Refugee MaritalAttitudes in America" (William H. Meredith and George P. Rowe); (9)"Resolving Sexual Assault: Hmong and the American Legal System" (BethL. Goldstein); (10) "The Hmong in Isla Vista: Obstacles andEnhancements to Adjustment" (Catherine Stoumpos Gross); (11) "FactorsContributing to a Split within a Clientelistic Needlework CooperativeEngaged in Refugee Resettlement" (Nancy D. Donnelly); (12) "The AmongResettlement Study: A Symposium" (Includes six short pages by severalauthors); (13) Cooperative Engaged in "Two-Word Expressives in it*iteamong" (Martha Ratliff); (14) "Verb Serialization in Hmong" (LaurelOwensby); (15) "The Morpheme 'Rom': A First Analysis and Look atEmbedding in Huang" (Annie Jaisser); (16) "Zero Anaphora and TopicProminence in Mang" (Judith Wheaton Fuller); (17) "Investiga"cingLiteracy: Approaches, Tools, tnd Their Consequences for Inquiry"(Gail Weinstein); (18) "Factors in Individual Acquisition of English:A Longitudinal Study of Hmong Adults" (Karen Reed Green and StephenReder); (19) "Shamanism in the Context of Hmong Resettlement" (JaquesLemoine); (20) "Guidelines for Mental Health Professionals to Helpamong Clients Seek Traditional Healing Treatment" (Bruce ThowpaouBliatout); (21) "among Perception of Illness and Traditional Ways of

Healing" (Xoua Theo); (22) 'Sleep Disturbances and Sudden Death ofHmong Refugees: A Report on Field Work Conducted in the Ban VinaiRefugee Camp" (Ronald G. Munger); (23) "A Cross-Cultural Assessmentof Maternal-Child Interaction: Links to Health and Development"(Charles N. Oberg, Sharon Muret-Wagstaff, Shirley G. Moore and BrendaCumming); (24) "Undue Lead Absorption in Hmong Children" (Karl Chunand Amos S. Deinard); (25) "Attitudes of Hmong toward a MedicalResearch Project" (Marshall Hurlich, Neal R. Holtan, and Ronald G.Munger). (KH)************************************************************************Reproductions supplied by EDRS are the best that can be made **from the original document. ************************************************************************

AzMR emurneeNT 101WATIONOleos et Edmore* Remene08, t end loprovementRONAL RESOURCES INFORMATIONCENTER (ERIC)document MO been reproduced esreceived from MR onion or oreeettattondriteneting it0 Mina ottenpe neve been made to improvereproduction qualityPointe of thew or opinions UMW In Me daceleantdo not newsier* represent °Mw eowe moon or policy"PERMISSION TO REPRODUCE THISMATERIAL IN MICROFICHE ONLYHAS BEEN GRANTED BYdio F; Toni asiC.44 eaTO THE EDUCATIONAL RESOURCESINFORMATION CENTER (ERIC)."

THE HMONG IN TRANSITIONedited byGlenn L. HendricksBruce T. DowningAmos S. Deinard1986Center for Migration Studies of New York, Inc.The Southeast Asian Refugee Studies of the University of Minnesota3

The Center for Migration Studies is aneducational non-profit institute founded inNew York in 1964 to encourage and facilitatethe study of sociodemographic,economic, political, historical, legislativeand pastoral aspects of human migrationand refugee movements.The opinions expressed in this workare those of the authors.THE HMONG IN TRANSITIONA joint publication of theCenter for Migration Studiesof New York, Inc. and theSoutheast Asian RefugeeStudies Project of theUniversity of MinnesotaFirst EditionCopyright © 1986 byThe Center for Migration Studies of New York, Inc.All rights reserved No part of this book may bereproduced without permission from the publishersCenter for Migration Studies209 Flagg PlaceStaten Island, New York 10304ISBN 0-913256-94-3 (hardcover)ISBN 0-913256-95-1 (paperback)Library of Congress Catalog Card No. 85-47918Printed in the USA4

TABLE OF CONTENTSPREFACEGlenn L. HendricksMap of LaosxixvPART 1HMONG CULTURE AND CHANGE 1INTRODUCTION 3Glenn L. HendricksSTAGES OF HMONG CULTURAL ADAPTATION 7William A. SmalleyThe Stages, 7; The People of Moos Plais, 9,Typical Innovations of the Stages, 15THE HMONG OF LAOS: ECONOMIC FACTORS 23IN THE REFUGEE EXODUS AND RETURNRobert Cooper*Introductory Note, 23; The Thesis, 23; The HistoricalEvidence, 27; The Observational Evidence, 30; TheEmpirical Evidence, 32; Summary, 37; Conclusion,38PROCESSES OF IDENTITY MAINTENANCE 41IN HMONG SOCIETYTimothy Dunn IgenIntroduction, 41; The Oppositional Process, 42; TheHomeland, 45; Language, 46; Religion, 47; Kinship,49; Identity Maintenance, 50CULTURE AND ADAPTATION:55HMONG REFUGEES IN AUSTRALIAGary Yia LeesThe Traditional Productivity Base, 56; Religion andSocial Structure, 57; Impact of Resettlement, 59;Cultural Consciousness and Adaptation, 61; HmongResearch and Social Change, 66; Conclusion, 69* Indicates invited Conference keynote speaker

viTHE HMONG IN TRANSITIONTHE MIAO IN CONTEMPORARY CHINA: 73A PRELIMINARY OVERVIEWLouisa ScheinIntroduction, 73; Ethnic Classification in China, 74;The Setting, 75; The Miao as a Nationality, 76;Language and Education, 78; Some Aspects ofTradition and Change, 81; Conclusion, 83GEOMANCY AS AN ASPEdT OF 87UPLAND-LOWLAND RELATIONSHIPSNicholas TappPART 2 ADAPTING TO A NEW SOCIETY 97INTRODUCTION 99Glenn L. HendricksTRAINING HMONG WOMEN:FOR MARGINAL'WORK ORENTRY INTO THE MAINSTREAMSarah R. MasonIntroduction, 101; Federal Assistance to RefugeeJob Training,103; Development of IndochineseWomen's Programs, 104; Employment Training forSoutheast Asian Refugees in the St. Paul-MinneapolisArea, 104; Employment Training for HmongWomen in the Metropolitan Area: An Assessment,106; Conclusion, 118101CHANGES IN HMONG REFUGEE MARITAL 121ATTITUVES IN AMERICAW iliam H. Meredith and George P. RoweMarriage in Laos, 121; Method, 124; Population, 125;Results, 126; Conclusion and Discussion, 129RESOLVING SEXUAL ASSAULT: HMONG AND 135THE AMERICAN LEGAL SYSTEMBeth L. GoldsteinTHE HMONG IN ISLA VISTA: OBSTACLES AND 145ENHANCEMENTS TO ADJUSTMENTCatherine Stoumpos Gross6

TABLE OF CONTENTSviiIntroduction, 145; Methodology, 145; Obstacles toAdjustment, 146; Factors Enhancing Adjustment,150; Outlook, 154FACTORS CONTRIBUTING TO A SPLIT WITHIN A 159CLIENTELISTIC NEEDLEWORK COOPERATIVEENGAGED IN REFUGEE RESETTLEMENTNancy D. DonnellyBackground, 160; Patron/Client Relationships, 162;Implications, 165THE HMONG RESETTLEMENT STUDY: 175A SYMPOSIUMINTRODUCTION 175Stephen RederPOPULATION TRENDS 179Douglas OlneySECONDARY MIGRATION TO CALIFORNIA'S 184CENTRAL VALLEYJohn FinckLANGUAGE ISSUES 187Bruce DowningHMONG EMPLOYMENT AND 193WELFARE DEPENDENCYShur Vang VangyiHMONG YOUTH AND THE HMONG FUTURE IN 197AMERICAMary CohnECONOMIC DEVELOPMENT ANDEMPLOYMENT PROJECTSSimon M. Fass202PART 3LANGUAGE AND LITERACYINTRODUCTION 213Bruce Downing

viiiTHE HMONG IN TRANSITIONA NOTE ON HMONG ORTHOGRAPHY 217TWO-WORD EXPRESSIVES IN 219wine HMONGMartha RatliffVERB SERIALIZATION IN HMONG 237Laurel OwensbyTHE MORPHEME "KOM": 245A FIRST ANALYSIS AND LOOKAT EMBEDDING IN HMONGAnnie Jalsser"Kom" as a Higher Verb, 245; "Kom" as aComplementizer, 250; The IdiomaticPhrase "Kom Tam", 258ZERO ANAPHORA AND TOPIC 261PROMINENCE IN HMONGJudith Wheaton FullerSentence-Level Zero Anaphora, 261; Discourse-Level Zero Anaphora, 262; Topic 1Prominence and Zero Anaphora, 268;Conclusion, 275INVESTIGATING LITERACY:APPROACHES, TOOLS, ANDTHEIR CONSEQUENCES FOR INQUIRYGall WeinsteinBackground, 279; Perspectives on the Role ofLiteracy, 280; Some Social Consequences ofLiteracy: Reflections of a Backyard Ethnographer,286279FACTORS IN INDIVIDUAL ACQUISITION 299OF ENGLISH: A LONGITUDINAL STUDY OFHMONG ADULTSKaren Reed Green and Stephen RederMethods, 302; Measuring Acquisition, 307; FactorsAffecting Acquisition, 314; Conclusions, 3238

TABLE OF CONTENTSixPART 4 HEALTH CARE ISSUES 331INTRODUCTION 333Amos S. DeinardSHAMANISM IN THE CONTEXT OFHMONG RESETTLEMENTJacques LemoineIntroduction, 337; Shamanism: A Hmong Way ofPsychotherapy, 339; Spirit Helpers and Shaman'sVocation, 340; The Shaman at Work, 343;Shamanism and the Psychomentil Complex of theHmong Refugees, 345337GUIDELINES FOR MENTAL HEALTH 349PROFESSIONALS TO HELP HMONG CLIENTSSEEK TRADITIONAL HEALING TREATMENTBruce Thowpaou BilatoutHmong Traditional Beliefs About Mental Health Problems,349; Types of Traditional Hmong Healers,355; Factors to Consider When Referring a HmongClient to Traditional Healers, 359; Conclusion, 363HMONG PERCEPTION OF ILLNESS AND 365TRADITIONAL WAYS OF HEALINGXoua ThaoHmong Shamanism, 366; Herbal Medicines, 366;The Concept of Soul Loss, 367; Ways of Calling theSoul Back, 368; The Mandate of Life, 370; NaturalCauses, 370; Organic Causes, 371; Traditional Useof Herbs, 371; Magical Causes, 372; Supernatural orSpirit Causes, 373; Ways of Dealing With Spirits, 373;The Ritual Called "Ua Neeb", 375. A Turning Point,376; Conclusion, 377SLEEP DISTURBANCES AND SUDDEN DEATH OF 379HMONG REFUGEES: A REPORT ON FIELDWORKCONDUCTED IN THE BAN VINAI REFUGEE CAMPRonald G. Munger3

xTABLE OF CONTENTSSudden Deaths of Southeast Asian Refugees in theUnited States, 380; The Ban Vinai Refugee Camp,383; Research Methods, 385; Case Reports, 386;Discussion and Conclusions, 393A CROSS-CULTURAL ASSESSMENT OF 399MATERNAL-CHILD INTERACTION:LINKS TO HEALTH AND DEVELOPMENTCharles N. Oberg, Sharon Muret-Wagstaff,Shirley G. Moore and Brenda CummingMethod, 399; Results, 403; Discussion, 410UNDUE LEAD ABSORPTION IN HMONG 417CHILDRENKarl Chun and Amos S. DelnardIntroduction, 417; Background, 417; Methods, 419;Results, 420; Summary, 422; Discussion, 423ATTITUDES OF HMONG TOWARDA MEDICAL RESEARCH PROJECTMarshall Hur itch, Neal R. Ho itpn andRonald G. MungerIntroauction, 427; Background, 427; Effects ofPrevious Research Experience on Current Projects,433; The Ban Vinai Experience, 436;Discussion, 439427CONTRIBUTORS 44710

PREFACEThe second Hmong Research Conference, held at theUniversity of Minnesota on November 17-19, 1983, drewfour hundred people. While this number was far beyondour expectations, perhaps we shouldn't have been surprised.The presence of more than sixty thousand Hmongin the United Statesall engaged in the struggle to existin a culture whose social, political and economicstructures are so different from their ownhas attractedattention from many groups. Among them are educators,academic researchers, shop owners, social workers, churchcongregations, appreciators of the arts, health care providers,landlords and neighbors. Their interests rangefrom seeking solutions for immediate resettlement problemssuch as providing medical care to theoretical considerationsof the structure of the Hmong language.The conference, entitled "The Hmong in Transition,"had as its focus the formal presentation of the researchpapers collected in this volume. It also encouraged informalexchanges between those involved in helping theHmong resettle in all parts of the country. How much helpis too much help, they asked of each other. Can theHmong adapt to a new culture and also preserve their traditions?What will become of Hmong children who arebeing taught one set of values at home and another atschool? Will the Hmong be caught in the vicious cycle ofpoverty that afflicts some other minority groups in thiscountry?When this book is published, it will be almost tenyears since the first mass exodus of Hmong from Laos,precipitated by the communist Pathet Lao takeover of theLaotian government. Scholars have now had almost a decadeto study the effects of this upheaval on the Hmongnot only in the United States but also in other places oftheir exile around the world. This volume includes informationabout Hmong in France, Australia and Thailand aswell as the United States. Particularly noteworthy isthat the Hmong, whose history was borne for centuriesonly in the hearts and minds and spoken words of its people,have now been surrounded by the written word.xi11The

x11THE HMONG IN TRANSITIONpapers prepared for this conference are but cne exampleof their incorporation into mankind's written records.The international scope of the Hmong dispersion wasreflected in the backgrounds of the three invited speakersto the conference: Jacques Lemoine is associated withthe School of Advanced Studies in the Social Sciences ofthe University of Paris; Robert Cooper is a British citizenworking for the United Nations High Commissioner forRefugees in Geneva; and Gary Yia Lee, a Hmong trained asa hropologist, is a leader of the Hmong community inAustralia.The conference was sponsored by the Southeast AsianRefugee Studies Project, a research oriented group of academics,established to encourage, coordinate, and supportresearch related to refugees from Southeast Asia whohave been resettled in the United States. It is fundedprimarily by the Center for Urban and Regional Affairs atthe University of Minnesota. Additional support for theconference came from the College of Liberal Arts. A generousgrant from the National Endowment for the Humanitiesenabled us to bring the three invited speakers tothe United States.The editors are particularly indebted to the editorialassistance of Ruth Hammond, who helped to bring consistencyand readability to the twenty-five papers includedin this volume. They were written by individualstrained in a variety of disciplines, each accustomed tousing widely varying writing formats and language in thepresentation of their research. Since w were aware thata large number of readers of this book would not be academic:,one of our chief aims in editing was to makereadable the sometimes esoteric, if not idiosyncratic,language some consider the hallmark of academia. RuthHammond's considerable experience as a writer and journalisthas been instrumental in helping us to achieve ourgoal of readability.This volume, a joint production of the Center forMigration Studies and the Southeast Asian Refugee StudiesProject is also a departure from usual production methodsin publishing a book. SARS not only provided the editedversion of the papers but also prepared the papers incamera-ready format. This demanded more than typicaltypist skills and we are indebted to Pamela Anderson,without whose abilities we could not have given this12

1PREFACEx111type-set appearance to the book. In addition, DouglasOlney provided a great deal of the technical knowledbe ofcomputers and word processing which allowed us to attemptthis format. The cartographic skills of Carol Gersmehlprovided us with the map indicating areas of Hmong settlementin Southeast Asia and the location of the majorrefugee camps in Thailand in which Hmong have beenhoused..For those seeking more background information aboutthe Hmong other easily available sources include the proceeding:,of the First Hmong Research Conference, publishedas The Hmong in the West (1982), published by theCenter for Urban and Regional Affairs, University ofMinnesota, 301 19th Avenue South, Minneapolis, Minnesota55455, plus these additional publications:Barney, G. Linwood. 1967. "The Meo of Xieng KhouangProvince, Laos." In Southeast Asian Tribes, Minorities,and Nations, edited by Peter Kunstadter, pp.271-294. Princeton: Princeton University Press.Center for Applied Linguistics. 1979. "Glimpses of HmongHistory and Culture." Indochinese Refugee EducationGuides, General Information Series #16. National IndochineseClearinghouse. Washington, D.C.: Centerfor Applied Linguistics.Dunnigan, Timothy. 1982. "Segmentary Kinship in an UrbanSociety: The Hmong of St. Paul-Minneapolis. AnthrologicalQuarterly, Vol. 55(3), pp. 126-136.Geddes, William. 1976. Migrants of the Mountains: TheCu !tura! Ecology of the Biue Miao (Hmong Njua) ofThailand. Oxford, Clarendon Press.Also, a discussion on Hmong orthography can be found inPart Three, Language and Literacy, on pages 217-218.- Glenn L. HendricksMinneapolis, MinnesotaAugust 198513

\ \ Sichuan / d\ I /Guizhou ' ) Ir 4 Hunan\1,,iYunnani ..,--,, II 4. % fr i'../ isi1 ') fit- n Pi ir 4: 1101111"- 1) r A-474r'f CHINA %-l.aro%L - 1 \ - -ITiJ, II... ..."/__.. .GuangxiIleb)?).C.) 'SAv.0Hmong HomelandsandRefugee CampsBURMABAN TONG,Chlengra.I,o Hue XayIoLuaam Neua0VIETNAMTonkin- National boundary- - - Provini a boundaryHmong homelandsRefugee campsNAM VA°oChuengmalSOP TUANGA0BO 4".IP le 0 ion KhouangSayabulyT HAOILANDOKI%hen laneNONG KHAI14200 Mi.360 gra.

PART ONEHmong Culture and Change15

INTRODUCTIONGLENN L. HENDRICKSA common theme that emerged throughout the conferencein both the papers and subsequent discussions wasthat of the identification of what is representive ofHmong culture and how it has changed in the period of thediaspora. Too frequently some observers and commentatorsof the Hmong have failed to understand that the conceptof culture implicitly assumes it is dynamic and that overtime varying degrees of change will take place in any society.The rate that the change takes place is a functionor the historical situation as well as attributes ofthe culture itself. The failure to understand this leadssome to think in terms of before and after, that therewas a traditional almost unchanging way of Hmong livingthat has been severely altered by the events of the periodof flight and subsequent resettlement. That Hmonglife has been radically altered cannot be disputed butthe change must be seen as a matte. of degree. The papersin this section provide a needed corrective in assistingus to gain a longer perspective of Hmong society, particularlyin the nineteenth and twentieth centuries.It should be noted that these papers, while focusingon aspects of Hmong society and culture, touch on somebasic issues in the social sciences. Namely, what is thenature of culture? What constitutes cultural change?What is the relationship between culture and ethnic identity?Can change be directed in such fashion that it becomesdisruptive of neither culture nor ethnic identity?We need not discuss these issues here but the carefulreader will discern distinct, if not opposing, viewpointsamong the authors of the papers that follow. But the issueof change is not necessarily the exclusive arena ofthe academic. A central question asked by most of thosewho are involved in the resettlement of Hmong refugeeshas been how to accomplish the process while allowingthem to maintain their distinctive identity as Hmong.These papers provide food for thought for those responsiblefor making decis;ons about the immediate issues316

4 THE HMONG IN TRANSITIONsuch as providing housing or deciding who can speak forthe Hmong.William Smalley, a linguist who first became involvedwith the Hmong over thirty years ago while workingon a project to provide a written form of the Hmong language,sketches out a series of stages that he sees aspart of the transformation of that part of the Hmong societythat migrated to the West. He points out thatchanges had begun to evolve long before the ultimateflight from Laos. Varying forms of individual and groupbehavior in the new setting can be understood, at leastpartially, as a function of events in the past. Contrastingpatterns of community leadership that have emerged invarious areas of resettlement around the United Statescan be traced to both the introduction of education inLaos and the way in which refugee camps in Thailand wereorganized.Robert Cooper, a British anthropologist who hasworked among the Hmong in both Thailand and Laos, presentsdata arguing that in addition to the commonly understoodpolitical reasons for the Hmong flight from Laosthere were also economic motivations for the move. Histhesis is that in examining historical, empirical and observationaldata there are grounds for concluding thatconditions in Laos were such that the land could nolonger support the traditional agricultural economicbasis of Hmong life. This paper proved to be the mostwidely discussed presentation of the Conference, withmany of the Hmong present disputing his interpretation ofthe facts and events cited.Timothy Dunnigan, an anthropologist who has beenkeenly interested in the Hmong since they first arrivedin the Twin Cities in 1975, carries our discussion of theimpact of resettlement on Hmong culture further. He discussesthe process by which they are maintaining theirethnic identity even while some of the attributes whichare said to be markers of being traditional Hmong are inthe process of change. In some cases what appear to besignificant changes in group organization, for example,he identifies as latent attributes of Hmong culture whichhave manifested themselves in previous situations ofHmong history when their ethnic identity was threatened.Gary Yia Lee, part of the small Hmong group that migratedto Australia, was educated there and earned his17

INTRODUCTION 5doctorate in anthropology at the University of Sydney.Subsequently he has been a leader of the Hmong communityof Australia. In both his presentation and his commentson others' papers he was able to give a valuable insider/outsider's perspective to our discussions. He arguesthat, in the Australian case at least, there are indeedsignificant changes in the economic basis of Hmong life,social structure and religious beliefs. However, thisneed not portend a destruction of their identity asHmong, but rather must be seen as an adaptation tochanging circumstances.A particularly valuable contribution to the participantsof this Conference was the presentation by LouisaSchein about the Miao (Hmong) in contemporary China. Itwas from this group, or at least parts of this group,that the Hmong of Laos, Thailand and Vietnam migratedduring the eighteenth and early nineteenth century.Nicholas Tapp, another British-trained anthropologist,who has worked for several years studying Hmong innorthern Thailand, examines geomancy, the practice ofdivination based upon geographical and spatial features,and its historical place as an attribute of Hmong culture.He points out that while the practice has its rootsin China, it is but one of many factors to be taken intoaccount in any discussion which assume a "ideal" or traditionalHmong culture against which any change might bemeasured.18

STAGES OF HMONG CULTURAL ADAPTATIONWilliam A. SmalleyAmericans who have had contact with Hmong in theUnited States are often aware that massive change hascome to these new immigrants as they have settled here.However, many of these same Americans are unaware of thechanges in Hmong life and culture which took place inLaos and in the Thailand camps over the last generation,and before.1, for example, had had some minimal contact withHmong in Laos in the early 1950s, but was puzzled at whatseemed to be very different leadership patterns exhibitedin Lao Family Community and its layers of Hmong organizationin the Twin Cities. I was slow to realize how muchthe Hmong learned during the years of military action,wartime migration and resettlement in Laos and camp lifein Thailand. On the other hand, I occasionally see theopposite extreme of this limited understanding in otherAmericans who write about aspects of Hmong life that developedin the disruptions of the sixties or seventies asthough they were "traditional."THE STAGESIn what follows I would like to suggest severalrather self-evident stages in the modern historicalchanges which are part of Hmong experience. Each stagemanifests certain typical innovations, some of which willbe pointed out later.'1. Laos penetration stage: The time when the Hmong weremoving into Laos, when they were oriented more toChina and/or North Vietnam, when there were relativelysmall numbers, when they had to begin to learn tocope with new ethnic groups and new power structuresaround them, notably those of the Lao and the French.7 19

8 THE HMONG IN TRANSITION2. Laos traditional stage: The period when the Hmonghad become a significantly large ethnic group inLaos, but when they kept to their mountains as muchas possible, and largely avoided contact with morepowerful groups.3. Laos adaptive stage: The period when a somewhatstable relationship had been estabished between theHmong and other ethnic groups, when they had learnedways of coping with such groups, arid when they wereseeking more aggressively to take a significant placein the larger life of the Lao state.4. Laos resettlement stage: The period when increasingpercentages of the Hmong population were forced outof their villages into resettlement conglomerationswhere they typically were mingled with peoples fromother ethnic groups, where the exercise of some traditionalcultural patterns was curtailed, and wherenew cultural forces were brought to bear on them withgreat suddenness and intensity.5. Thailand camp stage.6. United States resettlement stage.I have deliberately not set dates for these stages.They are not discrete periods of time, nor did they takeplace at exactly the same time for all Hmong in all partsof Laos. Some Hmong people, some villages, some familieswere in the Laos adaptive stage when other families, othervillages, other people were still in the Laos penetrationstage. In 1974, when he was an official in the Laocoalition ministry of planring, Yang Dao (personal communication)found Hmong in border areas of Laos who stillknew Chinese but not Lao. He discovered this case of theLaos penetration stage at the same time that many Hmongelsewhere were in the Laos resettlement stage.Most Hmong who have reached the United States belong,or did belong, to groups which went through all ormost of these stages. Presumably, very few of these individualsremember when their community was in the Laospenetration stage, but all the other stages have beenexperienced by the older people, and all Hmong in the20

STAGES OF HMONG CULTURAL ADAPTATION 9United States are products of this history, with someindividuals and groups having been more strongly influencedby developments in one stage or another.THE PEOPLE OF MOOS PLAISIn the early 1950s there was a cluster of four Hmongvillages two days travel (but only sixty miles directflight) north of Luang Prabang, the Royal Capital ofLaos. This cluster is usually called Moos Plais1 by itsformer inhabitants now in the Twin Cities. The villagesat Moos Plais were relatively stable and prosperous.There was enough land so that people did not have to migrate.They had a surplus rice crop, raised animals andgrew opium. I would like to illustrate briefly the stagesof adaptation by reciting a few bits of these maple'sexperience.At least three developments epitomized the transitionof Moos Plais from the Laos traditional stage to theLaos adaptive stage. The earliest to develop was regularand frequent trade relations with the Lao. The traditionalLaos period included extensive trade, of course.Bu in the Laos adaptive stage, Hmong did not depend soexclusively on traders coming to them; they also went toLao towns in very much larger numbers and with greaterfrequency.Next was cooperation with the French/Lao militaryeffort. Initially it was simply a matter of Hmong menpatrolling under the leadership of their own headmen toambush occasional bands of Pathet Lao and to defend theirvillages, but during the early 1950s the French/Laoforces set up a multi-ethnic garrison at Moos Plais, withten French officers led by a colonel. Lao officers andtechnicians were also stationed there, as were troopsfrom other mountain peoples in addition to Hmong. Helicoptersbegan to come with increasing frequency.An airstripwas built, shortening the trip from Luang Prabangto a few minutes. Eventually the Lao and French officersat Moos Plais were replaced by Hmong under the commandof General yang Pao, and the garrison was suppliedby American pilots.The third major indication of the Laos adaptivestage was in the establishment of a Lao school in Moos21

10 THE HMONG IN TRANSITIONPlais. This provided only two years of education, butfor the first time almost all of the boys and many of thegirls began to learn the rudiments of reading and writingLao, together with a little arithmetic and Lao civics. Asmall number of boys who showed aptitude and were otherwisesuitable then went to live with Lao contacts orfriends in nearby Lao towns to continue their education.The Laos Resettlement StageSome of the older men from Moos Plais now living inthe Twin Cities look back on the late period of the Laostraditional stage and the Laos adaptive stage as theirgolden age. They were prosperous; they sensed progressand change, but they were able to cope; they had a senseof accomplishment and of power.But by about 19673 the military garrison at MoosPlais was forced to withdraw, and most of the villagers(about 160 people) went with them. They left everythingthey could not carry or lead. After a day's walk theyreached Mom Phuv, a large Hmong village south of MoosPlais, which also had a military garrison and an airstrip.Their military garrison from Moos Plais joinedthat of Mom Phuv, and the civilians camped in tents suppliedby the military. They were completely dependent;all of their food was airlifted in.Thus began, on a small scale for these Hmong fromMoos Plais, the Laos resettlement stage. Other refugeesfrom nearby areas joined them, fleeing their own villages.Their settlement was a small example of whatAjchenbaum and Hassoun (1979) have called an "agglomerationde guerre," and what I will call a conglomeration.It was characterized by the intermingling of Hmong groupswho themselves did not know each other, in the same communitywith still other different ethnic groups (Khmu',Mien, Thai Dam, Lao). The military provided what overallorganization there was. In this particular case therainy season, the uncertainty about the future and thedagger made farming and other normal economic activityimpassible, so that all food and other supplies wereflown in.In scarcely two months the whole Mom Phuv conglomerationhad to move on again as the military situation oncemore became impossible and the combined garrison with-22

STAGES OF HMONG CULTURAL ADAPTATION 11drew. Their move formed a new conglomeration at TriosTub, another Hmong village with a garrison strategicallylocated in the mountains above the confluence of the NamOu (Lao: Ou River) with the Mekong, only twenty milesnorth of Luang Prabang. This conglomeration lasted sixmonths, during which time it was regularly surrounded byheavy fighting and suffered many casualties. Its numberscontinued to be swelled by more diverse groups whichjoined as they fled their earlier settlements and conglomerations.Finally, the civilians were flown out of Toos Tub toa safer destination, leaving the military forces to helpdefend Luang Prabang. They settled along the Nam Poui(Lao: Poui River), in an area that was across the MekongRiver from Moos Plais, toward the Thai border and aboutthirty miles southwest of the provincial capital ofSayaboury.4 The people flown in from Toos Tub joined asmall population of Lao and Hmong already in the area,which had not been appreciably affected by the war.Through further influx of displaced people, the populationin this new conglomeration eventually increased toeight thousand or nine thousand.Nam Poui was more systematically and more permanentlybuilt than the previous conglomerations in which itsnew residents had settled. Lao and United States aid officialslaid out the town in blocks.5 People from differentethnic groups lived in different parts of town.Within the Hmong section, the people originally from MoosPlais lived together and preserved the social structuresthey brought with them, under the same family and villageleaders.United States aid provided some animals for breedingand seed to plant both irrigated and mountain ricefields. Within a few months the people were growingtheir own food and were beginning to develop cash produce,primarily in pigs. The government and the Americanaid program extended a road into the community, helped inthe devel pment of irrigated lands and built publicbuildings, including a school. Lao, Chinese and Indianmerchants built stores and moved in.In many important ways the conglomeration at NamPoui was very different from the constricted and badlyoverpopulated larger conglomerations such as Long Cheng,in the heart of the war zone. Nam Poui had a similar23

12 THE HMONG IN TRANSITIONmixture of population (without the resident Americans)and a similar dominant military umbrella, but stresslevels were lower, and it became economically self-suff i-dent. Ultimately it did not have to be supplied withfood by the government, and it began to prosper.In terms of adaptation, Nam Poui shared with theother semipermanent conglomerations an increased exposureto nontraditional technology and material goods, greateropportunity for Lao education of the children, significantinteraction with non-Hmong, and an overarching politicalstructure which was supplied by the military, butwhich allowed traditional village and family structures(when they were still intact) to operate on a lower level.In this mixed environment, spurred by the needs createdthrough displacement, many Hmong rapidly learnedthings and became used to conditions they had never knownbefore.Nam Poui was settled about 1968. In 1975, someweeks after General Vang Pao left Laos, most of the conglomerationof Nam Poui left en masse for Thailand. Thecaravan of several thousand people crossed the border,was disarmed by the Thai, and was resettled in a locationthat became the camp called Sop Tuang.The Thailand Camp StageAt Sop Tuang, life was in some respects a continuationof the Nam Poui conglomeration. Most of the originalrefugees moved in together, and as they built SopTuang, they followed the organizational patterns they hadbeen using at Nam Poui. The same Hmong major who hadbeen in charge of the garrison at Nam Poui and who hadbeen the overall leader of the conglomeration was alsothe Hmong person in charge here under the Thai authorities.The same settlement by ethnic groups took place.These ethnic groups continued to be organized under theleadership each had brought with it. New refugee groupswere added from time to time, as had happened in NamPoui. They came as villages or other sizable, previouslyexisting groupsnot as individuals, small family groupsor parts of families.What was different about Sop Tuang, as compared toNam Poui, was that now the groups had constraints resultingfrom their status as refugees in a different country.2404:.;.4

STAGES OF HMONG CULTURAL ADAPTATION 13Placed above the social structure imported from Nam Pouiwas a level of authority responsible to the Thai government.A Thai military officer presided over the camp,and Thai soldiers were stationed there.Furthermore, in Nam Poui the Lao and United Statesgovernments had provided money, goods and personnel tohelp lay out and build the community and get the peoplestarted. But the people of the conglomeration had beenable to outgrow the need for extensive economic help. AtSop Tuang, after an initial period of inaction, the Thaigovernment, the United Nations High Commissioner for Refugees,and numerous religious and charitable agenciesgathered to provide food, medical care, education, sanitation,opportunities for economic development and essentialsupplies of material goods. Here was a loosely-knitgroup of organizations, sometimes cooperating, often competing,which the refugees had to learn to depend on forsurvival.The most significant difference from Nam Poui, sofar as the people were concerned, was that there was insufficienteconomic opportunity for them. Some men wereable to take advantage of the limited opportunities foremployment within and outside of camp, and women couldsell their handiwork to some extent,b but neither endeavorwas adequate to solve financial needs.The experience of the villagers from Moos Plais atCamp Sop Tuang was in several important ways differentalso from that of Hmong in the much larger, more denselysettled and more important camp of Ban Vinai. Refugeeswho arrived at Ban Vinai tended to come in small clusters,accompanied by only fragments of their socialgroupings. Unlike those at Sop Tuang, they often had totravel through Pathet Lao areas, had to hide out in thejungle for long periods, started out malnourished andarrived starving. They had to break up into small groupsto avoid detection, and many groups and individuals whoheaded for that part of Thailand never did arrive. Whenpeople did reach Ban Vinai there may have been members oftheir social groups already there, but the newcomers hadto take whatever housing location was available, andcould not live in groupings to which they were accustomed.So the lower rungs of leadership that existed underthe top levels of Thai control and the formal organiza-25

2614 THE HMONG IN TRANSITIONtion of Ban Vinai, were set up somewhat differently fromthose at Sop Tuang. Both camps were divided into sectors.In Sop Tuang the sectors were natural ones (interms of preexistent groupings) with people clustered insomewhat familiar ways.' In Ban Vinai people belonged towhatever sector they happened to be assigned housing in.8As families or villages were reunited at Ban Vinai, theydid, of course, reestablish older relationships, and aconsiderable part of the informal structure of the campwas made up of such groups in spite of their scatteredresidence. The formal structure, however, did not matchthe informal structure as well as they matched at SopTuang. An important related difference between Sop Tuangand Ban Vinai was in the part played by the educatedyouth at Ban Vinai. In the years immediately precedingthe initial evacuation of 1975, there had been about sixhundred Hmong students in Vientiane. They had formed astudent association as a mutual assistance group. Asthey evacuated to Thailand and eventually ended up in BanVinai they were very important in organizing and runningsome of the necessary functions of camp life. They assistedin food distribution, camp cleanup and sanitation,and they took over education of the children, as well asEnglish classes for everyone.It seems that they were in a strategic place for tworeasons: their education better prepared them for sonicaspects of this new situation, and the shredded socialstructures perhaps left a vacuum in leadership and potentialfor action on some levels of camp life. In BanVinai, as at Sop Tuang, the Hmong military provided theleadership at the top. Below that, in Ban Vinai, thefunctions served by the youth group were eventuallyabsorbed into the more formal structure of the camp or,in the case of education, were taken over by the Thai.Furthermore, the contribution was temporary because theeducated youth often went on to other countries soonerthan many others. But for a while they were critical tothe functioning of the camp.9At Sop Tuang, on the other ;land, the youth as agroup were of little importance in the operation of thecamp. They occupied a much more traditional relationshipto the community, I am : ,igesting, because community patternshad not been shredded in the migration, and alsobecause there was not as large a nucleus of students moreacculturated than their eiders as there was at Ban Vinai.

STAGES OF HMONG CULTURAL ADAPTATION 15The United States Resettlement StageThe only aspect of the United States resettlementstage on which I'll take time to comment is one regardingthe evolution of the role of the youth at Ban Vinai. Itis the difference in Hmong leadership and social organizationbetween Dallas-Fort Worth, Texas, and Minneapolis-St. Paul, Minnesota, reported by Mason and Downing (Mason1983; Downing 1983). In Dallas-Fort Worth, a strongyouth group operating under young leadership performsadaptive and supportive functions somewhat parallel tothose carried out by the youth groups of Vientiane, NamPhong and Ban Vinai. There is no mutual assistance associationcontrolled by elders. In the Twin Cities ofMinneapolis-St. Paul, however, although some of the leadersof the same group which was active in the camp arepresent, there is no significant youth leadership for theHmong community. Instead, the primary leadership comesfrom a variety of older types, including the former militaryand former people of influence in Laos, as well aspeople emerging in the American situation.As for the people of Moos Plais who are now in theTwin Cities, certain aspects of the social structurewhich they maintained throughout moves to Mom Phuv, ToosTub, Nam Poui and Sop Tuang still govern their relationshipshere. This is supplemented by tenuous ties to theLao Family Community and its network of relationships,all under the umbrella of an American system about whichthe people of Moos Plais are learning, but still understandvery little.TYPICAL INNOVATIONS OF THE STAGESHere follows, then, a sketchy list of the innovationswhich I see as being characteristic of the stages.I do not know much about the Laos penetration stage, butpresumably those Hmong who knew an outside language spokea Chinese language or Vietnamese, depending on where theycame from. They were small in number, kept much to themselves,and had infrequent contacts with the Lao orFrench.10In the Laos traditional stage Hmong had establishedtrade patterns with Haw, Lao and Chinese traders who came,,27

16 THE HMONG IM TRANSITIONto their villages, but there was minimal travel to Laotowns, minimal learning of the Lao language, and minimalcontact with the Lao/French military and government. Thecontact they did have included paying tribute that consistedof mountain and jungle products (Yang 1975:45).Yang Dao reports that the first low-level diet. ct officer(tasseng) was appointed as early as 1896, although itwas not until much later that this practice became wide-Hmong numbers, however, were increasing rapidly,spread.and some Hmong did experience some severe conflicts withthe French/Lao, notably the revo:e. of 1919 to 1921.Hmong cultivation of the poppy for the opium trade becameeconomically very important in the country.In the Laos adaptive stage, major changes, whichsignificantly began to alter the position and role of atleast some Hmong in Laos, took place. The Hmong presencenow became very noticeable as they traveled in groups toLao provincial centers to trade, becoming regular vendorsin the markets (Yang 1975:122-127), and as some of thembegan to live closer to these Lao towns. The provincialcapital of Xieng Khouang became an important Hmong c'ter. The establishment of Lao schools in Hmong villages'',and the practice of sending a few Hmong studentsto Lao towns to study, began in this period. Virtuallyall Hmong men now spoke some Lao, and Lao vocabulary borrowingslaced Hmong discourse. A few Hmong went toFrance to study, and a tiny educated Hmong elite began toemerge.A national Hmong leadership was epitomized at firstwith Touby Lyfoung, who became officially recognized asleader of the Hmong by the French/Lao goverment in 1947(Yang 1975:51), and by his rival Faydang Lobliayao, wholed the communist faction (Yang 1975:54). Later therewere Yang Pao and others. During this period the cooperationwith the French/Lao military and civilian authoritiesbecame strong among some of the noncommunist Hmong,and a sense of participation in the Lao state began toemerge. Hmong district heads (tasseng) were now frequentlyappointed.Christianity, both in Roman Catholic and in Protestantforms, made solid headway for the first time amongthe Hmong in Laos during this stage. Missionary effortshad existed long before that, but for some Hmong, conversionto Christianity now became a desirable option28

STAGES OF HMONG CULTURAL ADAPTATION 17(Barney 1957a,b).Some Hmong also began to make use of Western medicineas administered in Lao hospitals to a much greaterdegree than before, although Western medicalstill treatmenttended to be a solution of last resort, andqualitytheof the medical treatment available was not good(Dooley 1958:192ff; Halpern 1961).In the Laos resettlement stage, modern war came tolarge apercentage of the Hmong population in Laos withcruel force and disruption. The Hmong participation inthe war became intense. Hmong suffered terrible casualties,many villages being uprooted, some several times;and large parts of the population were resettled in ref u-gee conglomerations, often around military centers.12Normal agricultural and economic activity were often impossibleon any adequate scale, and in many of the centersHmong were forced to live by external aid plus whatevernew economic skills they could learn. Traditionalsewing skills were adapted to make items directly salableto Westerners.13For many Hmong, war and conglomerate life broughtparticipation in some forms of modern technology. Instrumentsof warand in some areas, those of transportationand agriculturebecame more widely known.tary Mili-participation and life in the conglomerationsbrought new social structures that overlaid thetionaltradi-ones still in place. Many military learned anEnglish-based code for communication with American officersand civilian pilots, while some Hmong learned tocommunicate in everyday English.-Westernization and adaptation to Lao culture wasintensified for some Hmong as they participated with Laoand Americans in the war effort, lived in the same conglomerationswith Lao and other ethnic groups, and in afew cas,ss became more educated in Lao.Hmong literacy increased slightly. In addition tothe rnmanized orthograthy, two Lao-based writing systemswere in use during the period. One was developed by theHmong who sided with theparts Pathet Lao and the otherof bythe Protestant church (Smalley 1976; Lemoine1972a). As economic and cultural deprivationtaryand mili-stress increased, there developed the Chao"Lord of the Sky") MessianicFa (Lao:movement, an attempt to recreatea golden past mixed with elements of the changingpresent, something which was uniquely Hmong. With it was. .t40On

18 THE HMONG IN TRANSITIONinvented a writing system unlike any of the others in usefor the language (Lemoine 1972a:142-146).The Thailand camp stage was in many ways an extensionof the conglomerate life of the resettlement areasin Laos. There were some important differences, ofcourse. Camp life brought contact with a different governmentand numerous foreign agencies. It brought artificialrestriction on movement, and even more limitedpossibilities for economic self-help. It also broughtsome relief from war pressures, although the pressurecontinued for those Hmong who slipped back across theborder to continue fighting or who had relatives in dangerin Laos.The degree to which traditional social structureswere shredded in camp life depended on the family historyof disaster. Some camps preserved the spatial proximityof traditional relationships from earlier village settlement.Ban Vinai in particular did not. For many Hmongthe camps brought the trauma of family separation of anew kind. Parts of extended families would leave totravel unimaginable distances to fearful and legendarydestinations in other parts of the world.The camps brought Thai education for children, andintensified aesire and opportunity for Hmong literacy.It also brought high motivation and increased opportunityfor learning English. Yang Dao (personal communication)states that at Nam Phong, the Thai camp to which- thefirst Hmong refugees were taken, there was an Englishclass taught by the few who knew some English "underevery tree."- Eventually some of this adult education wastaken over by social agencies from outside the camp.The Hmong who came to the United States, then, camewith a far more varied and complex experience than areading of Lemoine (1972b), set in the Lao adaptivestage, or Geddes (1976), set in an analogous stage inThailand, would lead us to expect. Some people had beenswept along by the changing events, without learning todo much more than minimal coping for survival. Othershad responded to the new situations by learning newskills and adapting creatively to new opportunities.The innovations of the United States resettlementstage are beyond the scope of this paper, even if theywere to be presented only in the sketchy form with whichI approached the other stages. But for many Hmong this30

STAGES OF HMONG CULTURAL ADAPTATION 19has been the most traumatic stage of all. This is thestage, of course, in which many Americans have come toknow the Hmong. The attempt to understand something ofwhat is behind the adaptations emerging in the UnitedStates requires us to look at the innovations which tookplace in the camps and, before that, in the resettlementconglomerations, in addition to examining traditionallife. Adaptation to life in America continues the processwhich has been going on through the lifetimes of theHmong who are here.ACKNOWLEDGEMENTSThis work was aided by National Endowment for theHumanities grant #R0-20310-82 to the Southeast AsianRefugee Studies program of the University of Minnesota.I have profited by comments on an earlier draft fromBruce Downing, Lysao Lyfoung, Vang Vang and Yang Dao.NOTESII am here concerned only with modern history of theHmong, and only with developments which took place inLaos and among Hmong who were dispersed from Laos since1975. The centuries-long experience of the Hmong inChina is not included.2This is probably a Hmong pronunciation of the Lao MuongMai. The Hmong name is Zos Key Xov Hlau. The villageswere located about one third of the way between MuongSai and Muong Ngoi. Principal informant on theexperience of the Moos Plais people was Vam NtxhaisXyooj (Vas Say Xiong), headman of one of the villagesthroughout the total experienced described. His informationwas supplemented by that provided by other membersof the group.3This date is estimated by working back from Vam NtxhaisXyooj's memory of approximate times spent in subsequentlocations.4A move of about one hundred direct-line miles.31

20 THE HMONG IN TRANSITION5Note that rectangular blocks and houses oriented to themare not in keeping with traditional Hmong houseorientationpatterns.6Doris White lock reported (personal communication) thatat nearby camp Ban Nam Vac:, men sometimes worked withwomen at the embroidery and became fairly good at it.7"The constraint of space, so common in the other refugeecamps, is not a problem in Sop Tuang. The camp isspread over approximately 1,000 acres .... [This has]allowed the residents to group themselves into naturalvillage communities." (Committee for Coordination1982:52).81 have not had time to find out how conglomerations likeLong Cheng were organized, and whether (as I suspect)some of that organization carried over into the camps.9lnformation about Ban Vinai and the youth group there isprimarily from Yang Vang, who was one of the youths involved.10For a map of nineteenth century Hmong migration intoLaos, see Yang 1975:10.11 Yang Dao reports that the first such school in XiengKhouang Province was established in 1939 (1975:134).12For an indication of the distribution of Hmong resettlementareas in southern Xieng Khouang Province in 1974,see Yang 1975:144.13The square or oblong pi ntaub decorative pieces sold byHmong women in the United States date back to 1964, duringthe resettlement stage in Vientiane. The sewingFoil Is are very much older, of course, as are most of.he designs typically to be seen in examples of thiscraft. However, the square form was adapted for sale totourists out of the traditional baby carrier shaped tohold the baby on the mother's back, and with strapsattached (Lemoine, personal communication).32

STAGES OF HMONG CULTURAL ADAPTATION 21REFERENCESAjchenbaum, Yves, and Jean-Pierre Hassoun. 1979. Histoiresd'insertion de groupes familiaux Hmong reen France. Association pour le Developpement dela Recherche et !'Experimentation en Sciences Humaines.Barney, G. Linwood. 1957a. "The Meo--An IncipientChurch." Practical Anthropology 4.2:31-50.. 1957b. "Christianity: Innovation in Meo Culture."Unpublished MA Thesis, University of Minnesota.Committee for Coordination of Services to Displaced Personsin Thailand. 1982. The CCSDPT Handbook: RefugeeServices in Thailand. Bangkok: du MaurierAssociates.Dooley, Tom. 1958. The Edge of Tomorrow. New York:Farrar, Straus and Cudahy.Downing, Bruce (assisted by Bruce Thowpaou Bliatout).1983. "The Hmong Resettlement Study Site Report:Dallas-Fort Worth, Texas." (Draft prepared in theSoutheast Asian Refugee Studies Program of the Universityof Minnesota.)Geddes, William R. 1976. Migrants of the Mountains: TheCultural Ecology of the Blue Miao (Hmong Njua) ofThailand. Oxford: Clarendon Press.Halpern, Joel. 1961. "Laotian Health Statistics." LaosProject Paper No. 10. Mimeographed.Lemoine, Jacques. 1972a. "Les ecritures des Hmong."Bulletin des Amis du Royaume Lao 7-8:123-165.. 1972b. Un village Hmong vert du haut Laos.Paris: Centre National de la Recherche Scientifique.Mason, Sarah. 1983. "The Hmong Student Association ofDallas and Fort Worth". In "The Hmong Resettlement33

22 THE HMONG IN TRANSITIONStudy Site Report," edited by Bruce Downing (seeabove).Smalley, William A. 1976. "The Problems of Consonantsand Tone: Hmong (Meo, Miao)." In Phonemes andOrthography: Language Planning in Ten Minority Languagesof Thailand, edited by William A. Smalley,pp. 85-123. Canberra, Australia: Lingustic Circle ofCanberra.Yang Dao. 1975 Les Hmong du Laos face au develloppement.Vientiane, Laos: Edition Siaosavath.34

THE HMONG OF LAOS:ECONOMIC FACTORS IN REFUGEE EXODUS AND RETURNRobert Cooper*INTRODUCTORY NOTEThis paper is concerned only with the economic aspectsof Hmong exodus from and repatriation to Laos. Itis not suggested that economic motives for fleeing an intolerablesituation preclude or dominate other motives.That the great majority of Hmong refugees fled because ofa genuinely-felt fear of reprisal or persecution from thenew regime is not called into question. I suggest onlythat there were additional economic reasons for the Hmongto leave Laos, and to leave when they did. These economicfactors in the Hmong exodus are important (or arelikely to become important) to any discussion of futurereturn to Laos, however and under whatever political ormilitary conditions that return is envisaged.I am fully aware that the suggestion that economicfactors played a role in the Hmong refugee exodus, albeita subordinate role, will transgress some Hmong sensitivities.I hope that my Hmong friends, on both sides of thefrontier, will understand that my intention in makingthis suggestion is only to contribute to the debate onthe future welfare of all Hmong, particularly those currentlycaught up in the no man's land of refugee camps.Naively, I ask the Hmong reader to set aside the intoleranceand suspicion bred through years of tragedy anduncertainty and to approach this paper with as much as heor she can muster of the tolerance and sincerity thatcharacterizes traditional, peaceful, village Hmongety.soci-THE THESISSo far, 190 Hmong have repatriated voluntarily toLaos since the United Nations High Commissioner for Refugees(UNHCR) began a voluntary repatriation programme in1980 (see Table 1; all figures are correct as of end of1983).2335

24 THE HMONG IN TRANSIT1011Table 1VOLUNTARY REPATRIATION TO LAOSYear Lao Hmong Yao1980 1931981 276 130 1331982 802 59 2081983 526 1 801,797 190 421 TOTAL: 2,408Not very many. But the importance of the programmecannot be judged simply by the numbers involved. A greatmany more Hmong are at this moment (January 1984) sittingit out in refugee camps in Thailand, waiting to see whathappens to the 190 before deciding which way to go: onto America or back to Laos.One-hundred-ninety people is, if you put them alltogether, only the size of one Hmong village, and not avery big one at that. One-hundred-ninety returnees comparedto more than 60,000 Hmong who have chosen resettlementin third countries; 190 returnees compared to morethan 40,000 Hmong currently in the camps; 190 returneescompared to more than 100,000 Hmong who have fled the LaoPeoples Democratic Republic.And 190 returnees comparedto an estimated 200,000 Hmong remaining in the country.The number of Hmong who have decided to turn aroundand go back home is tiny, yet the movement back to Laoshas a significance far greater that its size. Their decisionto return poses questions which must be answered.36

THE HMONG IN LAOS: ECONOMIC FACTORS 25Why did Hmong leave Laos? Why did the exodus takethe apparently erratic form evident in the annual andmonthly statistics? (Tables 2 and 3). Why are there suchremarkable differences between departure patterns of lowlandLao and highland Hmong (Table 2, 3 and 4). Why didthe majority of Hmong remain in Laos? Why did some Hmong,having left, decide to return to Laos?Reference to political, factors alone cannot fullyanswer these questions or fully explain patterns of behaviour.I am going to suggest that by adding an economictheory to the well-known political facts, we can comenearer to an explanation.In this paper, I shall argue that the environmentthat supported traditional Hmong swidden (slash-and-burn)cultivation in Laos had ganged significantly before thegreat exodus of 1975, and that because of this changemany of those who fled at that time might have died ofhunger had they remained in Laos. I shall also considerthe possibility that the very magnitude of the exodus reestablishedsomething of an ecological equilibrium, furtherreducing the urgency of the exodus rate and raisingthe possibility that resources in Laos could support amovement of Hmong refugee farmers back to the mountainsof Laos.I am also going to suggest that Hmong who chose toleave Laos early in the year probably had sound economicmotives, in addition to any other motives, for going atjust that time.I find it convenient to arrange evidence and argumentsto test the following thesis:In Laos, a direct correlation exists between theHmong population/resources ratio and the rate ofexodus and return.I shall consider evidence to support this thesiswithin three categories: historical, observational andempirical.37

Table 2ARRIVALS IN THAILAND BY YEAR1975 1976 1977 19711 1979 1980 1981 1982 1983LAO10,195 19,499111,065 48,786 22,045 28,96716,4283,2034,521HILL-TRIBE44,659 7,2663,873 1,013 23,943 14,9014,3051,8162,870Table 3ARRIVALS IN THAILAND BY MONTHA. HILL-TRIBE'Year San Feb Mar Apr May Jun 3u1 Aug Sep Oct Nov Dec TOTAL19816514031,221662358228 5628415114241454,30519821321713713191910 416216412710817816191323559323934017119 35420601121142202,870B.LOWLAND LAO19816514031,221662358228 5623415114241454,3051982643612423196189227 21525079120961483,2033ri

THE HMONG IN LAOS: ECONOMIC FACTORS 27THE HISTORICAL EVIDENCEDuring the years 1850 tofought a 1880, the Hmong in series Chinaof wars against the Manchu Government,which was at the time committed to a policyation in order to of heavy tax-pay off the heavy indemnities by the British after demandedthe Opium War andways of making money.was looking forOpium had caused theseand it seems likely indemnitiesthat the Manchusof paying off saw opium as a waythese indemnities. Certainlypolicy of banning opiumthe Chinesecultivation was now reversedproduction seems to have andbeen encouraged.sittingWith the Hmongon some of the best opium land inis logicalthe world, itto assume that thetime wanted Chinese authorities ofcontroltheover it and revenue from it.seems reasonable therefore Itto say that the Hmong-Manchuwars were predominantly economic in character.To say that the wars had economic motives doesof course, deny that not,there was repression by the Manchuauthorities or that manythe thousands of Hmong fledborder. acrossFor all we know, the nineteenth centuryHmong exodus from China could have beengreat as the twentieth every bit ascentury exodus from Laos.even greater.PerhapsOf course, repressionin spite of was not absolute andthe great number of refugees who fledcountry, the themajority of Hmong remained incentury China, as the majoritynineteenthof Hmong remained intieth century Laos.twen-Thus, the history of the Hmong isleast seen to contain atone precedent of large-scale refugee exodus.Hmong, unable Theto continue their way of lifeterferencewithout in-from the Chinese authorities, fled innumbers from one nationlargestate to re-establishsuccessfully in anotherthemselvesnation state.This nineteenth century refugee movementonce persecution,slowed downor the fear ofexist.persecution, ceasedHowever,tothe movement did notdid it reverse.stop. Still lessSignificantly, it continued. Why?I can see only one reason to explain the persistentmovement from China, throughand into Laos: the the northern part of Vietnameconomic certainty thatmuch better in theresources werenew location than in China.It seems likely that a situation ofimbalance or, to callenvironmentalit by another name, "resource scar-39

78THE HMONG IN TRANSITIONwith the French, this was only after the rn' 'ion wasunder way. Anyway, the Hmong in Laos probably extractedfrom the French a much greater degree of political autonomythan they could hope to obtain in Thailand.city," existed in some of the Hmong areas of nineteenthcentury southern China and that the Hmong solved thisproblem by relocating a large number of their people intoan area of fresh resources. Once ar 'vironmental balancewas reestablished in China, or the imbalancebecame less acute, the exodus slowed t r,As far as an be guessed, the 0,ation from Chinaseems to have taken two forms: a large-scale refugee exodusand a small-scale calculated relocation of individualfamilies or groups of families.Exactly the same forms of migration are evident inthe post-1975 Hmong exodus from Laos. This fact in itselfsuggests a likely similarity of causes, which Iwould summarize as fear of persecution plus the need toescape a situation of resource scarcity.By suggesting the existence of historical precedent,I do not wish to imply that the Hmong collectively orindividually copied actions which had proven effectiveresponses to problems encountered in the past. I am onlysaying that the Hmong have demonstrated an ability toface economic and political problems in a certain way,that way being large-scale migration.After the exodus from China into Vietnam and Laos, afurther movement of Hmong took place from Laos into Thailand.This gradual migration took the form of householdrelocation. It seems to have begun soon after the Hmongleft China and to have led to a sizable Hmong presence inThailand by the early twentieth century. Why did thismovement take place?A politica! motive seems to have been absent. Althoughthe Hmong in Laos did get into some skirmishes40

THE HMONG IN LAOS: ECONOMIC FACTORS 29I would suggest, therefore, the likelihood that theswidden use of mountain sides for opium cultivation(which progressively reduces soil fertility until landceases to be productive) created a resources/populationimbalance in certain Hmong areas of Laos throughout thefirst half of the twentieth century and that the Hmongresponded to that situation in exactly the same way theyhad responded to the same problem in China: by moving aproportion of the populationnot all of the populationinto a situation of resource plenty.Thus, it can be argued that two historical precedentsof Hmong economic migration across national frontierspreaated the beginnings of the Laotian civil war inthe 1950s.The thirty-year-long civil war was undoubtedly responsiblefor removing further portions of land in Hmongareas of Laos from the possibility of exploitation. Manymountain areas, previously open to cultivation, werestripped of vegetation by the effects of bombing and chemicals.During the fighting of the 1960s and early 1970s,the Hmong tended to group in certain safe villages, indisplaced person camps in the lowlands and in the "secretarmy" mountain town of Long Cheng, which had been establishedas early as 1961. Such large population groupingscould not hope to supply all subsistence needs by traditionalswidden cultivation, and a great many Hmong familiescame to rely increasingly on food drops by aircraft,handouts in the population centres or the soldier's payearned by adult males. Most estimates of the number ofHmong on some form of "welfare" during this period totalover one hundred thousand.In addition to these one hundred thousand we shouldremember that there were also Hmong in uniform on thePathet Lao side. They earned little money but it is trueto say that they also gained at least part of their basicsubsistence by nonagricultural means.The picture of Hmong areas of Laos in 1975 is one inwhich the pinch of resource scarcity had long promoted asteady movement of Hmong migrants into Thailand; thissituation of resource scarcity was compounded by a civilwar which had many of the effects of the industrial revolutionin Europe, driving farmers from their lands whilst41

30 THE HMONG IN TRANSITIONat the same time offering alternative means of livelihoodin centres of population.It is also worth noting that as far as we know comparativelyfew Hmong fled war-torn Laos in the decade before1975. This is surprisingeven amazingconsideringthat the Hmong were the principal casualties of the war.This low level of refugee outflow during the war suggestsone or both of two things: either that the Hmong stayedin Laos to fight against communism or that the provisionof aid and paid employment acted as an economic incentiveto remain, even when remaining was not only dangerous butrequired adaptation or abandonment of traditional lifestyle.(Similar economic incentives induced many thousandThais to leave the safety of Thailand and fight asmercenaries in Laos.)When, in 1975, the alternative means of livelihoodcame to an abrupt end, tens of thousands of Hmong foundthemselves abruptly face to face not only with the fearof the enemy's revenge but also with a situation of accumulatedresource scarcity.I do not doubt that the Hmong who fled Laos at thattime did so from a genuine fear of persecution followingthe change in regime. However, I feel it is worth makingthe point that they also had an economic motive for fleeingwhen they did. Had they remained in Laos, it is difficultto see how they could have avoided large-scalefamine. Certainly the Pathet Lao government, had it beeninclined to do so, could not have offered any assistance,since the rice harvests in the years following the changeof regimes were insufficient to feed the lowland population.Thus, whatever the political reasons for leaving(which are not in doubt), the convulsive exodus from Laoscould be said to have served economic functions, both forthe leavers and the stayers. The leavers were removedfrom a situation of resource scarcity and potential faminewhilst the stayers had far less competition foravailable resources.THE OBSERVATIONAL EVIDENCEMy argument that the Hmong of Laos suffered from avery adverse population/resources imbalance is supported42

THE HMONG IN LAOS: ECONOMIC FACTORS 31by my personal observations made over the period of December1980 to March 1983. During this time, I lived inVientiane and travelled very widely by aeroplane, helicopter,truck, jeep and boat, and on foot. During thesetravels around the country I was able to visit Sayaboury,Xieng Khouang, Luang Prabang, Vientiane, Savannakhet,Pakse and Attopeu provinces and got to see a number ofHmong villages.I was particularly interested in Xieng Khouang forfour reasons:1. The majority of Hmong refugees in Ban Vinai Camp (theHmong refugee camp in Thailand) seem to come fromthat province.2. Only one of the 190 Hmong repatriates to Laos returnedto Xieng Khouang. A few others had originatedfrom that province but chose not to return there.3. Many of some if teen hundred Hmong who live in thelarge settlement at KM 52 north of the town ofVientiane said sy had left Xieng Khouang becauseof the difficulty of finding enough land to farm inthat province.4. Several thousand Hmong from Xieng Khouang had resettledin several groups around Yang Viang. Some hadarrived there before 1975, but most had come sincethe change in regimes. All of those I met statedthat it wrl very difficult to make a living in XiengKhouang.I was able to fly extensively by helicopter throughoutXieng Khouang Province. Everywhere, the mountainswere covered with grassnot trees. And it was the kindof tough grass known technically as imperata, a grassthat grows where little else can and puts down long rootswhich make cultivation by hand hoe extremely difficult ifnot impossible.I had flown around northern Thailand ten years previouslyand seen much the same kind of thing in areas ofknown resource scarcity. However, even in the worstareas of northern Thailand, it had still been possible tosee some trees and some secondary growth on the mountain-31 43

32 THE HMONG IN TRANSITIONsides and there was the feeling that resources recoverywas at least possible. Throughout most of Xieng Khouang,however, I could find trees only in the folds betweenmountains, where the slope was usually too steep to permitcultivation.I did see Hmong villages and fields but far fewerthan I had expected to see. There were, by contrast,many Hmong in Phone Savane, the capital of Xieng KhouangProvince, including the president of the province. I wasable to speak with them without restriction and all gavea rather gloomy picture of the province's economy. Theiraccounts suggest that it is still possible for some villagesto get by by growing opium, and Xieng Khouangmains the major opium - producing province of Laos, but"getting by" means living next to poverty and, as thingsstand, there is no reason to assume that agriculturalprospects in the province will improve very much in thefutureṪhe situation in Luang Prabang, Sayaboury, and Vientianeprovinces was less severe than in Xieng Khouang.Several villages I visited seemed to be getting by reasonablywell in terms of fields cultivated, food stocksheld, household livestock and standard of housing, dressand so on. However, there is certainly no cause for complacency.Even in the best of areas it would be verydifficult to find a single location containing adequateresources to support installation of a village-sizedgroup of, say, two hundred Hmong wishing to follow onlytraditional techniques of cultivation.Many other foreigners travelling to various parts ofLaos have confirmed my impression that the mountain areasof Laos, in spite of very low population ratios, sufferfrom a situation of resource scarcity that could best bedescribed as varying between serious and chronic. It is,I feel, most likely that the roots of this situation extendback far beyond the great divide of 1975.THE EMPIRICAL EVIDENCEThe empirical evidence to support the thesis of directcorrelation between resource availability and therate of exodus is, I must admit, not 100 percent convincing.This is mostly because of the difficulties involved44

THE HMONG IN LAOS: ECONOMIC FACTORS 33in doing any really objective research in Laos and in thecamps of Thailand. I have therefore had to rely on statisticscollected for various official reasons, none ofthem collected specifically to test the thesis. It mustalso be borne in mind that these figures represent onlymovement in and out of camps. It is most likely thatthousands more Hmong than get into the statistics havemoved across the border into Thailand without passingthrough a camp and equally likely that several thousandhave "spontaneously repatriated" to Laos.The figures I am giving here do not conclusivelyprove my thesis. On the other hand, they certainly donot disprove it. To play it safe, I would say that areasonable interpretation of available statistics tendsto support the thesis.First, annual arrivals of Hmong refugees in Thailandsince 1975, as set out in Table 2, invite analysis a-gainst an economic background. (Figures are actually forhill-tribe arrivals and include a few thousand Yao andsome members of other minority groups, but the great majorityare Hmong.)Two peaks are evident: 1975 and 1979. The reasonsfor the 1975 exodus are obvious enough: fear of retaliationagainst the Hmong by the victorious Pathet Lao andthe very sound economic reasons for leaving that I havealready gone into. But why should figures peak again in1979? If fear of such factors as "yellow rain" were theonly reason for leaving, why did 23,943 flee the countryin 1979 and only 1,816 in 1982? The real answer, I feel,has a lot to do with economics and the economic historyof the Hmong.For a long, long time Hmong have lived near larger,more dominant ethnic groups. Speaking as we must in generalizations:the majority ethnic groupsChinese, Vietnamese,Lao and Northern Thaiwere firmly entrenched inthe lowlands where they practiced irrigation methods togrow rice and lived in communities typified by permanenteconomies, permanent villages, permanent authority structuresand so on. The Hmong, on the other hand, werehopping around from mountaintop to mountaintop, practicingshifting cultivation in communities typified byshifting economies, temporary villages and temporary authoritystructures. During thousands of years of quietsubsistence farming, the Hmong lived near a majority cult45

34 THE HMONG IN TRANSITIONture but had little reason to interact with that cultureand experienced no real competition for available resources.Then, along came cash-crop opium.Starting sometime in the nineteenth century, theHmong began to specialize in opium. They continued topractice economic specialization even when this meantthat sufficient rice or other foods could not be grown tomeet all domestic needs. It was more profitable to sellthe opium and buy the extra food required, or to exchangeopium directly for food.A situation developed in which, although very fewHmong actually owned irrigated rice fields, a great manyHmongperhaps most Hmongwere dependent for at least apart of their basic food supply on lowland fields in Laohands. The Lao were, in turn, dependent on Hmong (andsome other hill-tribe groups) for opium. But it is possibleto live without opium, whereas it is much more difficultto live without rice, as was dearly evident inLaos in 1977 and 1978.In 1977, Laos experienced a disastrous drought.This was followed in 1978 by an even more disastrousflooding at harvest time. In both years Laos was forcedto appeal for international assistance to prevent famine.There was, of course, very little rice from any sourceavailable to sell to the Hmong.The effects of this economic situation on the exoduspattern of ethnic Lao is seen very clearly in Table 2.More left in 1978 than in any other year. For the Hmong,the effect was a little delayed for two reasons. Firstly,the Hmong upland rice crop, insufficient for allneeds as it was, was relatively unaffected by the floods.The October harvest therefore was enough to keep everybodygoing until after the January opium harvest of 1979.By that time, however, it would have been clear that sup-,plementary rice would not be available and it seems reasonableto assume that this has something to do with the1979 Hmong exodus peak.The years 1981, 1982 and 1983 have seen the bestrice harvests in Laos for many years. (Laos can now beconsidered self-sufficient in rice cultivation.) Thesegood harvests, like the bad, seem to be reflected in theannual exodus rate for both Hmong and Lao in 1982 and1983, which are the lowest recorded since 1975.'46

THE HMONG IN LAOS: ECONOMIC FACTORS 35A second economic reason for the delayed effect onHmong exodus patterns is suggested by the monthly figuresfor Hmong refugee arrivals in Thailand over the years1981, 1982 and 1983, which are set out in Table 3.(Monthly statistics are not available before January1981.)Figures for the first four months of each year aremuch higher than for the other months of each year andpeak around March in all three years. Why should thispattern emerge and what does it suggest?A facile explanation often heard and credited, isthat the Mekong River, which most Hmong must cross beforearriving in Thailand, is low during the first half of theyear and at its height towards year's end. If this factis to be used to explain the least number of hill-triberefugee crossings (only four) in November 1981, it cannotexplain the high figure of 1,413 lowland Lao who crossedthe same river in the same month of the same year.certainly Itcannot explain why no Hmong crossed in June1982, when the river was at its lowest.Thus, whilst it is recognized that the Mekong inflood presents a fearsome obstacle to mountain people unusedto water travel, we must look beyond the rivertheory for an explanation of the monthly patterns of exodusevident in Table 3. I suggest that these patternsare less a result of river level than they are evidenceof the dictates of the agricultural year.Hmong harvest rice around October, sometimes intoNovember. This harvest is followed by Hmong New Yearcelebrations, which in turn are followed by the opiumharvest.The opium harvest usually takes a single familythree to four weeks to complete. The precise date it beginsvaries between localities and depends on soil conditions,type of seeds used and weather conditions in anyparticular year. At the very earliest, it begins inearly December and at the very latest it is completed inthe first days of February. The two months following theopium harvest, February and March, are used to cut newfields, repair the house and do various tasks to preparefor the coming agricultural year. Families intending tomove to a new location do so at this time of year. They,,47

36 THE HMONG IN TRANSITIONuse the revenue from post-harvest opium sales to helpthem move and to tide them over in the new location untilthey become reestablished.The hill-tribe refugee statistics of Table 3, inThai-showing a significant increase in Hmong arriving inland around March are, to an extent, simply reflectingtraditional patterns of movement (movement in these casesbeing across a national boundary). This suggests that,at least since 1981, economics has played a big part inthe decision to leave at a certain time. The main reasonfor leaving might be fear of persecution, but I canthink of no reason why that fear should be greater in theearly months of the year. If fear of persecution werethe only reason for fleeing the country, we would notexpect quite so many to wait around for the opium harvest.There is a further economic reason for Hmong in Laosto leave the country as soon as the opium is harvested,especially those in Xieng Khouang. If they get outquickly enough, they can avoid having to sell a largepart of the harvest to the government. The Laotian governmentnow operates the same kind of opium monopoly thathas always caused problems for the Hmong under the Chinese,French and Thai. The main problem involves fixingan appropriate price for the year's supply. Opium is acrop which varies dramatically in supply and price betweenyears. A few days' poor weather can greatly reducethe size of the harvest. When this happens in a freemarketsituation, the price rises to compensate somewhatfor the shortfall. However, when the only buyer is thegovernment, and the price is fixed by the state, smallbureaucratic adjusts its in the official price (usuallymade months after the crop has been sold) cannot providethe same level of compensation.Also rel'vant are the facts that the government buysin kip, the national currency of Laos, which can be reliedupon t:', devalue as the year progresses, and pays aprice which is much lower than the free market withinLaos, where payment is made in silver, dollars or Thaibaht. And the Laotian free market price is in turn muchlower than the price obtainable across the border inThailand.The dictates of the agricultural year together withthe marketing procedures currently in force in Laos might'48

THE HMONG IN LAOS: ECONOMIC FACTORS 37have very little to do with the total availability of agriculturalresources and the total annual rate of exodus,but statistics certainly suggest a link between resourceexploitation patterns and patterns of monthly exodus.They also suggest that at least a proportion of Hmongrefugees are sensitive to economic factors when selectinga time to flee the country.SUMMARYSupporting evidence for the thesis that a directzorrelation exists between availability of agriculturalresources and rate of exodus follows:1. Historical precedents exist. The Hmong have behavedin the same way before, moving across national boundariesin very large numbers when threatened by resourcescarcity.2. The Hmong exodus from LE,oa existed prior to the beginningsof civil war in 1950.3. The civil war disguised the population/resources imbalanceby providing alternative means of livelihood.4. The civil war promoted large-scale settlement and dependency.Economic survival required large-scalesettlements to disperse rapidly in 1975. Thousandsfaced famine; thousands fled Laos.5. The drought of 197; and floods of 1978 meant thetraditional supplement of rice could not be purchasedfrom the lowlands in 1979.6. Vast areas of Hmong lands hive become unsuitable forcultivation by traditional methods. This is particularlytrue in the former Hmong stronghold ofXieng Khouang.7. The majority of Hmong refugees in the camps in Thailandare from Xieng Khouang.'.49

38 THE HMONG IN TRANSITION8. Departure rate figures suggest that Hmong respond toeconomic factors when deciding a particular time toleave the country, in many cases completing an agriculturalcycle before departure.CONCLUSIONAt this moment many a Hmong is sitting in Ban Vinairefugee camp wondering whether to go back to Laos, go toAmerica or remain sitting in Ban Vinai.In the camp, or in America, this Hmong person can befairly sure his family will get enough to eat. His familyis also likely to benefit from the kind of medicaland educational facilities which would be almost unavailablein Laos.If he goes back to Laos, he will receive a certainamount of assistance in the form of rice and salt to helphim over the first year. But even providing this verybasic assistance causes tremendous problems. There is nobudget for the kind of grand-slam air drops of rice thattook place before 1975, and that kind of blanket assistanceis not appropriate for programmes aimed at achievingself-sufficiency within the minimum time period.Rice in the 1980s is delivered by trucks, boats,horseback and man-back. Delays in delivery occur and arenormal in the more remote areas. And there is a Catch-22aspect in the provision of rice assistance to Hmong: ricecan only be delivered to the location to which a Hmongchooses to return, yet the Hmong person in question ismost likely to move around looking for a piece of land tosupport his family. If he stays put in one place he willprobably receive rice but may not become self-sufficient;if he moves around he stands a better chance of becomingself-sufficient but he is likely to miss out on the riceassistance.Apart from rice assistance, a Hmong returnee familygets a one-time present of a "resettlement kit." Thiscontains clothing and black cloth to make Hmong clothingand the basic items of domestic assistance: pots andpans, bowls, spoons, blankets, mats, water buckets, soap,etc. It also contains a set of tools to assist in clearingthe ground and building a house: an axe, machete,hoe, hammer and so on. With this kit and a first in-50

THE HMONG IN LAOS: ECONOMIC FACTORS 39stallment of rice, the family istaken to a point as nearto home or the chosen area of return as it is possible toreach with available transport; then off they go onfoot. Back where they started from. The rest is up tothem.One of the most pleasant parts of my job in Laos wasvisiting returnees (Hmong, Yao and Lao) several monthsafter they had returned home. I was always impressedwith the Hmong ability to take up the old life againusually with the help of relatives or clansmen. It wasalso very nice to see how quickly most camp refugees recoveredtheir pride once back in the mountains in anewly-built Hmong house, wearing Hmong dress, sitting ona little stool by a big fire, smoking a ying -tong ordoing some embroidery.These impressions of the Hmong returnee at home havelittle to do with economics. They are of course subjectiveand are no doubt coloured by the fact that I am botha romantic Westerner and an anthropologist. But, thelast thing I would wish to do is romance about a Hmongreturn to Laos. I would much prefer such a movement tobe based on sound economic facts, and some of these factsI have tried to present in this paper.As things stand at the moment, I would say that the(economic) utilitarian balance swings against the returnof Hmong to Laos in large numbers.Most certainly, Hmong can go back to the mountainsand live in Hmong houses in Hmong villages. However,they won't be able to gain much of a livelihood if theygo back to Xieng Khouang and start to grow opium, andthis I feel is one of the major factors preventing alarger movement back to Laos.For the economic balance to swing in favour of returnto Laos, some form of agricultural change and developmentis required. Alternative uses of available resourceswould need to be promoted.A start has already been made on this development.As yet, a very small start. Some pilot projects havebeen established in Hmong areas by UNHCR and by theUnited Nations Development Programme (UNDP). These projectscentre on some form of permanent cultivation andresidence but are unlikely to require great changes inthe Hmong way of life.Mostly they involve creating newirrigated rice fields and rehabilitating old ones, pro-51'

40 THE MOM III TRANSITIONviding livestock and introducing or improving basic educationaland medical facilities. All of these projectsare geared to assisting the local population, i.e. thosewho never left the country, whilst creating opportunitieswhich will be reserved for returnees.Even if all the projects currently being undertakenby UNHCR and UNDP in Laos are successful, conditions willhave been created for the return of, at best, two orthree thousand Hmong. Clearly much more needs to be donebefore anybody can realistically talk about having laidthe groundwork to support return to Laos as a real economicalternative to resettlement in and assimilation intoa non-Hmong world.The economic conditions to support a large-scale returnto Laos will not suddenly appear tomorrow. They canonly be created by money and hard work. And the availabilityof money from international sources depends onpolitical rather than economic factors. It depends onprecisely those factors that Ipaper.have not mentioned in this*NOTEThe author was Programme Officer for the United NationsHigh Commissioner for Refugees in Laos betweenDecember 1980 and March 1983, and is currently GenevabasedResettlement Officer concerned with Indochineserefugees. He wishes to make it quite clear that, whileobservations made during his residence in Laos helpedform the ideas this paper contains, he is writing in apurely personal capacity. His views do not necessarilyreflect those of UNHCR and he should not be quoted in hiscapacity as a UNHCR official.52

PROCESSES OF IDENTITY MAINTENANCEIN HMONG SOCIETYTimothy DunniganINTRODUCTIONThe long-term survival of an ethnic minority does notdepend upon the retention of specific cultural characteristics.Edward Spicer (1980) points out that the YaquiIndians of northwestern Mexico have endured as a peopledespite more than 350 years of dramatic change. In 1617,Catholic priests began establishing missions extensivelyamong the Yaqui. European-derived economic institutionsgreatly affected Yaqui subsistence and settlement patternsthrough the colonial and post-independence periods(1617-1910). After losing a protracted and nearly genocidalwar to the Mexican government in the nineteenthcentury, many Yaquis were deported to southern Mexico.Others became refugees in the United States. Politicaland economic pressures of the twentieth century reducedthe indigenous communities almost to extinction. Increasedgovernment support of Indian economic rights overthe last several decades has encouraged Yaquis to resettlein their original territories, and a number of Yaquitowns are beginning to thrive once again.Considering all the changes that have occurred in theYaqui way of life, it is difficult to say in what sensethe contemporary Yaqui can claim continuity with thepast. According to Spicer, Yaqui history reveals the persistenceof a collective identity system, a unique set ofsymbols that embody group beliefs and sentiments.An identity system...develops independently ofthose processes by which a total culture pattern,a set of particular customs and beliefs constitutinga way of life, is maintained. The continuityof a people is a phenomenon distinct fromthe perftence of a particular set of culturetraits. (gicer 1971:798)The Hmong have survived challenges to their existencesimilar to those faced by the Yaqui. In both cases,4153

42 THE HMONG IN TRANSITIONthere have been long periods of adjustment to unstableenvironments. Although each group has remained sociallyand culturally distinct, popular conceptions of the Hmongand Yaqui as being very traditional distort reality."Tradition" usually implies a lack of change in respectto some vague preindustrial baseline. In this sense, theterm is not useful in describing the ways in which ethr.::minorities endure while undergoing culture change forcenturies. The issue is not whether the Hmong refugeesin America can remain culturally conservative or must acceptassimilation. More to the point are questions regardingthe future form and meaning of Hmong identitysymbols and the evolving structure of social relationsthrough which these symbols will be expressed. Ratherthan attempting to analyze the extremely complex Hmongidentity system as it exists and is being created inAmerica, will discuss some of the processes of identitymaintenor,_ at will affect relations between the Hmongand other is, oups for some time to come. My purpose is todescribe how dynamic identity elements such as ethnicpolitics, religion, language and kinship reinforce Hmongsocial boundaries as adaptations are made to changingcircumstances.'THE OPPOSITIONAL PROCESSGroup identity presupposes social boundaries definedby "we vs. they" criteria. Whatever the nature of thesecriteria, they are brought into sharper focus by the oppositionof etitsiders. In his essay on persistent peoplesin North America and Europe, Spicer (1971:797) characterizessuch opposition as a "continued conflict betweenthese peoples and the controllers of the surroundingstate apparatus." As an enclaved minority, the Hmonghave had to contend with the opposing interests of variousnationalities, The movement of Hmong from China intoSoutheast Asia was to some extent impelled by pressurefrom other ethnic groups (Savina 1924; Weins 1967;Larteguy 1979). Under conditions of extreme threat, revitalizationmovements led by Hmong prophets have resultedin armed conflict. A more typical Hmong reactionto external political authority has been accommodation.In sixteenth century Kweichow, China, the Hmong consti-

PROCESSES OF IDENTITY MAINTENANCE 43tuted a distinct but fully integrated segment , a broaderpolitical order.Tribal chieftains, including Miao [Hmong] chieftains,were appointed to hereditary administrativepositions, adopting Chinese surnames and enteringinto a kind of feudatory arrangement withthe imperial government on behalf of the tribesmenwithin their jurisdiction (Le Bai et al. 1964:70).In Laos, the Hmong sought similar arrangements withFrench colonial authorities and various Lao politicalfactions. They tried to maximize their autonomy by becomingvalued supporters of the established government oran emerging power.Hmong external relations in Laos before the collapseof the Constitutional Monarchy in 1975 and during the recentperiod of refugee resettlement, suggest the followinggeneral assumptions and principles. Accommodation andcompromise usually accomplish more than direct confrontation.The interests of the Hmong will not be dependablyserved by any dominant majority, and it is necessary tomaintain the positions of leaders capable of rallying theHmong in times of crisis. Since a Hmong political leaderachieves and holds his position by continuously demonstratingan ability to obtain help for others, he musthave access to resources not only within Hmong society,but also within the broader political order than thatwhich encompasses the Hmong. Persons of influence belongingto the dominant majority should be cultivated aspotential allies who will understand and sympatheticallyrespond to Hmong needs. The insertion of Hmong personnelinto majority-controlled institutions that affect theHmong begins with the establishment of such alliances.The Hmong were relatively unknown to the Americanpublic when they began arriving in the United States fromThailand ref dgee camps during the spring of 1976. Hmongleaders who spoke English moved quickly to establish contactswith American citizens willing to publicize how theHmong had supported United States interests it Laos. Theresulting media coverage helped to identify the Hmong asa special group of refugees to whom the United Statesowed a considerable debt. Exceptionally able Hmong wereJCyt.,.:-,L)

44 THE HMONG IN TRANSITIONpromoted by the group as candidates for jobs offeredthrough resettlement agencies. Although relatively young,most of the Hmong hired by the agencies to do caseworkhad already held respcnsible positions in Laos.Despite the hiring of Hmong by organizations servingrefugees, misunderstandings occurred between Hmong leadersand some agencies over control of government fundsallocated for resettlement programs. Experience hadtaught the Hmong that security depended upon solidarity,and the refugees were unlikely to remain united if theironly choice was to turn to outsiders for help. The Hmongwanted to be primary providers of some services to refugeefamilies. They hoped to negotiate arrangements similarto those that had been obtained in Laos, i.e., thedistribution of resources through Hmong political structures.Largely due to Hmong persuasion, Hmong mutual assistanceassociations were given limited public and privatefunding to carry out educational, legal service,housing and job development programs.Participation in mutual assistance associations increasedrapidly during the early years of resettlement,and some of the local self-help groups formed a looselystructured national organization. Families looked to associationofficers and board members for guidance. Thelarger, more intact kinship groups were expected to bethe most active in establishing policy. Dissenting groupswere co-opted in the interests of unity. Persons advantagedby education and training felt obligated to contributeconsiderable time to association affairs.As the threat of living in a strange country graduallylessened, the number of families regularly participatingin association functions declined somewhat. The needto support a protective alliance that overarched all ofHmong society no longer seemed imperative. As a result,many Hmong cut back on community involvement in order toconcentrate on matters of direct consequence to their kingroups. This has not signaled a breakdown in ethnic groupsolidarity. The Hmong appear to be in a new phase of acycle wherein reduced hostile opposition allows for a reductionin political hierarchical organization.The history of Hmong politics indicates that refugeeorganizations like the mutual assistance associations andother supra-kinship modalities, such as athletic and socialclubs, student associations, women's support groups,56

PROCESSES OF IDENTITY MAINTENANCE 45etc., are unlikely to disappear even if assimilationpressures are minimal and outside fina,ncial assistance isunobtainable. These groups will continue to be supportedbecause they are needed to exert leverage when it becomesnecessary to influence majority policy toward the Hmong.These organizations also have a number of other importantfunctions relating to the mahtenance of ethnic groupidentity.THE HOMELANDCommunity self-help organizations do much more thandeal with local problems. They serve as clearinghousesfor political news from Thailand and Laos. Members contributemoney to help finance the efforts of their nationalleaders to lobby government officials in Thailandand the United States on policy matters affecting theLaotian government. The continued interest in Laos isunderstandable. The concept of a homeland remains centralto the symbol system of a people even after controlover a territory has been lost (Spicer 1971). The LaotianHmong trace their roots back to China, but recognize Laosas the place where they attained a special political i-dentity. Refugees raised in Laos feel particularly closeto those who come from the same province, district orvillage. Regional loyalties are de-emphasized by mostrefugee political leaders, who prefer to focus on thespecial role that the entire Hmong population played inthe emergence of Laos as an independent nation.Most refugees intend to stay in America, even if areversal of the present political situation eventuallyallows the Hmong to resettle in Laos. The unifying powerof Laos as a symbolic homeland will likely remain strongbecause of Hmong historical awareness. As Spicer (1980:360) so elegantly states,In their sense of identity every people molds avessel into which they pour from generation togeneration the meanings of their historical experiences.Each such creation is for every enduringpeople a summation of their customs; insuch a listing the vital dimension is not present.This dimension is the people's feeling57

46 THE HMONG IN TRANSITIONabout the way they have performed and the valuesthey have stood for in the course of their lifehistory as a people. Such associations with realevents in which people have suffered and triumphedare poured into the mold of identity andso come to have significance.Knowledge of Hmong sacrifices in the context of thereal events of Laotian history will be perpetuated in thegeneral accounts known to the entire refugee community,and in the more detailed and specific oral histories ofindividual lineages.LANGUAGEThe concentration of Hmong from many areas of Laosinto refugee settlements during and after the recent conflictin Southeast Asia greatly increased in terdialect iccontacts. Over the past several decades there has probablybeen some leveling, but the two major dialects ofLaos known as Hmoob Dawb ("White Hmong") and HmoobNtsuab2 ("Blue Hmong' remain distinct, and each includesa number of regional variants. Refugees often switch dialectsdepending upon the situation. Some young adults,for instance, speak to their parents in one dialect, anduse other dialect forms when communicating with friends.Hmong leaders support the idea of continued dialectalvariety, even at the risk of reinforcing political divisionswithin the ethnic group. It is in regard to writtenHm..mg that political leaders have encouraged attemptsto establish a single standard. Working under the aegisof the Hmong Association in St. Paul-Minneapolis, respectedmembers from influential kin groups representingdifferent dialects were able to produce a consensus Hmongtranslation of junior high school curricular materials(Dunnigan and Vang 1980a). All participants agreed touse an orthography that had been developed in the early1950s for Hmoob Dawb. Subsequent attempts to modify thisorthography in ways acceptable to some Hmoob Ntsuab critics,who understandably want equal status for their dialect,have dissatisfied speakers of both dialects. I seelittle prospect for a resolution of this problem, but thepolitical consequences should not be very great.58

PROCESSES OF IDENTITY MAINTENANCE 47Research on factors affecting language survival hasbeen inconclusive. Fishman (1974) notes that there areexceptions to even the best documented generalizationsabout linguistic acculturation. Notwithstanding the manydifficulties associated with predicting the future statusof any immigrant language, Dunnigan and Vang (1980b) assertedthat language shift in the American Hmong communitycould be expected to reach the stage of stabilized bilingualismdue to the persistence of certain kinshipstructures that will provide the necessary separation ofdomain use between English and Hmong.More important for the maintenance of the Hmonglanguage is its function as an essential part ofritual processes by which social relations areestablished and intensified. This is particularlytrue of the extended family (Dunnigan and Vang1980b:44).The most elaborate expressions of ritual language calledpj. lug or "flowery speech" will probably be lost throughinterregional leveling. The death of many senior leadersof kin groups during the war seems to have acceleratedthis process. The younger Hmong generally wish to avoidthe ritual language contests that were an important partof making social contracts in Laos (Dunnigan and Vang1980b). Special ritual language will still be needed tonegotiate and formalize crucial kin group alliances.RELIGIONReligious belief and ritual observance, like language,has the potential both to reinforce and to dissolveHmong social cohesion. A majority cf the Hmong inLaos practiced a form of animism) Others were convertedby Catholic and Protestant missionaries. Approximatelyhalf of the Hmong refugees in the United Statesidentify themselves as Christians. A Hmong revitalizationmovement called Chao Fa, a Lao term meaning "Lordof the Sky," began in the last years of the Laotian war,and continues to be a force in Laos and in Thailand refugeecamps. Hmong political and military leaders were alwayscareful to honor Buddhism as the state religion.59 "

48 THE HMONG IN TRANSITIONBuddhist symbols are still used to express nationalisticsentiments.Hmong religious beliefs and practices are diverse,but there is a strong tendency toward uniformity withinthe patrilineal kin group. This occurs for several reasons.The membership must make collective decisions aboutsharing resources, and divided loyalties of any kindcould undermine cooperation. The group's charter is itshistory of descent. Religious beliefs define relationshipsbetween living members and their ancestors. Sacredrituals provide a means of memorializing these relationshipsand strengthening group bonds. For the non-ChristianHmong, knowledge of lineage ceremonies helps determinewho qualifies as a member. It is not surprisingthat reconstituted refugee kin groups have been known toreduce internal heterodoxy by pressuring members toswitch their allegiance from Christian back to animism.Conversion has also occurred in the opposite directionfor the sake :of group unity.A much debated subject in the Hmong refugee communityhas been the effects of Christian missionary work on kingroup relations. One concern is that Hmong solidarityhas been weakened because some fundamentalists shun non-Christians. According to my observations, the increasein Hmong religious diversity over the past three and ahalf decades has not seriously impeded the formation ofcrucial kin group alliances. Some Christian groups objectto members marrying outside of their faith, but such marriagesdo take place frequently. Because the couple normallylives with the husband's parents, the wife usuallyadopts her husband's religion. Thus the capacity of refugeeHmong to form suprakinship political structures hasnot been reduced by religious sectarianism.Rather than forcing the throng to assimilate, Christianaffiliation has helped them to maintain their culturaland social distinctiveness. Even when they are officiallypart of a larger, heterogenous church congregation,most Hmong prefer to meet separately, use their nativelanguage during services and worship under the directionof Hmong ministers and deacons. Christianity, aswell as animism, has a special ethnic significance forthe Hmong.6 0,

PROCESSES OF IDENTITY MAINTENANCE 49KINSHIPMost studies of Hmong kinship4 describe a set ofnested structures defined by patrilineal descent. The emphasison descent sometimes obscures important differencesbetween what Keesing (1975:9-10) calls culturalcategories and social groups. An example of the former isthe xeem or "patri-clan." Hmong who have the same clanclassification may choose to recognize reciprocal obligationsand form a group, but this is not required simplybecause a clan relationship exists. Those who trace theirdescent back to a known ancestor are of one sai ces,i.e., lineage or branch. This connection provides a muchstronger justification for common group membership. Patrilinealdescent, however, is only one of a number offactors that determine the composition of Hmong kingroups.A common adaptive unit of the Hmong is an associationof closely related extended families. There is noHmong term that refers exclusively to this kind of association,but the expression ojiv neeg or "close group"applies in certain contexts. Brothers and lineage cousins,uncles and nephews readily form such groups.Theirfamilies constitute a perdurable core to which are addedmore distant lineage and clan relatives. Occasionally, amarried male prefers to join the pawl ma of his moresuccessful and/or more congenial in-laws. The willingnessof the Hmong to recognize even the most distant ties ofconsanguinity and affinity has been crucial in reestablishingfamily mutual assistance groups that were decimatedand scattered due to war, refugee flight andresettlement.The assumptions and principles mentioned earlier inregard to external political relations apply in modifiedform to Hmong kinship organization. Every family grouptries to expand the range of its alliance network by accommodatingthe needs of other groups who will reciprocatewhen asked. Some kinds of help, e.g., sharing incomeand housing, can only come from within the familygroup. A family group will be in a better position tohelp and protect members if it has capable leaders whounselfishly promote the advancement of the entire group.A unified association of families is the most effectivesocial arrangement for adapting to a changing and some-61

50 THE HMONG IN TRANSITIONtimes hostile environment.American Hmong worry that the young are becomingmore independent and less committed to group interests.As resettlement problems abate, and Hmong grow up exposedto values that elevate personal achievement over socialduty, reliance on and devotion to the kin group may weaken.I anticipate that any reduction in the size and influenceof Hmong kin groups will be interpreted by non-Hmong observers as a late ste;e of assimilation. The expansionand intensification of internal coalitions historicallyhave varied in direct relation to externalthreat. A less intensive oppositional process in Americawill result in smaller corporate kin groups, but largenetworks of groups related by descent and marriage willbe maintained as security against an uncertain future.IDENTITY MAIN TENANCEStudies of Hmong refugee adaptation must present morethan "before" and "after" pictures that show the Hmongsuddenly crossing the threshold from a stable, traditionalexistence to a situation of rapid development in amodern society where ethnic differences are reduced totrivial proportions. Extensive culture modification andreplacement are certainly taking place, just as occurredin China, Laos and Thailand during earlier periods ofHmong history. It is important to understand thesechanges by analyzing their causes. We should also befocusing on how the Hmong are using their acculturationexperiences in America to reinforce Hmong social and culturalidentity.Political standing, gender, kinship and other kindsof social roles are being redefined to fit circumstancestotally new to the Hmong. Borrowed elements are beingadded to symbols of ethnicity and kinship. Constraintsimposed by majority society are forcing the Hmong to reviseimportant social rituals. It is also true that Hmongadjustment strategies in the United States replicate approachesfollowed in Southeast Asia for managing crossculturalcontacts and preserving Hmong ethnic integrity.The processes of Hmong culture change appear to be bothlinear and cyclical.

PROCESSES OF IDENTITY MAINTENANCE 51I began by presenting a brief overview oftory as an illustrationYaqui his-of the difference between the retentionof cultural traits and the maintenance of anidentity system. The Hmong, like the Yaqui, have survivedextremely difficult times by changing without sufferingidentity loss. This kind of resilience teachessome very important lessons. I again quote Edward Spicer(1980:362).Perhaps a people which endures, whether for onlya century or for millennia, embodies the most importantkind of social unit which men create.The persistent identity systems which characterizesuch social units are living, cumulative interpretationsof the significance of human life.Each is unique and unrepeatable. Each is notonly the summation of one people's accumulatedunderstanding of their purpose and destiny; itmay also be a window for other peoples on thegeneral human purpose. When peoples of differentidentities learn to cross one another's ethnicboundaries, they come closer to the fullest possibleappreciation of the meaning of human existence(1980:362).NOTES1Linnekin (1983:241) observes that:... tradition is a conscious model of past lifeways that people use in the contruction of theiridentity. The inheritance of an authentic traditionand the naivete of the folk are illusory.The use of a defining tradition exemplifies theobjectification ... and the invention ... of asymbolic construct. As a self-conscious category,tradition is inevitably "invented" ... the selectionof what constitutes tradition is always madein the present; the context of the past is modifiedand redefined according to a modern significance.A bibliography relevant to the issue of ethnic definitionis provided by Linnekin.6a

52 THE HMONG IN TRANSITION2The preferred self-designation in the Laotian provinceof Sam Neua is Hmoob Leeg.3Animism refers to a belief in spirits that reside innature and sometimes interact with people.4See Olney (1983) for a complete list of references onHmong kinship.REFERENCESDunnigan, Timothy. 1982. "Segmentary Kinship in an UrbanSociety: The Hmong of St. Paul-Minneapolis." AnthropologicalQuarterly 55:126-134.Dunnigan, Timothy, and Chia Vang. 1980a. "Hmong AttitudesToward Language Maintenance: A Community SponsoredTranslation Project." Paper given at the 40th AnnualMeeting of the Society for Applied Anthropology.Denver.Dunnigan, Timothy, and Tou Fu Vang. 1980b. "NegotiatingMarriage in Hmong Society, An Example of the Effectof Social Ritual on Language Maintenance." MinnesotaPapers in Linguistics and the Philosophy of Language6:28-47.Fishman, Joshua A. 1974. "The Sociology of Language: AnInterdisciplinary Social Science Approach to Languagein Society." In Current Trends in Linguistics andAdjacent Arts and Sciences, edited by Thomas A.Sebeok, 12:1627-1784.Keesing, Roger M. 1975. Kin Groups and Social Structure.New York: Holt, Rinehart, and Winston.Le Bar, Frank M., Gerald C. Hickey, and John K. Musgrave,1964. Ethnic Groups of Mainland Southeast Asia. NewHaven: The Human Relations Area Files Press.Larteguy, Jean, avec la collaboration de Yang Dao. 1979.La fabuleuse aventure du peuple de l'opium. Paris:Presses de la cite.64

PROCESSES OF IDENTITY MAINTENANCE 53Linnekin, Jocelyn S. 1983. "Defining Tradition: Variationson the Hawaiian Identity." American Ethnololist10(2):241-252.Olney, Douglas P. 1983. A Bibliography of the Hmong(Miao) of Southeast Asia and the Hmong Refugees inthe United States, 2nd ed. Minneapolis: Center forUrban and Regional Affairs. Southeast Asian RefugeeStudies Occasional Papers, No. 1.Savina, F. M. 1924. Histoire des Miao. Hong Kong:Imprimerie de la Societe des Missions-Etrangeres.Spicer, Edward H. 1971. "Persistent Cultural Systems: AComparative Study of Identity Systems that Can Adaptto Contrasting Environments." Science 174:795-800.. 1980. The Yaquis: A Cultural History. Tucson:University of Arizona Press.Weins, Herold J. 1967. China. New Haven: The ShoeString Press.C5

CULTURE AND ADAPTATION:HMONG REFUGEES IN AUSTRALIAGary Yia LeeIn an article on Hmong refugees in Sanfornia,Diego, Cali-Scott reports that they respond toversity "byexternal ad-becoming more Hmong rather than less(1982a:154).so"This is manifested by theirsettle inpreferencethetosame geographical area and tomostly with membersinteractof their own ethnic group. The resultis a heightened sense of ethnicity which,reinforced althoughby environmental factors, basesstrong itself onprimordial sentiments such as "alitical tradition ofautonomy po-and social self-reliance" and a "stronghomogeneous community" with no major differences inguage, religion or customs.lan-These attributes, whatever theirrefugees,relevance totell a casualthereader very little"being about whatHmong"isor, particularly, what isHmong.""being moreTo a Hmong, what distinguishes the Hmongother people is fromtheir Hmong way ofis centeredlife. This way of lifearound shifting agriculture, amutually language withintelligible dialects, a strongtorbelief in ances-worship and animism, a division ofto family membership anzlabour accordingsex, a social structure basedkinship ties onthrough the patrilineage and clanpatrivirilocal systems, apattern of residence, ationhistory offrommigra-southern China and a long tradition ofstateless.beingLet me elaborate on some ofHmong society.these components ofHere I will focus on economicity, social productiv-structure and religion. Thisbetterwill allow us aevaluation of Hmong culturaltralia, articulation in Aus-and may help in determining whether orrefugees can becomenot themore Hmong in the West or whetherthey have to change in order to accommodate to the newenvironment.5566

56 THE HMONG IN TRANSITIONTHE TRADITIONAL PRODUCTIVITY BASENo Hmong lifestyle is possible without subsistencefarming, supplemented by foraging, hunting, some fishingand handicrafts. Agriculture is the dominant form of economicproduction; it is closely related to the Hmong'sreligion through such practices as ritual offerings toappease fieid spirits and "first fruit" ceremonies forthe dead members of one's lineage in order to seek theirspiritual protection. The "prestige sphere" identifiedby Dalton (1971:14-16) in another context is also evidentin that the more agricultural goods a person can produceand display, the more his position in the community isenhanced. Agriculture, thus, gives rise not only to ritualsbut also to the ways in which social roles and thedivision of labour within the household are observed. Tosome extent, the Hmong come close to the Marxist contentionthat the economy determines the structures and insatutionsof society (Marx 1965:121).The role of family members, for instance, is definedby their order of birth as well as by their abilities tocontribute to the survival of the family as a whole.Thus, a husband's duties are to act as head of his householdand to provide for its members' physical and spiritualwelfare. He is responsible for the selection offarming sites and the felling of trees when clearing newswiddens (farmland prepared by the slash-and-burn method)but is rarely concerned with domestic chores, apart fromperhaps getting firewood. A wife, on the other hand, isgiven the tasks of caring for children, preparing meals,feeding chickens and pigs owned by the family, sharing inall agricultural activities with her husband and consultinghim on family needs or major decisions.In addition to the division of labour, the socializationof children is also affected when farming keepsfamily members close to one another with little or nooutside contacts and influences. Grandparents, parentsand children are together at nearly all times, whether inthe home or in the field. The norms of primary group cooperationand mutual assistance are thus instilled earlyin life through the labour requirements imposed by theeconomic base. Furthermore, agricultural traditions affectthe allocation and consumption of goods. While everyable-bodied person participates in the growing of crops,67

CULTURE AND ADAPTATION: HMONG REFUGEES IN AUSTRALIA 57these commodities are for the consumption of the whc,:ehousehold. Individual allocation of goods occurs onlywith jewelry, some domestic animals, clothing, opium ormoney.It is obvious that subsistence cultivation plays amost significant part in shaping social roles and relationswithin the household.' In order to achieve theireconomic ends, the Hmong have evolved their own rules offamily behaviour with rights and obligations that must beadhered to. As stated by Cohen (1968:2), in trying tosatisfy their needs, people are brought into relationshipwith one another in an institutionalized framework withcommon &a:Lenges and socially prescribed expectations.This is true of Hmong, as of other human groups.RELIGION AND SOCIAL STRUCTURETwo other dominant features of Hmong society are itssocial structure and religion. Hmong social groupings aregenerally the product of kinship organization and itsmanifestations through the rituals of ancestor Wt. rFtlip,two aspects of Hmong culture which are intricately interrelated.The fortune of an individual or family dependson close observation of ancestral reverence, geomancy andkinship networks. A Hmong's religion cannot be separatedfrom his social groupings, and his relations with otherHmong are meaningful only in terms of whether or not theyshare similar ancestral rites. Therefore, he cannot dowithout his kinsmen and a good knowledge of their ritualsin order to carry on his Hmong existence.When two Hmong meet for the first time, their immediateconcern is to establish their clan identities sothat they can relate to each other. It is simple to discoverone's clan through one's surname. If they belongto the same clan, the next question will be which subclanthey originate from. This is none by inquiring whetherthey perform similar rituals in relation to funerals, the"door ceremony" and the "ox ceremony"; whether the gravesof their dead are of the same construction. If these commonfactors are established, membership to a subclan isconfirmed. From this point, they may try to learn whetherthey descend from the same ancestor and, thus, belong tothe same lineage. A lineage is known as a "cluster of68

58 THE HMONG IN TRANSITIONbrothers" (ib cuab kwv tip or "one ceremonial household"(ib tus dab qhua. Members of a "ceremonial household"can die and have funerals in one another's houses, asthey would share similar ancestral rituals. People merelybelonging to a clan or subclan cannot be granted thisceremonial privilege.What this means is that the social structure dividesHmong into different groupings united under clans, sub -Gans and lineages. For a man, the clan makes it possibleto know which group to identify with and which womanhe can or cannot marry. After marriage, he is expectedto observe the residential prescription laid down by hiskinship system and reinforced by his ancestral cult. Patrivirilocality is preferred, at least until the couplehas two or three children and can establish its own nuclearhousehold away from the extended family. This doesnot prevent the man from maintaining close relationshipswith relatives of his wife. With the exception of materialsupport, there are no ritual bonds between a marriedman and his affinal relatives because religious ceremoniesare reserved for agnatic ancestors.The Hmong do not have elaborate ceremonies for birthor during various stages of life. Explanations for themeaning of life are to be found in funeral rites, themeeting point between the living and the dead. This relationshipbetween ancestors and their descendants, betweenreligion and social structure, gives the Hmong aclearly defined way of life. As postulated by Radcliff e-Brown (1941:40), such a belief system gives incentives tothose taking part in them to have a sense of dependenceon heir ancestors and to commemorate them for havingthem their life, while they are spurred on to bringgivc..uptheir descendants to whom they will one day also becomerevered ancestors. It is this sense of duty thatcreates a direct association between a Hmong's religionand his social structure, particularly his lineage.In their world view and ceremonial practices, theHmong see an explanation for their social groupings; intheir kinship ties, they find reason to maintain theirreligion. Religious ceremonies renew group sentiment,strengthen lineage solidarity, and inspire members tocarry out their obligations to the living, the dead andthose yet to be born. These factors, in Geertz's phrasing,form the group's ethos, and are "rendered intellec-69

CULTURE AND ADAPTATION: HMONG REFUGEES IN AUSTRALIA 59tually reasonable by being shown to represent a way oflife ideally adapted to the actual state of affairs theworld view described[,] while the world view is renderedemotionally convincing by being presented as image of anactual state of affairs peculiarly well arranged to accommodatesuch a way of life" (Geertz 1966:3).IMPACT OF RESETTLEMENTHaving discussed what I consider to be the most salientaspects of Hmong culture, let us now see how theyhave been maintained or adapted to respond to the demandsand needs of a Western environment. To do this, we willneed to examine first the major factors which make Hmongcultural adaptation necessary. Here I will use the Australianexperience of the Hmong refugees as illustrationof the issues involved. Let me add that the Hmong situatic,in Australia is by no means typical of other Hmonggroups, although some common features may be shared bythem.In contrast to their subsistence farming in Laos,where family members all w3rk together, the complex wageeconomy of Australia only accepts skilled adults into itswork force. For the refugees, this involves those whohave some degree of training and who can speak basic English.The division of labour is no longer based on familymembership, but on outsiders' decisions as to who is chosenfor work and who to be made idle. This means thatfor the 344 Hmong in Australia, only 50 are gainfully emplojed,representing 44.6 percent of the total employablepopulation of 112. Thus, a total of 62 adults are notemployed. There are 208 Hmong children in Australia, ofwhom 61.5 percent, or 128, are in primary and highschools.The old tradition of joint economic productivity byall household members is seriously disrupted with thosewho work going their separate ways each morning while theelderly and the unemployed remain home with or withoutchildren to attend to. The wage-earners are often awayfrom their families for eight to twelve hours a day, fiveto six days a week, spending only weekends and eveningsat home. Children of school age are left to find companywith Mickey Mouse or Sullwinkle on television in the mor-70

60 THE HMONG IN TRANSITIONning, and return from school to watch Sesame Street orthe Brady Bunch in the afternoon. While at school, theyread Walt Disney books or Shakespeare, interact withEnglish-speaking children, and learn from Anglo-Europeanteachers using a curriculum which makes no mention oftheir culture. The parents often actively support thisacculturation process, so that the children quickly identifywith the new society due to their ease in acquiringEnglish.This almost daily fragmentation of the family byschool and work has eroded the cherished idea of a cooperativehousehold united by common interests and economicparticipation. With 35 of the 67 family heads withoutwork, many parents are no longer productive. They oftendo not speak English and have to depend on their childrenfor simple tasks involving contacts with non-Hmong outsiders.This reversal of roles and the consequent lossof self-esteem on the part of the older family membershave made many of them feel despondent. The demands ofthe West have precluded them from fulfilling their traditionalroles as providers and decision-makers for theirfamilies. Thus, social values regarding parental authority,filial respect, clan obligations and reverence ofancestors become gradually irrelevant or unenforceable.This erosion of Hmong social values and norms withinthe family is also felt in the area of interpersonal relationshipsbetween members of the Hmong community. Thesocial structure, as I have discussed previously, is alreadygreatly weakened by the fact that many familiesshare only clan membership but no other kinship features,and a few families are the sole representative of theirclan in the new country. This means that there is littleor no similarity in rituals between them, since the sharingof rituals is restricted to lineage and subclan members.Of all the clans in Australia, the Lee has nineteenfamilies with five lineages, the Yang sixteen familieswith four lineages, the Vang thirteen families and threelineages, the Thao ten families and three lineages, theXiong and Vue three families and one lineage each, theMoua two families from one lineage, and the Chang onefamily from one lineage. Families belonging to the samelineage are often young and number no more than five tosix, because only households with at least one skilled or71

CULTURE AND ADAPTATION: HMONG REFUGEES IN AUSTRALIA 61educated member are admitted by the Australian Governmentfor resettlement. Furthermore, not all members of a lineagewere able to escape from Laos or to resettle in theWest, thereby significantly reducing the traditional networksof kinship and mutual assistance.The young and skilled families preferred by Australiaare generally ignorant of religious beliefs and practices.Most do not have with them older members who arewe!1 experienced in Hmong customs. Unless they are relatedto those who have such older members, they do notknow where to turn in time of spiritual need. They existin a religious limbo, neither as Hmong with knowledge andobservance of ancestor worship, nor as Australians withfaith in Christianity.2 Some appear to be content withthis state of affairs, filling the voids of their liveswith material objects. Sooner or later, however, theyhave to turn to other Hmong, because cultural and languagebarriers prevent them from reachingof the host society.out to membersCULTURAL CONSCIOUSNESS AND ADAPTATIONIn the face of this constant intrusion on their traditionsby incompatible external forces, the Hmong obviouslyfeel their very identity threatened. This is especiallytrue when their unity as a household, lineage orclan is adversely affected by conflicting claims of thenew country. Like any human groups in a state of uncertainty,most of those in Australia have reacted by tryingto regain direction by improving themselves with educationand economic productivity, or by joining force as agroup iu order to forge for themselves a new identity.At the individual level, the cultural value of selfsufficiencyis still strongly followed, despite many hurdlesto overcome. The Hmong are now given the chance tostand on their own feet, following years of surviving onhandouts in refugee camps in Laos and Thailand. They knowthat the material comfort enjoyed by members of the newcountry can be theirs if they strive hard enough for it.Many of those who are able to find factory work see eventhis as a big step forward compared to the farming theywere used to in Laos, for they are no longer at the mercyof nature and war. As a measure of their adaptability,72

62 THE HMONG IN TRANSITIONforty-two of the fifty -six households now own cars, andeighteen have bought their own houses. Two families arein the process of buying land for market gardening. Allthis has been achieved within the space of four to sixyears of resettlement.The formal group, for its part; also encourages memberstowards self-reliance and towards articulation oftheir culturebut in a manner acceptable to a Westernsociety. I am referring to the Hmong-Australia Society,which was first set up in the State of Victoria in 1978.At present, it has branches in New South Wales, the AustralianCapital Territory and Tasmania, with a federalbody to make policies and to coordinate the activities ofthe state branches. Since I am more familiar with thesituation in New South Wales, I will discuss mainly thesociety's activities in that state, which has 204 of thetotal Australian Hmong population of 344, in contrast toVictoria, which has 113, and Tasmania with 27.3In a similar vein to other ethnic-based organizations,the Hmong-Australia Society has as its main objectives:(1) to assist Hmong refugees in their resettlementin Australia, (2) to foster mutual acceptance betweenHmong and other ethnic groups, (3) to uphold Hmong culturaltraditions, and (4) to safeguard Hmong interest ingeneral. Membership is not restricted to Hmong and isentirely voluntary. Nevertheless, nearly all the Hmongin the country are financial members, except for twofamilies.To achieve its objectives, the society works throughcommittees covering such areas as welfare assistance, financialloans and donations, . hospital and home visiting,education, public relations, sports and recreation, women'saffairs, culture and religion, and fund-raising. Itis clear that most of these concerns reflect a consciousattempt to put Hmong cultural values into practice in anew environment where the kinship networks which traditionallyoversee these activities are virtually absent.The society thus replaces the social structure by being afocal point for members to fall back on in tine of celebrationor crisis.One of the first tasks undertaken was to get membersto understand the Australian government's policies onmulticulturalism and to steer the group's activities tobe in line with these policies. For example, ethnic73

CULTURE AND ADAPTATION: HMONG REFUGEES IN AUSTRALIA 63groups are now encouraged to maintain their cultures"without prejudice or disadvantage" and "to embrace othercultures" (Galbally Report 1978:4). This is so long asthere is acknowledgement of "common values which give allcitizens a sense of being Australian," with importancegiven to the English language (Commonwealth Schools Commission1983:31). Hmong are, therefore, urged to considerthemselves as Australian Hmong with a unique culturalcontribution to make to their country of resettlement andwith an obligation to identify with it.Faced with only a residual social structure amongits members, the society tries to promote mutual supportacross clan boundaries by insisting that Hmong refugeesassist each other on the basis of their common ethnicbackground rather than membership to a lineage or clan.Overall, the majority of members have shown a genuinedesire to help lessen clan and lineage consciousness bydevoting time and energy for the common benefit of thegroup without regard to kinship ties. There are, ofcourse, a few families which are still reluctant to takepart in a common New Year celebration because of theirreligious beliefs, but this is a minor problem.A good example of the society's involvement in culturaladaptation is its Ethnic School. Formed in 1980,it was designed to conduct weekend classes on Hmong languageand culture for Hmong children who are growing upin Australia and who are not exposed enough to their ownmother tongue and their parents' culture. The aim is toenable the students to read and write Hmong, and to appreciatethe customs of their older family members sothat they may understand and assist the latter in theirsettlement. With operation costs met from federal governmentfunding, the school has a current enrollment offorty-three, with an attendance rate of about thirty-twoin two classes. Two Hmong teachers alternate in theteaching of each class.After using an old primer from Laos for a year, theteachers found it to be too difficult, since many of thewords refer to traditional objects no longer found inAustralia, and therefore convey little meaning to therefugee children. This difficulty was compounded by thelack of suitable texts on Hmong culture. With a grantfrom the state government, the society has attempted tofill this gap by developing its own teaching materials74

64 THE HMONG IN TRANSITIONwith more relevance to the Australian context. Two booksfor learning Hmongone for young children and one forolder studentshave been completed, along with a bookof traditional songs (for learning old forms of poetry),a collection of modern songs (for learning new poetry), abook of proverbs an,1 a collection of folk tales. A Hmonggrammar and a text on the changing way of life of Hmongrefugees (for social studies) are being completed. Thenew books, in very simple form, have been used experimentallyin class during 1983, and have drawn much enthusiasticresponse from children and parents.Cultural materials are taught not only through theuse of these texts, but also through simulated performancesof simple ceremonies, colour slides, recordings andsamples of relevant artifacts. Many young men have successfullylearned the mechanics of common rituals throughrole-playing. The learning of more complex rituals hasbeen offered to older individuals according to their personalinterest, with the elders in the Hmong communityacting as teachers.Realising that the teaching of Hmong language andculture may offer the refugees a sense of cultural continuitybut will not be sufficient to help them integratein the Australian setting, the society also teaches themEnglish, or arranges to have English and simple skillscourses provided on a short term basis by government instrumentalities.This applies in particular to the adultThere have beenHmong, many of whom were illiterate.classes on sewing and dressmaking, Australian laws andsociety, basic mathematics, general health education andhomemaking. The primary aim is the learning of English,but these language skills are acquired through other subjectsdeemed useful to the social adjustment of theHmong. Similar orientation information is also publishedin both Hmong and English in the society's newsletter.Initially, most of the Hmong were reluctant to carryout any religious ceremonies in Australia, but they havenow gained enough confidence to perform readily such ritualsas "soul-calling," "wrist-stringing" and shamanictrances. Cultural expression outside the home or at publicfunctions is limited to the display of handicrafts,the playing of reed-pipe music and flute, and traditionalsinging. Simplified versions of the New Year Celebrationand funeral proceedings have been adopted, mainly from75

CULTURE AND ADAPTATION: HMONG REFUGEES IN AUSTRALIA 65lack of time and manpower rather than the fear of beingfound objectionable.Unlike some of those in the United States of America(Scott 1982b:63-85), the Hmong in Australia have neverquestioned the relevance of traditional beliefs to theirnew life, even when rituals are restricted to those notinvolving killing of animals. Several explanations forthis disparity between the two groups could be suggested.Firstly, the Australian Government's refugee program doesnot depend entirely on church and community sponsorship.Most of the Hmong were admitted under this official programwhich provides all necessary support services, withminor assistance from voluntary agencies. They do not,therefore, come into early contacts with people who mayinfluence them to doubt their traditional religion and toembrace a new one.Secondly, the Australian Government actively supportsthe idea of a multicultural nation, as mentionedpreviously, and has made funds available to variousgroups to test out the concept, especially in the fieldsof culture, arts and education. The Hmong happen to bein Australia at a time when this experiment is only atits beginning stage. By joining in, they are thus encouragedto carry on with their changing traditions whilelearning to adapt to the host community.Thirdly, and most importantly, the leadership andthe very small size of the Hmong population have beencrucial factors in this process of cultural adjustment.The informal decision-makers, whether of a traditional orWestern orientation, have been most accommodating to eachother, with decisions based on the democratic process insteadof the dictates of a few individuals. The old leadershipacknowledges the need for change, and the youngerleaders know that change has to be gradual, with culturalelements acting as both an impetus and a buffer. Theseleaders and their followers have equal roles in guidingthis change towards the common good of the Hmong andtheir new country.This is, of course, easier said than done, for manyproblems have to be faced both within and outside theHmong community. Overall, however, I do believe that 97.7percent of them share in this long-term vision. Withmutual support between themselves and the continuing assistanceof such organizations as the Lao Community Ad-76

66 THE HMONG IN TRANSITIONvancement Co-operative, local welfare agencies and variouslevels of government, the Hmong may yet achieve thisadaptation through the use of their culture as a meansand an anchor. After all, they are a universal minoritygroup, and the adage of "home is where you make it" hasalways applied to their situation, even if home will takea bit longer to make in the West.HMONG RESEARCH AND SOCIAL CHANGELest I am taken to task for being too optimistic andfor talking about hopes rather than reality, let me nowbriefly justify my arguments.As I have read literature on the Hmong in Thai, Lao,Hmong, French and English, I have been struck by a lackof perspective in many of these works. Too often, sweepingnegative statements or generalizations are made frominformation obtained on the basis of single encounters orinterviews. This is true especially of politically motivatednewspaper reporters, travellers and arm-chair writerswho often base their comments on second- or thirdhandsources of information. Even government officialsand serious scholars have not escaped this tendency torefer to old, outdated accounts of the Hmong, and to makefurther sweeping statements after a few short visits tothem.Let me illustrate what I mean. Geddes (1970), followinga few years of field work among a group of GreenHmong in northern Thailand, came to the conclusion thatthe Hmong economy is a cash economy based on opium, withtheir whole migration pattern and way of life affected bythe need to search for opium fields. He (1976) explainedHmong polygamy on the grounds that the men marry morethan one wife in order to obtain more labour for theiropium growing. Geddes did not take enough account of thefact that thousands of other Hmong in other places do notgrow opium; and that polygamy is strongly condemned inHmong society, whether or not opium is produced.Recently, Cooper (1978) speculated that "almostevery Hmong family will have some experience of patriuxorilocalityat some time." This is because five of theeighty-seven households he studied in northern Thailandexhibit this residential pattern. How valid is this in-7 7

CULTURE AND ADAPTATION: HMONG REFUGEES IN AUSTRALIA 67terpretation when only 5.7 percent of the Hmong in thesample live with relatives of the wives? Again, otherfactors such as the rules of the ancestral cult and thekinship system have not been given adequate consideration,for the Hmong hold patriuxorilocality as being sociallyand religiously unacceptable.More relevant to our concern here are statements byScott I referred to earlier. In his conference paper twoyears ago (Scott 1982b:66-67, 81), he discussed in detailhow "the relocation of the traditional Hmong religioussystem into an incompatible environment ... has entailed(1) the questioning of belief, much of which now seemspainfully inappropriate in an environment with which itis no longer resonant, and (2) the abandonment of ritualin ail but a few of the most conservative families ...."The annual New Year Celebration is the only occasionthese rituals are taken out of their closets to be displayedout of context as a reminder of a "rapidly recedingpast." In an article on ethnic solidarity among thesame group of Hmong in San Diego a few months later,Scott wrote that they have become "more Hmong" and interact"less and less with outsiders, even with those whosetask it is to help them" (1982a:154). Could the "rapidlyreceding past" have come reging back in such a shorttime, like strips of a movie film being pulled backward?It is true that more of the San Diego Hmong have nowjoined in formal meetings or formed ethnic-based organizations,but this does not mean that they have becomemore Hmong. It is only that they have changed and havelearned the value of formal organizations in Western societies.The formalized approach to the requirements oflife is in itself un-Hmong, a Western ritual quicklyseized upon by the refugees. The traditions based onsubsistence agriculture, as I discussed at the beginningof this paper, are mostly now inapplicable or unavailable,and many of them have to be abandoned or changed.If the Hmong react to an alien environment by turning toeach other or by migrating, we would not have five millionHmong in China today, but only a few thousand inneighbouring countries where they were supposed to seekrefuge from Chinese oppression. If they cannot be assimilatedin Laos, why do so many Hmong women in the Weststill wear Lao skirts while nearly all the men use Laowords that have been integrated into the Hmong language78

68 THE HMONG IN TRANSITIONwhenever they speak Hmong?I would argue that, given time and opportunities,the Hmong cannot help but adapt themselves to a new society,change their way of life, and make inroads intovarious social strata and employment structures. The keyfactors, however, are time and open doors. This changecannot be achieved quickly. It has to be slow and gradual,from the physical to the abstract, from the tangibleto the intangible, beginning now perhaps with cars,housing and clothes, but eventually ending with the adoptionof Western values and languages. Despite opinionsto the contrary, the Hmong objectively are not the autonomouspeople some of us like to make them out to be:they are usually accommodating, keen to become acceptableto other groups.The above comments are not designed to discourage.On the contrary, I would like to see more research anddebates on Hmong adaptation in the West. May I urge thatin our search for knowledge, we should be more consciousof what is hidden from us, what is not revealed butshould be known before useful generalizations can bemade. The United Nations Crop Replacement Project foropium farmers in Thailand failed, because there is more,much more to the Hmong economy than the need for cash.To avoid such pitfalls in social science research, Ibelieve it would be more fruitful to take both the "emic"and "etic" approaches in which the researcher's observationsand perceptions are combined with those of the researchsubjects, with information checked and crosscheckedbefore it is used. This was well stated byGoodenough (1970:104) when he said that: (1) whenever wewish to know what people are doing and why, or what theyare likely to do, we must know what kind of things theysee and respond to, and (2) we must know what they believeto be the relations between these things and whatthey see as the possible course of action for dealingwith them. As we try to unravel the complexity of theHmong resettlement experience in the West, I hope thatthese premises will help yield information, interpretationsand ideas on the Hmong in a much wider perspective.79

CULTURE AND ADAPTATION: HMONG REFUGEES IN AUSTRALIA 69CONCLUSIONhave tried to compare the new Hmong experience inIthe West with their traditional existence in the hills ofLaos in order to see whether or not they can maintaintheir former way of life in the new social environment.I have used the case ...:f the Hmong in Australia as illustrationof the fact thi no matter how hard an uprootedethnic group tries to k lld onto its links with the past,those links will gradually be severed and forgotten unlessthe group's culture can be adapted to fit into thepresent. Culture is not an impediment, but an instrumentof social change. Carefully guided, its transformationmay lead to a new form of cultural adjustment. Neglected,it will be doomed to extinction. It is argued here thatthe active articulation of their culture will help therefugees maintain their personal identity in the face ofperplexing external demands, while they attempt to adjustto these demands and to acquire new cultural elements togive themselves a new identity befitting life in a Westernsociety. The cultural past can act as a buffer tothe upheavals of the current social change, and can givedirection to the uncertainties of the future. Given somepatience and the opportunities, this adjustment will certainlybe achieved, because most Hmong want to adapt, tobecome self-sufficient and to enjoy peace and the comfortof life. Where else can they get all of that, but in theWest?ACKNOWLEDGEMENTI wish to thank the "The Hmong in Transition" conferenceorganizers for financial assistance to presentthis paper, and the Lao Community Advancement (N.S.W.)Co-operative in Sydney, Australia, for allowing me timeoff work to attend the conference and to visit Hmongrefugees in the United States.So

70 THE HMONG IN TRANSITIONNOTESIHmong economic cooperation rarely goes beyond the householdlevel. Inter-household or village relations arebased on kinship ties and other factors rather than thenecessity for economic productivity.2There has been no conversion into Christianity so farbecause of the group's cohesion.3No Hmong live in the states of South Australia, WesternAustralia, Queensland or the Northern Territory.REFERENCESCohen, Y., ed. 1968. Man in Adaptation: the CulturalPresent. Chicago: Aldine. Commonwealth SchoolsCommission, Australia.1983. Report on the Ethnic Schools Programme.Canberra: Australian Government Printing Service.Cooper, R.G. 1978. "Unity and Division in Hmong SocialCategories in Thailand." Studies in ASEAN Sociology,edited by P.S.J. Chen and H.D. Evers. Singapore:Chopmen.Dalton, G. 1971. Traditional Tribal and Peasant Economies.Mc Caleb Module in Anthropology. Reading,Mass.: Addison-Wesley.Geddes, W.R. 1970. "Opium and the Miao." Oceania,XLI(1): 1-11.. 1976. Migrants of. the Mountains. Oxford:Clarendon Press.Geertz, C. 1966. "Religion as Cultural System." InAnthropological Approaches to the Study of Religion,edited by M. Banton, pp. 1-46. London: Tavistock.Goodenough, W. 1970. Description and Comparison in CulturalAnthropology. Chicago: Aldine.81

CULTURE AND ADAPTATION: HMONG REFUGEES IN AUSTRALIA 71Marx, K. 1965. Precapitalist Economic Formations, editedby E.J. Hobsbaum. New York: International Publishers.Radcliffe-Brown, A.R. 1941. "The Study of Kinship Systems."Journal of the Royal Anthropological Instituteof Great Britain and Ireland 71:1-18.Report of the Review of Post-arrival Programmes and Servicesfor Migrants (known as the Galhally Report).1978. Migrant Services. Canberra: Australian GovernmentPrinting Service.Scott, G.M. 1982a. "The Hmong Refugee Community in SanDiego: Theoretical and Practical Implications of itsContinuing Ethnic Solidarity." AnthropologicalQuarterly 55 (3):146-160.. 1982b. "A New Year in a New Land: ReligiousChange Among the Lao Hmong Refugees in San Diego."In The Hmong in the West: Observations and Reports,Papers of the 1981 Hmong Research Conference, editedby B:uce T. Downing and Douglas P. Olney, pp. 63-85.Minneapolis: Center for Urban and Regional Affairs,University of Minnesota.84

1 THE MIAO IN CONTEMPORARY CHINA:A PRELIMINARY OVERVIEWLouisa ScheinINTRODUCTIONOf the more than five million Hmong and Miao scatteredaround the world today, the vast majority stilllive in south and southwest China, the land they have occupiedfor centuries. From the perspective of Hmong inother parts of the world, this region has been regardedas a remote ancestral homeland, a repository of archaiccultural forms. The result has been a tendency to overlookthe fact that development and political changes havebeen a reality for ethnic groups in China just as theyhave been in Southeast Asia. Although many Miao in Chinaare still living in the mountains acid engaged in agriculture,there have been dramatic changes in their livesover the course of the twentieth century and especiallysince 1949. It is thus both important and illuminatingto update and put into context our present knowledge ofChina's Miao.To begin, this paper will describe some characteristicsof the Miao group in China, with examples to illustratethat a thorough ethnographic understanding entailsan examination of change as well as "tradition."1 Thepaper will review relevant aspects of minority policy toprovide a background of recent developments affecting allethnic groups in the People's Republic of China. It willdiscuss the classification of the Miao as a nationalityand the region they occupy. Some features of language andcustom will be described along with recent innovations asa result of education. The paper will attempt to showsome aspects of unity in an ethnic group that exhibits agreat deal of apparent diversity both in language and incultural traits.The information presented here is a result of ongoingwork among Lao Hmong refugees in the United Statesand a year-long comparative study of the Hmong/Miao groupin France, Thailand and the People's Republic of China.'Research in China was conducted over a se' en-month periodfrom March to October 1982. Work was primarily under the73 83

74 THE HMONG IN TRANSITIONauspices of the Central Nationalities Institute inBeijing where I resided as visiting scholar for a totalof three and a half months. The remainder of the timewas spent traveling, with short-term research in NationalitiesInstitutes in the southwest provinces.Information was collected through interviews withMiao and Han Chinese scholars and researchers, and withMiao who were students or otherwise members of NationalitiesInstitutes. Regular meetings were held at which anindividual or a group of scholars presented relevant materialfollowed by questions and discussion. In the caseof "cultural exchange" meetings I, in turn, made presentationsconcern...g the Hmong of Southeast Asia or theUnited States. In addition, published material in Chinesewas collected and translated. This paper reliesprimarily on these Chinese sources, leaving treatment ofthe extensive body of Western literature for future work.Where there are no specific citations, the informationgiven is cumulative, based on personal communications andinterviews in China.ETHNIC CLASSIFICATION IN CHINAThe population of China suroassed one billion in1982. Of that population, 6.7 percent are the officiallyrecognized "minority nationalities." These include theMongolians and Tibetans, the many Islamic peoples, andnumerous smaller groups concentrated especially in thesouth and west. In contemporary China, the term "minzu"3or "nationality" is used for ethnic groups, including HanChinese, of all sizes and socioeconomic "stages of development."This usage differs significantly from the WesternEuropean connotations of the term "nation-state" (Fei1980:97-98). Since 1949, the People's Republic of Chinahas carried out a systematic effort at identification andclassification of minority groups for the purpose of devisingappropriate policy, delimiting areas of "autonomous"government (Weng 1950:7), and allocating seats forrepresentation in provincial and national government.This effort has undergone several phases and continues tothe present day.Between 1949 and 1953, a process of formal selfidentificationtook place, resulting in the registration84

THE MIAO IN CONTEMPORARY CHINA 75of more than 400 minority groups. Chinese ethnologistsconsidered this number misleading and, as a result, activeresearch began in 1953 with the aim of discerningwhich names were actdally local variants or designationsof subgroups (Fei 1980:94). At the outset of this project,experts estimated that there were probably aboutsixty groups that could accurately be classed as distinctnationalities (Tsung 1954:17; Wang 1955:5). Using language,religion, customs, dress style and other aestheticcriteria (De Vos 1982:15), researchers gradually identifieddiscrete groups for official recognition, arrivingat the present-day figure of fifty-five minority nationalities.Of these, the Tujia, initially thought to be abranch of the Miao group, received separate nationalitystatus in the 1950s, and the 3inuo, the most recentlyrecognized nationality, received status only in 1979.There are still more than sixteen groups under considerationfor nationality recognition (Fei 1980:102).THE SETTINGThe southwest provinces of China occupied by theMiao are varied in topography, with altitudes rangingfrom four hundred to twenty-five hundred meters (Nationalities1981). The Yunnan-Guizhou plateau, characterizedby precipitous mountain slopes and deep ravines, haspeaks of two thousand to twenty-five hundred meters inelevation. Where river valleys or plains exist, they arefertile agricultura/ areas due in part to the temperateclimate and high level of precipitation. A very largeproportion of the area is rugged and steeply sloped, withsignificant mineral deposits of coal, mercury and tin(Nationalities 1981:445). But this proportion is suitableonly for the cultivation of such grains as buckwheat,millet and maize (Nationalities 1981:446). Even thoughthe region has valuable and untapped natural resources,survival has been a constant struggle for its indigenousinhabitants. Early estimates (Cressey 1934:369) suggestthat as little as 5 percent of the total area of theYunnan-Guizhou plateau is arable, a figure that has likelybeen outdated by improvements in agricultural technology(Modern 1977:91; Peng 1983:57). Miao and other mountaindwellers have become known for their utilization of

76 THE HMONG IN TRANSITIONeven the apparently inaccessible mountain slopes by meansof skillful terracing and irrigation.In addition to the Han Chinese in these areas, Miaohave had neighbors from different ethnic groups dependingon their location. In Guizhou, initial surveys after 1949estimated that 30 to 40 percent of the total populationof the province comprised as many as thirty non-Han minoritygroups (Fei 1951:289). These groups have been reclassified,as described above, but the number illustratesthe context of ethnic diversity in which the varietyof Miao groups developed. Some of the most importantnationalities in the area, for whom a significant degreeof coresidence and contact with Miao can be assumed, includethe Tujia in Hunan; the Yi, Buyi and Dong inGuizhou; the Yao in Guangxi, and the Zhuang and Hani inYunnan. Division by province, however, is misleadingsince ethnic distribution in these areas is in the formof a complex mosaic which, until recently, had no specificcorrespondence to administrative districts nor anyparticular constancy over time With the establishmentof regions of autonomous government and the gradual sed-'ntation of formerly semi-nomadic peoples, increasingstabilization of interethnic relations will undoubtedlybe the trend.Since 1949, policy for regional autonomy has beenimplemented on the basis of minority concentration inwhat are called "compact communities." Where minoritypopulations are of certain density, regardless of theirpopulation in proportion to Han and other nationalities,they have a right to autonomous government (Liu 1954:11;Yin 1977:24-25). In some areas, where more than one minorityis represented in compact communities, the regionis jointly administered, as in the Xiang Xi Miaozu-Tujiazu Autonomous Prefecture in western Hunan. In allcases, the minorities involved have the right to determineth,. form of government, to have majority representationin government organs, and to run the affairs of theregion in their own language (Liu 1954:11-12).THE MIAO AS A NATIONALITYThe designation "Miao" has had .nany different connotationsthroughout the history of its usage (Ma 1876:86

THE MAO IN CONTEMPORARY CHINA 77102; Lin 1940:328-329), and thus the classification ofthe group to which it refers has been somewhat ambiguous.In the first half of this century, attempts at comprehensivecounts of Miao subgroups arrived at totals of seventyto eighty group names, figures which undoubtedly representedsome overlap or subcategorization. It is likelythat some of the groups then referred to as Miao are nowclassified as separatE nationalities.Based on primarily linguistic criteria, the Chinesesurvey teams of the 1950s determined that the huge varietyof self-identified Miao subgroups were appropriatelyclassed as one ethnic group, with one language. Althougha striking degree of cultural diversity obtains withinthis group, the subgroups are now said to recognize oneanother as coethnics, undoubtedly in part a , result of recentimprovements in communication, transportation andeducation.Although the name "Miao" had formerly been derogatory,used mostly by Han and other outsiders, the decisionwas made to adopt it without its original connotations.It was considered the only appropriate term toembrace the various subgroups that had been found to belinguistically similar enough to be considered coethnics.Thus, unlike any previous era in Chinese history, thename "Miao" is now widely used for self-identification bymembers of that nationality and there is significant evidencethat negative connotations have indeed been dispelled.When asked to identify their nationality in theirown language, several Miao, speaking various dialects,used the Chinese term "Miaozu" first, giving their subgroupname only when prompted. Given these indicationsit is prcbable that, increasingly, subgroup names will beused only for internal identification among Miao groups.This represents a remarkable shift, one that will undoubtedlyhave far-reaching consequences for the Miaosense of group affiliation, particulary if it takes rootin remote villages as well as among the highly educated.The Miao, numbering 5.03 million as of the 1982 census(Renmin Ribao), are the sixth largest nationality inChina, composing 7.5 percent of the total minority population.They are widely distributed over the south andsouthwest provinces but their largest numbers and greatestdensities are in Guizhou province where 54 percentare located. Hunan and Yunnan provinces also have rela-8 7

8878 THE HMONG IN TRANSITIONtively large populations, eachtotal. Smaller groupings areinces of Sichuan, Guangxi,Anhui and Jiangxi, and inShanghai.with 5 percent of thescattert.d over the prov-Hubei, Guangdong, Fujian,the cities of Beijing andThe degree to which geographic isolation has innuencedcultural diversification, sometimes for centuries,must be stressed at this point (Barth 1969:12-13). Miaoare commonly characterized as seminomadic hill peoples,swidden (slash-and-burn) cultivators of maize and uplandrice. However, even within what is referred to as southand southwest China, the groups called "Miao" have beenwidely scattered into varying ecological contexts withdifferent neighbors and particular local histories. Forinstance, some of the Miao of southeast Guizhou, formerlyof water and fertile valleys has allowed many to en-referred to as the "Hei" or "Black" Miao, live along theQingshui River and have developed an elaborate river culturewhich informs not only their folklore and their customs,but also their economic activities. The availabilitygage in lowland cultivation. Paddy rice is their primarycrop and staple food. This group, accounting for a largeproportion of the total number of Miao in Guizhou, cannotnecessarily be considered any less "Miao" for its divergencefrom common stereotypes of Miao ethnic identity(Leach 1954:29 -41).LANGUAGE AND EDUCATIONAccording to contemporary classification, the Miaolanguage can be divided into three major dialects and onemajor subdialect. There are also several regional subdialectsand vernaculars (Miao 1962:2-3; Lemoine 1972:17). To the untrained ear the major dialects are mutuallyunintelligible. This was demonstrated by the observeduse of Mandarin Chinese between Nationalities Institutestudents of different Miao dialect groups. Linguistsspecializing in Miao language, however, mair min that,although the dialects differ in vocabulary as well aspronunciation, 40 percent of the vocabulary is sharedamong the three and that the names used by the Miao forthemselves are cognates, or variations of a single word.The dialects, named for the regions in which they arespoken, break down as follows:

Chuan Qian DianTHE MIAO IN CONTEMPORARY CHINA 79Spoken primarily in Sichuan, Guizhou and Yunnan bynearly half the total Miao speakers, the "Western" dialectis thought to have retained the most elements ofwhat is considered to be the "ancient" Miao language.Chuan Qian Dian speakers call themselves "Hmoob" (Barney-Smalley romanization), and their speech is clearly intelligibleto a speaker of the Hmoob Dawb (White Hmong) orHmoob Ntsuab (Blue Hmong) dialects of Southeast Asia.The Man Dong Bei subdialect of Chuan Qian Dian, spokenin northeast Yunnan, bears less resemblance to SoutheastAsian Hmong dialect.Qian DongMost speakers of the "Central" dialect are inGuizhou Province, although they are also found in theneighboring areas of Hunan and Guangxi. Qian Dong speakersmake up about one third of the total Miao speakersand pronounce their names as variations of "Hmu" or"Hmo."Xiang Xi,Less than a quarter of the total Miao speakers usethe "Eastern" dialect of western Hunan and easternGuizhou. This dialect differs from the previous two inthat it hits six tones instead of eight. Xiang Xi speakerscall themselves variations of "Qhov Xyooj" (Barney-Smalley romanization).In addition to speakers of the three major dialects,there are at present more than two hundred thousand Miaowho don't speak their own language. The majority of theseare Han speakers, but some are also speakers of the Yaoand Dong languages. This point raises some questionsabout ethnic classification.No evidence has been found of ancient scripts forthe Miao language. However, the legend that there wasonce such a script is widespread among all groups ofMiao, as among many other ethnic groups of this region.Before 1949, some writing systems had been created bymissionaries working in Miao areas. In Panghai, East89

80 THE HMONG IN TRANSITIONGuizhou, a romanized script was devised for Qian Dongspeakers, and in the Zhaotong-Weining district on theYunnan-Guizhou boraer, Samuel Pollard developed a scriptwhich was learned by an estimated twenty thousand tothirty thousand Dian Dong Bei speakers.During the 1950s, Chinese policy for developmentwork in minority areas placed a high premium on the creationof written languages for all minorities that had noscript of their own. The Institute of Linguistics andPhilology of the Chinese Academy of Sciences sent fieldteams to the southwest provinces to investigate minoritylanguages (Fu 1957:28). Scripts were developed for eachof the three major dialects and for the Dian Dong Beisubdialect of Miao. All are romanizations. While thescripts were intended for use in schools and in the affairsof autonomous regions, they have not been as widelyemployed as was first expected. This is largely attributedto the desire on the part of many nationalities tostudy and work in Chinese so as to be able to communicatewith members of other nationalities.In addition to development of minority written languages,education for all nationalities has been extensivelysupported. Primary and secondary schools in minorityareas were established at a remarkable rate after1949 (Yin 1977:74-5). In mountainous regions or whereinhabitants were nomadic, boarding schools were set upwith students' living expenses absorbed by the State(Huang 1982:7). Ten Institutes for Nationalities, includingthe Central Institute. in Beijing, now operate in additionto local higher education institutions in minorityareas. Students receive an education closely comparableto the standard college curriculum in Chinese universities.One can major in such fields as math and srience,history and politics, language or arts, with remedial educationavailable where necessary, especially in the caseof students for whom Chinese was not a first language.Preferential admissions policies exist in the form oflower qualifying scores required on entrance examinationsand priority over Han Chinese with comparable qualifications(Huang 1982:5). After completing their collegeeducation, most minority students return to their homeprovinces to occupy teaching or government positions.90

THE MIAO IN CONTEMPORARY CHINA 81SOME ASPECTS OF TRADITION AND CHANGECertain traits and orientations were regularly identifiedby Miao and Han Chinese in such a way as to beconsidered "symbols" in the identity system of the Miaoin China (Spicer 1971). For instance, members of bothgroups noted the historic Miao reputation for opposingexternal domination and fighting fiercely for autonomy(Castile 1981:xix). Miao are said to differ from Han inthe extent of "respect" accorded to women in their society.As with most of the minority nationalities, womennever had bound feet and customarily worked alongside menin the fields. With regard to dress, there are as manyvariations in style as there are Miao localities, but asa whole the Miao are known for their intricate needlework(Tsung 1954:18), their wax-resistant dyeing techniquesand the characteristic pleated skirts worn by the women.Men are renowned for their skill in crafting elaboratesilver ornaments (Peng 1983:60).Certain forms of music and dance are considered distinctiveelements of Miao culture. A great number of annualfestivals and ceremonial occasions require specificperformances which differ by subgroup. Important instrumentsinclude gongs, drums and leaf-blowing, depending onregion. All Miao groups are renowned for their skill andversatility in singing. The improvised dialogue style ofsong is quite widespread. Dance styles are also characteristicespeciallycourtship dances performed by unmarriedyouth at festival times. One dance common among Miaoof southeast Guizhou involves a delicately embroideredribbon of cloth which the unmarried girl uses as a symbolof her affection, and the lusheng or qeej (kheng), amouth organ of bamboo pipes. A suitor, playing thelusheng, asks the girl for her love by asking for theembroidered ribbon. She dances, coyly withholding ituntil he has pursued her enough to persuade her of hissincerity. Then, if she accepts his offer, she ties theribbon on his lushlen and they dance together in celebration.A carefully choreographed performance of this dancewas observed at the annual pan-Miao festival held on thecampus of the Nationalities Institute. Every spring, onor about April 8 of the lunar calendar, Miao in theBeijing area gather for a festival in honor of a Miao91

82 THE HMONG IN TRANSITIONhero from central Guizhou who died defending his peoplefrom conquest by another tribe. Among the songs and dancesperformed in 1982 was the one described above.The festival gives some insight into the types ofnontraditional vocations in which Miao are engaged andinto the kind of attention paid to minority customs inthe People's Republic of China. The dancers, both of theMiao nationality, were professionals who had been trainedin several genres of dance. One is a member of a Beijingbasednationalities song and dance troupe; the other wasin Beijing for a year of additional training before returningto the home province to work either as a teacheror a r !dormer. Choreography for this type of performanceis a result of research in minority provinces. Instructorsat the Nationalities Institute spend some months outof each year traveling to minority areas both to recruitstudents and to study indigenous music and dance. Theseinstructors, often minorities themselves, use the resultsof their research to design dances which students of allnationalities will then study and perform.The bamboo pipe organ, called lusheng (the Chineseterm meaning reed pipe) by many of the Miao in China, isalso considered a symbol of the Miao nationality. Thereis an extraordinary range of variation both in the styleof the instrument itself and in the traditions for itsuse. In the more easterly locations in Guizhou and northernGuangxi, the bamboo tubes are arranged vertically andare sometimes as tall as twelve to fourteen feet. Differentvarieties of the instrument are used for differentfestival and ceremonial occasions, sometimes accompaniedby elaborate, almost acrobatic rancirg, In the past thelusheng was also said to have a martial function, used tocall members of neighboring villages to battle, to boostthe morale of the soldiers and to frighten the enemy.In western Guizhou, Sichuan and Yunnan provinces,lusheng tubes are horizontally arranged. They are shorterand are used for fewer purposes than is standard farthereast. In some of the western areas, the use of the instrumentis restricted for the most part to funerals.Again, great differences in style and usage have notchanged the importance of the lusheng as a symbolic instument of all Miao.92

THE MAO IN CONTEMPORARY CHINA 83CONCLUSIONI have attempted here to give a brief introductionto the Miao of contemporary China. Diversity is one ofthe most salient characteristics of the group that ispresently classified as Miao in the People's Republic.in language and agriculture, as well as in dress and festivalstyles, a great variety of cultural traits amongMiao subgroups appears to have resulted from geographicseparation and from differences in context. Further, directedchange, in such efforts as development and education,has had a far-reaching impact on the lives of allChina's nationalities, especially since 1949. Thesepoints are intended to indicate the importance of investigatingthe group studied in the context of the timesand milieu which inform their daily lives. Toward thisend a great deal of work remains to be done.What I have presented here is some basic data andpreliminary conclusions to suggest possibilities for furtherresearch. There is a significant and growing literaturein both Chinese and Western languages which canprovide insight into the contemporary situations of Hmongand Miao in various parts of the world, as well as intotheir role in the history of China and Southeast Asia.In addition, primary ethnographic research can continueto be a rich source of new information. It is my hopethat ongoing work will give us more perspective on theHmong and Miao, not in terms of static cultural forms,but .-Lther, according to varying contexts, in terms of adynamically emerging identity.NOTESI"Tradition" is used here to refer to a concept the Hmongthemselves have of their cultural origins. Although itis a problematic construct in the con text of ongoingchange, it is a powerful image for Hmong who have migratedfar from their Chinese homeland.2This latter study made possible by a Samuel T.Arnold Fellowship from Brown University.3The "Hanyu Pinyin" system of romanization is used herefor Chinese terms and the Barney-Smalley orthography for93

84 THE HMONG IN TRANSITIONHmong terms (cf. Ernest Heimbach, White Hmong-EnglishDictionary, Ithaca, New York: Cornell University, 1969).REFERENCESBarth, Fredrik. 1969. Ethnic Groups and Boundaries.Boston: Little Brown and Company.Castile, George Pierre, and Gilbert Kushner, eds. 1981.Persistent Peoples: Cultural Enclaves in Perspective.Tucson: University of Arizona Press.Cressey, George Babcock. 1934. China's Geographic Foundations.New York: McGraw-Hill Book Co.De Vos, George, and Lola Romanucci-Ross. 1982. EthnicIdentity: Cultural Continuities and Change. Chicago:University of Chicago Press.Fei Hsiao-tung. 1980. "Ethnic Identification in China."Social Sciences in China 1:94-107.. 1951. "The Minority People in Kweichow." ChinaMonthly Review 121, 12:289-94.Fu Mao-chi. 1957. "Written Languages for China's Minorities."People's China 3:25-31.Huang-ying. 1982. "A Brief Introduction to China'sMinority Education." Beijing: Cultural Palace ofNationalities (unpublished).Leach, Edmund R. 1964. Political Systems of HighlandBurma. Boston: Beacon Press (first published,T755737Lemoine, Jacques. 1972. Un Village Hmong Vert du HautLaos. Paris: Centre National de la RechercheScientifique.Lin Yueh-Hwa. 1940. "The Miao-Man Peoples of Kweichow."Harvard Journal of Asiatic Studies 5, 3-4:261-345.94

THE MIAO IN CONTEMPORARY CHINA 85Liu Chun. 1954. "National Minorities Enjoy RegionalAutonomy." People's China 1, 1:9-14.Ma Touan-Lin d'Herve' de Saint-Denys, trans. 1876. PeuplesEtrangeres a la Chine. Geneva: H. Georg. (1972by Gregg International Publishers, England).Miao Language Team, Chinese Academy of Social Sciences.1962. "A Brief Description of the Miao Language."In Miao and Yao Linguistic Studies, edited byHerbert C. Purnell. Ithaca, New York: CornellUniversity, Southeast Asia Program, Data PaperNumber 88, 1972, p. 1026.Modern China Series. 1977. China's Minority Nationalities:Selected Articles From Chinese Sources. SanFrancisco: Red Sun Publishers.Nationalities Affairs Commission, editing Committee ofthe Publishing Series for Five Kinds of NationalitiesProblems, "China's Minority Nationalities"Editing Group. 1981. China's Minority Nationalities(in Chinese). Beijing: People's Press.Peng Jianqun. 1983. "A Miao Nationality Village." ChinaReconstructs 32 (3):57-63.Renmin Ribao. 1982. People's Daily. (in Chinese). 29October 1982.Spicer, Edward H. 1971. "Persistent Cultural Systems."Science 74, 40! l:795-800.Tsung Yun. 1954. "China's National Minorities." People'sChina 11: 17-23.Wang Shu-tang. 1955. China: Land of Many Nationalities.Peking: Foreign Languages Press.Weng Tu-chien. 1950. "China's Policy on National Minorities."People's China 1, 7:6-7.Yin Ming. 1977. United and Equal: The Progress of China'sMinority Nationalities. Peking: Foreign LanguagesPress.95

GEOMANCY AS AN ASPECTOF UPI. AND-LOWLAND RELATIONSHIPSNicholas TappThe remarkable practice of geomancy by the Hmong isthe focus of this paper. Geomancy literally means asystem of divination by reference to the forms of theearth. This is a system for the siting of villages andthe graves of ancestors and relatives according to thecontours of the mountains and watercourses formed intheir valleys. Feuchtwang (1974) has provided the mostcomprehensive account of this system as it is practicedby the Chinese, and Lemoine (1972) was the first to noteits practice among the Hmong. Geomancy, or loojmem, asthe Hmong call it, provides a very complete and articulateidiom for the metaphorical expression of socialchange and conflict. It has much in common with landscapepainting and the arts of gardening, which alsoorginated in China. Unfortunately, I do not have spacein this paper to go into the details of the system aspracticed by the Hmoob Dawb with whom I worked in thenorth of Thailand. Instead I have concentrated on thestrange anomaly of the practice of this system, which hasalways been seen as a peculiarly Chinese one, by theHmong who constitute an ethnic minority within China.Feng-shui, as ti.. Chinese call loojmem, is mostwidely practiced in the hilly, mountainous regions ofsouth and southwest China, where the Hmong and certainother minority nationalities have settled. While most ofthe imagery and directions for the selection of a siteassume the presence of a mountainous landscape, feng -shuiis practiced in many non mountainous areas, where trees,rocks and other natural phenomena must suffice to symbolisethe great mountains and watercourses which thesystem properly demands.According to Feuchtwang,shui's concentration on mountains has never been adequatelyexplained. Since expressions of mountainousimagery are far more common in the classification systemsof upland-dwelling minorities such as the Hmongl thanthey are in the classification systems of the great wet-879

88 THE HMONG IN TRANSITIONrice cultivating mass of the Chinese population, andsince the Hmong practice feng-shui with no apparentawareness that it may be of non-Hmong origins, the systemcan be interpreted as being at least as much a Hmong oneas it is a Chinese one.Some historical confirmation of my point comes fromtheories which hold that feng -shui itself did not originatein China, but rather came up from the south, togetherwith iron-working and other practices which resultedin the "distinctly Chinese" custom of double burial.But this sort of evidence is not what I wish to basemy argument upon. I base it rather upon the very extensiveanthropological work on the theory of ethnic boundariesthat has been conducted in the region (by which Iinclude the mountainots area of Burma, Laos, Vietnam andThailand as well as southwestern China).Very often it has been found that whole communitiesmay be ambiguous about their ethnic status, and considerthemselves, for example, as both Shan and Kachin (Leach1954), or both Lahu and Lisu. (Durrenberger 1970). Verymany cases of individuals changing ethnic status to be inaccord with changing economic, marital or other situations,have been noted over the years. Lehman (1979) hasremarked that an ethnic category must never be confusedwith the various genetic-linguistic groups with which itmay be identified at different times. The ethnic term"Miao" is a prime example of this, since at differenttimes (and particularly before the t:-!rteenth century[Ruey-Yih-Fu 1962]) it has been applied to widely differinggroups. (And today again in China since the Revolutionit has taken on a different context.) The real issue,as Lehman emphasises, is whether an ethnic labelsuch as "Miao" or "Karen" has remained constant overtime, since it is likely to have retained a constantmeaning (as opposed to usage). This is the case with theChin people, where "Chin" was originally a Burmese termfor "allies." So it becomes irrelevant, from a historicalperspective, whether the term "Miao" in the past includedthe actual ancestors of the present-day Hmong.What is important is that it conveyed certain connotations(of rebelliousness, for example) and that it isstill widely used in this derogatory sense, at least inThailand, to refer to the Hmong today.9 73

GEOMANCY AS AN ASPECT OF RELATIONSHIPS 89Even the ethnic label "Han Chinese" is notoriouslyvague in the genetic-linguistic sense, particularly inthe southern parts of China where the history of theHmong extends. Many of the present-day population ofsouthern China who call themselves "Chinese" are the descendantsof people who did not call themselves "Chinese"at all, but were the members of quite different groupssuch as the Norsu.2 Francis L.K. Hsu studied a wholevillage of this type (1949). They have "become" Chinese,anddating back to Leach's work on the Kachinthere aremany examples of people "becoming" members of other ethniccategories. So I am arguing that the real issue isnot whether feng-shui, or loojmem, is Hmong or Chinese.Rather, I wish to demonstrate that the system has evolvedout of the context of interethnic relations in south China.Members of the expanding Chinese state came intocontact with the members of other genetic-linguisticgroups only gradually, over a period of some two thousandyears, often becoming members of those groups themselves,or adopting members of those groups as "Chinese." Thus,the practice of feng-shui as a distinctly Chinese systemarose out of an accommodation with the members of manydifferent groups.From this point of view I would also argue againstthe common tendency to view the Hmong as an isolated,autonomous group which traditionally lacked relationswith other groups (Hinton 1969). On the contrary, it isclear from all their stories and many of their ritualsand customs, that they have had long and sustained contactswith the members of other cultures throughout theirhistory. The myth of the self- sufficient tribe has beenfostered by colonialists and early anthropologists asuseful administrative tool and a pretext for exploitingthe people so categorised (Helm 1968). But such a viewdoes not do justice to the extensive interdependencywhich in the past, before the nineteenth-century adventof European colonialism, characterised the relations betweenthe uplands and lowlands. In many cases the peopleof the mountains and the remote regions were dependent onthe people of the lowlands for rice, which grows best inthe lowlands. The mountains and their forests, in turn,were full of valuable products which could be exchangedfor the rice of the lowlands. From this point of viewanother controversyover whether the Hmong were origi-93'

90 THE HMONG IN TRANSITIONnally wet-rice cultivators settled in the lowlands, orhave always been shifting cultivators based in the mountainsalsobecomes irrelevant. Very probably differentgroups, communities, even families, of the ancestors ofthe Hmong did both at different times and in differentregions. The question of the "original ecology" of theHmong becomes senseless once one realises the danger ofover-reifying ethnic categories, as well as the risks ofassuming that the ethnic labels "Miao" or "Chinese" alwaysreferred to culturally distinct groups over the past(or that ecological distinctions such as the one between"shifting" and "permanent" agriculturalists were alwaysisomorphic with ethnic distinctions). Also senseless isthe related question of who preceded whom in the actuallands of China. What one wants to emphasize, therefore,is the adaptability of the Hmong and groups like them,their capacity to adjust to different ecological situationsand the relations they have always held with themembers of many other different ethnic groups. Thesefactors are, of course, of some relevance to their currentsituation as refugees.The geomantic system which the Hmong practice istruly a remarkable one. On the welfare of the ancestorsdepends the fortunes of their descendants. There aremany accounts of the most ideal burial site, specifyingthat mountains must be arranged in a particular wayaround the site, with the rivers flowing in a certaindirection down to a dragon's pool where rebirth begins.The deceased must be buried at a certain time, along theloojmem or veins of the dragon, which run through themountains. (The recurrent image of the mountains is thatof the dragon, which has connotations of natural strengthand energy and is associated with underwater creatures,royalty and the ancestors.) If a person were buried insuch a place, it is said, he would "catch" the loo'memrather as one catches a train, and ride it as one ridesan ox, to the place where he enters the otherworld, wherehe would become a king. And when the father is king inheaven, as was once graphically told me, the son will becomeking on earth. This is because the ancestor watchesover the welfare of his or her descendants in this worldof yang, and thus it is in the interests of descendantsto assure their parents and other ancestors of the bestpossible burial places. The Hmong in Thailand assume99

GEOMANCY AS AN ASPECT OF RELATIONSHIPS 91that one can never find the absolutely perfect burialplace, since if someone had, he would be a king and rulethe land, and it is not the Hmong who rule the land ofThailand. Such places might exist in China, they said,but usually one does the best ore can, and tries to avoidthe more inauspicious conjunctions of natural features.While I was seeking the origins of loojmem I collecteda most remarkable story that illustrates the precisecontours required for the best, the absolutely idealburial site. The story is about two brothers who separatelyrequested their respective descendants to bury themvery close together, in the same, ideal spot. The elderbrother was buried with gold beneath his head for a pillow,the younger had only a stone pillow. But when thesons of the elder brother came to pay their respects atthe grave three years later, they found that flowersasign of the decay necessary for reincarnationhadsprouted first on the grave of their uncle, the youngerbrother. So they dug up the stone and gold pillows andexchanged them. The elder brother was thus enabled to bereincarnated first, traveling along the lines of thedragon's veins to the pool which is the entrance to theotherworld, before his younger brother. When the sons ofthe younger brother turned up and discovered what hadhappened, they realised how wicked their cousins were,and the story concludes that "since that time we Hmonghave never got on well together with the Chinese. WeHmong moved away and refused to speak the same language."Now clearly this refers to a time in the Hmong imagination,when the ethnic categories of Hmong and Chinesewere not clearly demarcated, and this alone is some justificationof the historical viewpoint I have adopted.But more important is that here the Hmong are using thegeomantic idiom itself, the idiom of fenq -shui, which hasalways been seen as a peculiarly Chinese one, to explicatetheir own differences from the Chinese. And this isnot the only case where this is done. A Lolo story employsa similar idiom in a similar context.3 Similarthemes occur often in the Hmong stories of the variousincarnations of Tswb Tchoj, the culture hero who periodicallyarises to unify the clans and establish hegemonyover the land. In these stories, contests take place betweenthe Hmong and the Chinese for the ideal site inwhich to bury their respective ancestors. The usual rei.10 0

92 THE HMONG IN TRANSITIONsuit is that the Hmong lose out, owing to the trickeryand deception of the Chinese. Very often the Chinesebury some form of metal in the grave of the Hmong. Becauseit does not putrefy, the metal prevents the decaywhich must take place if effective reincarnation is tooccur. Often, too, in the stories, the Chinese desecrateHmong graves in other ways, just as the Imperial dynastiesdesecrated the graves of ancestors of pretenders tothe throne. The imagery is clearly one of sovereigntyand rebellion (Feuchtwang, personal communication). Itis highly improbable that the Hmong merely adopted theideology of a conquering people and used it to explainhow they lost control of the land to those same conqueringpeople, the Chinese. As a result of these stories, Imuch prefer to see geomancy as being as distinctivelyHmong as it is Chinese. I would also suggest that geomancyevolved (as did the Chinese state itself in thesouth) out of southern China's complex ever-changingpolitical situation in which interethnic relations wereforged around such ideal terms as "Han," "Miao" and"Lolo." This seems truer to both the historical and theethnographical evidence.A similiar case can be made for the patrilinealclan-based kinship system of the Hmong, which some authoritiesbelieve the Hmong borrowed from the invadingChinese. It is far more likely that what seem to be"Chinese" surnames, but which in fact are shared by manyother peoples of the region (such as the Lisu, Yao,Kachin and Shan), actually form part of a common systemwhich such people share with the "Chinese." We are rightto consider geomancy, the patrilineal kinship system andthe burial system together, since geomancy is above all asystem for the siting of ancestral graves, while Hmongsurname groups may be differentiated internally as"Hmong" or "Chinese" according to the way each groupaligns graves to the slant of the mountain, and whetherstones or biodegradable materials such as branches areused to mark the site. Indeed, other stories account forChinese mastery over the land by their demarcation oftheir territory by stones, in contr. s' L3 the grasses andtwigs used by the Hmong, which were blown away by thewind. It is particularly interesting that geomancy itselfshould be above all a system concerned with acquisitionof certain types of territory which assure the pros-101

GEOMANCY AS AN ASPECT OF RELATIONSHIPS 93perity and welfare of one's descendants, and that itshould be this very system which has been chosen by theHmong to express their ethnic rivalries with the Chineseand other peoples. We know that throughout the recordedhistory of southwest China fierce conflicts took placebetween the expanding Chinese statewhich was attractedby the mineral resources, salt wells and forest productsof the mountainous regionsand the people who livedthere. The written accounts provide a very clear recordof the long historical struggle for control of materialresources. It may even be that metal is felt to be particularlyunlucky because it was through knowledge of theprinciples of ironworking, at about 600 B.C., during themid-Chou period, that the early Chinese states began toestablish their hegemony. This point is speculative, butleads me to my conclusion.Apart from contributing my findings to some of thecontroversies about the origins of the Hmong, their kinshipsystem and technology, this paper makes a more importantpoint. This is that history is not always to beread in books or written documents, or even archaeologicalremains, and that the oral legends of the Hmong abouttheir past have much to teach us about real history; thatis, a history which is being lived and felt now. What Ihope to have emphasised is the continuing power of theHmong oral tradition to encapsulate and deal with changingconditions through the use of traditional symbols,even in the context of what McLuhan called a post-literatesociety. Memories of the past are always selective,and can never be accurate or comprehensive in every detail.Thus one is looking for a new kind of history, notone divided into "true" or "false," but one arising froma more phenomenological concern with historical consciousnessas it affects current behaviour. Real historyis what is remembered.102

94 THE HMONG IN TRANSITIONNOTESthe Hmong language, this mountainous imagery is found1Inin such upslope/downslope dichotomies as pem (uphilleast,from) and nram (downhill from); sab hnub tawm (t-.e.the sidt_ the sun leaves) and sab hnub poob (the west,the side the sun falls), and ib nrab (one half of ahorizontal measurement) and ib ntav7c7ie half of a verticalmeasurement).2There is even some evidence to suggest that the Hmongterm Suav, used to distinguish the Chinese, referredmore to social than ethnic differences, being alsoapplied for example to their "Lolo" (Norsu) landlords.3See Graham (1955).REFERENCESDurrenberger, E.P. 1970. Lisu Project: A Socio-MedicalStudy of the Lisu of North Thailand. Tnbal ResearchCentre, Chiangmai.Feuchtwang, S. 1974. Chinese Geomancy, An AnthropologicalAnalysis of. Vientiane, Laos: Vithagna.Graham, D.C. 1955. "A Lolo Story: 'The Great God of0-1-1-Bi-Zih' by Lin Kuang-Tien." Joumal of AmericanFolklore 68.Helm. 1, ed. 1968. Essays on the Problem of Tribe.Proceedings of the 1967 Annual Spring Meeting of theAmerican Ethnological Society, Seattle, University ofWashington.Hinton, P., ed. 1969. Tribesmen and Peasants in NorthThailand: Proceedings of the First Symposium of theTnbal Research Certre, Chiangmai 1967. Tribal ResearchCentre, Chiangmai.Hsu, Francis L.K. 1949. Under the Ancestors' Shadow:Chinese Culture and Personality. London: Rout ledge &Kegan Paul.103

GEOMANCY AS AN ASPECT OF RELATIONSHIPS 95Leach, E.R. 1954. Political Systems of Highland Burma.Lehman, F.K. 1979. "Who Are The Karen, and If So, Why?Karen Ethnohistory and a Formal Theory of Ethnicity."In Ethnic Adaptation and Identity: The Karen on theThai Frontier with Burma, edited by C. Keyes. Institutefor the Study of Human Issues, Philadephia.Lemoine, Jacques. 1972. Un Village Hmong Vert du HautLaos. Paris: Centre National de la Recherche Scientifique.Ruey, Yih-Fu. 1962. "The Miao: Their Origins and SouthwardsMigrations." Proceedings of the InternationalAssociation of Historians of Asia. Taipei, Taiwan.101

PART TWOAdapting to a New Society

INTRODUCTIONGlenn L. HendricksIn the previous section the papers focused on someglobal issues of Hmong history, culture and the processof change and adaptation over time. In Part Two the papersprovide concrete cases of these processes at work.Hmong women are faced with possibilities for establishingnew roles in the United States. But their lackof education has made the search for employment opportunitiesquite difficult. Sarah Mason examines training programsthat have been designed for Hmong women and findsmany of them lacking in their ability to provide adequateeconomic mobility for the future. Rather she sees themas slotting the women participants into marginal low-wagebackwaters of the :mpioyment market.William Meredith and George Rowe report on a studyof shifts in Hmong attitudes towards aspects of marriageas a consequence of their contact with western ideas andexperiences. Echoing some of the authors of Part One theysuggest caution in generalizing about cultural commonalitiesin such practices as polygyny, bride price and clanendogamy of the Hmong in their Lao homeland. They suggestthat in some cases changes were already underway long beforetheir arrival in the United States. However, theirdata indicates that in certain areas, particularly forthe educated and female, under the influences of the newenvironment, changes are reported in attitudes towardgender equality in family decision-making and the acceptabilityof polyandrous relationships.Beth Goldstein's study of a case of sexual assaultdetails the social implications when American values andtheir legal expression conflict with Hmong values andtraditional forms of dispute adjudication. She points outthat in the American view such cases are seen as conflictbetween individuals and consequently resolution pits oneagainst the other. From the Hmong standpoint the emphasisis upon the effect the assault has upon the family andclan system and therefore settlements represent negotiations that heal the breach in group relations.Catherine Stompus Gross provides a case study of theadaptation of Hmong to a specific community, Isla Vista,99106

100 THE HMONG IN TRANSITIONCalifornia. She outlines what she finds to be both inhibitingand facilitating factors in the adjustment of thisgroup of Hmong to the community.Nancy Donnelly's description of a Hmong needlecraftsales cooperative provides a sociological analysis of thetenuous relationships that exist in a joint American-Hmeng economic venture. More importantly she providesthose individuals and groups who have an interest in similarendeavors with a framework by which they might examineaspects of their own activities. As she points out,the very nature cif the refugee situation places the Hmongin the subordinate role of client to others, typicallyAmericans, who assume the super-ordinate roles of brokeror patron.The final item in this section is a report introducedby Stephen Reder of a symposium organized to discussmajor issues that emerged in a government fundedstudy of Hmong resettlement in the United States. DougOlney reviews available demographic information. JohnFinck discusses the significance of the large migrationof Hmong to the Central Valley of California. BruceDowning outlines some issues relating to the acquisitionof English. Shur Vang Vangyi reviews the employment situationand the related problem of welfare dependt.tricy,while Mary Cohen offers a hopeful view for the future asHmong have come to view education as a key to their eventualccommodation to life in the United States. FinallySimc . Fass takes an economist's viewpoint in a report onthe various efforts that have been made to assist theHrnong in gaining economic self-sufficiency.10 'I

TRAINING HMONG WOMEN: FOR MARGINAL WORKOR ENTRY INTO THE MAINSTREAMSarah R. MasonINTRODUCTIONSince the American withdrawal from Southeast Asia in1975, approximately 650,000 refugees from Vietnam, Cambodiaand Laos have reottled in the United States. Whilethis refugee population as a whole has adjusted well toAmerican life, serious problems remain unresolved forsubgroups such as the elderly, older youth, preliteratepeople and women. Recently the situation of refugeewomen has become a focus of serious concern among resettlementofficials, service providers, volunteers andwomen's advocacy groups. The problems of Indochinese andother refugee women have also been addressed by severalinternational conferences, including the United NationsDecade for Women (Copenhagen, July 1980), the IntergovernmentalCommittee for Migration Seminar on Adaptationand Integration of Permanent Immigrants (Geneva, April1981), and the United Nations High Commissioner for RefugeesWorkshop on Integration of Refugees from Indochinain Countries of Resettlement (Geneva, October 1980).A major issue at these international meetings hasbeen the low economic status of refugee women, a disproportionatenumber of whom are household heads with largefamily responsibilities. In the United States, a studyconducted in 1980 (Walter 1981:17) of 347 Indochineserefugee women sponsored by Lutheran Immigration and RefugeeService, showed that one-third of the 23R householdswere headed by women. Despite the stipulation of the1980 Refugee Act that women must have the same employmenttraining opportunities as men, training programs accessibleto women have been inadequate and inequitable.Generally job training programs targeted for male headsof households are designed to prepare them for employmentin the economic mainstream, while women are channeledinto programs leading to marginal work such as housecleaningand cottage industry sewing. Income-generatingprojects for refugee women have also focused on pieceworksewing, craft production, crop picking and other low-101108

102 THE HMONG IN TRANSITIONpaid, exploitative labor. Although this type of employmentprovides refugee women with much-needed income, ittends to stabilize their economic status at a low levelwithout providing them with the skills necessary for economicmobility. Recent research (Pass 1983: Table 1)indicates that sewing projects organized in the UnitedStates as employment training and income-generating programsfor refugee women, for example, have provided anaverage annual income per participant of only $117 inpro' ,,cts with minimal external funding and $303 in thosewith substantial external funding,The purpose of the present study is to survey theemployment training programs available to Hmong womensince 1975 in Minneapolis-St. Paul, Minnesotawhich hasone of the largest urban concentrations of Hmong in theUnited Statesand to evaluate the effectiveness of theseprograms in resolving the problem of low economic statusamong refugee women. The focus of the study is on thetypes of training provided, accessibility of the programsand economic outcome for the women trained.The method of the study is, first, the collection ofstatistical data, annual reports and other printed materialsavailable from program administrators. The secondstep is the interviewing of administrators, instructorsand Southeast Asian counselors and interpreters involvedin refugee employment training to gain additional informationnot included in printed materials. This includesdata on support services provided (including child careand transportation), location of training sites, tuitioncosts, priorities in enrollment and tuition assistance,educational prerequisites and potential for economic advancementin various areas of employment. The third stepis the evaluation of the training programs available toHmong women within the larger framework of training programsprovided for Southeast Asian refugee:.The author views this as an exploratory study thatwill provide preliminary data for a more comprehensivestudy of employment training programs for Southeast Asianwomen throughout the United States. The larger study willinclude interviews with Southeast Asian women who haveparticipated in these programs, thus providing individualizeddata on refugee women's work experience. This pipeof data has not been available for the present study.109

TRAINING HMONG WOMEN FOR MARGINAL WORK 103The conceptual framework of the study is based onrccent research in the growing field of Third World women'seducation. New studies in this field have developedfresh perspectives on the effect of educational programson Third World women's lives as workers, citizens,mothers and wives, whereas earlier research focused onwomen's education as a means of advancing the economicdevelopment of Third World countries. Basic issues includeaccessibility of educational programs, type of educationprovided, content, quality of instruction and thesocial, political or economic outcome of the training.The author views these issues as basic to the study ofeducational programs for Third World women moving intonew socioeconomic orders, whether in developing countriesor in immigrant/refugee communities in the United States.FEDERAL ASSISTANCE TO REFUGEE JOB TRAININGFederal educational assistance for refugees in theUnited States has been available since the influx ofCuban refugees following Fidel Castro's rise to power in1959, but until 1980 these benefits have varied widelyamong nationality groups. The Cuban Refugee Program,authorized under the Migration and Refugee Assistance Actof 1962, provided for adult education and employmenttraining, as well as refresher courses for medical professionalsto assist them in meeting certification requirements.The 1962 act remained in effect for overtwenty years, with no time limit on eligibility for theapproximately 800,000 Cuban refugees who arrived in theUnited States during that period.The Indochina Migration and Refugee Assistance Actof 1975 extended these benefits to Southeast Asian refugeesbut imposed a limit of three years on eligibility(reduced to eighteen months in 1982). The 1975 act expiredon 30 September 1977, and nearly a month elapsedbefore new legislation authorized the continuation ofeducational programs and other forms of assistance. FromOctober 1977 to March 1978 an interruption of five monthsoccurred, causing confusion in the refugee community andfrustration among administrators of refugee training programs.It was not until the enactment of the Refugee Actof 1980 that permanent, comprehensive refugee legislation1 i 0

111104 THE HMONG IN TRANSITIONbecame effective. By early 1981, retrenchment in federalspending on social programs caused drastic cuts in refugeeeducational programs, despite the provisions of the1980 act. These discontinuities have had a devastatingeffect on refugee women's programs, looked upon by manyadministrators as secondary in importance to other refugeeprograms.DEVELOPMENT OF INDOCHINESE WOMF.N'S PROGRAMSIndochinese refugee women's programs have evolved asa complex network of English and cross-cultural orientationclasses, support groups and employment trainingprograms. Funded by various combinations of public andprivate funds, they have been sponsored by voluntaryagencies, adult education centers, technical and vocationalschools and Indochinese mutual assistance associations(MAAs). While many publicly funded women's programshave ceased to exist since 1981, employment trainingprograms have in some cases continued to receive federaland state funding and others have continued with thehelp of private resources. Some income-generating projectsfor refugee women have also developed training components.EMPLOYMENT TRAINING FOR SOUTHEAST ASIAN REFUGEEIN THE ST. PAUL-MINNEAPOLIS AREAMinnesota has been a major area 3f Southeast Asianrefugee resettlement since the late 1970s. In the fallof 1981 the refugee population react- NJ a peak of twentysix thousand, 85 percent of whom lived in the St. Paul-Minneapolis metropolitan area. The largest group werethe Hmong, who numbered nearly twelve thousand andwiththe exception of a small settlement in Duluthvirtuallyall of this group resided in the metropolitan area. Duringthe winter of 1982, which was a period of unusuallyharsh weather and increasing unemployment in Minnesota,several thousand Hmong left the state in hopes of findingjobs and a warmer climate elsewhere. With the return ofwarmer weather and an upturn in the state's economy, theHmong population stabilized in mid-1983 at approximately

TRAINING HMONG WOMEN FOR MARGINAL WORK 105eighty-five hundred out of a total of twenty-three thousandSoutheast Asian refugees in Minnesota. About twentythousand Southeast Asians continued to live in the metropolitanarea.Employment training for Southeast Asian refugees inthe St. Paul-Minneapolis metropolitan area has included abroad range of programs, the majority supported by stateand federal refugee assistance funds. At one end of thespectrum is the University of Minnesota's degree and certificationprogram for bilingual teachers. Orly the mosteducationally advantaged Southeast Asians have qualifiedfor this program. Employment training for refugees atsix metropolitan-area technical and vocational instituteshas also been provided but, due to the sheer number ofrefugee applicants, only on a selective basis. Vocationalschools with large Southeast Asian enrollments haveestablished special bilingual preparatory classes forthose lacking English proficiency.At the other end of the spectrum is the publiclyfunded Refugee Homemaker Program, where large numbers ofSoutheast Asian refugee womenparticularly Hmong women- -are enrolled in prevocational programs. There are noprerequisites for these classes, and bilingual interpretersare employed to assist in conducting the classes.Some classes have long waiting lists.Several small-scale employment training programshave been supported by both public and private funds.While these programsmany of which are sewing projectstrain a relatively small number of women, some have developedinnovative training programs that may prove usefuli- generating new ideas.Two agricultural projects for Hmong familiesoneprivately funded and the other supported by public andprivate fundshave proviaed English classes and trainingfor adult members of participating families. One of theprojects, sponsored by a voluntary agency, has thus farprovided training for the men only, although many of thewomen are enrolled in Refugee Homemaker classes and maybe trained in record keeping. The other project has providedwinter classes for men and women together, althoughthe men have far outnumbered the women during their firstseason.4.,

106 THE HMONG IN TRANSITIONHmong women have participated in training programsorganized for refugees and nonrefugees, programs initiatedspecifically for Hmong refugees and programs thatenrcll predominantly Hmong women.EMPLOYMENT TRAINING FOR HMONG WOMENIN THE METROPOLITAN AREA: AN ASSESSMENTWhile Hmong women in the metropolitan area have participatedin a broad range of employment training programs,most of these programs have trained them for industrialor cottage-industry sewing, housecleaning orfood service, labor that is generally low-paid and seasonalor part-time. Despite the initial availability offederal funds for training refugee women on an equitablebasis, these programs have tended to channel Hmong womeninto conventional sex-typed labor that represents a directextension of household activities. This policy canonly reinforce the traditional pattern of exploitation ofimmigrant women's work that has prevailed since the turnof the century. Programs departing from this conventionhave generally trained only a few women but have playedan important role in the development of a core of leadersamong Hmong women in the metropolitan area. A handful ofthese had prior education in Laos, but most did not. Acrucial factor in gaining admission to all but the lowestlevels of training has been access to English as a SecondLanguage (ESL) classes. Adequate time for ESL has notbeen available to all Hmong women on an equitable basisdue to priority lists favoring heads of households, andlack of support services such as child care and transportationȦt the post-secondary level, the University of Minnesota'sbilingual teacher training program includedfourteen Southeast Asian refugees, among them three Hmongmen and one Hmong woman, during its first academic year,1982 to 1983. Although only one Hmong woman participatedduring the first year, more may be trained in the futureas increasing numbers of Hmong students graduate fromhigh schools in Minnesota. The first Hmong woman enrolledmay also serve as a role model for female Hmonghigh school students, providing an example of a Hmongwoman who completed high school and continued her educationat the university level.113

TRAINING HMONG WOMEN FOR MARGINAL WORK 107This program is administered by the University ofMinnesota's College of Education and is funded by bilingualteacher education grants from the United States Departmentof Education. Participants receive financialassistance for tuition, insurance and books. To qualifyfor admission, applicants are required to pass the MinnesotaBattery Test, designed for students who did notgraduate from high school in Minnesota and have lived inthe United States for less than three years. Studentsadmitted to the College of Education or College ofLiberal Arts are also eligible.Most of the participants during the first year ofthe program were already employed as bilingual teachersin the St. Paul and Minneapolis public schools; by specialarrangement they were paid at the same rate asteachers with B.A. degrees and certification. However,a degree and teacher certification will enable the bilingualteachers to attain higher pay levels or move intoother areas of employment such as administration orresearch.At the six metropolitan area vocational schools, approximatelyeight hundred fifty Southeast Asian refugeesenrolled in employment training programs (including remedialand prevo-cational classes) during the fall term of1983. Of the total eight hundred fifty about three hundred(35 percent) were Hmong, but only sixty (7 percent)were Hmong women. Most of the Hmong women enrolled ineither the St. Paul Technical Vocational Institute (TVI)or the Minneapolis Technical Institute (MTI). Nine enrolledin a part-time prevocational program administeredby the Special Intermediate District 916 VocationalTechnical Institute (916 Vo-Tech) at the Galt School inWhite Bear Lake, an eastern suburb of St. Paul.Southeast Asian refugees applying for employmenttraining at these three schools are required to take theStructured Tests in English Language (STEL). Those withsatisfactory scores are admitted to regular vocationalclasses, and a limited number with lower scores areplaced in remedial or prevocational classes.In many ways vocational training for Southeast Asianrefugees at these three schools is reasolably accessible.All have Southeast Asian counselors, ar.:: two have bilingualassistants in some of the classes in which Hmongstudents are enrolled. Most Southeast Asian students are114

108 THE HMONG IN TRANSITIONeligible for Pell grants and guaranteed student loans,and some are eligible for public assistance while enrolledin employment training courses. All three schoolscan be reached by public transportation.For many Hmong women with preschool children, however,a major obstacle is the lack of adequate child-carefacilities. Although some have a mother-in-law in thehome who can care for the children while the youngerwoman attends school, in many instances this is not thecase. MTI students may use the child care services atthe adjacent Minneapolis Community College (MCC), but thecostup to $65 per week for a full-time student with onechildis prohibitive for most Hmong women. Neither 916Vo-Tech nor TVI provides child-care facilities.The largest number of Hmong women have enrolled intraining programs at TVI. About thirty-two enrolledduring the fall term of 1983. The convenience of TVI'slocation in the Summit-University areaan area of concentratedHmong resettlementhas made it a popularchoice among Hmong women. More important, TVI is the onlyvocational school in the metropolitan area that has developeda full-fledged bilingual program. Because Hmongwomen generally received less education than men in Laos,and have often had lower priority for dnrollment in ESLclasses in Minnesota,' this is an important factor inmaking vocational training more accessible to this groupof women.The Bilingual Program of TVI is the outgrowth of anearlier program initiated for Hispanic students in 1975.The latter program was established after the AFL-CIO andaffirmative action groups urged the Minnesota Legislatureto provide employment training for Hispanics who arriveeach year as migrant farm workers and often remained tolook for work in the winter. When the Southeast Asianrefugees arrived in the state soon after the program wasinitiated, the program was extended to meet the needs ofboth groups. Bilingual services for Southeast Asian studentsat TVI include bilingual counseling, bilingual assistantsin classes as needed, and bilingual supportservices, including tutoring, for students in all vocationalprograms. In addition, some bilingual courses areoffered each term for students with a low level of115

TRAINING HMONG WOMEN FOR MARGINAL WORK 109English proficiency. These prepare them for jobs that alsorequire little English, such as wall painting or apparelarts.While a few Hmong women who have graduated from Minnesotahigh schools achieve a high enough level of Englishproficiency to enter programs in such subjects asoffice skills or bookkeeping, the majorityof Hmong womenat TV1 have enrolled in apparel arts, a class in industrialsewing that requires an English proficiency levelof only 200 (the minimum for regular vocational classes)as measured by the STEL tests. Often there is a waitinglist of Hmong women for this class. At the end of eachterm recruiters from the garment industry and other manufacturingcompanies come directly to the school to signup Hmong women who have completed the training. Theserecruiters usually offer the women about $3.70 per hour,or just over minimum wage, and also offer them piecework.The hourly wage Hmong women can eventually earn in thisindustry may reach $5 or $6, but layoffs are frequentduring the winter months, and while some companies offerbenefits, others do not. Several companies have hired asmany as forty Hmong women to work under a Hmong foreman.With this arrangement the women need only a minimum levelof English proficiency.A much smaller number of Hmong women have enrolledin special programs in commercial foods that also requireonly a minimum of English proficiency. During the early1980s eighteen Hmong women were trained in food servicefor work in a restaurant at the Twin Cities InternationalAirport. Twelve of the eighteen women were hired by therestaurant manager as salad makers at minimum wage. Inearly 1984 a bilingual class for Southeast Asian studentswith English proficiency levels as low as 100 was initiatedto train kitchen assistants for restaurant work.Four Hmong women and four men signed up for the coursewhich will prepare them for work at just over minimumwage.A housecleaning class for refugees with low Englishproficiency, which was offered in the summer of 1983,also enrolled a majority of Hmong women (although mostHmong women training for the workplace seem to prefer toenter sewing programs). Only about 40 percent of thestudents got jobs after completing the course, and forthat reason the class was not repeated. Women who did116

110 THE HMONG IN TRANSITIONfind employment earned $4.50 or slightly more per hour,but worked part-time without benefits. After a number ofmen joined the class, however, instruction in janitorialwork was added for them, while the women received trainingin cleaning homes. Some of the men got janitorialjobs at $5.50 to $7.50 per hour, although they too workedon a part-time basis.A comparison of the types of training and economicoutcome for Hmong men and women at TVI illustrates theinequities of Hmong women's training opportunities invocational schools generally. If a woman trained in foodservice works full-time year around in a minimum wage jobat $3.35 per hour, she can earn $6,968 per year. A womantrained in industrial sewing can earn $4 per hour or$8,320 per year if she works full-time year around in agarment factory (most are laid off during the wintermonths). If her pay reaches $5 per hour she can earn$10,400 if she is not laid off. Women trained in housecleaningearn far less in part-time work. All the aboveincomes are well below the poverty level for a family ofsix, the average size of Hmong families in St. Paul andMinneapolis.2In contrast, Hmong men at TVI, many of whom alsolacked formal education when they arrived in the UnitedStates, have been channeled into fields such as automechanics, welding and machinists' training. Weldersmake from $8 to $15 an hour or more and machinists earnfrom $6 to $9 an hour. Wages for auto mechanics varyaccording to the number of skills mastered, and rangefrom just over minimum wage to $12 an hour. The economicbenefits increase if an individual opens his own shop..While much of the training of Hmong women at TVI andother vocational schools has been conventional, sex-typedjob training, it should be pointed out that other typesof training for Hmong women have existed in the past andstill occur from time to time. For example, from 1980 to1982, TVI staff developed one of the most innovative refugeeprograms in the metro area, the Health InterpretersTraining Program. About half the forty-five interpreterstrained in this program were Hmong women. Supported byfederal and foundation funding, the primary goal of theprogram was to train Southeast Asian health interpretersto provide effective communication between local healthcare providers and refugee patients.117

TRAINING HMONG WOMEN FOR MARGINAL WORK 111Initiated by the Ramsey County Child Health Consortium(comprising public and private health, education,welfare and social service agencies), the Health InterpretersTraining Program served both the East and WestMetro Refugee Health Projects. The projects were organizedin 1980 in response to the growing recognition bylocal health care providers of the need to expand theirservices to the large number of Southeast Asian refugeesarriving in St. Paul and Minneapolis in 1979 and 1980.With the guidance of the Refugee Program Office, TVIstaff members developed curriculum and teaching materialsfor a short-term certificate training course for healthinterpreters that included both health education and ESL.The two main criteria for acceptance into the programwere prior health care experience, either in SoutheastAsian or the United States, and sufficient proficiency inEnglish (tested at TVI). Trainees for the first classwere recruited through CETA and contacts in resettlementagencies and refugee mutual assistance associations.After the first class was trained, recruitment was nolonger necessary as refugees applied for admission afterhearing about the program, or were sent by resettlementagencies in Minnesota and neighboring states.More than half the trainees were Hmong, due to thelarge size of the Hmong communities in St. Paul and Minneapolisas compared to other refugee settlements. Onlythree or four Hmong men were accepted for the program.About twenty-two Hmong women were accepted and completedthe health interpreters training course.Health interpreters were generally placed in metropolitanarea clinics or nutrition programs, or returnedto agencies in other states immediately after certification.Initially their pay at the entry level was $5 anhour, but those employed by city or county agencies becamecivil servants and usually received substantialraises after their jobs were evaluated by the civil servicesystem. Those employed by voluntary agencies alsoreceived raises, and most were eventually earning between$8 and $10 per hoth by the time the program ended in1983. While most hearth interpreter positions becamepermanent, those who lost their jobs after the programended had no difficulty finding other jobs in the healthcare field.118

112 THE HMONG IN TRANSITIONMTI and 916 Vo-Tech have enrolled fewer SoutheastAsian students than TVI. However, the pattern of employmenttraining for Hmong women is the same at MTI, and 916Vo-Tech enrolled only one woman in regular vocationclasses in the fall of 1983. In 1983 approximately 200Southeast Asian students enrolled at MTI each term.About 60 of these were Hmong, but only 18 were Hmongwomen. The majority of the women enrolled in apparelarts services, a class in industrial sewing. There wasno Hmong counselorthe only bilingual counselor wasVietnameseand remedial and prevocational classes wereno longer available at the school. At 916 Vo-Tech approximately154 Southeast Asians enrolled in the fall of1983. About 90 were Hmong (including those in remedialand prevocation classes). Only 10 of these were Hmongwomen, and only one attended regular vocational classes.Nine were enrolled in a prevocational clas. A Hmongcounselor, who assists in classes when needed, and hisHmong assistant provided support services to Hmong students.The Refugee Homemaker Program in St. Paul and Minneapolis,supported by state and federal funding, enrollsthe largest number of Hmong women of any refugee educationprogram in the metropolitan area. While the homemakerprogram has been described as prevocational, it isfar less formal in format than prevocational classes atvocational schools. While the latter require testing foradmission and are concerned mainly with improving thestudents' English and math, homemaker classes have noprerequisites, and bilingual interpreters are availablefor all sessions. Tuition is free. Classes meet for twohours, once or twice a week.In St. Paul, classes are located in public housingprojects, and Hmong women living in the projects arewithin easy walking distance of the class sites. Childcare is provided without charge. In Minneapolis, classesare held in schools, churches and community centers.While they are generally located near refugee residentialareas, the women must provide their own transportation.Only one site provides child care.The St. Paul homemaker program dates back to 1964,when the residents of public housing projects werelargely immigrants or migrants from the South. The initialgoal was to enable low-income, homebound women to

TRAINING HMONG WOMEN FOR MARGINAL WORK 113conserve their resources by efficient home management.Curriculum included food and nutrition, family sewing,child development, consumer education and home management.The first Hmong enrolled in homemaker classes inSt. Paul in 1977. As the number increased,tions were setspecial sec-aside for them because of language problems,and interpreters were provided for these sections.Very quickly the classes became predominantly Hmong.Because the Hmong women wanted to learn employmentskills, an attempt was made to provide prevocationaltraining in classes like home sewing and housecleaning.Other classes like coat relining and knittingskills forcottage industry projectshave been added more recently.Nearly four hundred Hmong women were enrolled in homemakerprograms in St. Paul in the fall of 1983.An innovative spin-off of the St. Paul program hasbeen the Training for Hmong Liaisons, a project initiatedin the fall of 1983. Initially liaisons were selectedfrom women in the classes and trained informally by theinstructors to serve as intermediaries between Americanteachers and Hmong women students. The main criteria intheir selection was proficiency in English and knowledgeof the subject, although most were women with no prioreducation in Laos. Formal training in professional interpretingand general work skills will enable thesewomen to find comparable jobs if funding for their presentpositions is cut off, or to seek full-time, yearroundwork from which they can earn more than in theirpresent part-time, seasonal work (no classes are held inthe summer). Ten liaisons were enrolled in the Trainingfor Hmong Liaisons program, which included ESL as well asprofessional training. Initial training sessions wereheld at TVI with the help of TVI staff. These werefollowed by classes in homemaker classrooms with instructorsfrom the homemaker program. While pay for liaisonshas varied according to length of experience, trainedliaisons will earn up to $7 per hour. They may be ableto find higher pay outside the homemaker program.The Minneapolis Refugee Homemaker Program was initiatedin January 1983. While the St. Paul programserved as a model, the Minneapolis classes have been moreobviously shaped by the need of Southeast Asian refugeewomen to learn a skill for employment. One of the Minneapolisprograms, the Regina Project, which began as a1 2 0

114 THE HMONG IN TRANSITIONcommunity education class before the Minneapolis homemakerprogram was established, is a full-fledged employmenttraining course in industrial sewing and housecleaning.The Regina Project, begun in the fall of 1982,enrolls forty Hmong women per semester in two eighteenweekterms, beginning in September and ending in May.From Tuesday to Friday, two groups of twenty women eachattend one hour of English and two hours of sewing (includinginstruction in the use of industrial machines).On Mondays, all students attend two hours of housecleaninginstruction (four hours each Monday during thelast six weeks of the class), taught by an employee of aprofessional cleaning service.Women enrolled in the program are allowed to useboth the industrial and home sewing machines after hoursand on Saturday to do piecework on commission from localbusinesses. This provides the women a means of earningsupplementary income. By working very quickly, a few canmake up to $5 an hour, but most earn just over the minimumwage of $3.35 an hour.While Hmong women trained at TVI have had littledifficulty finding employment in the garment industry,Regina graduates often have not found work in industrialsewing, particularly in winter. A few have found sewingjobs elsewhere at $3.50-an-hour entry wages. Many havecontinued to do piecework at home if they own a sewingmachine. Apparently employers are looking for Hmongwomen with a higher level of English and sewing skills,such as those trained at TVI.Cleaning jobs have been available at $4 an hour,part-time, with no benefits. Many women trained in theRegina Project have found such jobs, although most preferpiecework sewing. Training at Regina has not includedHmong women could gethotel or janitorial work thus far.full-time hotel jobs at minimum wage, or part-time janitorialwork in the evening at $5 to $10 an hour. Thelatter includes no benefits, although full-time hotelwork may include them.Although no child care or. transportation is providedby the Regina Project, it has drawn Hmong women from allover the metropolitan area and 100 are on the waitinglist. A survey of women in the English class indicatedthat most were able to leave their children with unemployedspouses or other relatives, and that many had121

TRAINING HMONG WOMEN FOR MARGINAL WORK 115formed car pools from St. Paul or north Minneapolis toreach the Regina Project in south Minneapolis. Whilethis shows the resourcefulness of Hmong women and theirneed for employment training, it may also mean that womenat Regina are a self-selected group that has greater resourcrtand more relatives to turn to than many otherHmong women may have available.Other Minneapolis homemaker classes besides theRegina Project have included instruction in family planning,consumer education, foods and nutrition, use ofcommunity services, home management, office skills andorientation to employment. Most of the classes haveinterpreters available. About 250 Hmong women, as wellas other refugee women, were enrolled in these programsin the fall of 1983.An experimental program was also initiated in 1983by the coordinator of the Refugee Homemaker Program.While still in its earliest stage, and already threatenedby cuts in federal funding, the program provides statevocational part-time teaching licenses for SoutheastAsian instructors in health and nutrition. The instructorsare employed as interpreters by metropolitan areahealth agencies, but are paid for their part-time teachingby the homemakers program. Eventually classes willbe taught by Southeast Asian instructors in other homemakersubject areas as well.Three Southeast Asian instructors--all Hmong womenhave been licensed thus far. Licensing is obtainedthrough MTI, and job training is provided by sponsoringagencies. The three aiready licensed were trained in theHealth Interpreters Training Program at TVI, and thenemployed by the Ramsey County Nursing Service, the St.Paul Public Health Department and the federal food programfor Women, Infants and Children (WIC). They teachweekly classes in health and nutrition at McDonough Homesin St. Paul and were paid $14.23 an hour in 1984.While the main purpose of the Refugee HomemakersProgram has not been to provide vocational or prevocationaltraining to refugee women, several innovative projectsin interpreter and teacher training that haveemerged from the Homemaker Program provide evidence ofthe possibilities for Hmong women's employment trainingbeyond the conventional areas of industrial s_ ding anddomestic housecleaning.122

116 THE HMONG IN TRANSITIONFive small training projects for Hmong women in themetropolitan area have been sponsored by a privateagency, an MAA, a church organization and two enterprisingindividuals. Three are sewing projects that are primarilyincome-generating, but have added a trainingcomponent. In two of these the women work for minimumwage or less while receiving training in handwork ormachine sewing. Some earn just over minimum wage onpiecework, if they work very quickly. In the third,three women who were trained in the shop earned $4.50 perhour in the fall of 1983, and continued to learn otherbusiness skills in the course of their work. In two projectsthe women work part-time only, while one providesfull-time work for four or five women a year. None ofthese projects provide child care or transportation, andnone receive public funds.Family Service of St. Paul, a private agency supportedlargely by United Way funding, has provided bilingualinstruction for Hmong women interested in licensingtheir homes for day care as a means of supplementingtheir income. Since early 1981 more than 100 Hmong womenhave participated in the agency's Family Day Care TrainingProgram. This training includes twelve to fifteenhours of training in application procedures, county regulationson health and safety, nutrition, first aid andchildren's activities such as crafts and games. There isno charge for the training. No child care or transportationis provided.Although many Hmong women have participated in theprogram, a large number have not completed the programfor a variety of reasons. Often their homes did not passlire inspection or they could not afford the requiredequipment. In early 1984, twenty-seven Hmong homes werelicensed for day care.Many of the clients of Hmong day-care homes areSoutheast Asian refugees attending ESL or employmenttraining classes. If they are on public assistance theMinnesota Department of Public Welfare pays for day carewhile they attend classes. The Department of Welfaremakes day-care payments directly to the care provider atthe rate of $1.40 per hour or a maximum of $9.75 per dayper child. No more than aye children, including a maximumof two infants, can be cared for by one woman. However,the majority of clients leave their children only a123

TRAINING HMONG WOMEN FOR MARGINAL WORK 117few hours each day, and in the summer tar fewer childrenare brought to day-care homes, with the result that mostcare providers' incomes vary considerably according tothe season. If a Hmong woman cares for three childrenfor four hours a day, five days a week, she can earn $336a month, from which she must subtract any expenses involvedin providing the care.Often Hmong women who provide day care are themselveson public assistance, and welfare regulationsrequire that 40 percent of the earnings of self-employedrecipients (providing day care) must be deducted fromtheir welfare grants. If they live in public housing,their rent will be raised proportionately to the amountearned.Training for family day-care licensing does notqualify an individual for work in a day-care center, andjobs in such centers are often difficult to obtain. Thuswhile Hmong women may gain some knowledge of Americanexpectations in caring for children, the training theyreceive does not provide directly transferable skills orqualifications.In the fall of 1983 a housecleaning training programfor Southeast Asian men and women was initiated by a publiclyfunded refugee employment service. Of the sevenwho completed the ten-week course in December, three wereHmong women. All are working for $4 an hour on a parttimebasis with no benefits. Because they are welfarerecipients they must work under twenty hours a week toavoid jeopardizing their benefits. No support servicesare provided by the training program; those receivingwelfare may be eligible for child-care and transportationassistance.All four of the small-scale projects described haveprovided training in traditional areas of women's work,leading to low-paid, part-time or seasonal work. Businessskills learned in one project could be transferableto another job or used for self-employment in a businessor cooperative enterprise. Neither child care nor transportationhas been provided by these projects.In the spring of 1983 the Agricultural ExtensionService of the University of Minnesota launched a trainingprogram for thirty Hmong families as part of theMinnesota Agriculture Enterprise for New Americans, knownas the Hmong Farm Project. Participating Hmong families124

118 THE HMONG IN TRANSITIONhave remained on public assistance during the trainingperiod, but eventually will be owner-members of a Hmongcooperative:. While no households headed by women havethus far been included in the project, Hmong women haveparticipated in winter training classes on an equal basiswith their husbands during the first year of operation.Training has included ESL and agriculture courses. Womenhave participated in machine maintenance, woodworking andother work traditionally performed by men. During thewinter of 1983-1984, these classes comprised thirty-twomen and seven women. The number of women participatingis expected to increase after the first year. The projectpays transportation and child-care allowances forparticipating families. The economic outcome cannot yetbe determined.CONCLUSIONThe data from the present study indicates that mosttraining programs for Hmong women in the St. Paul-Minneapolismetropolitan area have reflected a conventionalnotion of immigrant women as needleworkers and housedeaners,a notion more relevant to an era that is longpast in the American economy than to the 1970s and 1980s.While such labor provided immigrant women with a stablebut meager income at the turn of the century, currentpolicies of training Hmong women in the same areas ofmarginal, low-paid work can only create a permanentlydisadvantaged group of women.On the other hand, several innovative training projectsthat have emerged in the fields of teaching, interpretingand health care suggest that, given the opportunityfor adequate preparation in ESL and employment training,Hmong women can successfully enter the economicmainstream of American society. A program still in theplanning stage proposes retraining hea th interpreters,and others with health-care backgrounds in Laos or theUnited States, as health professionals. This mightinclude training as nurses, nurses' aides, medical techniciansand dental assistants. Other areas not fullyexplored include training Hmong women in administrativeskills, business management and organization of cooperatives.125

DataTRAINING HMONG WOMEN FOR MARGINAL WORK 119from this study also indicates that the lack ofsupport servicesparticularly day care and transportationhasbeen a serious obstacle for many Hmong womenwho want to study ESL and attend employment trainingcourses. The lack of these services in some cases hasbrought about a process of self-selection whereby onlythose women with relatives they can turn to for childcareassistance can attend classes.The question raised by this study is how resettlementpolicy can be restructured to provide more innovationin training programs for Hmong women, and to assurethe availaoility of support services necessary to givethem a fair chance to participate in the employmentopportunities in the United States.ACKNOWLEDGEMENTSI owe a great debt of gratitude to all the administrators,teachers, bilingual counselors and students whopatiently answered my questions about Hmong women'straining programs in the St. Paul-Minneapolis metropolitanarea. Among them are May Beecham, Sandra Becker,Father Cherian, Gloria Congdon, loyce Dewey, SandraDuvander, Sharon Washburn-Dreyer, Karen Gensemer,Patricia Hattiberg, Lucy Hartwell, Colleen Holverson,Shirley Hume, Laurel Janssen, Kathleen Johnson, SisterMarie Lee, Herbert Murphey, Janice Pederson, Ninh Phan,Naida Reda, Darlene Ross, Rebecca Storlie, RichardTeachout, Francisco Trejo, Blang Vang, Constance Walker,Kia Moua Yaj, Blang Yang, Gaoly Yang and Hoa Young.NOTES!In the early years of resettlement, heads of householdsin the metropolitan area were often given priority inattending ESL and employment training classes. This wasdiscontinued following a lawsuit. (Patricia Hattiberg,interview with author, St. Paul, Minnesota, 26 January1984.)126

120 THE HMONG IN TRANSITION2Douglas Olney's Hmong Community Survey, Center for Urbanand Regional Affairs, University of Minnesota, unpublished,1983, indicates the average size of Hmong familiesin St. Paul is 5.9, and in Minneapolis 5.7; 1981Bureau of Labor Statistics.REFERENCESFass, Simon M. 1983. "Development Dynam;,7s: HmongRefugees in America." Unpublished.Walter, Ingrid. 1981. One Year After Arrival: TheAdjustment of Indochinese Women in the UnitedStates, 1979-1980. Lutheran Immigration and RefugeeService, New York, N.Y. (Available at no chargefrom Lutheran Center, 360 Park Ave. S., New York,N.Y. 10010.)127

CHANGES IN HMONG REFUGEE MARITALATTITUDES IN AMERICAWilliam H. Meredith and George P. RoweHmong refugees from Laos have been tossed from anisolated, simple environment into a technologically modernsociety with customs and language very strange tothem. The Hmong are known for their strong, highly cohesive,extended families. How have their family, and marriagepractices in particular, been able to adjust tosuch dramatic changes? What traditions have they maintainedin this strange land? Which have they been forcedor chosen to drop?This article will examine the impact a new culturehad on the institution of marriage for one ethnic groupof refugees that began entering this country in 1975.willItalso review marriage customs the Hmong practicedwhen they lived in Laos. The adaptation of their maritalrelationship in the United States will then be examinedbased on the results of a research study of 134 Lao Hmongadults.MARRIAGE IN LAOSHmong society was a patrilineal clan system. Thesociety was divided into social groups, or clans, and achild at birth was automatically placed into the clan towhich the father belonged (Clarke 1973; Barney 1967).Therefore, a modified brother-sister taboo was observed.It was not proper for a young man to express interest ina girl bearing his clan name. However, the child of aman could marry the offspring of that man's sister becauseinthe eyes of the Hmongthey would not have beenof the same clan.After puberty, a young man attempted to gain the attentionof a girl of his liking. This was generally doneduring the New Year's Festival season when different villagesjoin each other for feasting. Girls proudly woretheir colorful skirts as a display of their ability to121128

122 THE HMONG IN TRANSITIONsew and 'mbroider.Young men demonstrated their prowesswith hones and in contests. They also serenaded thegirls with various musical instruments (Barney 1967).If a relationship between a boy and a girl developedinto a serious mutual interest the boy found excuses tovisit the girl's village. Girls who had attained pubertyslept apart from the rest of the family so young menmight be expected to visit during the night when the restof the family was asleep. Premarital sex, however, wasclosely controlled and disliked by parents. A girl whobecame pregnant was considered a disgrace to her family(Barney 1967). Men married between eighteen and thirty,while women married between fourteen and eighteen.Marriage could only be realized after considerablemaneuvering and bargaining. The young man secured a "gobetween,"usually a relative, to carry on negotiationswith the girl's parents (Clarke 1973; Barney 1967).When the negotiations between the go-between and thegirl's parents resulted in consent to the marriage, awritten agreement was signed. (Hmong informants told usthat every village had at least two or three people whocould write, and most marriage contracts were put inwriting from 1972 on. Prior to that time, contracts wereabout evenly split between oral and written. If oral, theleaders of the clan and the go-betweens were the contractwitnesses as long as they lived.) As part of the marriageagreement the man would pay the prospective wife's parentsa bride price of at least four fifteen-ounce silverbars. Under recent silver prices that would be the e-quivalent of $600 in the United States. The agreementwould also include provisions concerning the possibleeventualities such as divorce. In addition, the husbandgenerally agreed not to say bad things about the girl'sclan of origin. Prior to any use of physical force, theman would also agree to take his wife back to her familyfor "counseling and assistance" if she did not prove tobe a good wife. To be a good wife, the woman must listento her husband and know her responsibility to him. Shemust accept his decisions although she need not agreewith them. She also was expected to be active in farmproduction and to open her house to guests.A feast with a roasted pig was then held over twodays as the official announcement of marriage. It alsoserved to provide a bond between different clans. Some-129

CHANGES IN HMONG REFUGEE IN AMERICA 123times, for political reasons, young men and wome., wereforced by their families to marry mates they did not wishto narry.If a settlement could not be reached, the couplemight resort to an elopement. In such cases settlementswere made afterwards by a panel of neutrals. Before the1960s, elopement was not possible because of tight parentalcontrol. However, young people gradually startedlearning about customs in other countries. Occasionallya girl would not wish to marry a boy. While the parentswould act upset, it was considered acceptable for theboy, with the help of his friends, to kidnap the girl.Generally the threat of kidnap was enough to encouragethe parents to negotiate an agreement.Marriage had very important effects on the girl'srelationship with her parents (Barney 1967). In theory,when she left her father's family, household and clan,she became fully identified with her husband's family.Should he die, she would become responsible to his family.However, she could visit her family of origin frequently.Sometimes an additional wife was secured to add tothe family work force. In some cases, the first wifewould request her husband to marry an additional woman.Having more than one wife was a common custom but wasusually the result of the levirate, which is the practicewhereby a man's widow automatically becomes the wife ofhis brother (Barney 1967). This practice increased rapidlyin the 1960s as the result of the war. Wealthy menwere able to have several wives. This was a sign of theirwealth. Hmong leaders would also take more than one wifeto gain political support from rival clans. One of themwas considered to be the more important "big wife" anddirected the others in terms of household duties. Allwives usually shared the same sleeping area. Divorce waspossible, but was avoided at all costs (Barney 1967).Disagreements between couples were heard by clan leadersand strong words of advice were given. If a woman lefther husband she would lose all rights to whichever oftheir children the husband's family wished to keep. Thehusband's family would generally chooseto keep all ablebodiedmales and girls who could be expected to bring agood marriage price. In earlier times the wife's familyhad to return the bride's price to the forsaken husband.130

124 THE HMONG IN TRANSITIONIn more recent times, the wife had to pay the husbandwhat he had given for her. Unhappy wives were left withthe choice of fleeing the village or, in rare cases, committingsuicide, at which time the husband would have topay damage to the wife's parents.The family structure after marriage was patriarchalThe father was head of the family and was(Barney 1967).in strict control of the family. The mother was in chargeof housework. The father made all final decisions andwould lot tolerate strong disagreement. Women's influencevas more subtle and indirect and they generally heldan inferior position to men. Although families varied,the ideal role of the wife was to have devotion and respectfor the husband (Yang 1979). If a wife came from afamily of power or wealth she would expect more influencein decision-making. Because of the husband's pleasure inbeing connected to her family he would usually give her agreater role in family decisions. In other cases wherethe wife was clearly more intelligent than her husband orin those few cases where she earned more money than he,her power was increased. Others in the village, though,generally disapproved of the man who could not "control"his wife.METHODIn order to determine changes in marital and familylife attitudes, twenty-six statements were selected thatincluded both values commonly held by the Hmong in Laosand values commonly held by Americans in the UnitedStates. The statements were selected with the help offour Lao Hmong consultants who also conducted the interviews.The interviews were handled in small groups. Afterhearing the question each respondent marked his orher own answers on an answer sheet. The questionnaireswere administered orally to the refugees because the majoritywere illiterate. The language used for the oraldelivery was Hmong.With the exception of the first four questions concerningthe respondents sex, age, education level andlength of time in the United States, the remaining thirty-one131

CHANGES IN HMONG MARITAL ATTITUDES IN AMERICA 125statements all involved the same response format. Thiswas done to make the questionnaire easier for a populationnot acquainted with surveys to understand.Two response formats were included as one: theLikert scale and the Faces scale (Andrews and Withey1976). Each of the categories of the Faces scale correspondedto a rating level on the Likert scale. The interviewersread the titles c,f the faces to the respondentsseveral times. After that point, the illiterateHmong, in particular, basically answered to the Facesscale. As many of the illiterate Hmong do not know numbers,it would have been unrealistic to expect them tofollow an answer sheet where questions were labeled withnumbers. therefore, a picture was substituted for eachstatement number. The interviewer read each question andtold the illiterate refugee to answer on the line labeledby a certain picture. The Faces scale and this nu.nberingmethod allowed illiterate respondents to answer in a confidentialmanner.In order to better understand changes in attitudetoward marriage over time and culture changes, a retrospectivemethod was used. Respondents recorded both whatthey remembered their attitude toward the statement tohave been in Laos and their attitude at the time of thesurvey.Data were analyzed in terms of percentages and interms of significant differences (based on the Chi-squaretest) between the retrospective and present view. Also,significant differences for the present view were correlatedwith demographic data. Significance was set at .05or lower.POPULATIONThe population studied consisted of every Lao Hmongrefugee, age sixteen and above, living in Omaha, Nebraska.One hundred thirty four persons fell in this agerange. Most were young adults between sixteen and thirty.In terms of education, the formal educational level attainedwas limited. Forty percent of the 134 participantshad no formal education while another 23 percent had somegrade school education only. High school graduates andthose with some college background or specialized train-132

126 THE HMONG IN TRANSITIONing accounted for 18 percent of the refugee adults. Nearlyone-quarter of the Lao Hmong refugee adults had livedin the United States for between five and six years. Fewerthan 19 percent had resided in this country less thantwo years.There were seventy females in this study and sixtyfourmales. Since the total population was studied, allthe married participants in the population were marriedto another participant.RESULTSThe results of the marriage attitude survey with 134Lao Hmong refugee adults will be discussed under fourtopical areas: mate selection, marriage preparation, marriageroles and divorce. Significant differences in responsesbased on demographic data between the retrospectiveand present views will be indicated when present.See Appendix 1 for the present attitude responses for thetotal group of Hmong refugees only.Mate SelectionWith the exception of the attitude toward polygynousmarriage, past attitudes concerning mate selection havegenerally been maintained. Slightly more than half of therespondents felt it was not proper for a Hmong person tomarry someone who was not also Hmong. Most of the remainingadults were unsure. Only 7 percent supported remarryingsomeone from the same clan, although a significantdifference was found based on education (p=

1CHANGES IN HMONG MARITAL ATTITUDES IN AMERICA 127approve of a boy taking a girl for marriage when the parentsdisapproved. Nearly half reported they had thoughtthat was proper in Laos.Interms of age of marriage, 87 percent stated thatit is best if a woman waits until she is eighteen to marry.This was significantly higher than the reported retrospectiveview of attitudes in Laos (p=

128 THE HMONG IN TRANSITIONMarriage RolesWhile greater agreement with the idea of equality asa general principle exists now as opposed to theseadults' perceptions of their past view, still clearly thehusband holds the dominant role. In fact, 83 percent ofthe respondents agreed with the statement, "A good wifeshould do what her husband tells her to do." However,those who were educated were significantly less likely toagree with the statement (p=

CHANGES IN HMCNG REFUGEE ATTITUDES IN AMERICA 129sions. In spite of the confusion present, both men andwomen were significantly less likely to believe presentlythat the children belong with the father's family thanthey did in Laos (p=

130 THE HMONG IN TRANSITIONcation of wealth and influence. Those who were educatedwould more likely be in positions of power and wealth.They would more likely feel robbed of a badge of influenceby our laws on marriage. Polygyny found very littlesupport from the Lao Hmong women.Past material concerning the Hmong has been of ahistorical, anthropological or journalistic nature. Somewritings suggest a strongly cohesive culture. Marriagecustoms are discussed as being held by all Hmong. Suchdiscussions do not include the differences of opinion andvariability in attitude that this study shows to exist.Past writings appear to have problems with generalization.The present paper has examined the current attitudestoward marriage of Lao Hmong refugees in the UnitedStates.Retrospective views were also examined as werethe influences of various demographics. The paper hasalso shown the differences of opinion that exist withinthis population.The Lao Hmong can anticipate further change in attitudestowards marriage as their residence in the UnitedStates lengthens and they come under increasing Westerninfluence through the educationai system, the media andpersonal relationships with those of the new culture.Increasing financial independence will also create greaterfreedom from the Lao Hmong community. Such factorswill allow many Hmong the ability to closely examinetheir past customs, as well as the new dominant culture'sattitudes. Many will choose to blend the new with theolder customs to meet their specific needs.REFERENCESAndrews, Frank M., and S.B. Withey. 1976. Social Indicatorsof Well-being. New York: Ptenum Press.Barney, G.L. 1967. "The Meo of Xieng Khouang Province."In Southeast Asian Tribes, Minorities, anu Nations, Iedited by P. Kunstadter, pp. 271-294. Princeton: 1Princeton University Press.137

CHANGES IN HMONG MARITAL ATTITUDES IN AMERICA 131Clarke, R. 1973. Peoples of the Earth. Verona: DanburyPress.Vang, Tou Fu. 1979. "The Hmong in Laos." In Introductionto Indochinese History, Culture, Language andLife edited by 3. Whitmore, pp. 93-192. Ann Arbor:Center for South and Southeast Asian Studies, Universityof Michigan.

132 THE HMONG IN TRANSITIONanEd1091Appendix 1Pewit Attitude Responses to the limong Marriage Inventoryfor the Total Group of limns Refugees in Omaha, Nebraska(in randed-oli psr-unagenNei34I. It Is not right for a liaiong person toMelly amescoe Ms Is net also limn&2. te s wr dtotarry imam fromStrongly Meetly2419P--- AirsNotbMMostly Strongly_bum Zama!33% 39% 13% 6% 1%7% 0% 9% 14% 69%3.wa weednum tor greaterkwtawar* power Lsp.4. The huebend anti We shad! love eachother in manage.3. If a boy mate to marry a girl andher prams &approve, it Is allrhOtt for Ms to take her stymy.6. A omen should have the final say annem the mats to marry.7. Yeas people should met or dateseveral people before deciding Montthey mat is marry.S. A girl should net have to get marrieduntil she vents to.9. %mem *add wait Until they my atlent eighteen years old to merry.10. If it ea a legal, 1 believe that aman shorn lave the freedoms to RAMMOM thin the Vast If he nuns to.A1119MIEBISkat11. It is proper fora men to pay a brideprice for his mite.12. A marriage works best If a contractbetween the Walls is agreed uponin advance.13. It is ben to hem go-betmensamnia the marriage, rather thin theboy and the girl themselves.Merriam Roles14. It is the mother's lob to take careof the children.51% 28% 16% 4% 1%76% 13% 6% 1% 4%21% 17% 19% 26% 17%51% 29% 10% 4% 4%13% 31% 18% 10% 6%49% 24% 9% 8% 7%64% 23% 5% 2% 5%14% 7% 21% 11% 44%34% 24% 21% 11% 10%42% 28% 14% 6% 5%47% :3% 15% 11% 5%33% 39% 13% 6% 10%13JBEST COPY AVAILABLE

CHANGES IN HMONG MARITAL ATTITUDES IN AMERICA 133U. The husband Ind left should be equalIn power in marriage.M. The eau, not the ovema, *add makethe nosier decisions In the for*.D. A meths, with smaU Minikes shouldnot lark at a jai amide el the homeIII.A geed wife amid de stet how hellions*tells her le de.19. Noland, should help their wives withhausewerk.10. A father *raid help mks ass el hischildren.II.the men should noise the decides u towhether his wife 0101610111 her educations.22. It is all right for a seem to woe*outside of the hoses once her childrenare green.StronglyAareeMostlyNotAgog_ laMostlyStronglyNuns_ D_Inimi53% 20% 13% 6% 10%24% 311% 23% 11% 3%38% 24% 15% 11% 12%53% 30% 7% 4% 6%54% 27% 10% 7% 1%63% 22% 7% 2% 4%zn 21% 24% 21% 5x57% 79% 2% 1% 13%Divorce23. in one of deems. possessionsthet.4 be divided In half between thehusband end wile, regardless of whorefault It is.Se.In awe d divorce, the childrenbelting with the father's badly.73. if the father aft the father'sfull, should decide what happens tothe children.16% 16% 35% 18% 16%23% 13% 36% 16% 13%30% 22% 21% Et% 13%BEST COPY AVAILABLE140

RESOLVING SEXUAL ASSAULT:HMONG AND THE AMERICAN LEGAL SYSTEMBeth L. GoldsteinMuch of the research on sexual assault in theUnited States either focuses on the person, seeing sexualassault as an act of violence of one individual againstanother, or focuses on power relations between gendersand how they are acted out in our jural and social systems.The American legal system itself emphasizes theprotection of individual rights and punishes individualsfor violation of others' rights. Institutionalized assistanceto both victims and assailants also centers onthe individual. Men and women participate in the privateand public negotiations which take place.Hmong mores about gender, male-female interaction,and family and social control contrast sharply with Americanones. Within Hmong society, sexual assault is a matterinvolving the family, kinship group and intergrouprelations, rather than centering on individuals and genderrelations. Hmong emphasis on the group over the individualassigns responsibility for actions, includingindividual ones, to the group. In the private, homecenteredsphere, women participate in decision-making,but the public sphere is dominated by men. Therefore,Hmong values and actions are often at odds with the waythe American system assists in resolving sexual assault.When assault of a Hmong occurs in the United States, thedifferences between American and Hmong social structures,conceptions of personhood and approaches to conflict resolutionbecome critical.The situation is further complicated by the currentperiod of acculturation to life in the United States. Ashas been documented for various immigrant ethnic groups,individual members of the group go through a process ofselective acculturation during which their behavior ismodified by and for the majority culture, based on patternsof interaction and conflict (Clark, Kaufman andPierce 1976; Melville 1980). The Hmong adolescents andyoung adults typically involved in sexual assault cases1354 1

136 THE HMONG IN TRANSITIONhave been exposed to both Hmong and American attitudes.Their emerging adult identities are a composite of Hmongand American values, perspectives and expectations. Asthe following case study illustrates, their attempts tocope with the emotional, jural and social aspects of anassault are entangled by personal internal conflicts aswell as public external ones between the two socioculturalsystems. As DeVos and Romanucci-Ross (1982) suggest,to understand continuity and change in ethnic identity,we must study people in social settings and examine howthe subjective experience of an ethnic identity is relatedto adaptive behavior.This paper presents a case study of a Hmong girl assaultedby a Hmong man to examine how different culturallenses influence the resolution of an assault. One afternoonafter school, a fourteen-year-old Hmong highschool girl was offered a ride home by two Hmong boys whowere fellow students. Although she usually rode the bus,she accepted a ride on this occasion because one of thetwo was her cousin. The first person to be dropped offwas the cousin. Then the driver turned onto an unfamiliarroute, over the protests of the girl, stopped in anisolated wooded area and assaulted the girl. She eventuallyescaped from the car and began to walk home. Policeon routine patrol passed her, noticed the tornclothing and bruises and picked her up. Soon after, theassailant was located and arrested, and a statement wastaken. The victim was brought to the hospital for a gynecologicalexam, at which time a counselor from a rapecounseling center and members of her family were broughtin. The assailant was booked for first-degree sexual assaultof a minor and later released on bail. The casewas handed over for prosecution to the district attorney'soffice, and a public defender was assigned to theassailant. Eighteen months later, following a plea bargain,the assailant was sentenced to probation.The assault victim, a tenth grade girl whom we shallcall Joua, had lived in the United States for threeyears. Born in Laos, she spent two years in a Thai refugeecamp before being resettled together with her motherand older brother in the United States. She was sponsoredby her brother-in-law, wno had arrived two years142

RESOLVING SEXUAL ASSAULT 137previously. She lives with her mother, an older womanwho speaks no English and seldom leaves home withoutanother family member as escort.The assailant, whom we shall call Pao, was a nineteen-year-oldhigh school junior. He had lived in townwith elderly parents for more than four years. Becausehis English is quite fluent, while his parents are illiterateand speak only Hmong, Pao has become their primaryinterpreter of American society. Pao has no other familyin the immediate vicinity. As a legal adult, Pao receiveswelfare assistance payments independently of hisparents and is therefore more financially mobile thanmany of his classmates.Joua's and Pao's mothers were acquainted with eachother in Laos. However, the two children had had no contactwith each other until they coincidentally enrolledat the same high school in the United States. Until theday of the assault, Pao and Joua were never alone together.He had repeatedly offered rides to her, which shealways refused. In Hmong traditions, unrelated males andfemales have no contact with each other after puberty unlessin the company of other people. Married couples donot touch each other in public, and teenage boys do nottouch even the women in their immediate family. It wasonly because of her cousin's presence that Joua agreed toaccept a ride on this one occasion.Cultural conflicts arose in the days immediately afterthe assault. The girl's adult male relatives gatheredat her home to await a personal call from the assailant'sfamily. The culturally appropriate Hmong responseto assault would be for Pao to publicly acknowledgeresponsibility and for his family to pay restitutionto Joua's family in lieu of the bride price thatcould have been expected from a marriage. As becameclear later, the money was most important as a symbol ofthe re-establishment of balance between the two clans andthus as a public peacemaking gesture. However, the assailantand his family did not contact the girl's family.Through the community grapevine, her family was informedthat because she went to the police and sought the interventionof an external authority, his family felt theywere no longer bound to follow Hmong tradition. Instead,the assailant went to a state-level Hmong employee to requestthat the government worker extricate him from the143

138 THE HMONG IN TRANSITIONcharge on the grounds that a marriage was pending, Thisrequest was denied on the basis that they lived in theUnited States now and that, since the police were alreadyinvolved, American legal procedure must be respected.Only if satisfactory settlement was not reached withinthe court would a limong council of elders be willing tobecome involved.Among the issues raised in this incident is that ofethnic boundary maintenance (Barth 1969). At stake forthe Hmong is how to balance their desire to re-establishHmong social organization and control through clan networks,with the fact that they must simultaneously livewithin American structures and controls. Whereas thegirl's family expected Hmong procedure to be followed,the assailant's family argued that the boundary betweenHmong and American tradition had already been crossed.But then, when Pao violated Hmong protocol, Joua's familydecided to use the American court system to achieve itsdesired ends. While the senior members of the two familiesadvocated settling within the Hmong community, theyounger members, all of whom interacted daily with bothHmong and American communities, vacillated as to whichsystem could best settle the dispute. Hmong marriagecustom calls for courting during public celebrations,such as the New Year. Marriage arrangements are subsequentlymade by the two families or a form of "elopement"occurs. "Elopement" is normally preceded by mutual consentof the couple, with the girl leaving her home tojoin her suitor. Within days of the "elopement," publicacknowledgement of the marriage is made, and the brideprice is settled between the families. When Pao went tothe Hm^ng government employee, he chang i his account ofthe incident from the admission of assault given to thepolice, to overtures of marriage. By claiming that Jouahad consented to accompany him, he ref ramed the incidentin terms of traditional marriage"elopement"--procedures.Before the preliminary hearing; Joua's senior malerelatives gathered to have a representative of the rapecounseling center explain the legal proceedings to them.She explained the court system and prosecution process,the presumption of innocence until proof of guilt, thecourt delays that could be expected in reaching a resolutionand the difficulties that Joua could expect giving144

RESOLVING SEXUAL ASSAULT 139testimony. She tried to explain why Joua was considereda victim an i not a guilty party by American law and bythe counseling center. According to American socialnorms, Joua would be able to lead a normal life as a girland later as a woman. The counselor perceived Joua'smost pressing need at the moment to be for psychologicaland emotional support from family members. The men respondedthat the assault may be of little long-term consequencein American society but that while marriage wasstill possible, no self-respecting Hmong man would marryJoua now. And the family certainly did not want her tomarry a non-Hmong. The meeting allowed the men to understandthe options available to the family in the Americancourt system and to decide on their action. This entireconversation took place in English and Lao, languagesthat the women standing on the periphery of the groupcould not understand. When it was requested that the mentranslate it into Hmong for the women, the three-hourdiscussion was summed up in one sentence. What mightlater have been conveyed to the women is not known.Through this process, Joua became increasingly confusedabout her identity and self-worth. At this earlystage in her selective acculturation to American society,Joua had assumed many of the material-culture characteristicsof American teenagers. For example, her physicalappearance was that of a typical high school girl. Shehad acquired commercialized teen notions of romance andmarriage. These superficial characteristics led theAmerican lawyers and counselors to a false expectation ofher sexual attitudes and ethnic values. School had becomevery important to her. She had come to judge herpersonal value according to accomplishments hi school (anew and still uncommon phenomenon for Hmong women). But,following the assault, particularly given the presence ofthe assailant in the same school building, she could notconcentrate on her schoolwork and her grades dropped.Thus, at the same time as her family's estimation of herdeclined, her own self-image was falling.Because of her strong bonds with family members,Joua wanted support from them.She said she had nobodywithin the family to talk to, since her mother did notunderstand what was happening and she could not discusssuch personal topics with her male relatives. In Laos, anetwork of kin women would have lent the girl personal145

140 THE HMONG IN TRANSITIONShe would have been able to talk to them, asksupport.them about how the incident might affect her future anddiscuss the physical pain she'd endured. However, thisnetwork of support had been scattered by war and resettlement.Joua became confused by the contradictory messagesshe was receiving from male relatives and fromAmerican counselors. She took to heart family criticismof her behavior and attempted to suppress her anger overthe assault and the slowness with which the courts werehandling it. She became increasingly concerned that thefamily be satisfied with the court decision, since thatwould relieve her of family pressure.Pao was also suffering from cultural disorientationand misunderstanding of American sexual mores. His interpretationof what he saw and heard of American highschool males' behavior initially led him to expect leniency,if not support, from Americans for his own behavior.He perceived his sexual aggressiveness as a permissibleblend of American and Hmong actions. He was takingadvantage of the context of the weakened Hmong socialcontrols and experimenting with his own version of a"liberated" American male.At the preliminary hearing many of these issues cameto the fore. All the officials present were male, whichintimidated the girl. Testimony was arranged through atranslator, which proved difficult for several reasons.It was impossible to find an available translator competentin Hmong and English who was unrelated to either ofthe two parties. Therefore, a Hmong social worker whowas related to both was used. But he was concerned thathis neutral position in the community would be destroyedby his participation in the hearing. Only with the consentof community elders did he agree to translate. However,the translation itself proved hard because of itssexual content. It was inappropriate for him to be discussingsuch topics with the girl; verbatim testimony andtranslation were therefore hampered.It became clear that Joua's family did not want tobe held responsible for any greater disruption of Hmongand kin networks than necessary. They tried to use thecourt to achieve the desired ends of public punishment ofthe youth and financial settlement. They wanted it toappear that the decisions were the court's and not theirown since they knew the financial burden on Pao's family14d

RESOLVING SEXUAL ASSAULT 141would be severe and therefore potentially divisive. Becausefinancial settlement is not usual in assault cases,the court would have required a public statement fromJoua's family that it would be appropriate according toHmong custom. Since her family did not want to publiclyacknowledge this claim, the issue was dropped. Joua'sfamily decided it would instead be satisfied with a jailsentence to be served in the local community so all wouldsee justice achieved.This single incident has raised many of the culturallydistinct conceptions of gender and interpersonal relations.The American perspective was represented by twodifferent approaches: the rape counseling center and thecourt system. The center is oriented toward treatingrape victims as individuals through one-to-one and peercounseling in groups. The woman is seen as victim andthe damage understood to be her emotional and psychologicaldisruption. Healing occurs through the support ofother women who think women should be independent, incontrol of their own destiny and able to move aroundfreely. With minors, this counseling includes work withthe parents. It is the decision of the rape victim todetermine the extent of the counseling to be provided.Because Joua was a minor, the court automaticallyentered a charge of first degree sexual assault, the premisebeing that a minor cannot consent to sex. The courtsystem also views assault from an individualistic perspective.Though it will bring charges for a minor, theburden of proof still rests with the victim. Though attemptingto be culturally sensitive, the court must operatewithin its mandate designated by local law. The assailantis free until proven guilty. Testimony relies onaccuracy of translation. Family and Hmong community concernsmust be excluded until a verdict is reached.In contrast, the Hmong view rape as a family andcommunity problem. The girl is the site of the trouble;she symbolizes her family position and honor. Althoughconcern about and love for the girl were evident throughout,family actions centered on kin group balances, reestablishmentof which would also heal the victim. Littleattention is given to the immediate personal or psychologicalneeds of the woman. Once she is past puberty,the girl is no longer a minor by Hmong standards and isconsidered responsible for her actions. Concern is pri-147

142 THE HMONG IN TRANSITIONmarily with kin group honor, symbolic financial settlementand male public control of individual behavior inthe community. Although women provide emotional supportfor each other, they are not in authority positions withinthe Hmong social structure. Decision-making and actionsare controlled by groups of men. Individual identityis formed as a member of a specific family group,with females' positions formally represented by the malehead of their households. Intervention of the police andlaw is viewed as interference that can impede justice.The case also illustrates the clash between externalpublic and internal private means of settlement. Americancourt systems operate in a public sphere, whereas theHmong turn to private resolution between kin groups beforeresorting to public assistance from a council ofelders. The American court isolates individuals and pitsthem against each other in public, instead of allowingfamily units to negotiate appropriate settlement and punishmentin private. In the process, not only does theAmerican system isolate individuals and bring outsiders(police, lawyers, counselors) into a dispute between kingroups, it also violates gender roles as they divide betweenthe public and private domains. Hmong women's networkshave traditionally operated in the private sphereof action and influence, but the American court demandsthat Hmong women take individualized public action.ACKNOWLEDGEMENTSI wish to thank the Hmong who permitted and encouragedthis discussion of a personal, painful event.Thanks also to Susan Bo lyard Millar for her comments onan earlier version of the paper.REFERENCESBarth, Fredrik, ed. 1969. "Introduction". In EthnicGroups and Boundaries, pp. 9-38. Boston: Little,Brown and Company.148

RESOLVING SEXUAL ASSAULT 143Clark, Margaret, S. Kaufman, and R. Pierce. 1976. "Explorationsof Acculturation: Toward a Model of EthnicIdentity." Human Organization 35(3): 231-258.De Vos, George, and Lola Romanucci-Ross, eds. 1982. Introduction.In Ethnic Identity: Cultural Continuitiesand Change. Chicago: The University of ChicagoPress.Melville, Margarita B., ed. 1980. "Selective Acculturationof Female Mexican Migrants." In Twice aMinority. St. Louis: The C. V. Mosby Company.1 4 J

THE HMONG IN ISLA VISTA:OBSTACLES AND ENHANCEMENTS TO ADJUSTMENTCatherine Stoumpos GrossINTRODUCTIONWithin the college community of Isla Vista,nia, Califor-exists an enclave of approximately eight hundred Indochineserefugees.' Comprising half of this group arethe Hmong of Xieng Khouang Province in northern Laos. Anumber of articles about the Isla Vista Hmong havepeared in ap-both the local and national press, almostvariablyin-referring to them as an illiterate, animistic,independent mountain tribal society, whose men werestandingout-guerrilla fighters and whose people in generalhad never seen an electric switch or door lock, oraliftedpencil prior to their arriv ere. Other than thispress image, little is known ch.. 4t the Hmong, even bytheir Isla Vista neighbors.The information in this paper arises from an analysisof the findings oxof an ethnographic study of the HmongIsla Vista, which was undertaken in 1982 (see Gross1983). One purpose of this study was to recordthe Isla some ofVista Hmong people's recollectionscific of their spe-historical and cultural background. The studyattemptedalsoto investigate to what degree thesetraits have changed culturalor continued since flightand tofrom Laos,excl.me the effects of cultural changeyouth,on Hmongadults and the elderly, and on familyships.relation-METHODOLOGYFocused, open-ended interviews were conducted betweenJanuary and October 1982, of eleven EnglishspeakingHmong representing approximately 13 percent ofthe Isla Vista Hmong families and 20 to 25 percent of thehouseholds. Informants varied in age (from sixteen tothirty-eight), sex, education, marital and parental status(see AppendiA), and included two community leaders.While no elderly were interviewed, most informants145150

146 THE HMONG IN TRANSITIONreferred repeatedly to the problems and feelings of theelderly. The interviewer posed more than fifty questionsreferring to specfic aspects of life, including housing,employment, diet, dress, daily life, leisure activities,education, ceremonies, and religious and traditionalpractices, both as they existed in northern Laos and asthey occur presently. Interviews also explored thedegree of interaction with Americans and others, specificadvantages and difficulties in adjusting to life in theUnited States, Hmong thoughts regarding actions Americanscould take to ease their difficulties, and individual andcommunity mechanisms for adjustment and accommodation.Also investigated were changes in parent-child relationshipsas children assume some roles normally reserved foradults, such as translating for elders and representingthe family in dealings with the larger community. Amongthose consulted were social service and welfare workers,high school and English as a Second Language (ESL)teachers aid administrators, and volunteer organizationrepresentatives concerned with the Indochinese community.What emerged from the data is an illumination ofsome of the obstacles which inhibit the adjustment of theHmong, as well as an illumination of the factorsparticularlythose in the Isla Vista environmentwhich enhanceadjustment and encourage greater interaction withAmerican society.OBSTACLES TO ADJUSTMENTThe Hmong in Isla Vista feel that their greatestbarrier to adjustment is language; they are certain thatmastery of the English language will lead to better jobs,higher income and improved housing. Language, however,is only one of the obstacles facing the Hmong. Withoutsome awareness of the anxieties and apprehensions thatbesiege the Hmong, one cannot begin to understand theirdifficulties in adjustment. Almost everything that alarmsthe Hmong can be considered an obstacle to their adjustment.A chief anxiety of the elderly is their concern overdifferences in burial practices. Most elderly Hmong inLaos traveled from mountain to mountain, carefully selectinga burial site "high on the slopes and with a good151.

THE HMONG IN ISLA VISTA 147view." The bodies were never buried side by side. Thesubject of burial is of considerable interest toHmong.theWhen a local mortuary representative came to talkto the community, he drew the largest turnout ofVista HmootIslaever seen up to that time. Despite his explanations,,.onfv4ton and concern continue. One informant,for example, asked me if the grave location is nowdecided by the church, and many old people are reportedto "feel sad" at the prospect of adjacent burials, aswell as the possibility of bodies being placed on top ofone another and of bodies disinterred to make room forothers. In addition, while it is permissible to cut aperson when alive, it is objectionable to do so afterdeath. This objection apparently refers to both embalmingpractices and autopsies.The elderly are aware of their high vulnerability tocrime. Fearing robbery and rape, they will not go outalone. Crime was much lower in Laos and, in any case,informants report, the elderly feel unable to call forhelp in the English language. Many are not eligible forESL instruction, which is officially limited to employableadults between ages eighteen and fifty-five. Thefew elderly receiving such instruction are findinggreater difficulty than other age groups in learning andin retaining vhat they learn,Still another area of concern for the elderly is theimpending changes in timong culture, which they see as adverselyaffecting c,cisting family structure. Among otherthey fear an escalated divorce rate as the re-concerns,sult of increased education and freedom for Hmong females.Such apprehensions are not without foundation.Some student teenage brides are finding it difficult tomanipulate both school homework and duties traditionallyrelegated to the youngest married woman in a household.In addition to learning about alternative lifestyles insociology classes, these brides often receive sympatheticmoral support from classmates. A few such teenage bridesare leaving their husbands and returning to their families.These areas are more a matter of concern for theelderly than for younger people. Nevertheless, in viewof the continuing strong tradition of respect and concernfor the well-being of elders, the unhappiness of theelderly affects the entire community. In fact, every152

148 THE HMONG IN TRANSITIONperson who answered my questions about what Americans cando to help more, replied that we should help the oldpeople to learn English or to find a job. The elderlyneed to be useful, the informants said.Isolation brought on by cultural and language differencesis a principal factor inhibiting adjustment.which IDespite the advantages of such a support system,will address titer, the cohesiveness of the Hmong :ummunityeffectively retards acquisition of the Englishlanguage and limits interaction with the larger society.High on the list of obstacles are economic difficulties.With welfare assistance essentially phased out,many Hmong are having trouble supporting large familiesand finding housing they can afford on their meager salaries.The little-known requirement to repay the IntergovernmentalCommittee on Migration (ICM), for transportationcosts from Thailand to the United States, imposesone further economic and psychological burden upon theHmong. Families receive a bill each month for the entireamount, and have been advised to send a monthly paymentto "show their good faith," according to an IndochineseCommunity Project (IC P) representative.Hmong sensitivity to misrepresentations and inaccuraciesabout then in the press also serves to inhibit interactionwith the outside community. Contrary to themyth perpetuated in the press, some Isla Vista Hmong didattend school in Laos and are literate in Laotian. A fewinformants are also literate in French and one informantreads Chinese. Most Isla Vista Hmong speak Lao and someThai in addition to two Hmong dialects. Neither is thepersistent report that the Hmong have no written languageaccurate, since the Hmong language was endowed with aRomanized script by missionaries in Indochina. Thisscript is being read in some of the Hmong households inIsla Vista. A Hmong elder concerned with 1.i,. preservationof Hmong culture and language has been :onductingreading classes for mothers in this script.Not only is such misinformation and underestimationdangerous in terms of the direction of official policy,but such irresponsible reporting further inhibits interactionwith the outside community. For example, a leaderwho was at first very responsive avoided a further interviewwith me, mer.tioning the negative press about theHmong. This occurred shortly after the publication of153

THE HMONG IN ISLA VISTA 149one such poorly researched article in the Los AngelesTimes (3apenga 1/82).The desire of some Hmong to return to Laos appearsto reflect a combination of anxiety and loneliness forthose left behind and a sense of despair over the seeminglyinsurmountable obstacles to successful survival inthe new society. Reunification with family and friendswas the most common reason given for the desire to return.One informant cited the problems of the elderly asanother reason.While a number of refugees hope to return to Laossomeday, many fear being sent back by a president who mayat any time decide that Laos is now safe and cancel theirrefugee status. The abrupt reduction and termination ofwelfare benefitsperhaps seen as reflecting a change inpolicy toward refugees -- probably caused anxiety thatreached beyond monetary concerns. Various cases of Hmongreluctance to call upon the police may stem from similarunderlying apprehensions. In one instance, an informantreceived three visits from an intoxicated male who usedthe family shower, made long-distance calls from its telephone,and gave orders to family members. Yet neitherthe informant nor anyone in her family tried to contactthe police. After the third such visit, family memberscalled a leader, also a relative, who contacted theapartment manager, who then notified the police.An unwillingness by most Isla Vista Hmong to rememberformer non-Christian religious practices may be seenas yet another manifestation of what they feel is theirtenuous position in this society. Even recent convertsto Christianity appear sincerely not to remember anyformer traditional religious practices. A question thatmust be raised, therefore, is to what extent the conceptof Christianity is considered by the Hmong in a nonreligiouscontext; that is, how many convert as a matter offaith and how many because they believe it's a necessarystep in adjustment? Additionally, the danger exists thatthis reluctance to remember may adversely affect the oraltransmission of cultural information and folk history tooffspring and, hence, to succeeding generations.Also revealed in the interviews is the custom ofconferring a different, or "old" name upon a male when hebecomes a father for the first time or soon after histwenty-fifth birthday. His original first name is con-1 5 4

150 THE HMONG IN TRANSITIONsidered his child name, which is usually erased thereafter.His mature name, given to him by his father-inlaw,becomes his given name. Most Hmong do not legallychange their names, both because of the difficultiesinvolved and the confusion it might create in the workplace,at school and elsewhere. More and more men,therefore, are known by one first name within the Hmongcommunity and by anothertheir child namein the largersociety. While some men are reported to feel badly aboutthis, others says that it is "no problem." Nevertheless,the difficulties encountered in retaining this customforge another link in the chain of obstacles separatingHmong and Americans.In addition to concerns expressed in interviews andthrough other sources, the Hmong in the United States undoubtedlyexperience anxieties which they can never discusswith outsiders, ranging from horrifying war experiencesto the forbidden existence of a second wife who hasmanaged to enter the United States. Such wives and theirchildren are obliged to live a lie in their dealings withthe American community throughout their lifetimes. Theypossess one identity within the Hmong community andanother outside of it.Finally, family heads have experienced the loss ofsome authority, prestige, self-esteem and self-confidenceas the result of forced dependence on outside support andof shifting roles within the family. Being corrected orcriticized by offspring, however gently, is a source ofembarrassment and sadness. One informant reports a suicideby a Hmong father in another community as the resultof such loss of authority. It is to be expected that,especially among older people, some depression exists asa result of these factors.FACTORS ENHANCING ADJUSTMENTOne factor easing the Hmong through the adjustmentpr -cess is a number of similarities between Hmong andAmericans in important cultural valuesamong these thehard work ethic, an emphasis on the importance of education,a strong desire for upward mobility and economicadvancement, an independent spirit, high self-esteem andan emphasis on money. These values are held for somewhat155

THE HMONG IN ISLA VISTA 151different reasons. While Americans value the materialaspects that money can provide, the Hmong have traditionallyused moneyin the form of both bride-price andfinesto solidify marriages and maintain social order.Further, the historic Hmong qualities ofity,flexibil-adaptability and resourcefulness, which were amplyexhibited throughout this study, have undoubtedly helpedthe Hmong cope with their new surroundings. Flexibilityhas been the keynote in most facets of Hmong life, fromdress and division of labor to courtship, marriage andnaming practices. Change and adaptation to different environmentsand peoples are not new to the Hmong but havelong been a part of their way of life. Secondary andtertiary migrations within the United States to improvehousing and income may not be seen by the Hmong as particularlydisruptive, in view of their historical traditionof migration. It should be remembered that theHmong have not come from a point of stability directly tothe United States, but have endured warfare, upheaval andfamily separations since about 1960.One example of Hmong resourcefulness that wasbrought out in the study is the organization in theUnited States of neighborhood or clan lending pools, towhich each member contributes a small amount monthly andfrom which he is able in return to draw out large amountsin order to purchase expensive items, such as a refrigeratoror car. No interest is charged, and the borrowerrepays when he can. In Isla Vista, however, very fewHmong belong to such an .Jrganization.Expectations of the welcome and assistancereceived in to beAmerica appear to be low. This is predictable,considering that, even in good times, the Hmonghave been tolerated rather than welcomed in the areasthrough which they have migrated. Historically, minoritystatus has been their lot, so their present situation isnot so very unfamiliar.Although the Hmong people's organization as a communitycould be considered a disadvantage in terms of theextent to which it retards acculturation, the presence ofthe Hmong community provides a substantive vein of continuityand stability during a period of great disruptionand change, and probably ranks as the foremost factor enhancingthe Hmong people's adjustment. This finding issupported by other studies and surveys which suggest that1.50

152 THE HMONG IN TRANSITIONIndochinese refugees be settled only in clusters (Trillin1980; Bureau of Research 1979).Despite some shifts in family power relationships, adegree of filial rebellion is expected, and parents areconfident that they remain in control of their children.Young informants report a continued desire and obligationto care for parents "until the end of their days."Finally, the socioeconomic climate of Isla Vistaitself appears influential in easing adjustment and promotingfriendly relations between Hmong and Americans.Competition for factory employment is minimal; neitherthe student community nor the adjacent affluent communityof Santa Barbara seeks factory employment. As a result,75 percent of the employable Indochinese in the area areemployed. This compares favorably with larger urban communities,such as the Twin Cities area, where only 20percent of Hmong adults are employed (Southeast AsianRefugee Studies Newsletter 1983). Competition for housingis limited to university students, who realize theyhave no permanent stake in the community and generallyreact with indifference to the Indochinese presence. TheHmong in Isla Vista, therefore, are not subjected to thedegree or intensity of hostility from other segments ofthe community that is being experienced by Hmong in somelarger, problem-ridden communities such as Long Beach andMinneapolis (Calderon 1982; Kershaw 1982).On the contrary, in their contacts with Americans,the Isla Vista Hmong are exposed primarily to sentimentsof concern, sympathy, admiration and friendliness. Thesecome from teachers, welfare workers, students and otherAmerican community volunteers.In addition to ESL and vocational programs, a numberof other forms of American community assistance are beingextended to refugees. An American woman organized a"Mama's Class," at which children are welcome, for thoserefugees who were unable to attend regular ESL classesbecause of child-care problems. Most of the students areHmong; their numbers 'include elderly women ineligible forregular ESL classes. This informal Mama's Class, taughtfive days a week by volunteers, emphasizes practicalEnglish for home management. A university student organizedstudent volunteer tutors in a program funded by theuniversity. The Isla Vista Youth Project, frequented byHmong and other refugee children, is also staffed by157

THE HMONG IN ISLA VISTA 153volunteers and provides tutorial services as well as anafter-school activities center for children of workir3parents. The Indochinese have found such private lessonshelpful, and there are not enough volunteers to providethe hours of tutoring sought by the Hmong. Some Americanwomen have attempted to initiate a cottage industry bycombining Hmong embroidery and Western dress, and twoother American women are active in exhibiting and sellingHmong handicrafts at local fairs and craft shows. A smallplot in community gardens has been made available to eachHmong family in Isla Vista.A nutritional program called the "Let Isla Vista Eat(LIVE)" Project was initiated recently. It provides freecold breakfasts five mornings a week and distributes bulkfoods three mornings a week. A large percentage of thosetaking advantage of the breakfasts and food distributionare Hmong children and Hmong women, respectively.Recently, the school district jointly sponsored aworkshop for Indochinese parents, emphasizing (1) differencesbetween Indochinese and American educational systemsand the roles of parents in each, (2) what childrenlearn and how parents can assist and encourage them, and(3) career awareness. Educators stimulated a dialogue tohelp themselves discover what particular questions andfrustrations these parents might have. While anothersizable Indochinese group was absent, the Hmong attendedthe workshop in large numbers.The favorable employment rate is not the only indicationthat the Isla Vista Hrnong have had a comparativelyeasier adjustment experience than Hmong resettled inother communities. While Hmong mutual assistance associationshave sprung up around the country, Isla VistaHmong seem to have found it unnecessary to form or belongto one. Neither is there significant nembership in clanassociations, as there is in other Hmong communities. By1982, at least seven Hmong out of this small communityhad entered college. This compares favorably with othercommunities, such as the one at Long Beach, which hadapproximately ten times the Hmong population but stillhad no Hmong students in college.Another likely indication is the increasing degreeof Hmong participation in community activities. For example,in 1982, Hmong men and women in traditional dressparticipated for the first time in Santa Barbara's Summer158

154 THE HMONG IN TRANSITIONSolstice parade. No other Indochinese group was represented.In addition, Hmong handicrafts are displayed andsold from a booth operated by Hmong women at numerouslocal art and craft exhibits.It is not the intention of this paper to minimizethe overwhelming obstacles faced by the Hmong in IslaVista or to imply that the forms of external assistancedescribed above are in any way sufficient to meet theeconomic needs of the Hmong in this high-cost area.However, the obvious desire of Americans in this communityto help, and the expressions of warmth and welcomewhich are transmitted by Americans with whom the Hmongare in direct contact, are undoubtedly instrumental ineasing the adjustment difficulties of the Isla VistaHmong.OUTLOOKThe Hmong are survivors. While they miss theirformer way of life even more than their homeland, mostfeel their new home is the United States, and they arestriving to accommodate and adapt to American society asquickly as possible. They are least concerned abouttheir children, who seem to have no adjustment problems.Neither do the Hmong seem concerned with retention ofauthority over their children, in spite of the way relationshipswithin the family have shifted as childrenlearn English more rapidly than their parents. Nevertheless,change in the desira1:1 age for marriage and inexisting extended family structtre can be predicted.While their resourcefulness, flexibility and adaptabilitysupport them in this latest challenge for survival,the Hmong people's internal community network isunable to dispel negative rumors or allay fears and misunderstandingswhich arise concerning American motives,values and bureaucratic regulations. The rigidity andimpersonality that now characterize management of therefugee problem are in need of re-evaluation. Clearly, ameaningful, continuing dialogue with the Hmong, on botha community and governmental level, is needed to dealwith sometimes easily solved but frustrating issues whichplague the Hmong. The recent education workshop forIndochinese parents was a step in the right direction.L159-Isommommillmimimminimmil

THE HMONG IN ISLA VISTA155Other such dialogues should be implemented.representativesPolicecould address the Indochineseto explain thecommunitypeople's rights as well astions.their obliga-Another discussion might investigatebilitytheoffeasi-changing Hmong men's first nameslengthy legal proceedings,withoutperhaps by changing themcollectiveon abasis each year. If small garden plots weremade available to refugees in other communitiesare inas theyIsla Vista, it would provide a degree of selfrelianceby means of a familiar form of labor and wouldbring a small measure of economic assistance.tinuingSuch con-efforts can ease the adjustment of theand,refugeesin the process, serve to reduce the social distancebetween Americans and Hmong.ACKNOWLEDGEMENTSThis study was supported by a President's UndergraduateFellowship from the University of California, SantaBarbara. I am grateful to theand support: following for their adviceDr. Ruth L. Moore andthe University of Debreana McGrath ofCalifornia, Santa Barbara, and R. GreetKershaw of the University of Long Beach. I am alsodebted toin-the Hmong people of Isla Vista who graciouslyreceived me into their homes or came to mine, and toleratedthe multitude of questions.The ethnographic study on which this p.i, er is basedwas conducted in the Isla Vista community between Januaryand October, 1982. While some information has beencluded or in-clarified as the result of continued researchthrough October 1983, the ethnographic present of thispaper remains the time within the original study period.NOTE1Refugee population figures were provided by an IndochineseCommunity Project (ICP) representative during thestudy period and supported by a Hmong leader, also connectedwith ICP at the time. These figures, I discoveredrecently, were underestimated to the public in anattempt to minimize the Indochinese presence in thecommunityan example of fear once again, the time on160

156 THE HMONG IN TRANSITIONthe part of non - Hmong personnel concerned with the refugeecommunity. This underestimation resulted in reducedrefugee assistance efforts. A year-end 1983 estimate isapproximately one thousand Hmong and Laotian refugeescombined in the Isla Vista area. This higher figurealso reflects the recent one-way flow of Hmong refugeesinto the area. Only one Hmong family has left IslaVista since April 1983.REFERENCESBureau of Research and Training. 1979. National MentalHealth Needs Assessment of Indochinese Refugee Populations.Prepared by Office of Mental Health, PennsylvaniaDepartment of Public Welfare. July 1979.Calderon, Eddie. 1982. "The Impact of Indochinese Resettlementon the Phillips and Elliot Park Neighborhoodsin South Minneapolis." In The Hmong in the West: Observationsand Reports, Papers from the 1981 HmongResearch Conference, edited by Bruce T. Downing andDouglas P. Olney, pp. 367-386. Minneapolis: Centerfor Urban and Regional Affairs, University ofM inn esota.Gross, Catherine Stoumpos. 1983. "Transition in Perspective:A Study of the Hmong of Isla Vista." DISCOVERYUCSB Journal of Undergraduate Research, Spring 1983.University of California, Santa Barbara. Vol. 6, pp.1-34.Japenga, A. 1982. "A Meeting of Young Minds in 24 Languages."Los Angeles Times, Part V. 13 October 1982.Kershaw, Greet. 1982. "Hmong Families in Long Beach1978-1982: Between Incorporation and Rejection."Paper at Southwestern Anthropological AssociationConference Meetings. Sacramento. April 1982.SARS (Southeast Asian Refugee Studies) Newsletter. 1983.Minneapolis: Southeast Asian Refugee Studies Project,University of Minnesota. January 1983, Vol. 3, 2,P. 1.161

THE HMONG IN ISLA VISTA 157Trillin, Calvin. 1980. "U.S. Journal: Fairfield, Iowa.Resettling the Yangs." The New Yorker. 24 March1980, pp. 83-100.162

APPENDIXINFORMANT DATAN ISLA VISTAIN LAOSYrs. inRefugeeCampArrivedin U.S.Approx.AgeSexMaritalStatusNo. ofChildrenNo. ofBed-roomsNo. ofFamiliesin HHNo. inHouseholdNo. inHouseholdNo. ofBedrooms*Subtribeof Familyor Couple3 1976 24 F Widowed I 5 2 2 5 2 Unknown4 1978 35 M Married 5 IT I vs," I a Blue3-4 1978 33 M Married 4 14** 2** 2 5 3 Blue4 1978 24 M Married 1 A I mill 3 Blue/White3 1978 38 F Married 7 9 I 3 15 4 White3-4 1978 32 F Married 4 14" 2** 2 5 3 Black/Blue2 months 1976 36 M Married 2 a 3 4 12 4 Blue /White3 1978 16 M Married 0 10 2 4 9 5 or 6 Blue/White3 . 23 1978 19 F Single 0 6 I baths 6 I WhiteI 1976 25 M Married 2 9 2 3 22 6 Blue3 . 23 1978 32 F Married 4 II 2 baths 6 3 Blue/WhiteIn addition, the living room was often used as a bedroom for single family members.* Two members of same household inadvertently interviewed. Household size normally 6, but swelled temporarily with migrationof 8 relatives to Isla Vista from another state.Information not obtained. 163NOTE: Two informants, both leaders, live outside of the physical boundaries of Isle Vista.BEST COPY AVAILABLE

FACTORS CONTRIBUTING TO A SPLIT WITHIN ACLIENTELISTIC NEEDLEWORK COOPERATIVEENGAGED IN REFUGEE RESETTLEMENTNancy D. DonnellyRefugees who resettle in countries very differentfrom their land of origin face difficult problems of adjustment.Not only must they cope with physical illnessattributable to changes in die* and climate, and withpainful memories of loss, but they may also be expectedto find a productive niche in a society whose economy,political structure, language and customs of social interactionare unfamiliar to them. Researchers who havestudied refugee responses to their new situations oftenhave drawn on concepts from the literature of psychologicaladjustment. For instance, Bar-Yosef (1968) describesa process that she calls "dematuration" undergone by immigrants(some of them refugees) to Israel; even in Israel,where newcomers were very welcome, immigrantsfloundered for a time in their new setting because theyhad lost the underpinnings of their former identitytheir jobs, material possessions and familiar round ofdaily activities. Bar-Yosef presents a model in whichthe Israeli bureaucracy assumes a parental role, whilethe immigrants perforce occupy the status of childrenuntil they remature by one means or another.For Hmong in the United States, Bar-Yosef's modelmay seem to describe a situation analogous to their own,arriving as sponsored refugees only to find a societythey did not expect, so that they must depend on othersto teach them how to live. This paper seeks to understandthe process by which some Hmong have tried to reassumecontrol of their own lives. Thus it describes aparticular instance of what Bar-Yosef terms a "copingmechanism" (1968:27). Instead of drawing on the conceptsof the psychological literature, however, the process isdescribed here in terms of patron/client interactions.The patron/client model permits delineation of the dialecticalnature of resettlement; the actual situation describedin the paper is presented in a simplified and di-

160 THE HMONG IN TRANSITIONagrammatic manner, in order to clarify relationships thatmay have been experienced by many Hmong who have been resettledin the United States.BACKGROUNDIn 1980, Hmong and Mien refugee women in an Americancity asked their American teachers to help them selltheir needlework. The American women began to arrangeentry into street fairs and other markets, eventually organizinga nonprofit cooperative with themselves as officers,the many refugee women as members, and other Americanvolunteers as a sort of auxiliary. In fall 1981 theyincorporated as Asian Needlecrafters (AN).1 Soon theymoved into a small shop (Harris 1982). By the fall of1982, more than 100 women had sold o ndau and otherhandmade needlework through AN at one time or another.Sales in 1982 exceeded $40,000.When the store opened in the spring cf 1982, therewas rhetoric about putting the refugee women in charge ofsales, and in articles and public statements the refugeemembers were clearly implied to be running the operation.Officially, AN was under the control of a board which includedrefugee women. But in fact, AN continued throughthe summer of 1982 to be run by a core of American volunteers,who made such decisions as: (I) whether to acceptparticular pieces made by refugees, (2) how large an inventorytd retain undisplayed, (3) which fairs to enter,(4) which pieces to take to what fair, (5) what materialsand thread to order, '4) how to display merchandise, (7)how the shop would 1._: staffed, (8) what hours the shopwould be open, (9) what sorts of new products to try,(10) whether to attempt wholesaling, (11) what kinds ofpublicity to seek, (12) how to set up the bookkeeping,(13) what sort of receipts, price tags and brochures touse, and (14) what nonprofit grant funding to pursue.2The American volunteers decided that the general goals ofthe organization should be: "cultural preservation andsupplemental income" (E. Thomas 1982). To facilitatethese two goals they would engage in "production and saleof handcrafted art and needlework by Indochinese refugeesliving in [this city]" (Hafey 1982; emphasis mine).This position, that all the work was done locally, dif-165

'166NEEDLEWORK COOPERATIVE 161ferentiated AN from other groups that, frankly, importedneedlework. It appealed to buyers with charitable interests,permitted higher prices, and was also connected tothe goal of cultural preservation.The American managers made an effort to promote thesense that AN was a cooperative venture, that everyonewas helping in her own way to make a success out of it,and simultaneously, that everyone was helping each other.The American women made home visits and redistributedused clothing. Relationships between the American womenand the refugees were characterized by amity. Personalincome generated by AN's sales accrued only to the refugeewomen.Hmong members of Asian Needlecrafters were also makingsome decisions during the summer of 1982. These included(1) whether or not to sell through AN, (2) colorand style of the products they made (out of a range ofpossibilities suggested by the Americans), (3) how manyhours to spend sewing at home, (4) when to "donate" twohalf-days per month in "volunteer" work at the store(this was required), and (5) what price they wanted fortheir work (Americans often recommended that they raisetheir prices).If these lists are compared, it is clear that therewas some discrepancy between the public stance of AN asan enterprise operated by refugees and the actual operationalstructure. Notably, decisions made by Americanwomen affected the overall operation and deeply affectedthe income levels of individual refugee members, whiledecisions made by refugee women affected themselves almostexclusively. Refugees selling through AN were basicallyin the position of pieceworkers producing on consignment.This said, it is important to add that one decisionduring this period had been made jointly by Americanand refugee women. The amount of the store markup(20 percent) was the outcome of negotiation. Refugeeswould have preferred no markup at all, and Americans beganby suggesting a range from 10 percent to 40 percent.In discussion the comrromise figure was finally reached.The negotiation over marklir demonstrates what everyonewho has worked with nonliterate refugee women alreadyunderstands: there were compelling reasons for the divisionof responsibility that existed in AN during 1982,since the refugee women did not know ordinary American

162 THE HMONG IN TRANSITIONbusiness practices. They lacked skill in English, aswell as in salesmanship and in knowledge of their market.They were unable to plan for their own futures because oftheir inexperience with life in the United States, andthey could not have continued selling their needlework,especially through a retail outlet, without the help ofAmerican women.There were also other reasons for the division ofresponsibility stemming from the personal goals of theAmerican volunteers. By the summer of 1982, volunteersconsisted almost entirely of committed Christian women.None of them had had prior business experience. Theywere oriented instead toward service. Perhaps they weregiving the refugee women religious instruction. (This wasnot discussed with me, however.) Besides the novel experiencethey were gaining in running a business, theAmerican women developed friendships among themselves andreceived positive attention from members of the localcommunity working in refugee resettlement and from thelocal media, They were happy to receive the expressedappreciation and good will of the Hmong and ('ien womenand their families, and they felt they achie.ed a senseof greater closeness to God. They felt that Asian Needlecrafterswas what I will call a nursery of democracy,and that they were demonstrating Christian love. Sinceall these goals were accomplished by the status quo ofAsian Needlecrafters, the American volunteerswhe remainedin control of AN's structurehad little incentiveto push actively for greater autonomy among their refugeemembers.PATRON/CLIENT RELATIONSHIPSThe relationships outlined above are very similar toclassic patron/client relationships, as described in theanthropological literature. Much has been written aboutpatronage structures (see Eisenstadt and Roniger 1980 fora thorough review). Most anthropological work has concentratedon Mediterranean and Latin American societies,but recently there has been increasing recognition ofdientelism as part of the formal 'governmental structuresin Southeast Asian societies (Hanks 1977). Informal patronagenetworks have been noted within Euro-American so-167

NEEDLEWORK COOPERATIVE 163ciety; for instance, inside corporate structures (seeLande 1977). Pptron/client relationships have also beennoticed growing up between people from different societiesthat have economic interaction and great disparitiesin wealth. In this case, individuals called "brokers"may find profit in mediating between the two, when thereare communication difficulties between them (Hendriksen1971).The schematic and somewhat simplified drawings inFigure 1 will help clarify patron /client ties. "Patrons"are defined as persons in possession of material resourcesor favors, who distribute them to other people inreturn for loyalty and praise (and ultimately for profit,since a sizable following generally facilitates the aggregationof resources in various ways). "Clients," whohave fewer material resources, nevertheless can offer thepatron an intangible goodloyaltyneeded by the patronto continue in this superordinate position. Clients gaindirectly from the patron through the patron's disbursementof goods or favors. Thus both patron and client provideeach other with something of value which is lessuseful to them than what the other can provide to them.Friendship (or at least cordiality) is an importantelement of patron /client relationships. Still, it is easyto distinguish these relationships from true friendship,and to distinguish the patron from the client: whichgoods are transferred and when they are transferred, isin the hands of only one party, the patron. The patrontakes a position of leadership, while the client acceptsa position of followership. These are actually relationsof exchange. In Lande's (1977) view, the counters of loyaltyhave less value than the counters of material benefits,so that the clients are at a disadvantage, and haveto even the score by giving the patron an added incrementof obedience when necessary to the patron.1G3

164 THE HMONG IN TRANSITIONFigure 1EXCHANGE MODEL IN PATRON/CLIENT SITUATIONmaterial goods, accessto advancement, favorsAM/The Patron's FundPATRONr--1 \ICLIENT/I CLI CLIENTIloyalty, praise, support(obedience)STRONG VERTICALASSOCIATION, AMITYweak horizontal tiesbetween clientsThe Client's Fund(the increment)EXCHANGE MODEL IN PATRON/BROKER/CLIENT SITUATIONPATRONBROKERtCLIENT CLIENT CLIENTBroker commonly translates,interprets and/ormanipulates the intention,meaning or implicationof the Patron's largesseto the Clients andthe Clients' responses tothe Patron. Usually hasown personal goals. Oftenretains some goodsand some support/obediencefor own use.Brokerage can commonly arise when a Patron cannot or willnot contact Clients directly because of physical orsocial distance.ml=1... ml... Ob...---1 cd

NEEDLEWORK COOPERATIVE 165Where a broker intervenes between patron and client,there is probably a social or physical distance betweenthe patron and the clients. But a broker is not simply atransmitter of social and economic goods. Very likely, abroker has his own agenda and goals that are facilitatedby this intermediary role, and the broker commonly retainsfor his own use some of the goods that pass throughhim from patron to client, as well as some of the loyaltyoffered by the clients. Nontangible goods like loyaltyand demands for obedience are likely to be reinterpretedby a broker to suit his own purposes (see Paine 1971).Hmong society in Laos did not include the clientelismseen in Southeast Asian societies with more verticallevels (for instance, the Thai monarchy; see Hanks 1977).This is probably because each extended family (sublineage)formed an independent economic unit that could fillits own needs by farming or by trading, using connectionsformed along family lines (Lemoine 1972:158-9). Thoughthese family networks might depend on one central familymember for success, they could not be considered clientelessurrounding a patron, because each member householdwas self-sufficient. They did not display the chronicneediness and insufficiency associated with dependencythat seem to be prerequisite for the emergence of patron/client structures.IMPLICATIONSBut obviously most refugees enter the United Statesin a condition, however temporary, of dependency. Clientelismis a fact in the heart and essence of refugee resettlement.Clearly the largest patron is the federalgovernment, and there are doubtless many smaller clientelisticnetworks. I don't think clientelism in a settingof resettlement is a negative thing, because I don'tknow how else resettlement could be pursued. How to bringthese relationships to a close, however, may not be readilyapparent, especially to the patron and the broker,whose social power depends on the presence of a clientele,and who may like the rewards of being a patron (orbroker).In the case of Asian Needlecrafters, I would say theAmerican women were unaware of any difference between mu-17u

166 THE HMONG IN TRANSITIONtuality of enterprise and the operation of their business.The Mien members' attitudes are not known to me.The Hmong members to whom I spoke in the summer of 1982seemed to be exclusively interested in the organization'spotential for providing income. To me, AN looked (structurallyspeaking) like Figure 2. Here, money rewardsmove downward while needlecrafts and "volunteer" timemove upward. Though the vertical associations betweenthe American volunteers and the refugee members werecharacterized by friendship, horizontal ties between refugeemembers were much weaker (another characteristic ofpatron/client arrangements), and charged with competitionand hostility, especially between some Hmong women andsome Mien women. Volunteers arranging fairs needed tocall a dozen oh more women in small networks. Once at afair, certain refugee women were likely to bury otherwork beneath their own, and hostile scenes occasionallyoccurred. I think it would not be entirely far-fetchedto add God as the ultimate Patron of AN. Without belaboringthe point, Christianity provides a parallel model,with believers in the position of clients, the pastor asbroker, and the exchanges taking place with counterscalled "Love" (Paine 1971:15).Despite these various viewpoints, AN might have continuedunchanged for a while, except for a crisis precipitatedby an outside force, the federal funding cuts ofMay 1982. Any refugee who had been inside the UnitedStates longer than eighteen months lost federal supportpayments at that time. Nearly all the Hmong members ofAN experienced great financial difficulty. This suddenloss of income was combined with a deepening recession,so that jobs were scarce. Families began to leave forstates with more generous welfare programs or for stateswith jobs. Suddenly the remaining Hmong women had totake farm labor and house-cleaning jobs. Suddenly, theynoticed the value of their time. Before the funding cut,an embroidered pant leg panel that took two months tocomplete might sell for $50 (Harris 1982). Afterward, awoman saw that she could not retrieve the value of herlabor.171

NEEDLEWORK COOPERATIVE167Figure 2STRUCTURE OF ASIAN NEEDLECRAFTERS,SUMMER 1982PurchasesCustomers, DonorsAmerican Managers &Volunteers)41 ir -,A + 1kIHmong and Mien Members IThe Patron's FundPatronsBrokers (20 percentsales retained;respect in communityretained)ClientsFriendliness; NeedlecraftsVolunteer TimeThe Clients' FundThe Increment-- -- -- -- -- -- -- --- -- --The Hmong women solved this problem by starting toimport the time-consuming centers of appliques and embroideriesfrom their relatives in refugee camps; theyexpected to remit part of the profit from the sale of apiece to the person in the camp. But they discovered thatAN would not accept these imports, for AN was differentiatingitself in the local market by selling only locallyproducedneedlework. The American women felt that theycould not accept foreign-made work because of their previouslyannounced goal of cultural preservation. TheHmong women could not get the American leadership toagree that cultural preservation was identical with earningmore money.Faced with this dilemma, some Hmong women quit.Others began to present the imported work as their own,perhaps adding a border first. Thus a conspiracy grew todeceive the American women. It seems that out of thisdispute about the origin of pieces, the Hmong women arrivedat the conviction that in spite of appearances andrhetoric, the Americans 'owned" Asian Needlecrafters andwere ignoring the needs of the Hmong women in favor ofthemselves.172

168 THE HMONG IN TRANSITIONSome became quite assertive, even offering to meetcustomers in private to avoid the 20 percent markup. Disputesarose as various refugee women accused the Americansof favoritism (these accusations came from Mien aswell as Hmong women). Nobody could get enough money tolive on, it seemed. Everyone appeared desperate.Mien members, in contrast to the Hmong, behaved in agenerally more pliant, noncompetitive and passive manner.Perhaps they perceived the Americans to be truly actingin their behalf. The treasurer of AN had sponsored severalMien families. The coordinator had learned the Mienlanguage. One central volunteer told me that "the Mienare so cute," and that "whatever the Hmong are, they'renot cute."When federal Impact Aid funds became available inAugust of 1982, AN successfully applied for a traininggrant that provided for twelve part-time clerk-traineesto staff the store and two part-time translator/organizers,one Hmong and one Mien. The primary candidate forthe Hmong position was a woman who had never been amember of AN and who did not produce needlework. She hadgood business skills, but her main qualification was herposition as wife of a Hmong leader. In October 1982, inan Asian Needlecrafters meeting attended by about sixtypeople, this leader (who had never come to an AN meetingbefore), announced that the Hmong women had formed a neworganization, the Hmong Artwork Association (HAA).3 Themembers of this new organization included nearly all theHmong women currently in AN; the president of the neworganization was the leader's own wife. Presenting himselfas spokesperson for the Hmong women, who were tooshy to speak for themselves, the leader said that theHmong women demanded as a bloc that his wife should behired as AN's Hmong translator/organizer.The American women of AN felt that the leader's wifewould try to sabotage the success of any other candidateso they hired her. But after this meeting they were convincedthat the leader, through his wife, intended totake control of AN, eject the Mien and convert AN into avehicle for reinforcing his own position in the localcommunity.In this assessment they were partly correct. InNovember 1982, the leader was indeed speculating on tacticsfor taking over AN. But the American women failed173

NEEDLEWORK COOPERA fIVE 169to consider how much his actions were permitted and invitedby the women themselves. They were unaware of theHmong women's anger regarding the AN policy on importedpieces (among other things), because the refugee women intheir clientage position had felt unable to express thisopenly. They were ignorant of the traditional cohesionof Hmong society, with its processes of consensus. Theysaw the leader as an independent force bearing down fromabove on the hapless and bedeviled Hmong women. They paidlittle attention to the fact that the leader's wife reliedon her own charisma and strength to unify the Hmongwomen, instead preferring to think of her as a mere agentof her husband. Having little experience with refugeewomen who were already competent to survive fairly independentlyin American society, the American women of ANunderestimated the ability of the leader's wife to supplantthem, in her role as broker, as the principal mediatorbetween Hmong women who wanted to sell needlecraftand American buyers. Above all, they could not see thatthe way into their organization was open to her becauseof their own policies, which no longer met the suddenlyincreased financial needs of the Hmong women, and whichthey would not alter to accommodate their members.The winter of 1983 is a story of struggle betweenthe American women and their new Hmong translator/organizer,full of jockeying and distrust. She appears in mynotes as a wedge driving an ever-widening gap of misunderstandingsbetween the Americans and the Hmong women.The grant specified that classes should be conducted inmath; soon the teacher realized that during translationthe translator was providing the answers. The Americanwomen refused to tell her most financial information relatingto the operation of the store, and they would notlet her distribute fabric in case she gave it away. Thescope of her job became a point of contentionshe calledherself a "management trainee" while they called her an"employee." She asserted that she should be taught tooperate the store and should take it over. It was indeedone of the grant's goals to turn operations over to therefugee women. But the American women did little to facilitatethis goal, and the point was not discussed at ANmeetings. Hostility between Hmong and Mien members increasedduring this period. Hmong complaints directedthrough the translator/organizer did not reach the Hmong174

170 THE HMONG IN TRANSITIONwomen.By February 1983, the rift between factions had becomeunbreachable to all the women. The catalyst for actionproved to be very trivial: a Hmong baby hat that theAmericans would not hang in the store because it had notbeen made locally. Early in March, amid acrimonious debate,almost all of the twenty-six actively selling Hmongwomen followed their translator/organizer out of AN intoan independent Hmong Artwork Association.Patron/client ties are by their nature rather fragile,according to Eisenstadt and Roniger (1980), sincethey are subject to constant reappraisal and may be terminatedby either party if found wanting, especially if amore promising situation is available. Wells (1979), forinstance, describes a Mexican-American community in amidwestern American city, which found a willing broker(Wells says "patron") in a local foundry's Mexican-Americanassistant personnel manager. This individual couldprovide lobs as long as workers were "uncomplaining." Ashappened in Asian Needlecrafters, it was during a conflict(over job-related injuries) that the clients eventuallyperceived that their broker did not always act intheir behalf, but instead had been using their loyalty asa basis for pursuing his own goal (incorporation intoAnglo-American society). This discovery, combined withhard economic times and lay-offs, increased strain betweenclients and broker, at the same time that new brokers(Chicano representatives of La Raza, an ethnic organization)appeared on the scene. In the following crisis,the foundry workers shifted allegiance to the new brokers,who more closely shared their own cultural goals,and who could also deal effectively with the larger non-Mexican community.The Hmong defection from AN obviously displays similardynamics. In both cases an easily marked minoritysought more sympathetic linkages between itself and thedominant society during a financial crisis. What couldbe (and was) described as a retreat into "enclavement"actually constituted a tactic to maximize effective interactionwith the dominant group. In neither case wereclients unaware of their options. In fact, whether theHmong leader's wife has used her followers more or lessthan they have used her to achieve their own ongoing goalof financial adjustment to American life, is an interest-175

NEEDLEWORK COOPERATIVE 171ing question that might be pursued. During its first ninemonths of operation, HAA produced an impressive total insales, of which nearly three-quarters went to memberswhile none went to HAA's manager (the Hmong leader'swife); she, however, retained her superordinate position.Many anthropologists writing on the topic of clientagehave noted that the short-term mater -I rewards maygo predominantly to the clients themselves, while the rewardsof patrons and brokers (which a foLowing of clientscannot provide but can only facilitate) may exist inprimarily disembodied form as social opportunity and influence.Patron /client relationships from the client'sviewpoint are worthless without immediate rewards, whilethe patron/broker may intend long-term gains that can beeasily frustrated if clients melt away during times ofslender returns. Thus the risk of such relationshipsseems to accrue primarily to the patron/broker, at leastwhere the clients have other options, as is demonstrablein the cases of Asian Needlecrafters and also the HmongArtwork Association, whose true "winners" so far are theclients. Thus the goals of refugee resettlement (improvementsin the social and economic position of refugees)have been successfully attained so far by bothorganizations, the major costs being in the realm ofdisappointed expectations for some of the brokers. In anAmerican setting, it seems, patrons and brokers shouldseek short-term as well as long-term rewards to justifyestablishing relationships that might tend to be expensiveand full of potential disappointment. What theserewards may be, when they are not financial, is anothertopic of interest.ACKNOWLEDGEMENTSI want to express my appreciation to the members andvolunteers of both Asian Needlecrafters and the HmongArtwork Association for their great patience in answeringquestions and for accepting me as a mere watcher in theirorganizations. My thanks are also due to Dr. Marshall G.Hurlich of the Department of Anthropology, University ofWashington, for his continuing encouragement.176

172 THE HMONG IN TRANSITIONNOTES'This is not the organization's true name.2Hmong and American informants did not agree on who madedecisions regarding items 1, 2, 4, and 5. In this listI have followed the perception of my Hmong informants.Data collection began late in the summer of 1982, soearlier events are based only on informants' reports,which do not always agree.3"Hmong Artwork Association" is a name invented by theauthor to conceal the group's identity.REFERENCESBar-Yosef, Rivka Weiss. 1968. "Desocialization and Resocialization:The Adjustment Process of Immigrants."International Migration Review 2(3):27-43.Eisenstadt, S.M., and Louis Roniger. 1980. "Patron-ClientRelations as a Model of Structuring Social Exchange."Comparative Studies in Society and History22:42-77.Hafey, Pamela, n.d. [19821 "Southeast Asian Textiles."Flier.Hanks, Lucian M. 1977. "The Corporation and the Entourage:a Comparison of Thai and American Social Orgdiiization."In Friends, Followers, and Factions, editedby Steffen W. Schmidt et al., pp. 161-166. Berkeley:University of California Press.Harris, Jane. 1982. "Designs from Another W.,r1d."Bellevue (Wash.) Journal-American, 28 June 1982.Hendriksen, Georg. 1971. "The Transactional Basis ofInfluence: White Men Among Naskapi Indians." InPatrons and Brokers in the East Arctic, edited byRobert Paine. Toronto: University of Toronto Press.Lande, Carl H. 1977. "Introduction: The Dyadic Basis ofClientelism." In Friends, Followers, and Factions,177

NEEDLEWORK COOPERATIVE 173edited by Steffen W. Schmidt et al., pp. xiii-xxxix.Berkeley: University of California Press.Lemoine, Jacques. 1972. Un Village Hmong Vert du HautLaos. Editions du Centre National de la RechercheScientifique. Paris: Ecole Pratique des Haetes E-tudes.Paine, Robert. 1971. "Introduction." In Patrons andBrokers in the East Arctic, edited by Robert Paine.Toronto: Toronto University Press.Thomas, Elaine. 1982. "Hmong and Mien Textile Art: Motifand Meaning." Unpublished manuscript.Wells, Miriam J. 1979. "Brokerage, Economic Opportunity,and the Growth of Ethnic Movements." Ethnology18:399-414.

THE HMONG RESETTLEMENT STUDY: A SYMPOSIUMINTRODUCTIONStephen RederThis symposium is designed to present an overview ofthe Hmong Resettlement Study, a national project fundedby the Uni.ed States Office of Refugee Resettlement. Thestudy was the joint undertaking of Northwest Regional EducationalLaboratory (Portland), the University of Minnesotaand Lao Family Community (Santa Ana). The major purposesof the study were to examine closely the resettlementof Hmong refugees in the United States, focusing onrelated issues, and to answer the following questions:I. What has been the resettlement experience of theHmong?How are the Hmong faring in terms of employment,dependence and adjustment?Are there areas of employment in which the Hmonghave been particularly successful?What are the major problems that Hmong are experiencingin their resettlement and adjustment?What do resettlement workers and the Hmong regardas the major impediments to effectiveHmong resettlement and self-sufficiency?What role does secondary migration (the voluntarymove from the first city of resettlement toanother) play in the resettlement of the Hmong?What are the reasons for resettle;.;ent of theHmong? What are the reas "ns for secondary migrationamong this group? What are the implicationsfor resettlement strategies?II. What resettlement efforts and economic strategieshave provided effective results for the Hmong?17517 J

176 THE HMONG IN TRANSITION- How are problems being handled? What kinds ofsolutions are being tried by different resettlementcommunities and by the Hmong themselves?- What factors account for the effective resettlementof the Hmong in certain communities? Whichresettlement efforts have proven to be the mostpromising?How many and what kinds of entrepreneurial economicdevelopment projects involving the Hmongare currently in operation, e.g., farming projects,A ndau cooperative..? How were they developedand how successful are they?What kinds of Hmong employment strategies havebeen particularly successful?III. How might current strategies be changed to resultin more effective resettlement and long-term adjustmentof the Hmong?How might resettlement be conducted differentlyfor the Hmong? What new projects and approachesare being considered by those involved in Hmongresettlement? How would the Hmong want resettlementto be done differently?Flow can the Hmong be resettled in a way thatbetter utilizes their strengths and unique characteristics?What do the Hmong want for themselves? Whatconditions do Hmong view as essential for effectiveresettlement? What are their goals fortheir own future and for that of the next generationof Hmong?Research conducted in the project included analysis ofexisting data about the Hmong, compilation of informationgathered through numerous informal face-to-face and telephoneconversations with Hmong informants in nearly everyHmong settlement which could be identified across thecountry, and on-site observations, group meetings and per-1Su

THE HMOING RESETTLEMENT STUDY 177sonal interviews with Hmong individuals and families, aswell as resettlement officials, service providers and membersof the host communities. On-site studies of Hmongresettlement were conducted in seven selected locations:Orange County, California; Fresno, California; Portland,Oregon; Minneapolis-St. Paul, Minnesota; Dallas-FortWorth, Texas; Fort Smith, Arkansas, and Providence, RhodeIsland.These communities were selected to exhibit a range ofconditions pertinent to Hmong resettlement, including sizeof the host community, size of the local Hmong population,ability of the local economy to absorb the Hmong,and available welfare and social services.Staff from the participating institutions worked asa team to carry out the overall project and the sevencase studies.The Northwest Regional Educational Laboratory teamcomprised Stephen Reder, project director; Mary Cohn, JohnFine,: (also with State of Rhode Island), Mike Sweeney,Bruce Thowpaou Bliatout (also with City of Portland), KarenReed Green and Marshall Hurlich (also with University ofWashington).The University of Minnesota Center for Urban and RegionalAffairs team comprised Bruce Downing, Simon Fass,Doug Olney, Sarah Mason and Glenn Hendricks.Project workers from Lao Family Comrunity were ShurVang Vangyi, Dang Vang and Thongsay Vang.The results of the project are available to the publicas a series of reports published by the United StatesGovernment Printing Office (GPO). Copies may be orderedfrom:Dr. Allen GallOffice of Refugee Resettlement330 C Street, SWSwitzer Building, Room 1229Washington, D.C. 20201Mr. Bud TumyRefugee Materials Cntr.U.S. Dept. of Education324 E. 11th Street9th FloorKansas City, MO 64104161

178 THE HMONG IN TRANSITIONFinal ReportsVol. 1: SummaryVol. 2: Economic DevelopmentVol. 3: Exemplary ProjectsExecutive Summary (written in English)Executive Summary (written in Lao)Executive Summary (written in Hmong)Site ReportsOrange County, CaliforniaFresno, CaliforniaPortland, OregonMinneapolis-St. Paul, MinnesotaDallas-Fort Worth, TexasFort Smith, ArkansasProvidence, Rhode IslandThe purpose of this symposium is not to summarizedetailed findings of the studythe series of public reportsis available foi that purpose. Within the timeavailable, the participants in the symposium wish merelyto highlight a few issues about selected aspects of Hmongresettlement in the United States.Presentations will thus concern a number but certainlynot all of the critical issues observed in Hmongresettlement: population trends, secondary migration,language issues, employment and welfare dependence, youthand future issues, and economic development. Althougheach presenter served a key role in the study, the researchwas not divided in a parallel fashion. For thissymposium presenters have synthesized the results ontheir assigned topic from the range of case studies andother information on hand. !n many cases they relied oncase studies conducted by others and in some cases addedtheir own opinions and evaluations of the issues. Wehope the resulting diversity provokes a lively discussion.1 82

POPULATION TRENDSDouglas OlneyIt is difficult to obtain accurate population figureson the Hmong in the United States or, for that matter,in the rest of the world. The figures put togetherfor the Hmong Resettlement Study are from several sourcesincluding the United Nations High Commission for Refugees(UNHCR), the Federal Office of Refugee Resettlement(ORR), surveys done previously in Hmong communities anddata collected in the course of this study. The figuresare not to be taken as absolute, but they do provide ageneral picture of Hmong population trends.Population Prior To ResettlementLaos. Estimates of the prewar and preexodus populationin Laos vary a great deal. An early 1960 censusstated there were about fifty thousand Hmong, but a moreaccurate figure is that of Yang Dao, who stares chat theHmong numbered over three hundred thousand or approximatelyten percent of the total population of Laos.Movement to Thailand. Table I shows the movement ofhill-tribes to Thailand ("hill-tribes" or "highland. Lao"include other Lao ethnic groups, but 90 to 95 percent areHmong). In 1975, after General Vang Pao left, a greatexodus occurred. In the next few years the flow diminishedbut it surged again in 1979, when pressure on Hmongin Laos again increased. Thi; second wave peaked in early1980 and then dropped off quickly in 1981 and the followingyears.Table II shows the resettlement countries and numberof hill-tribe people who have moved out of Thailand.Over one hundred ten thousand Hmong have left Laos since1975, and there are still about fifty-two thousand inseven refugee camps in Thailand.179 133

184180 THE HMONG IN TRANSITIONTable IHILL-TRIBE MOVEMENTTO THAILANDTable IIHILL-TRIBE RESETTLEMENTTO OTHER COUNTRIES(August 1983)1975 44,659 United States 53,9501976 7,266 France 6,5641977 3,873 French Guiana 1,4131978 8,013 Canada 8061979 23,943 Australia 3171980 14,801 Argentina 2221981 4,356 Other 2891983 (May) 1,785TOTAL: 110,512 TOTAL: 63,561THAILAND REFUGEE CAMP POPULATION, May 1984: 52,374Movement to the U.S. Table III and Figure I showthe history of hill-tribe settlement in the UnitedStates. The peak flow to the United States was in 1979and 1980 following the second wave of exodus from Laosand corresponding to an increased awareness of the refugeeproblem in the United States. Since 1981 Hmong arrivalshave slowed considerably; it seems that the Hmongwho remain in the camps do not want to resettle to thirdcountries.Table IIIHILL-TRIBE RESETTLEMENTTO THE UNITED STATES1975 3011976 3,0581977 1,6551978 3,8731979 11,3011980 27,2421981 3,7041982 2,5111983 300TOTAL: 53,950

POPULATION TRENDS 181Figure I38808 -75 76 77 78 79 88 81 82 83YearPopulation In The United StatesGeneral. As Table IV indicates there are betweensixty-one thousand and sixty-five thousand Hmong currentlyin the United States; this includes a high level of185

182 THE HMONG IN TRANSIT:ONnatural increase in the number who were resettled here.California has about half of the Hmong population, andtwo-thirds of those live in the Central valley. The approximatepopulations of the sites visited in this studyare shown in Table V.Table IVHmong outside CaliforniaHmong in CaliforniaCentral Valley32,000 - 34,00028,000 - 30,00020,000 - 22,000Total 60,000 - 64,000Table VHMONG POPULATIONS IN STUDY SITESFresno 10,000Minneapolis-St. Paul 8,500Orange County 3,000Providence 2,500Portland 1,000Fort Smith 300Dallas-Fort Worth 300Communities. There are currently over seventy Hmongcommunities in the United States, ranging in size fromunder one hundred to over ten thousand. About fifty-sixcommunities have populations under one thousand whilethere are only two with populations over eight thousand.While there are several distinct clusters of Hmong, manysmall Hmong communities are scattered across the nation..4i factor that contributed to the building of Hmongpopulation centers has been the heavy sponsorship ofHmong by Hmong. In Portland it was found that eighty-fivepercent were sponsored by other Hmong and in the TwinCities of Minneapolis-St. Paul, fifty-six per :ent. Of186

POPULATION TRENDS 183course we know that once the Hmong arrived in the city inwhich they were sponsored, they often moved on to othercities.The wide variation in community size and the variationin other factors leads one to question whetherthere is a relationship between Hmong community size andeconomic and social self-sufficiency. Is there an optimumsize for a Hmong community? Do different cities havedifferent optimum populations which they can support?Table VIHMONG COMMUNITIES IN UNITED STATESSize (persons)Number0 - 299 28301 - 999 281000 - 1999 62000 - 2999 43000 - 3999 24000 - 4999 28000+ 2Birth rates. In the United States, the Hmong birthrate continues to be high, just as it was in the refugeecamps. The rate is about 50 per 1000; this may be comparedto the overall rate in the United States of 15.9per 1000. While actual birth rates may be slowing, theavailability of better medicine has reduced the infantmortality rate as compared to that in Laos. In the TwinCities seventeen percent of a sample of 1,805 individualswere born in the United States. In addition to the generaldesire of Hmong to have large families, the highrate of marriage among Hmong teenagers contributes to thehigh birth rate. In sevEral of the cities we visited itwas reported that eight-five percent to ninety percent ofthe teenage girls were dropping out of high school to getmarried. This rapidly growing population may lead to furtherproblems in attaining self-sufficiency. For example,continued large family size may drain family resources.On the other hand, it has been shown that the large Hmong187

184 THE HMONG IN TRANSITIONfamily can be a resource in itself.Age-sex characteristics. In the various samples collected,the Hmong population remains young: twenty-fiveto thirty percent of the Hmong in the United States areunder five and over fifty percent are under fifteen,while at the other end of the spectrum only about fourpercent are over sixty. This age distribution suggestsconsequences in several important areas of Hmong resettlement,particularly in the area of education. Schoolsystems in which there are substantial numbers of Hmongshould expect continued enrollment of Hmong children asthose under five reach school age. Subsequently we shouldsee a larger, historically unprecedented proportion ofHmong who will have had some education. The sex ratio ofthe Hmong seems to be fairly even; in the Twin Citiesfifty-four percent were male and forty-six percent female.In general, families seem to be intact although inthe Twin Cities twelve percent of Hmong households areheaded by women. (Sources: UNHCR Monthly Statistics, ORRHmong/Highland Lao Workshop, Hmong Resettlement Study).SECONDARY MIGRATIONTO CALIFORNIA'S CENTRAL VALLEYJohn FinckApproximately one-third of the Hmong in the UnitedStates moved from one city to another between the time ofthe first Hmong Research Conference, held in October1981, and this seccnd conference, held two years later.What drew twenty thousand Hmong to the Central Valley ofCalifornia in so short a period? In the space of twoyears, an estimated two thousand Hmong left St. Paul-Minneapolis, and half of the Hmong population of Portland,Oregon, about two thousand people, moved to CentralValley towns such as Fresno, Stockton and Merced. Fresnohas already replaced the Twin Cities as the largest resettlementsite of the tirrong inthe United States and isreported to have the Largest peacetime concentration ofHmong in history.188

SECONDARY MIGRATION TO CALIFORNIA'S CENTRAL VALLEY 185Although the concentration of so many Hmong in onegeographical area has permitted the reunification of manyfamilies and friends, it has not been without some problems.Fresno also has the largest number of Hmong on welfareof any county in the country. The numbers on cashassistance have grown exponentially. In July 1980, therewere 168 Hmong individuals on public assistance in Fresno.Three years later, Fresno had over seven thousandHmong on assistance. Unfortunately, Fresno's agricultureand service-based economy has not yet been able to absorbthe new arrivals. Some individuals are concerned aboutthe potential for Hmong becoming .permanently dependent onwelfare in the Central Valley.Migration to the Central Valley appears to have beenlargely an unplanned, spontaneous choice by dozens oflineage and sublineage groups who moved for a variety ofreasons. As Cheu Thco reported at the first Hmong ResearchConference, the Hmong have a long tradition ofwhat he termed "moving for betterment." With this understanding,we 'an conclude that the migration into theCentral Valley stems in part from family reunificationand in part from the perceived lack of opportunities inthe primary host communities. The apparent irony lies inthe fact that the Central Valley seems to be one of theleast desirable sites in the country at present for employmentand eventual self-sufficiency of large numbersof undereducated refugees. Their decisions to move tothe Central Valley appear to have been based largely onthe shortcomings of the sending communities rather thanon knowledge of what might lie ahead in Fresno or Merced.Interviews with Hmong informants in Fresno revealedseveral reasons for moving, including:a) The absence of jobs or a history of repeated layoffsin the host community. The national :ecessionlast in thetwo years hit the Hmong particularly hard. EvenHmong with employment histories were subject to on-again,off-again hirings and firings. Frustration with the jobmarket led many Hmong to consider moving. However, thejob situation in the Central Valley was blea'cer than whatmost Hmong left behind in other states. In 1983, about 5percent of the employable Hmong in Fresno were working.By comparison, about 20 percent indicated they had heldjobs before coming to Fresno.189'

186 THE HMONG IN TRANSITIONb) The desire for English classes and vocationaltraining. There is some evidence to suggest that refugeesin other parts of the country assumed that what wasonce widely available in Orange County, Californianamely, vocational training and English languageclasseswas also available in Fresno. Unfortunately,Fresno, as late as spring 1983, had no vocational trainingand only limited English as a Second Language (ESL)classes for Hmong.c) The dream of farming. Perhaps half of the Hmongcame to Fresno with the idea of farming. After seeingfirsthand the requirements for capital, knowledge of modernfarming techniques and marketing experience in theCentral Valley, less than 20 percent of those who camewith farming in mind indicated a continued interest. Thecosts and risks are great. The failures have been many.The minority of families who have successfully broken intofarming in Fresno have inspired many others to try.And some progress is under way. But agriculture willlikely support only a fraction of the Hmong populationpresent in the valley.d) The than: e in national ..lic for cash assistance.The federal government reduced the period of refugeecash assistance from thirty-six to eighteen monthsin the spring of 1982. Those states without welfare programsthat paid benefits comparable to refugee cash assistanceto unemployed two-parent families, were obligedto cut many refugees from their welfare ranks. A significantnumber of families without jobs in cities likeSeatilt and Portland therefore had to find other means ofsupport, which in most cases turned out to be a generalpublic assistance program that accepted intact familiesin a neighboring state.e) The snowball effect of family reunification.Large numbers of Hmong families in a single location usuallytend to attract additional families. This is thewell-known experience of secondary migration. Youngerfamily members find it difficult to resist the call oftheir elders either to move with a group or to join thegroup in a new setting. To some extent, migration to theCentral Valley took on a life of its own that soon spread190

LANGUAGE ISSUES 187to all parts of the country. Even Hmong leaders who advisedagainst moving saw their counsel go unheeded in thecontagion of the impulse to move to California.LANGUAGE ISSUESBruce DowningThe Hmong Resettlement Study addressed questionsconcerning the acquisition of English and of literacy bythe Hmong and its importance to more general economic andsocial adjustment. Some of these issues will be discussedhere briefly.1. The Value of English for Economic and Social Self-Sufficiency.As in virtually every previous study we found thatlearning English was considered both very important andvery difficul.: by the Hmong interviewed.The Hmong see language as the principal barrier toemployment as well :is other social adjustment. In theTwin Cities Hmong Comcaunity Survey (University ofMinnesota, August 1982,, respondents were asked to namethe problems facing them or their families. "English"was the problem most often named, by 58 percent of allrespondentsand by 71 percent of the women respondents.(The second most -often named problem was "no job," mentionedby 46 percent.)In Orange County it was found that "in spite of manyindividual successes in learning English in a short time,the community as a whole sees inadequate English as anoverwhelming problem. English is seen as necessary forentry into jobs, for promotion on the job, for making useof job training, for higher education and for acceptancetr: .he American community."While some considered job acquisition to be the mostimportant problem facing the Hmong, training and Englishwere seen as the keys to obtaining jobs, and ability tospeak and read English was considered essential in most191

188 THE HMONG IN TRANSITIONjob-training programs; thus, the primary factor in manycases turns out again to be English. In the Twin Citiesit was found that nearly two-thirds of unemployed Hmonghousehold heads were not actively seeking a job. Of thereasons given for not looking for employment, the mostfrequent (mentioned by 32 percent) was "limited English."In a few areas, such as Dallas, Texas, it has beenrelatively easy to obtain jobs without knowing Englishand without special job skills. But now many of thoseHmong who have had jobs are worried that if they are laidoff they will not be able to regain employment.Figure I (based on data from the Twin Cities HmongCommunity Survey) shows graphically the relationship betweenproficiency in English and employment.Figure I30B HOLDING IN UNITED STATESBY SELF-RATED ENGLISH PROFICIENCY:TWIN CITIES HMONG HOUSEHOLD HEADS70 66.76050.Percent41d-30-20 18.8 20.110- 7.70No Little IINellAbilityr.nglishto SpeakUeps.192

SECONDARY MIGRATION TO CALIFORNIA'S CENTRAL VALLEY 1892. Educational and Linguistic Background.A major concern in providing English language instructionhas been the educational and cultural experiencesthe Hmong had in Laos and in the refugee camps, andhow they affect their ability to learn English.Some data are presented in Figure II (based on theTwin Cities Hmong Community Survey).Figure IIEDUCATIONAL ATTAINMENT OF HMONG OVER 1223 -HenCI WomenPercent10I1-3 4-6 7-9 10 +Years of Education1931

190 THE HMONG IN TRANSITIONFifty-five percent of the men and eighty-nine percent ofthe women had no schooling in Laos.The importance of background factors such as literacyin Hmong, literacy in any language, knowledge of asecond language, and level of education to the learningof English was brought out very clearly in papers byBarbara Robson and Stephen Reder (1982) at the firstHmong Research Conference and need not be reemphasizedhere.3. The Extent of English Proficiency and Literacy of theHmong.An important question in this study concerned theextent to which the Hmong are learning the English neededfor economic and social self-sufficiency. Here are somecases.The Hmong families in Fort Smith, Arkansas, were alloriginally resettled in some other location. Nearly halfof the Hmong household heads have never received formalinstruction in English. Hmong adults in Fort Smith believethat their limited ability to communicate in Englishis a major barrier to employment, yet none of thethirty-five men who attended a meeting for the study hadattended ESL classes in Fort Smith. Asked if they wouldlike to have additional English instruction, all saidthey would.In Orange County a local study showed that eightysixpercent of the Hmong there had attended some type ofclass and in 1981 (the time of the survey and the peakyear of Hmong population in Orange County) seventy-threepercent of the Hmong were currently enrolled in trainingof some kind.In Dallas, where employment rather than training wasemphasized by resettlement agencies, only about two inthree of the Hmong originally resettled there receivedinstruction in English. Despite the insistence of someresettlement workers in Dallas that ESL is not necessary,because the Hmong can pick up English at work, both menand women interviewed indicated that they could not learnEnglish on the job.Surveys intended to determine levels of English proficiencyin the Hmong population have been limited toself-assessments by those surveyed. For example, in a194

SECONDARY MIGRATION TO CALIFORNIA'S CENTRAL VALLEY 191sample of Hmong household heads in the Twin Cities, wheremajor efforts have been made to provide appropriate languageinstruction for the large local Hmong population(Hmong Community Survey), self-rated English proficiencywas as follows. These responses are to the question "Howwell do you speak English?"Not at all 31.2%Very little 42.6%Well 25.0%Very well 1.2%Rated proficiencies were slightly lower for "reading English,"and slightly lower again for "writing English."The Hmong Resettlement Study Final Report summarizesthe situation in which the majority of Hmong find themselvesthis way: though not all jobs which the Hmong obtainrequire English or literacy skills, it is generallybelieved by the Hmong and educators and employers thatmost Hmong need to know more English to retain a job, tosuccessfully compete for new jobs in case of layoffs, andto be promoted from minimum wage positions.4. Practical Barriers to Learning English.The Hmong Resettlement Study interviews revealed anumber of important practical barriers to the learning ofEnglish by Hmong adults, distinct from the backgroundfactors already discussed. These include social isolation,difficulty in learning English on the job, unavailabilityof programs, and problems of access to the programsthat do exist.The assumption that if jobs are obtained the workerswill learn English on the job is not warranted. Hmongwho don't know English have obviously been hired only forthose jobs where communication throughlanguage is not apart of the work setting--for example, repetitious jobson the assembly line. Workers holding such jobs havelittle opportunity to use English. Where employers havehired a group of workers with one bilingual among them tohandle communication, the bilingual gets lots of practicein using English; the others get none. And if there is noimprovement in English on the job, then obviously there196

192 THE HMONG IN TRANSITIONis no possibility of advancement to a more skilled job ora supervisory position for which English is clearly requiredİnterviews with Hmong adults who have attended ESLcourses support the view that six months or a year oreven two or three years of ESL are not enough to get manyof the Hmong adults to a point of self-sufficiency, tothe point where they can compete for jobs on an equalbasis with workers who speak English v. ell, or to thepoint where they can handle jobs that pay enough to feedtheir families.Other factors frequently mentioned by the Hmong interviewedfor the Hmong Resettlement Study as barriers tostudying and learning English included transportation,child care (a major barrier to both training and employmentfor women), psychological problems, and family responsibilitieswhich leave little time for study. Finally,in many smaller communities and some larger oaes,there are simply no ESL classes or none at the appropriatelevel or with an appropriate focus.5. Improvement of Language and Literacy Efforts for theHmong.Hmong adults interviewed for the study had many suggestionsfor the improvement of ESL instruction, whichmust be listed without discussion here:a more concrete, structured, step-by-step teachingapproach, with regular periodic assessmentsand feedback on progress, so students can knowwhether they are making progress;more attention to practical language use;more preparation for employment and for jobtraining;more efficient use of class time including moreeffective use of bilingual classroom aides;beginning classes that start at a truly beginninglevel, especially in teaching literacy, perhaps196

HMONG YOUTH AND THE HMONG FUTURE IN AMERICA 193with Hmong instructors to provide explanation inHmong, or beginning classes taught by Hmongthrough mutual assistance associations;appropriately designed courses (rather than placingHmong in courses designed for more advancedSpanish speakers or Vietnamese);coordination of ESL with specific job training orjob experience;opportunities to practice English outside ofclass.HMONG EMPLOYMENT AND WELFARE DEPENDENCYShur Vang 7angyiOverviewIt has been almost eight years since the first waveof Hmong refugees arrived in the United States. Moststill have no idea about what types of jobs are available.Some want to work immediately after arrival; othersfeel they will never be able to get a job or it may takeyears before they can be ready to work. Among the manyreasons for these feelings are "culture shock," languageproblems, and the fact that the work environment in thiscountry is very different from that of Laos.By 1982, when we started this project, there wereabout sixty thousand Hmong in the United States. In tryingto document Hmong employment conditions, we have encounteredseveral problems:1. Hmong employment is not stable.2. There has been widespread secondary migration ofHmong.3. Those Hmong who are employed tend to have found theirown jobs and thus have dropped out of official fig-197

194 THE HMONG IN TRANSITIONures such as employment services registries and welfarerolls.This study used many informants and documentary resourcesto draw a picture of Hmong employment in the sevencommunities where case studies were done. Estimatesof data on local employment patterns were collected bytelephone from knowledgeable Hmong informants in almostevery other Hmong community in the country.Local DifferencesThe rate of Hmong employment differs widely fronicommunity to community. It depends on the number andtypes of jobs available and the competition for thosejobs. For example, the rates of employment of Hmong inOrange County and Fresno, California and the Twin Citiesof Minnesota are quite low compared with those of smallerHmong communities such as Providence, Dallas, and FortSmith.Hmong employment is also quite vulnerable to variationsin the local economies. In the case of OrangeCounty, for example, two years ago, over half of theHmong households had at least one adult working, but afterthe economy slowed down in 1981 and 1982, only aquarter of the families were working. Hmong, like otherpeople, are always looking for better opportunities.After changes occurred in refugee cash assistance policy,many states without adequate public assistance programsfor intact, unemployed families lost their Hmong populationto secondary migration. Families without jobs wereunable to stay in these areas. Most moved to Californiaeven though there were no jobs there; they wanted to moveaway from the frustrations and depression of waiting forwork. At least in the Central Valley, they thought, theycould start doing some farming.Individual DifferencesHmong are employed at a variety of jobs. Most of thejobs are entry-level positions in light manufacturing,requiring little previous training. Examples are elec-193

HMONG YOUTH AND THE HMONG FUTURE IN AMERICA 195tronic assembly, jewelry grinding, machine operation,sorting and collating, and food processing. Hmong arealso employed at minimum wages in other industries, particularlynursery work, carpentry and industrial sewing.Most of the jobs obtained thus far require ,ninimal Englisheither to learn or to perform.At the beginning of Hmong resettlement, during 1.978-79, employers would hire refugees as long as there wassomeone in the company who could communicate with them intheir native language. But that flexibility did not lastvery long. From 1980 to the present, because of thedownturn in the economy, many companies have been forcedto reduce operations or close down. As a result, employersbecame more selective and were reluctant to hire anyonewho could not speak English.People who took unskilled or semiskilled entry-levelpositions commonly did not encounter much opportunity foradvancement. Wages received for those entry-level jobsvaried between $3.50 and $5.00 per hour for men, and$3.50 to $4.50 for women. Few received any medical benefits.In California, many Hmong seemed to be laid offwithout adequate explanation when their wage reached$6.00.A minority of the Hmong had better experience in thelabor market. A small group of Hmong who spoke enoughEnglish took vocational training for positions as electronictechnicians and machinists. These workers receivedsomewhat higher wages, ranging from $7.00 to $10.00 perhour, and usually were offered health benefits. Anothergroup includes the Hmong bilingual social service andparaprofessional educational workers. The average wagefor this group is almost $4.50 to $6.00 per hour, occasionallyreaching up to $15,000 per year. Another evensmaller group included white-collar positions such as insurancesales and secretarial work.The general picture, then, has the majority of Hmongemployees working at minimum wage jobs and a small numberin more highly skilled labor and technical or whitecollarjobs. About 80 percent of the adult Hmong in Californiaare not employed at all, compared with less thanfifty percent of those who are outside of California.19J

196 THE HMONG IN TRANSITIONWelfare DependeriaIn some parts of the country, Hmong have had sponsorswho found jobs for them as soon as they arrived. Inother states where there were no individual sponsors, theHmong refugees have had to go onto public assistance.The resettlement agencies have tried to avoid such dependencyby providing cultural, social and vocational orientationto the refugees. Unfortunately, the training oftentakes six months to one year to be effective, longerthan most can survive without cash assistance.Many efforts have been made to find jobs for Hmong,but it has been difficult. There are not enough suitablejobs for them. Not only Hmong, of course, but many oftheir American peers, too, cannot get jobs. Across thecountry, wherever we met with Hmong groups and individuals,they expressed their concerns that the Americanpeople may think that Hmong are lazy. Hmong have a strongdesire to work, to get off welfare, to take any job withwhich they can support their families. However, becausethe cost of health care is so high for large families,many need subsidized medical care for their families.Policies linking medical assistance to cash assistance (anational welfare policy) make it very hard for largeHmong families to get off welfare. Transitions from welfareto self-sufficiency are further hindered by the arbitrary100-hour limitation on earnings regardless oflevel of need. Hmong who have found good jobs do not wantto quit, but for some there have been no other viable alternatives.Many Hmong feel the welfare system serves to keepthem on welfare and to keep them poor, and they recognizethe irony in that situation. In many states, the federalgovernment has spent a lot of money to help the Hmong andother refugees to become self-sufficient, but this assistancehas not been effective because cf too much red tapeand too many political clouds and arbitrary restrictions.Despite the many obstacles, the Hmong still sharethe major goal of refugee resettlement in this country:self-sufficiency.2 v0

21HMONG YOUTH AND THE HMONG FUTURE IN AMERICA 197HMONG YOUTH AND THE HMONG FUTURE IN AMERICAMary CohnThe Hmong population profile shows that the Hmong inAmerica are a young and rapidly expanding population,with the proportion of relatively younger members of thecommunity growing each year. What the future holds forthe young Hmong will eventually determine the outcomes ofHmong resettlement. As Hmong discussed the future of theyounger generation in the course of this study, the twointerrelated issues of education and cultural adjustmentemerged as crucial.Education IssuesIn a very short time education has come to be tremendouslyvalued by Hmong in the United States. ManyHmong who have given up hope of rebuilding their ownlives pin their hopes for the future of the Hmong inAmerica on the education of their children.Though there are many individual exceptions, studyinterviews indicate that there are three different needgroups of Hmong children and adolescents currently in theeducational systems.The first group is made up of children who are enteringschools with English skills and who thus canquickly participate in the mainstream of the school system.These children appear to be adjusting well toschooling and to be succeeding in their studies. Thesecond group is made up of young Hmong children, manyborn in the United States, who are entering schools withno English skills. These children apparently learn Englishquickly, but need special attention and recognitionof their limited English proficiency as they first enterschool. Besides English problems, some of the childrenwho have had no previous contacts with non-Hmong havecultural adjustment problems as well. The third group ismade up of older Hmong children and adolescents, agedabout thirteen to twenty-one, who arrived in the UnitedStates with no previous education, having spent most oftheir school-aged years in flight from war or in refugee

198 THE HMONG IN TRANSITIONcamps. These students, generally placed in grades by agerather than skill level, typically. have great difficultiescatching up to their classmates in time to graduate.The dropout rate for these adolescents is very high, andif they graduate they are often. still unprepared to lookfor a job or to continue their education.The very high dropout rate among Hmong teenagers isan issue of great concern among the young people, theirparents, leaders and teachers. Though the attrition ratevaries from locale to locale, being quite low for examplein Dallas, it remains high in most areas. Hmong youth,leaders and educators identify several contributing factorsto school dropout. These include academic difficultiesin school, inadequate direction and counseling, welfaredisincentives and early marriage and pregnancy.Students relate problems in reaching competenciesfor the exit examinations now required by several states,including California and Oregon. Faced with failure inthese exams, some students simply leave school. Evenstudents who stay in high school indicate they have greatdifficulty in Glasswork which involves more advancedreading and writing skills, such as social studies andAmerican history.Hmong teenagers regularly identify a need for counselingin school, particularly career counseling. Studentsfeel this direction must come from the schools,since their parents are still unfamiliar with the Americaneducational and work world. Many students also suggestthat their parents would benefit from some educationabout the American school system, since parents now havelittle idea of how best to participate and encouragetheir children in school.Some youth see slow changes taking place in traditionalmarriage patterns. Yet in spite of some effortsby youth groups and Hmong leaders to encourage youngpeople to marry a bit later, the Hmong traditional patternof early marriage and childbearing persists. Essentiallyat odds with the design of American school systems,early marriage and pregnancy mean that the dropoutrate for Hmong high school girls is very high. Nationwideit is probably at least fifty percent; in some impactedareas visited it is as high as eighty or ninety percent.(Again, Dallas exhibits a much lower dropout rate.) Ingroup meetings with researchers, teenagers indicated they202

199 HMONG YOUTH AHD THE HMONG FUTURE IN AMERICAare interested in but have very little knowledge of familyplanning and do not have the information to makechoices about the size or spacing of their families.Ironically, in many of the schools Hmong attend, sex educationor birth control information is not taught untilthe junior or senior year of high school, by which timemany Hmong girls have already dropped out, married andgiven birth to one or two children.Cultural factors and education are interrelated: forthose young Hmong men and women who are already experiencingdifficulties in school, and have little future orientation,the culturally accepted patterns of marriageand family become more attractive. Teenage marriage poseslong-term problems for integration into American life:young men and women with education far below the norm forthis society will continue to have difficulties competingfor jobs, and large young families without two incomesmay continue to live at tne margins of poverty.Hmong leaders see an intergenerational pattern ofpoverty developing that they want aborted now. In familiescurrently dependent on welfare, youth and their parentssee AFDC policies as a substantial disincentive forHmong adolescents from poor families to complete highschool, particularly in California. Older youth who arebehind in high school skills are often not able to graduateby their nineteenth birthday, when their AFDC grantsmay be cut. Under these conditions many young people feelt' .ty must quit school and go to work, only to find thatley cannot find a job without a high school diploma.Higher EducationIn spite of the difficulties for Hmong youth, manyhighly motivated young Hmong across the country see highereducation as their main hope for the future; a fewschool systems have also made laudable efforts to helpHmong young people of limited educational background.Among those Hmong students who have achieved success inhigh school and are able to go on to college, there is avery strong feeling of responsibility to share the educationthey receive by giving it back to the Hmong community.There are several youth groups in the country thattutor other Hmong students, advise them on educational203

200 THE HMONG IN TRANSITIONplans, and give Hmong-language workshops for their elderson such topics as American government and economics.Based on the nationwide telephone poll conducted aspart of this study, there are three hundred to four hundredHmong students attending colleges full-time. Thisnumber constitutes a tiny elitelesi than five percentof the Hmong population of college age, although manymore students are probably attending colleges part-time.The major barriers to higher education which Hmong studentsidentify are graduating from high school, financialproblems and f, oily responsibilities. For those studentswho are admitted into college, lack of funds seems to bethe largest problemmost Hmong college students use federalBasic Educational Opportunity Grants (BEOG) and liveat home with their families.Cultural AdjustmentHmong young people straddle two cultures. They oftenfeel conflicts between the values of independence in theAmerican society they encounter in school and in theirneighborhoods, and the traditional values of interdependencein their own families. Along with a strong attachmentto their families and to "being Hmong," young peoplealso voice a desire to be accepted as American. For example,teenage girls told us of problems they have "fittingin" with their peers. Many said they would like tojoin school clubs and go out with American friends, butthey have to be home to take care of younger siblings,and their family responsibilities come first. Some ofthe cultural adjustment problems that arise in familiesare due to the discontinuities between the experiences ofHmong parents and new experiences of their children. InAmerica, traditional family authority and boundaries nolonger seem clear-cut.The Future Of The Younger GenerationThe Hmong youth's current experiences with adjustmentand education have important implications for theirfuture. In the short-term future, the specter of welfaredependence looms large for those young people aged about204

HMONG YOUTH AHD THE HMONG FUTURE IN AMERICA 201fourteen to twenty-five who have incomplete schooling andinsufficient skills to compete for jobs. Early marriageand large families mean that this new generation of Hmongparents may also experience difficulty, not only in becomingself-supporting, but also in integrating socialand economic life into the mainstream resettlement community.The younger Hmong children are likely to experiencemore success in school than their older brothersand sisters, though they may continue to lag behind theirpeers in language skills. Whether or not this group willcontinue to marry young. drop out and start large familiesis uncertain, but the way this issue is resolved mayhave a large impact on their own self-sufficiency and adjustmentand on the economic and social adjustment of thenext generation of Hmong.One attitude clearly emerges in interviews withyoung Hmong people: in spite of the uncertainty abouttheir future, they feel responsible for helping themselvesand helping their community. Although they feelsome conflicts with the values of American society andtheir roles within their own families, they still intendto take on responsibilities for their families; theywould like to be able to support their parents in theirparents' old age; they wish to use the skills they gainthrough better education to help the Hmong. Most youngpeople see themselves as Hmong and they wish to remainHmong, but they also want to he armed with the skills andknowledge of the larger society in which they now lire.The choices that they face are sough and often confusing.As one young man told us:I think we know it's hard, and we all want a collegeeducation. We understand that we might goin the wrong direction and we are w3rried aboutthat. We want to go in the right direction.235

ECONOMIC DEVELOPMENT AND EMPLOYMENT PROJECTSSimon M. FassAgainst a backdrop of high unemployment and welfaredependency, Hmong across the country are attempting topromote self-sufficiency through a variety of innovativeeconomic development projects. About one hundred sewingventures, farm projects and small businesses are nowunder way. They touch the lives of twenty-five hundredfamilies, or 25 percent of the Hmong population.SewingBeginning with two ventures in 1978, the number ofsewing projects rose to twenty-eight by 1983, and thetwenty-three hundred women involved were generating annualsales of over $700,000 in fifteen states.American women initiated most projects. They viewedthem primarily as methods to preserve ethnic art or tohelp Hmong with psychological adjustment to the UnitedStates. As men's difficulties in securing employmentgrew evident after 1981, project participants found themselvesunder growing pressure to maximize their contributionto family income.Three strategies to increase earnings arose fromthis pressure. The first was to emphasize marketing oftraditional crafts produced at home or imported from relativesin Thailand. Participation in street fairs, churchbazaars and gallery displays, and advertising led to anincrease in sales. A few women succeeded in earning$3,000 a year or more, but the majority made less than$75.To address the needs of these women, as well as toexpand sales in general, the second strategy focused ondesign, production and marketing of inexpensive utilitarianitems, like belts, aprons, eyeglass cases or fashionclothing that incorporated ethnic decorations. Some productsdid not sell. Others, like eyeglass cases, becamequite successful. Earnings remained low, averaging $240a year for each x oman, but this constituted a sizable increasefor those unable to market traditional craftwork.206202

ECONOMIC DEVELOPMENT AND EMPLOYMENT PROJECTS 203The third strategy focused on training women in commercialsewing skills, such as dressmaking and tailoring,and then marketing their services, e.g., contract alterations.Initial results from the handful of projects thatpursued this approach were promising, with individualmonthly earnings of $100 to $200 not uncommon.In implementing these strategies the projects had toovercome many difficulties.Some refugees were afraid offailure and unenthusiastic about changing their traditionalmethods of craftwork. Americans with ethnic artand social service backgrounds found themselves illpreparedto transform sociocultural projects into whatwere essentially small businesses. They did not haveprior exposure to market assessment for new products,patterns or colors, or to methods of marketing. Moreover,most projects were underfinanced. After payingrefugees 60 percent to 80 percent of the value of sales,the American initiators had little left over to financenew product lines, to advertise or to acquire sewing machines.Eleven ventures secured some external funding, andfared better than projects entirely dependent on internalresources. Externally stipported projects, with resourcesto pay for full-time managers, materials and equipment,marketing expenses and training stipends, generated fivetimes more in sales and two and a half times more in refugeeincome than the others: $303 a year for each womanvs. $117. The better-endowed projects were not, however,financially viable. In all but two cases expenses exceededrevenues. Preliminary indications neverthelesssuggested that outside assistance, in terms of funds andstaff with business exrerience, could make a significantdifference to refugee income.FarmingFamilies began gardening soon after their arrival,and almost every Hmong community enjoyed at least oneseason of growing vegetables for family consumption. Agriculturalextension services, community organizationsand self -help associations assisted the Hmong in most ofthese farming projects. The gardens provided householdswith in-kind incomes of $100 to $700 in retail value eachseason.2 0 7

204 THE HMONG IN TRANSITIONPressure to secure alternative sources of income,noted above, and Hmong success at producing crops, ledAmericans and refugees to initiate eleven major projectsin five states. At the end of 1983, these initiators wereattempting to assist 230 households, and planned to coveranother 70 to 80 within two or three years.Project characteristics varied considerably. Participantsranged from three to sixty families, farm sizefrom five acres to thirteen hundred acres, and initialinvestment from $7,000 to $425,00C, In general theseprojects pursued one of two goals. The first type ofproject tried to create a financially viable source ofsecondary income: a seasonal venture on rented land toprovide wage income for unemployed adults, unemployablewomen and elders, and out-of-school children. One projectin Iowa collected $16,000 from private foundationsin 1982 and launched sixty families into commercial farmingon fifteen acres. It generated an average income of$200 per family, but lost $12,000. The project expandedto twenty-five acres in 1983, diversified its production,yielded $400 per family, and set aside funds to beginagain in 1984 without extensive recourse to outside support.The second type of project tried to provide completeself-sufficiency. A twenty-two-acre venture in Washingtonused almost $90,000 in 1982 to generate $13,000, or$1,100 each for a dozen participating families. In 1983,after infusion of another $70,000, it yielded $46,000 insales, and $3,000 per family in net earnings. Incomegoals for 1984 called for $9,000 per family.Another scheme, a four-year, $6 million effort forup to ninety families, organized by the University ofMinnesota Agricultural Extension Service, started in 1983with an initial investment of $472,000. The project successfullyproduced over $200,000 in crops, but could harvestand sell only $70,000. Among other factors, communicationdifficulties between Americans and refugees preventedHmong from providing the amount of field laborrequired for complete harvesting.Experiences in the Iowa, Washington and Minnesotaprojects were typical, and dampened the original enthusiasmabout agriculture as a rapid mechanism for selfsufficiency.Participants learned that commercial agriculturewas a more formidable challenge than consumption-2 J8

ECONOMIC DEVELOPMENT AND EMPLOYMENT PROJECTS 205oriented vegetable gardening. To greater or lesserextents the projects had to overcome constraints imposedby inadequate knowledge of commercial farming practices,insufficient financial resources, lack of contacts withneighboring farmers or extension agents, absence of marketingchannels, communication breakdowns, weaknesses infinancial planning and organizational management, problemsin community relations, and administrative complicationsarising from the status of refugees as welfareclients.The constraints proved surmountable, however, andexperiences of initial seasons served to improve subsequentperformance. Participants in the projects believedthat ultimate outcomes would prove successful. This beliefspread, and twelve new projects in eight more stateswere to begin in early 1984.As with sewing, none of the farming projects are oldenough to warrant judgments about effectiveness in promotingself-sufficiency. Rather, the ventures have thusfar served the basic purpose of pointing efforts in a directionthat might otherwise have been overlooked, particularlywith regard to assisting families whose adultmembers cannot easily find work in the regular job market.Small BusinessHmong entrepreneurship grew slowly. The number ofgrocery stores climbed from one in 1980 to over thirty in1984, but other business ventures tended to fail or stagnatemore often than they succeeded. Fifteen starts since1980 yielded only two clear instances of success by 1984.Most grocery stores were food-buying cooperativesestablished with the savings of kinship groups, and organizedby younger community leaders. Besides providingfamilies with Oriental foodstuffs at lower prices, cooperativesserved as social gathering places, as symbols ofcommunity cooperation, andat least in theoryas sourcesof financial return on community investment.Sixteen stores that functioned as cooperatives in1983 started with average investments of $20,000, andgenerated gross monthly sales of $5,400. Narrow operatingmargins permitted only six to offer paid employment,209

206 THE timoric IN TRANSITIONand five managed to generate small profits. The stores'contribution to self-sufficiency, therefore, took theform of providing families with expenditure savings, and,increasingly after 1982, of providing outlets for gardensand farm project output.The profit-oriented grocery stores were capitalizedfrom personal savings of employed refugees and their relatives.These stores began with investments 50 percenthigher than the cooperatives, had monthly sales fourtimes higher, employed more people, and usually yieldedprofits. The more viable status of these stores resultedfrom a variety of factors, including prior experience ofowners in working for other businesses, their higher educationlevels, and placement of the goal of profitabilityabove that of community service.Other types of businesses in operation in 1983 wereeither embryonic or stagnant. The embryonic businessesincluded a community credit union, a bakery, and a woodcraftsenterprise established with the help of two RotaryClubs. The stagnant operations included a jeweler, ahousecleaning service, a stuffed-animal subcontractor, asupermarket and a videotape store/theater. Earlier lotnturesno longer in operation included a restaurant, anauto repair shop, a medical equipment subcontractor, afishing gear subcontractor and a jewelry store.One success story began in 1981 when six Hmongpooled $80,000 in personal and borrowed funds to open arestaurant. They opened a second facility in 1983 withanother $60,000. By 1984 the two units employed eighteenpeople, and were training more for a third outlet underconstruction. Another success was a security guard servicethat began with a $350,000 loan to a Hmong accountantby American friends in 1981. After coming to withina week of bankruptcy in 1982, the service expanded rapidly.By 1984 the firm employed fifty-five people andplanned expansion to another state. Five stores and otherbusinesses received informal assistance from Americanor other refugee business people. Although most Hmongentrepreneurs also sought technical help from agencieslike the Small Business Administration, that help was notforthcoming. As in other projects, entreprer eurs movedahead at a pace dictated by the speed at which theylearned from experience about business operations.210

ECONOMIC DEVELOPMENT AND EMPLOYMENT PROJECTS 207Employment ProjectsA few Hmong may benefit from the enterprises mentionedabove in the near future, but most depend fortheir self-sufficiency on employment services that providejob training, development, placement and counseling.While these services have not served Hmong as well asthey've served others, in some cases, where innovativeapproaches were used, the employment services were quiteeffective.!n Minnesota a sponsorship agency worked with seventyvolunteers tied into one thousand churches. Besidesits ability to link with a large number of church-goingemployers, a major strength of the project was its assumptionthat every Hmong had a remunerative skill. Theproblem was to identify the skill clearly and then matchit with an employer. Placement costs in the state during1982-83 ranged from $600 to $1,200 per individual. Theproject placed 510 people at a cost of $270 per refugeein the same period.Hmong in Missoula, Montana, had high unemployment in1982 due to the recession. Working with a social serviceagency, a self-help association searched for job openingsin Billings, 350 miles away, and moved families there ata cost of $570 per job, including transportation subsidies.This regional approach to employment worked well,and about 80 percent of the families who moved were selfsufficientat the end of 1983.Focusing on the need to help Hmong become accustomedto the culture of the American work environment, a projectin the State of Washington combined on-the-job acculturationwith housecleaning and lawn-care work in a smallbusiness operation, and then placed "graduates" in regularjobs. Attention to the stresses imposed by workenvironments, a major impediment to Hmong employment,helped refugees retain their jobs. Organization of theproject as a service business kept net project costs to$870 per placement in 1983, about twenty-five percentless than the cost of other training placement efforts inthe state.Corporate leaders collaborated with media and serviceorganizations in Wisconsin to raise consciousnessamong employers of the benefits of hiring refugees whobelieved in the American work ethic and were more than211

208 THE HMONG IN TRANSITIONwilling to work diligently at the minimum wage. Businessmanagers, convinced by this argument, instructed theirpersonnel departments to open doors to Hmong and to helpthem adjust to the workplace. The project generated 140jobs in 1982 at an out-of-pocket cost of $21 per placementto the business people involved.While the cost-effectiveness of these employment effortsmust await future assessments, the important interimaspect is that creative talents are searching for formulasto benefit Hmong more quickly.ConclusionHmong and American project participants launchedthemselves into undertakings that were new to them. Somecame prepared with prior training or experience, but mostdid not. The Hmong had no exposure to the basics ofincome-generation in the United States. Most Americanparticipants knew little at the beginning about analyzingmarkets, merchandising, minimizing costs, using creativeplacement techniques, preparing plans and contingencies,etc.Lack of prior knowledge among participants suggestedthat, notwithstanding the slow pace of progress, recentefforts constituted important first steps in a difficultprocess of self-learning. Project difficulties seemedless important than the processes by which participantslearned from errors so that they did not repeat them, andlearned from successes so that they pressed these intoservice again.While learning from experience appeared essential,it was not enough. Projects needed outside assistance inorder to move forward more quickly. Unfortunately, manyparticipants were not able to pinpoint their own limitations,and could not describe the help they needed. Whenthey could, they did not know how to secure it. One compellingneed was for guidance in how to harness locallyavailable human resources for the benefit of a projectand its participants.This problem may be taking care of itself. Effortsof private and federal and state government agencies tobring together Hmong and Americans from projects aroundthe country, are yielding a network of experienced indi-212

ECONOMIC DEVELOPMENT AND EMPLOYMENT PROJECTS 209viduals who can assist each other over the course oftime. Workshops organized in 1983-84 brought participantsof sewing and farming projects together with others whowere planning new initiatives. A network of people involvedin different projects is beginning to take shape.This kind of process has historical antecedents.The National Alliance for Polish People provided insurancecoverage for members of its own ethnic group; Jewishgroups developed their own savings and loan associations.Hmong may face greater obstacles than their predecessorsin moving into the economic mainstream, but similaritiesin the paths taken by earlier migrants and now the Hmong,and the help being extended by Americans, provides causefor optimism. Optimism encourages continued effort, andeffort is the key to progress.REFERENCESReder, Stephen. 1982. "A Hmong Community's Acquisitionof English." The Hmong in the West: Observationsand Reports, Papers of the 1981 Hmong Research Conference,edited by Bruce T. Downing and Douglas P.Olney, pp. 268-303. Minneapolis: Center for Urbanand Regional Affairs, University of Minnesota.Robson, Barbara. 1982. "Hmong Literacy, Formal Education,and Their Effects on Performance in an ESL Class."The Hmong in the West: Observations and Reports,Papers of the 1981 Hmong Research Conference, editedby Bruce T. Downing and Douglas P. Olney, pp. 201-225. Minneapolis: Center for Urban and RegionalAffairs, University of Minnesota.2 ; 3

PART THREELanguage and Literacy214

INTRODUCTIONBruce DowningThe papers in this section reflect the interest of anumber of researchers in the Hmong language and the problemsHmong face in learning English.In the first area, the arrival in the United Statesof more than fifty thousand native speakers of Hmong openedup new opportunities for linguistic research. TheHmong language, virtually unknown in the West prior to1975, has provided a fertile field of investigation forlinguists interested in uncovering its structure and inadding a clearer understanding of this fascinating languageto our storehouse of knowledge about the whole arrayof extant human languages.On the more practical side, the fact that the Hmongcame to America in most cases with very little knowledgeof English and with little if any acquaintance with writingmeant that professionals in teaching English as asecond language, in adult hasic education, Ind in teachingpublic school children with limited proficiency inEnglish were faced with a new and challenging set ofproblems to be solved. These problems have drawn the attentionof both educational researchers and persons interestedin the social functions of literacy and of bilingualism.A good sampling of the work being done in these twoareas of research is represented here.The first group of papers, by Ratliff, Owensby,Jaisser, and Fuller, are all descriptive studies of Hmonggrammar. For non-linguists, reading about grammar mayseem as dull and as difficult as reading someone else'scomputer programs or trying to get through a study in astrophysicswhen you don't know physics and your only astronomyclass was 20 year ago. But the glimpses of theinner workings of the Hmong languagepresented here arewell worth the reading. Language is an aspect of culture,and we find here that the linguistic aspect of Hmong cultureis just as interesting, in its subtlety, its complexity,and its differences from a European language213215

214 THE HMONG IN TRANSITIONlike English as the other cultural characteristics of theHmong discussed elsewhere in this volume.Martha Ratliff's paper beautifully illustrates thelinguist's aim of discovering structurea system inwhat may at first appear to be only a welter of words.She takes a set of two-word expressions in Hmong, examplesof which are listed as "intensifiers" in Heimbach'sWhite Hmong Dictionary, and shows us that far from beingidiosyncratic and unpredictable, these expressions areconstructed according to very clear patterns. Like littlepoems, they follow strict rules for the arrangementof sound and syllables, and, also like poetic expressions,they serve to add nuances of expressive meaningreflecting sound, movement or the speaker's attitudestothe basic meaning of a sentence.Americans in many places are studying the Hmong languagein a practical way, hoping to be able to speak atleast a little of the language of their Hmong friends,co-workers, or students. A person doesn't get very farin the study of Hmong before encountering a grammaticalpattern very unlike English: sentences in which oneclause, with just one subject, contains two or moreverbs, not related as conjoined predicates with and or orto tie them together. Laurel Owensby calls this pattern"verb serialization," and her paper makes clear, withmany examples, what these sentences are and what theymean.Annie Jaisser addresses a puzzle in the analysis ofsome complex sentences of Hmong containing the word kom.(She refers to kom as a "morpheme," using the technicalword for a indivisible meaningful unit.) The question iswhether kom is a verb meaning 'tell' or 'command,' whoseobject is a clause, or whether it is a part of the subordinateclause, namely, a "complementizer" or subordinatingconjunction that begins that clause. She sorts outthe confusion with a series of neat arguments that leadto a clear answer: sometimes kom is a verb, but at othertimes it is in fact a complementizer introducing a newclause. The reader not only is led through this analysisstep by step, but along the way is shown quite a bitabout how "sentence embedding"--the construction of sentencesthat contain both a main clause and an embedded orsubordinate clause, is done in Hmong.216

INTRODUCTION 215The paper by Judith Fuller is concerned not onlywith description of Hmong but also with the question ofwhere this language fits with respect to a typology oflanguages that classifies them as "topic-prominent," likeChinese, or "subject-prominent," like English. In thiscase the answer is not so clear-cut; Hmong falls somewherein between, meaning that in Hmong the syntactic arrangementof a sentence is determined in part by the"topic"what a whole section of a discourse is "about,"and in part by consideration of which word is the grammatical"subject" of a sentence.The linguistic papers are illustrated with a largenumber of example sentences, written in the RomanizedPopular Alphabet (RPA), the writing system most widelyused by the Hmong, at least in the United States. Eachsentence is translated into English, first word by wordand then, on the next line, with a free translation intoidiomatic English. Grammatical terms used to identifyHmong words, e.g. comp. for "complementizer," are identifiedby each author in a footnote. The general correspondencesbetween the RPA orthography and the sounds ofHmong are summarized in the "Note on Orthography" at theend of this introduction.The last two papers of this section, although writtenindependently, form a set that can be read togetherfor a rather thorough introduction to questions of Hmongliteracy and in particular the learning and use of literacyin English.Gail Weinstein provides an overview of some of theissues concerning literacy as they have been discussed inthe literature pertaining to basic literacy for adults.This background material is related to the situation ofthe Hmong refugees through an appended report on herfirst-hand observations of the uses of literacy and the"social consequences" of literacy on the part of theHmong refugee community in Philadelphia.In the final paper, Karen Green and Stephen Rederpresent preliminary results of a detailed longitudinalstudy of the acquisition of both spoken English and Englishliteracy by Hmong adults in another part of theUnited States. Their analysis focuses on an attempt toidentify background factors and concurrent factors thatcan be statistically related to the rate of acquisition.Their results are important because so little of the pre-217

216 THE HMONG IN TRANSITIONvious research that informs theories of adult second languageacquisition has been concerned with the less educatedadults commonly found in third-world refugee populations.213

A NOTE ON HMONG ORTHOGRAPHYAll references to elements of the Hmong language inthis volume are presented in the Romanized Popular Alphabet(RPA), which has become the standard orthography forHmong at least in the West.The RPA was devised by American and French missionariesand introduced in Laos in 1953. It represents allof the consonant, vowel and tone distinctions of Hmongwith ordinary Roman letters, entirely avoiding specialsymbols and diacritics.Words in Hmong are generally monosyllabic. The orthographicrepresentation of any syllable has the followinggeneral characteristics:I. Each word includes one or two vowel letters (e.g., a;2, Li) ia) representing a simple or diphthongalvowel nucleus.2. Doubling of the same vowel letter (ee oo) indicatesnasalization of the vowel and possibly, depending onthe dialect, a syllable-final consonant sound likeorthographic -28. in English. (There are no other finalconsonants in Hmong.)3. Preceding the vowel there may be one to four consonantletters, indicating an initial consonant or consonantcluster. The letter h following another consonantindicates aspiration of the preceding sound,so that is similar to the English .2 in .at_ (incontrast with the unaspirated 2 in speak). An initialh (in the sequences hl) marks the soundit accompanies as voiceless. The letter n precedinganother consonant indicates prenasalization; it ispronounced as m before 2; n before 11 and _1:; andDI before k.4. A consonant letter as the last letter of a syllableor word represents the tone on which the vowel isproduced, as follows:21721,J

218 THE HMONG IN TRANSITION- b high tone-j high falling tone-v mid rising tonemid tone-s mid-low tone-g mid-low breathy tone- m low glottalized tone-d low rising tone (a predictable variant of m)Apart from this special use of final letters to marktone, the consonant and vowel letters generally havefamiliar international values, the principal exceptionsbeing:xszpronounced like English spronounced like English sh or the s in sugarpronounced like the English z in azurer a retroflex stop sound more like English t or dthan English rc a sound like the 1,-y. sequence in English fit youthq a sound like k or A but pronounced with the backof the tongue further back in the mouthw a vowel similar to that in English but or unstressedjustaw a diphthong similar to w but prolonged.One final caution to the reader: two distinct dialectsof the Hmong language are spoken by refugees in theUnited States. These are usually referred to as GreenHmong and White Hmong. In addition there are many regionaldifferences of pronunciation, vocabulary and grammar.The RPA orthography was designed to provide a singlestandard written form of each word; it will nottherefore correspond to an individual's actual pronunciationin many cases.220

TWO-WORD EXPRESSIVES IN WHITE HMONGMartha RatliffThe data which I have collected from three speakersof White Hmong and from Heimbach's dictionary reveal theexistence of a formally well-defined set of word pairsassociated with certain kinds of meanings. A formal descriptionof these word pairs is much easier to deliverthan an adequate account of their meanings. In general,they add immediacy and vividness to a description; theyusually reflect the sensory experience of the speaker orhis attitude toward what he is describing; and that sensoryexperience or attitude of te. determines the consonantsand tones and perhaps (but less clearly) the vowelsof the words chosen. A large set of related single wordsexists, such as dhev, 1.41. and Lithi, some of which participatein the word pairs described here. Both the wordpairs and the single words ace usually found in clausefinal position, and the functions and meanings of the twoseem to be similar. The single words are more abstract,however, and are much less likely to have a close connectionbetween meaning and sound.The term "expressive" is taken from the work ofGerard Diffloth (1972, 1976, 1979), whose insightful anddetailed descriptions of the expressives of some Mon-Khmer languages (Semai, Khasi, Sre and Bahnar) and theexpressives of Korean make it clear that a similar wordclass exists in Hmong.1 They have been called variously"descriptive words" or "aspect words" by Ts'ao (1958),"post verbal intensifiers" by Heimbach (1979), and "re:stricted modifiers" by Lyman (1974) in their descriptionsof Hmong. They are related to the "ideophones" of theBantu and other African languages; they exist in Japanese,some Dravidian languages and some Austronesianlanguages as well.Expressives are formally and semantically differentiatedfrom the language of propositions, the languageproper, where utterances can be reduced to predicate andarguments, and words participate in relations of synonymyand antonymy. They are more likely than nonexpressive21922i

220 THE HMONG IN TRANSITIONwords to reflect in their phonetic composition the speaker's(and his speech community's) perception of one aspectof the object being described: either its sound, itsappearance, another one of its innate characteristics, orhis attitude toward it. The relationship between wordand meaning is likely to be more iconic; that is, theword, to the speaker's mind, resembles the thing he isdescribing; zig zuag, because of the initial consonantsand the whispery tone, sounds itself like "walkingthrough high grass." The relationship between word andmeaning in the majority of words in any language is, onthe other hand, predominantly arbitrary and conventional:any string of sounds would suit equally well to expressthe meaning of "house," for example, and it is only historyand custom which determine that that string shouldbe tsev.Iconicity in Hmong, or the direct reflection ofmeaning in phonologial representation, is also apparent(1) in reduplication of nonexpressive words to convey theidea of augmentation; (2) in the vowel gradation in theclassifiers of Weining Hmong described by Wang (1957);and perhaps, too, (3) in the vowel gradation and tonepatterns in the second and fourth member of the foursyllablecoordinative constructions, or lus ua txwm("paired words") described by P'an and Ts'ao (1958),Mottin (1978), and Johns and Strecker (1982). Since inthese instances an effort is made to bend the prosaiclanguage to mirror the natural world, a language whichhas a well-developed expressive system can be expected tohave other kinds of "word play," or evidence of the languagecraftsman's skill: this is certainly true of Hmong,in which pig latins (secret languages) abound, and ahighly sophisticated and elaborate oral literature exists.The class of expressive words in Hmong is indeed aclass separate from noun, verb, adjective, adverb andother classes Westerners are comfortable with. Ts'aofelt Chinese scholars also needed to be warned about thenature of this word class which has no dear analogue inChinese, except perhaps in Chinese color-word modifiers.Expressives have their own roots, and are only accidentallyhomophonous, on occasion, with words of anotherclass. Expressives are usually found at the ends ofclauses, outside the Subject Verb Object (SVO) matrix and222

TWO WORD EXPRESSIVES IN WHITE HMONG 221hence isolated from the structure of the proposition.Onomatopoeic expressives in Hmong, at least, are introducedby what have been called in other languages "carriers,"words which effectually underscore this separationfrom the rest of the sentence: these are nrov("sound") verbs of sound-making, e.g. quaj ("cry"), hais("speak") and poob ("fall"). The two-word expressivescan be said to have their own morphology; that is, thetwo words have a characteristic shape which involves thebinding of their respective partstheir consonants,vowels and tones, which, despite the creative element inherentin all expressive language, is part of the systemand cannot be altered at the whim of the speaker. Inthese respects, Hmong expressives are like those of Semai(Diff loth 1976), Korean (Martin 1962; Diff loth 1972; Kim1977) and Bini (Wescott 1973). Most importantly, thesewords are part of every speaker's everyday speech; theyare not reserved for poetry, song or ritual, althoughsome individuals, those considered to be good storytellers,may be known to make more extensive use of themthan others. The meanings of the expressives I have collectedare overwhelmingly onomatopoeic, but I am not convincedthat this is a representative sample of the wholecorpus of Hmong expressives. Heimbach's appendix listsonomatopoeic phrases as a subclass of post-verbal intensifiers.According to Ts'ao, more Mhu expressives are"descriptive" than onomatopoeic. Based on his studies ofthe Mon-Khmer languages, Diffloth (1976, p. 262) has observedthat most expressives do not describe noises.Nonetheless, sounds are easy to demonstrate and discussin an interview, and therefore a disproportionate amountof time was spent collecting them. Although I probed forother kinds of expressives, I generally discovered themindirectly, i.e., when I was told that an onomatopoeicexpressive could also be used to describe the appearanceof something, or someone's behavior. For example, disdaws was first explained as a kind of continuous sound,perhaps because we had been talking about sounds, andsecondarily as the appearance of a long line of people.The multiple associations I was given for some ofthese expressives are fascinating, but are also a nightmarefor the lexicographer. These words do not have definitionsin the same way words in the prosaic part of thelanguage do, and they are almost impossible to translate.223

222 THE HMONG IN TRANSITIONThis seems to be true of expressives generally:The meanings of expressives seem to be extremelydetailed and idiosyncratic, describing a situationperceived as a whole, as an independentclause would. On the other hand, the same expres-situa-sive can be used to describe a variety oftions which at first glance seem to be quite differentbut share a common core which could be definedas a cluster of elementary sensations. Forinstance, /klimare:1/ is used to describe anarrow or knife stiffly vibrating after embeddingitself into a piece of wood; it can also describethe walk of a tall, skinny old man. The clusterof sensations common to both meanings (and recognizedby informants) are: stiffness, perpendicularity,and repeated small oscillations. Miff -loth 1976, p. 257]Accordingly, at this early stage in the study of Hmongexpressives, in Heimbach's dictionary and in the data inthis paper, either a very specific association will begiven, e.g., t..4x11 pcsi, "a rat or mouse crying out in asnake's mouth," or a very vague, general abstraction willbe given, e.g., suj. gees, "perseveringly." Neither givesenough information for the expressive to be well understood.There is a question as to whether the meaning ofan expressive can always be reduced to "a duster of elementarysensations," with all associations having equalstatus with regard to each other. Perhaps one sense isbasic and the others derived. For example, hawv huav hasthese associations: (a) sound of animals growling, readyto bite, (b) human panting, and (c) nervous, tense pacing.The sense of sound may be basic and apprehension ofthe related behavior may be secondary; similarly, the behaviorof the animals may be basic and the behavior ofthe humans may be perceived as animal-like. We knowenough of expro;sive meaning nonetheless to discontinueuse of the te..n "intensifier." These words particularize;they are eyewitness accounts of the passing sceneand, as such, give immediate apprehension to stories andcharacterizations. They do not intensify in the sense ofgiving more of some aspect of meaning that is already inthe utterance without the expressive. That is the func-22 4

TWO WORD EXPRESSIVES IN WHITE HMONG 223tion of reduplication in Hmong.The two-word expressives in Hmong have a characteristicmorphology which makes them easily recognizable.The two words have identical initial consonants or consonantclusters, but dissimilar vowels. Either the twotones will be the same, or the first will be high falling(-j) and the second any other tone. If the tones of thetwo words are identical, the vowel of the first word willbe /44; if the tone of the first word is high fallingand the tone of the second word is some other tone, thevowel of the first word will be /-u-I. From the semanticassociations I was given by my tutors, there seems to beclear tonal iconicity in the more well-represented tonalpatterns. For that reason, the data are organized bytonal patterns, and rough impressions of the meaningrange for each group are given.2 The pattern/ A isassociated with low-pitched, echoic, hollow, diffuse orairy sounds: di& du& "boiling of thick liquid"; RR&Ipaulyv "a bird rising from its nest on the ground"; vivv_%, "the roar of a fire; strong wind; fast traffic."-j/-j is associated with energetic, fast, short soundswhich are not necessarily sharp, high or falling: E1f_b* "a bird or plane whizzing by"; nthij nthooj, "cow'sor children's heads butting"; ntsij ntsiaj, "pushingpullingbolt on an M-l6"; tij tauj, "raindrops." -s/-sis associated with flat, continuous, unending sounds andsights: dis claws, "a continuous mixed sound; the appearanceof a long line of people"; lis loos, "bees buzzing."-j/-v has a double orientationback and forth, in andout: dhuj dheev_, "whimpering on and off; rising from andfalling back into unconsciousness; almost remembering,then forgetting again"; nyuj nyav, "whining; smile withdouble meaning; bittersweet feeling of longing"; gui gsb"the sound of a creaky door." -j/-s suggests both backand forth (-j/-v) and level and steady (-s/-s): lui laws,"a big, continuous humming sound; continual complainingto oneself"; luf vias, "unsteady walk; going back andforth on a swing." I have no sense of the nature of theiconicity, if any, associated with the minor tonal patterns.As a check to the validity of my generalizationsabout the meanings associated with particular tonal patterns,I asked my tutors if they knew of two-word expressivesfor certain meanings I proposed, which I guessed225

224 THE HMONG IN TRANSITIONwould be typical of certain tonal patterns. My predictionswere not always right, and usually there was no expressivefor the meanings I proposed, but if one did existit matched my prediction with more than chance frequency.For example, I predicted thunder would have azeri pattern (ail ntwi), the movement of a snake av pattern (Luj, xuav), hiccups either a -j/-j or apattern (and I was given both 11 au' and iv awv!).Needless to say, I did not tell my tutors which pattern Iwas expecting.The -j/ -v or "double orientation" pattern is especiallyinteresting because of its analogues in Semai andBini, a Nigerian Kwa language. In Semai, if there isvowel disharmony between corresponding morphemes in atwo-word expressive, the associated meanings have to dowith randomness, sloppiness, or imbalance, e.g., klicwbcklicwec"random kicking, as of a turtle in sand" (Dif f-loth, personal communication). In Bini, polysyllabicexpressives with alternating high and low tones are associatedwith "irregular shape or motion," for example theexpressives for "staggering," "fluttering," "twisted,""crippled" and "jerky" (Wescott 1973, pp. 200-201). Theiconic relationship here between irregular form and irregularmeaning can be said to be diagrammatic; that is,the relation of the parts of the object, motion or soundrepresented corresponds to the relation of the parts ofthe phonological string.In Hmong two-word expressives, then, tones are meaningfulnotonly in their power to distinguish morphemes,but in themselves. Linguists have been aware for a longtime that parts of morphemes can be meaningful, and canstand in relation to the other parts of the same morphememuch as words stand in relation to ont another in a sentence(Bolinger 1950, lakobson 1966, McCarthy 1983).When we look beyond the White Hmong dialect to theHmongic family, the details of expressive morphology heredescribed do not seem as important as the fact that thisword class, with these kinds of meanings associated witha pattern of some kind, exists. Ts'ao's description ofthe Mhu of Lushan in eastern Guizhou province in Chinareveals the existence of both one- and two-word expressiveswith meanings reminiscent of the expressives ofWhite Hmong. While the tones of the words in each twowordexpressive reported in Ts'ao's article are identi-226

TWO WORD EXPRESSIVES III WHITE HMOIIG 225cal, the words have identical vowels and dissimilar initials,the mirror image of the morphology of the WhiteHmong expressiveslThe connection between meaning and tone is often reinforcedand refined by consonantal iconicity. I can onlycite isolated examples of this, where my perception ofthe phenomenon being described matches some aspect of theacoustic or articulatory properties of the consonant involvedor where expressives with similar meanings can becompared. As yet, I see no system of consonantal icon-:city such as Diffloth describes for some Mon-Khmer languagesand Korean, but that does not mean such a systemdoes not exist. Zero-initials seem to involve cries andinvoluntary sounds originating from within with no oral(consonantal) impediment: ig avIA, "wild pigs fighting inclose combat, a gobbling sound"; ii al, "babies crying";ii a21, "hiccups"; iv awv "dogs' hunting bark; hiccups."The "liquid" consonant /-1-/ may be actually felt to beassociated with water; witness the contrast between kipauj, "fruit falling on the ground," and AIL ploovp"fruit falling in water."3 The contrast between Isi&gpa "monkeys jumping through the treetops, separateabrupt jumping noises," and lig zuag, "monkeys swishingthrough the treetops, continuous sound," lies only in theiconicity of the consonants the initial of the former isan affricate, which consists of a stop (hence the abruptjumps) and a fricative, whereas the initial of the latteris a simple fricative, one of the continuant class (hencethe continuous swishing). The idea of heaviness may beassociated with I- 9/, witness pl pooh, "bomb impact";tig tool, "heavy footsteps"; lig vamp "a herd of animalsstampeding"; and 21k ploov, "fruit falling in water.%Many such guesses can be made upon inspection of the data.The vowel of the second word has not yet been analyzed,but is also a likely factor in the iconic equation.The exceptions to the canonical form of the two-wordexpressive involve phrases which have characteristicallyexpressive meanings but which either do not have identicalinitials or dissimilar vowels, or do not have one ofthe tone patterns described above. It may be that the inventoryof tone patterns will have to be increased andrefined when more examples are collected. I doubt,though, that the observation that these words have iden-. 7 ' 227

226 THE HMONG IN TRANSITIONtical initials will have to be modified; in most instanceswhere the two words of a phrase had differentinitials, one or both of the words were found to be nonexpressivewords being used metaphorically.For example,nrov ncha ntws, a loud sound which "flows" or reverberates,contains ntws, which can also be a regular verbmeaning "to flow" and which my tutor believes to be averb in that context. I expect that we may find otherexplanations which will account for the few remainingforms which appear to be exceptions.In the expressive domain, where meaning is so elusiveand so intimately involved with culture and worldview, it is necessary to have the insight that only nativespeakers can bring to the problem of isolating theessence of each expressive and of describing the semanticrange of the class as a whole. As an outsider, I can onlycomment with confidence on those formal Patterns whichforced themselves on my attention while gathering thedata. An analysis of the one-word expressives will benecessary, too; for their own sake and because they oftenserve as constituents of the two-word expressives. To myknowledge, no work at all has been done on them. Heimbachhas performed an invaluable service in simply listingthem in his dictionary, wisely not attempting to analyzehow many of them alter the meaning of the sentences inwhich they appear. Native speakers have a hard time doingthis, too. It may be that the one-word expressives arevery old forms and whatever clear meaning they once hadhas been lost through constant use and increased abstraction.One clearly delimited expressive group serves tofine-tune colors; it would be a pleasure to have a preliminaryanalysis of the way this is done.ILLUSTRATIONS(The following are associations or [possibly incorrect]abstractions; they are not definitions. "+" indicatesthat an expressive is partially or solely associated witha sense other than sound. "*" indicates that an expressivemakes use of phonological contrasts not usually employedin the language.)-b/-b: (high level/high level) high pitched, shortsounds.228

TWO WORD EXPRESSIVES IN WHITE HMONG 227nplhib nplheebplib plebzib zeba) silverware rattling (or othermetal on metal);b) pin coming out of a handgrenadewood crackling on a firea big pig fightingzgrE, (falling "whispery" / falling "whispery") productive;low-pitched, echoic, hollow, diffuse orairy sounds.ig awgdig dugkig kuagmig mogmlig mlogntig ntwgpig poogplig plawgplig plogsig suagtig toogtsig tsuagvig vagvig voogvig vwgzig zawgwild pigs fighting in close combat(gobbling sound)boiling of thick liquid like oatmeal,corn mash (thick, ponderousbubbles)mice feet on an empty boxa) a cow mooing;b) tigers playing(associations unclear)thunderbomb impact (cf. n_t1 ntoj)a bird rising from its nest on thegroundsomeone jumping into waterrain on a thatch roofheavy footstepsa) a downpour;b) monkeys jumping through thetreetopsdiscontinuous (cf. zazuag)the sound a tree makes as itbrushes through other trees andgrowth while fallinga herd of animals stampeding,running scared (cf. nqaj riga)a) the roar of a fire;b) strong wind;c) fast traffic;d) small airplane motora) asthmatic breathing;b) snoring;c) panting (cf. hawv huav)229

228 THE HMOMG IM TRANSITIONzig zuaga) walking through high grass(cf. rhij rhuaj);b) drizzling rain;c) monkeys jumping through thetreetopscontinuous sound (cf.i& ga ):Li: (high falling/high falling) very productive; energetic,fast, short sounds (not necessarilysharp or high); perhaps surface contact asopposed to contact and penetration; not necessarilyfalling sounds.ij ajiii)babies (baby animals?) crying (cf.ij awjhiccups (cf. iv awv)dhij dhujbreathing with a full nosefij fwjbird or plane whizzing by (cf. mjjmew j)*gij gaj babies crying (cf. i1 j)hij haja) laughing;b) fightingmij mejmosquitoes or other insects flyingaround your ear*mij mwj bullet whizzing past (cf. fwj)nphij nphooj a) pop of bamboo in fire (cf. .gii.(=*bhij bhooj) Ac2bb) sound of a lot of popcornpopping (full pan, not much room)(cf. 211j. 2.j)nplij nploja) foot pulling out of tar (cf.(=nplij nplooj)txej);b) pop of chewing gum;c) clap of bamboonplij npluaj(associations unclear)ntij ntawjfast, sharp typirg(=nrij nrawj)ntij ntojlow impact sound, as cut lengthsof bamboo rolling down a mountain(cf. Ai.j. 212i, nqaj mug)nthij nthoojheads butting (cows, children...)ntsij ntsiaj pushing-pulling bolt on an M-16230

TWO WORD EXPRESSIVES IN WHITE HMONG 229pij paujhail or fruit falling to thegroundplij plej just a little popcorn popping (alot of room in the pan) (cf.nphij 1202o.)plij ploja) bullet impact (cf. ntil ntoj);b) bamboo bursting (cf. nphijnphooj)plhij plhLwj a) helicopter propellers;b) birds making short flights(e.g. pigeons flying from oneroof roost to another)+rhij rhejlight and sound combination!a) lightning (light and crack);b) small amount of water on greaseor vegetables frying (cf. rhujrhev):110 rhuaj foliage or grass rustling as aperson or animal walks (cf. zigzuag)tij taujraindropstxij txeja) rat or mouse crying out in asnake's mouth;b) foot pulling out of mud (cf.mill nplo(o)j)vij vawjdogs barking at something (cf. ivawv)-s/-s: (low level/low level) clear group, although notlarge; flat, continuous, unending sounds andsights.is wsis asthis chaus+dis dawsa) everyone saying "ws";b) speaking in a monotone;c) little pig running after itsmother(associations unclear)(associations unclear)a) a continuous mixed sound: e.g.,a group of monkeys chattering andjumping around, heard at a distance;many people walking on amixed surfacetwigs, rocks, soil;231

230 THE HMONG IN TRANSITIONlis loosmlis mlasb) the appearance of a long line ofpeoplebees buzzing (cf. ILA les)cat meowing (cf. mluj mlob)-v/-v: mid rising/mid rising) vowel of first wealso be /-aw-/.iv awvdhawv dhev+hdwv huav+khawv khuavpliv ploovqiv qawvqhuav qhawvtxiv txevxyav xyova) dogs' hunting bark (cf. Ifilva_ wi);b) hiccups (cf. !jam')whimpering; separate whimpers(cf. dhuj dhev, dhuj dheev)a) sound of animal growling,ready to bite;b) human panting (cf. zag)v ;c) nervous, tense pacingsteady runa) ducks diving underwater;b) submerged bottle filling upwith water;c) fruit falling in water (cf. pi jmid)manner of grinding teeth(associations unclear)rats or birds chirping (cf. ntujntiv)associations unclear)-j / -b: (high falling/high level).+chuj chiabm(uj mlob+ncuj nciabzuj zebworking absentmindedly (cf. nrhujnrhawv)a) whining (cf. nyuj nyav);b) cat meowing (cf. mlis mlas)grudginglylittle hungry pigs squealing(nigh falling/falling "whispery").232

TWO WORD EXPRESSIVES IN WHITE HMONG 231duj dog+nqaj nqugzuj ziaga) the sound of a number of treesfalling one right after anotherand their appearance, overlaid,on the forest floor;b) a big herd of cows stampeding(cf. yigloog);c) everyone talking at the sametime in small groups;d) bamboo rolling and slidingdown a mountain (cf. ntij ntoj)a) pulling a metal chain oversomething;b) a cicada singing-j/ -m:(high falling/low glottalized).duj duamqhuj qhemmanner of walking (?)(associations unclear)-j/-s: (high falling/low le\.e1) suggests both back andforth (-j/-v) and level and steady (-s/-s).+luj lawsluj les+nkuj nkaus+nkhuj nkhoos+nplhuj nplhoosa) big, humming sound, continuous(e.g. an electric generator)b) continual complaining to oneself;c) hais luj. laws = to speakwithout expression, but notwithout feelinga) bees buzzing (cf. lis loos);b) airplane;c) vacuum cleanerback and forth (e.g., nthos nkujnkaus = to snatch and put backquickly)a) sound of chopping dead orhollow tree (or log) with an axe;b) sound of a dog gnawing a bone;c) appearance of an old personwith a bent backa) sound of a boar grunting;233

232 THE HMONG IN TRANSITION+quj gees+vuj viasb) putting off duty, doing thingsto avoid performance of dutya) perseveringly;b) gradually building (e.g., ch:rngees)a, unsteady walk;b) going back and forth on a swing(movement)-j / -v: (highbackfalling / mid rising) double orientation:and forth, in and out, up and Gown.dhuj dhev+dhuj dheev+khuj khuavnruj nreev+nrhuj nrhawvntuj ntiv+nyuj nyavquj qev+rhuj rhevrhuj rhuavcontinuous whimpering (cf. dhujdheev, dhawv dhev);irtrhimpering on and off (cf.dhuj dhev, dhawv dhev)b) rising from and falling backinto unconsciousness;c) nco dhuj dheev = almost remembering,then forgetting againdeliberate, slow mannera) sound of tree popping (fast)before it falls (cf. nrhujn rhawv);Srsound of tight bow stringtwanging after arrow is releaseda) sound of tree popping (slow)before it falls (cf. ma' nreev);b) with difficulty, slowly;absentminded walking (cf. chujchiab)rats eating (cf. txiv txev)a) quaj nyuj nyav = whining (cf.mlob);b lung nyuj nyav_ = smile withdouble meaning;c) bittersweet feeling of longinga creaky doorlightning and thunder (light;pause; sound) (cf. rhij.a) sound of cutting vegetation;b) sound of bird shufflingthrough leaves looking for insects234

tuj tauvtuj tev+xuj xuavTWO WORD EXPRESSIVES IN WHITE HMONG 233a) water dripping from a leakyroof (inside the house);b) the sound of a voice engagedin flatteryunclear speech; muttering, butnot necessarily all low pitcheda) a long, easy, all-day rain;b) the undulatilg movement of asnake moving slowlySome apparent exceptions to the canonical form of thetwo-word expressive:+daws duam+nqhawv nqho+nrawv nroos+nrawg nroosqos qawvmus daws duam = manner of walkingITquaj nqhawv nqho = animalsgrowling and panting;b) eua nqhawv nqho = to doeverything in a hurryone after anotherequally welltorn hniav gos qawv = manner ofgna.;hing teetACKNOWLEDGEMENTSI am grateful for the suggestions of GerardDiffloth, who taught the seminar on expressives at theUniversity of Chicago in the spring of 1983, in conjunctionwith which I began th:s study. I would like tothank my two primary Hmong tutors, Xab Xyooj and KuamYaj, for their patience, hard work and good humor. Iwould also like to thank Nancy L. Dray, Miriam Lykke andDavid Strecker for their criticisms of an earlier versionof this paper.235

234 THE HMONG IN TRANSITIONNOTESIDif floth's intellectual debts, and therefore mine, canbe traced back through Jakobson (1966) to the Americanphilosopher Charles Sanders Peirce (1931), whose studyof the various relationships between .form and meaningincludes the notion of iconicity, which is of key importancehere.2It may be accidental, but onomatopoeia predominates inthe patterns where the tone is identical in the twowords, and other senses and attitudes, or multiplesenses, are much better represented in the patternswhere the tones are dissimilar.3See also 2J plop (someone's body hitting the water).Two other associations with pliv ploov also involvewater: the sound of ducks diving underwater and thesound of a submerged empty bottle filling up with water.REFERENCESBolinger, Dwight. 1950. "Rime Assonance and MorphemicAnalysis." Word 6:117-36.Diffloth, Gerard. 1972. "Notes on Expressive Meaning."Papers from the Eighth Regional Meeting of the ChicagoLinguistic Society, pp. 440-47. Chicago: ChicagoLinguistic Society.. 1976. "Expressives in Semai." AustroasiaticStudies, Oceanic Linguistics Special Publication No.13, pp. 249-64. Honolulu: The University Press ofHawaii.. 1979. "Expressive Phonology and Prosaic Phonology."Studies in Tai and Mon-Khmer Phonetics andPhonology, edited by Thongkum et al. pp. 49-59.Thailand: Chulalongkorn University Press.Heimbach, Ernest E. 1979. White HmongEnglish Dictionam,revised edition. Linguistics Series IV. SoutheastAsia Program Dc..a Paper No. 75. Ithaca: CornellUniversity, Department of Asian Studies.236

TWO WORD EXPRESSIVES IN WHITE HMONG 235Jakobson, Roman. 1966. "Quest for the Essence of Language,"Diogenes 51: 345-59.Johns, Brenda, and David Strecker. 1982. "Aesthetic Languagein White Hmong." In The Hmong in the West:Observations and Reports. Papers of the 1981 HmongResearch Conference, edited by Bruce T. Downing andDouglas P. Olney, pp. 160-69. Minneapolis: Centerfor Urban and Regional Affairs, University ofMinnesota.Kim, Kong-On. 1977. "Sound Symbolism in Korean." Journalof Linguistics 13:67-75.Lyman, Thomas Amis. 1974. Dictionary of Mong Njua. TheHague: Mouton.McCarthy, John J. 1983. "Phonological Featt.res andMorphological Structure." Papers from the Parasessionon the Interplay of Phonology, Morphology andSyntax, pp. 135-61. The 19th Regional Meeting ofthe Chicago Linguistic Society. Chicago: ChicagoLinguistic Society.Martin, Samuel E. 1962. "Phonetic Symbolism in Korean."American Studies in Altaic Linguistics, pp. 177-89.Bloomington: Indiana University.Mottin, Jean. 1978. Elements de grammaire Hmong blanc.Bangkok: Don Bosco Press.Pan Yuan-en and Ts'ao Ts'ui-yun. 1958. "Four SyllableCoordinative Constructions in The Miao Language ofEastern K weichow." In Miao and Yao LinguisticStudies, edited by Herbert C. Purnell, pp. 211-34.Ithaca: Cornell University Department of AsianStudies.Peirce, Charles Sanders. 1931. Collected Papers. Cambridge:Harvard University Press.Ts'ao Ts'ui-y.un. 1961 "A Preliminary Study of theDescriptive Words in the Miao Language of EasternKweichow." In Miao and Yao Linguistic Studies,237

236 THE HMONG IN TRANSITIONedited by Herbert C. Purnell, pp. 187-210. Ithaca:Cornell University Department of Asian Studies.Wang, Fu-shih. 1957. "The Classifier in the Wei NingDialect of the Miao Language in Kweichow." In Miaoand Yao Linguistic Studies, edited by Herbert C.Purnell, pp. 11145. Ithaca: Cornell UniversityDepartment of Asian Studies.Wescott, R. 1973. "Tonal Icons in Bini." Studies inAfrican Linguistics 4: 197-205.238

VERB SERIALIZATION IN HMONGLaurel OwensbyAs fellow linguistics students and I began studyingthe Hmong language, looking at folk tales (Johnson 1981)and working with a bilingual informant, I remember oftenbeing surprised at the way verbs seemed to be "strung together."My surprise at finding so many verbs within asingle sentence has developed into this report on verbserialization in Hmong.The phenomenon of verb serialization has been describedas a feature of many of the world's languages.Verb serialization has been defined by Isaac George(1975-76) as "essentially a surface sentence containing arow of two or more verbs or verb-phrases without any connectiveword between them."Charles Li and Sandra Thompson, in their work onserial verb constructions in Mandarin Chinese (1973:97),use the following definition:Serial verb sentences consist of a subject followedby two predicates where the first subjectis the subject of both predicates.INPSubjectV (NP)PredicatelV (NP)Predicate2William Welmers, in his studies of African languages(1973), found that serial verb constructions are used foractions which are closely related to each other, and thatserialized verbs are often translated by one English verband a preposition or sometimes just one verb.I have used three criteria in identifying examplesof verb serialization in Hmong: one subject, with two ormore verbs or verb phrases, and no conjunction betweenthem. Examples thus identified seemed to divide up intovarious distinct types.What at first glance seemed the simplest type ofserial verb construction involves what linguists call reduplication.This is when the same verb is repeated. Re-237239

238 THE HMONG IN TRANSITIONduplication occurs frequently iii Hmong, with adjectivesand adverbs, and with verbs as well.Reduplication is not true serialization, accordingto the usual more narrow definitions, because the twoparts make up a single, compound verb. Thus, it does notsatisfy the definition of Li and Thompson, cited above,which requires that there be two distinct predicates fora single subject.Reduplication of verbs in Hmong may serve a varietyof purposes. It may mark an action repeated over and overagain, an action in progress, or an action that occursoften. The function of reduplication of verbs may varyaccording to semantic classes of verbs.Certain verbs in English, like slap, jump, cough andshout, signify a single occurrence of an action. When wehear "I sneezed," for example, we assume one action, in abrief period of time, unless it is marked otherwise, like"I sneezed and sneezed" or "I was sneezing." Verbs ofthis type can be reduplicated in Hmong to show a continuingaction, rather than a single occurrence of an action,as in examples (1) to (3) below.(1) Mas nroj niam nplawm nplawm thiab ntaus ntausPart. weeds slap slap and hit hitnplej thiab pob kws.rice and corn"The weeds are fighting with the rice and corn."(From the folk tale, "The Fist Farmer")(2) Nws dhia dhia.she jump jump"She jumped many times."(3) Kuv tau gw.I past shout shout"I shouted (for a longReduplication may also be used to show that an actionis in progress, while something else occurs, as illustratedin (4).2,1ü

VERB SERIALIZATION IN HMONG 239(4) Thaum nws tseem cheb cheb tsev muaj ib tugthen she action in sweep sweep house have one clf.progressneeg muaj txiag tuaj txog ntawm nws lubman have money come arrive there her clf.qhov rooj.door"While she was sweeping, a rich man came to herdoor." (From the folk tale, "The Woodcutter")Some verbs are not usually reduplicated, like mi.'to eat.'(5) *Nws noj noj.He eat eatOur Hmong informant said this sentence would be acceptableif it had something else before it that happenedwhile he was eating, like nws txham 'he choked.' Thus nwstxham nws not rIcij, could be used to indicate "He chokedwhile he was eating."The reduplication of a verb can also mean that anaction is something that's usually done:(6) Nws ua ua paj ntaub.She make make pa ndau."She makes pa ndau (more than she does anythingelse)."Another type of construction, this time an instanceof true serialization involving two distinct verbs, isused with instrumentals. In Hmong, as well as in otherlanguages, two verbs are used with instrumentals, ratherthan a verb and a preposition, as in the English "do(something) with (something)." This construction is illustratedin (7) to (9).(7) Tus neeg caum nqaij xuas phom tua tus noog.clf. man chase meat use gun shoot cif. bird(hunter)"The hunter shoots the bird with a gun."241

240 THE HMONG IN TRANSITION(8) Kuv tus tij lauj siv hlua nce ntoo.My clf. older brother use rope climb tree"My older brother climbs the tree with a rope."The verb muab 'to take in hand' is used frequentlywith instrumentals also.(9) Muab hmuv rzit_gi ntiaj teb ...take spear stab earth"Stab the earth with a spear ..."This verb muab is found frequently in other serialverb constructions with an object, as in (10):(10) Ces nkawd txawm muab pob zeb dov scum roobso they2 then take stone roll top mountainlawm nram kwj ha.compl. down valley."So they rolled the stone down the mountain."It occurs before an object that is being manipulatedsomehow by first picking it up, which is then followed byanother verb. This is possible even when the object isnot expressed.(11) (they) tau muab txhoov ua tej daim colpast take cut up make many pieces takemus tso rau ghov txhia chaw.go put to clf. all place"They chopped it into pieces and put themeverywhere."Notice that in this last example there are two series ofthree verbs each, all associated with a single subject.(Examples (9) to (11) above are from the folk tale, "TheFlood.")The most frequently occurring type of verb serializationI've observed is the use of the verbs mus 'to go'and los 'to come' after verbs of motion. These verbsmark the direction of the action, toward or away from thespeaker or point of reference, as shown in (12) and (13).242

VERB SERIALIZATION IN HMONG 241(12) liga taw mus.carry wood go"Take the wood."(13) Nza taw los.carry wood come"Bring the wood."This use of mus and los tells us where the speaker is inrelation to the listener.A somewhat different example is (14).(14) Nws mus nyooj lawj ua av meet; nkaws los.He go growl make earth shake come"The tiger went and growled and made the earthshake."In this example, because it is a story being told, thepoint of reference is the participants in the story itself.The tiger went away from the other animals andgrowled. He made the earth shake and the shaking reachedthem, as indicated by los.Mus and los after a location are used to mark fromwhat place and to what place an action is going. Considerthe sentences (15) and (16).(15) Nws nisag hauvShe crawl inside"She crawled out(16) Nws nkag hauvShe crawl inside"She crawled intotxaj mus.room goof the room."txaj.roomthe room."This pair of examples indicates that mus after a locationmeans that the subject is leaving from that place. Withoutthat indication of direction, the location is assumedto be the goal of the action.Los means not only 'come' but 'come to a place whereyou belong or where you've been before.' Compare sentences(17) and (18):243

242 THE HMONG IN TRANSITION(17) Koj mus qhov twg los?You go where come"Where are you coming back from?" (to someone inyour household).(18) Koj Los qhov twg los?You come where come"Where are you from?" (to someone you don't knowvery well, but you knew a long time ago).This meaning of los is in contrast with tuaj which alsomeans 'come' but doesn't imply where something belongs:(19) Koj tuaj qhov twg tuaj?You come where come"Where are you from?" (to a stranger).My study of verb serialization has involved, first,finding examples of sentences with two or more verbs orverb phrases after a subject. Second, I have tried toclassify them by type: reduplication, and "true" serializationinvolving instrumentals and directional markers.Further study is needed of verb "compounds" that don'tyet fit neatly into a'tegories. Yet verb serializationcan definitely be seen as an important characteristic ofsentence structure in Hmong, with a variety of syntacticfunctions.ACKNOWLEDGEMENTMr. Tong Vang has been an invaluable consultant inthis data collection and interpretation.NOTE1NP indicates noun phrase and V = verb. Other abbreviationsused in the glosses of the example sentences canbe explained as follows: part. = participle, clf. =classifier, they2 = third person dual and comp!. = completiveaspect.244

VERB SERIALIZATION IN HMONG 243REFERENCESLi, Charles N., and Sandra A. Thompson. 1973. "SerialVerb Constructions in Mandarin Chinese: Subordinationor Co-ordination?" In You Take the High Nodeand I'll Take the Low Node: Papers from the ComparativeSyntax Festival. A paravolume to Papers fromthe Ninth Regional Meeting of the Chicago LinguisticSociety, edited by Claudia Corum, T. Cedric Smith-Stark and Ann Weiser, pp. 96-103. Chicago: ChicagoLinguistic Society.George, Isaac. 1975-76. "Verb Serialization in Kwa-TypeLanguages." Dissertation Abstracts International 36:3633.Johnson, Charles. 1981. Hmong Folk Tales.Linguistics Department, Macalester College.St.Paul:Welmers, William E. 1973. African LanguageBerkeley: University of California Press.Structures.245

THE MORPHEME KOM: A FIRST ANALYSISAND LOOK AT EMBEDDING IN HMONGAnnie JaisserThe intent of this paper is to give a syntactic andsemantic analysis of the morpheme kom in Hmong in an attemptto gain insights into sentence embedding.' In hissection on subordinate clauses, Mottin (1978) has describedkom as the con junction "so that, in order that"heading a resultative clause. Heimbach (1979) has describedkom as a verb meaning "mouse, bring into effect."Bertraisn979) listed kom as a verb meaning "make,want," followed by a subordinate clause. Yang Dao (1980)has described kom as e" a conjunction expressing "sothat, in order that" or verb, "want," followed by asubordinate clause. These descriptive statements (demonstratedby example sentences) afford no coherent syntacticanalysis. In an attempt to arrive at such an analysis,I looked at the example sentences given by theseauthors as well as data found in the folk tales (Johnson1981) which we used in a seminar on Hmong syntax. I thenformulated hypotheses which I tested out with my informants.Results from this study, to be presented below,show that kom can function as a higher verb meaning"tell, order (somebody to do something)"; as a complementizerintroducing an embedded sentence; and, when followedby the "can/able to" particle Au, as an element ofan idiomatic phrase. I will took at each of these, usingas a frame.1rk for analysis the "Standard Theory" ofgenerativegrammar developed in the work of Noam Chomsky andothers.KOM AS A HIGHER VERBLet us consider one of the examples Mottin .-ives(p. 130):245 2 4 6

246 THE HMONG IN TRANSITION(1) Kuv kom nws mus.I order him go"I ordered him to go/I told him to go."Mottin claims that "as in reported speech, the mainverb can be understood" and that "when this is the case,the understood verb is always qhia 'tell.' In otherwords, kom by itself can be translated by 'tell' (mytranslaticTrr There is no solid syntactic evidence inMottin's statement to disprove the hypothesis that kommight simply be the main verb in the example he gives.According to my informants, qhia could indeed be insertedbefore Ism this, however, would yield a different sentence,in which kom functions as a complementizer (seediscussion in the next section).My hypothesis that kcal could be a main verb wasstrengthened by sentences in the data I collected whichfollowed the same pattern as Mottin's example and by asentence Heimbach gave (p. 85):(2) Kuv tsis kom nws mus.I NEG make him go"I won't make him go/I won't cause him to go."2In Heimbach's example, the negative particle tsis occursbefore kom; this is a practical test for determiningwhether an entity is a verb or not. I double-checkedthis example with my informants and used the further syntactictests discussed below to check out the hypothesisthat kom is a main verb. All these tests help determineif an entity is a verb or not.Negative InsertionIn Hmong the particle tsis is placed in front of theverb to negate a statement, as seen in:(3) Kuv tsis mus.I NEG go"I don't go."The insertion of tsis in Mottin's example yields the sentenceHeimbach gave:247

(4) Kuv tsis kom nws mus.I NEG order him goTHE MORPHEME KOM 247According to my three informants, this sentence is grammaticallyacceptable and means "I didn't order/tell himto a."Question FormationInserting the particle puas before the verb turns astatement into a question, as illustrated in:(5) Koj puas mus?you Q go"Did you go?"The insertion of puas works in Mottin's example as well:(6) Kuv puas kom nws mus?I Q tell him go"Did I tell him to go?"Past Tense FormationIn Hmor; the appearance of the particle tau beforethe main verb makes it overtly past, as seen below:(7) Kuv tau mus.I past go"I went."The insertion of tau is acceptable in Mottin's example:(8) Kuv tau kom nws mus.I past tell him go"I told him to go."Future Tense FormationThe insertion of yuav before the verb puts the actionovertly into the future, as illustrated in:248

248 THE HMONG IN TRANSITION(9) Kuv yuav mus.I fut. go"I will go."This, too, works in Mottin's example:(10) Kuv yuav kom nws mus.I fut. tell him go"I will tell him to go."Negative Imperative FormationHmong insert the negative particle tsis followed bythe verb to turn a state-Tsis is optional and txhobthe particle txhob in front ofment into a negative command.is obligatory, as seen in:(11) (Tsis) txhob mus!NEG NEG go"Don't go!"This test is positive with kom:(12) (Tsis) txhob kom nws mus!NEG NEG tell him go"Don't tell him to go!"These tests strongly suggest that kom is a verb.However, all the particles used are preverbal and don'texclude the possibility of there being a deleted mainverb (qhia) before Los as Mottin claims. So (4) Kuvtsis kom nws mus could be analyzed as kuv tsis 2 kom nwsmus withO.- being the site of the deleted qhia. We needfurther evidence to support the hypothesis that kom canbe a main verb.The "Can/Able To" Modality TestInserting the particle tau after the verb adds themodality "can/able to" to it, as seen in:24)

THE MORPHEME ROM 249(13) Ku, mus tau.I go can"I can go."When applied to Mottin's example, the result is ungrammatical:(14a) *Kuv kom tau nws mus.I tell can him goAlthough we can't have (14a), we can have:(14b)Kuv kom nws mus tau.I tell he go canAccording to one of my informants, tau can go either withIsm, yielding "I could tell him to go" or with njus.,yielding "I told him he could go." The structure in (14b)supports the hypothesis that kom functions as a main verband suggest that tau can be attached at the end of thesentence to affect the higher verb. Further research isnecessary to confirm the latter point, however.Tag FormationHmong use a negative duplicating the main verb, asillustrated in the following example from Mottin (p. 88):(15) Koj mus los tsis mus?You go or NEG go"Are you going or not?"When this tag is attached to (1), kom shows up in thetag:(16) Kuv kom nws mus los tsis kom?I tell him to or NEG tell"Did I tell him to go or not?"If we put qhia before kom in (16), qhia shows up intag:the2,50

250 THE HMONG IN TRANSITION(17) Kuv qhia kom nws mus los tsis qhia?I instruct comp. he go or NEG instruct"Did I instruct him to go or not?"The fact that kom shows up in the tag in (16) is the mostpersuasive piece of evidence to support the hypothesisthat kom can be a main verb.The occurrence of the preverbal particles Isis,puas, tau, yuav and (tsis) txhob before kom (4, 6, 8, 10,12) and, most of all, the duplication of kom in a negativetag (16) confirm the hypothesis that kom can be amain verb. Mottin was right when he said that "kom byitself can be translated by 'tell'" (p. 130), but notbecause sraA is the understood main verb. The syntactictests mentioned above show that kom can function as ahigher verb meaning "tell, order" in its own right. Whenused as such, it implies a strong command. This is illustratedfurther in the following sentence from a folktale:(18) Ces Txiv Nraug Ntsuag txawm kom niag Niamand the Orphan boy then order NiaNkauj Kub Kaws mus ntov ntooNgao Kou Ker go cut tree"And then the Orphan ordered Nia Ngao Kou Kerto go and cut down trees."(From Nia Ngao Zhua Pa, 1st version by Lou Lee,p. 18)Semantic incompatibility between the modal tau "can/ableto" and the verb kom "tell, order" accounts for the ungrammaticalityfound in (14).KOM AS A COMPLEMENTIZERIn this section I will look at sentences in whichkom is inserted between a higher verb and an embeddedsentence and discuss the hypothesis that kom is a complementizerin that case. Mottin makes a distinction betweenthe followirg two sentences (p. 130):251

(19a) Kuv xav mus.I want go"I want to go."(19b) Kuv xav kom nws mus.want comp. him go"I want him to go."THE MORPHEME KOM 251In the first sentence we will assume that the subjects ofthe higher and the lower verbs are coreferential and an"Equi-NP deletion"3 rule controlled by the higher subjectdeletes the lower subject. According to Mottin, the insertionof kom in (19b) "implies an intention, a sense ofwill towards another party" (p. 130). From a syntacticstandpoint, kom seems to function like a complementizerintroducing an embedded sentence. Native speakers' intuitionalfeedback on constituent breaks is helpful indetermining where the higher sentence stops and the lowerone begins. When asked where (19b) could be "cut intwo," my informants said after xay. They always separatedsentences following the same pattern after the mainverb. Additional syntactic evidence to support the hypothesisthat kom is a complementizer is provided by applyingthe negative insertion, question formation and"can/able to" modality tests defined in the first sectionof this paper to the following sentence:(20) Kuv qhia kom nws mus.I instruct comp. him go"1 instructed him to go."Negative Insertion(21) Kuv tsis qhia (*tsis) kom nws mus.I NEG instruct comp. him go"I didn't instruct him to go."The negative particle tsis is inserted before the higherverb; it cannot occur before kom.252

252 THE HMONG IN TRANSITIONquestion Formation(22) Kuv puas qhia (*puas) kom nws mus?I Q instruct comp. him go"Did I instruct him to go?"The question marker puas can be inserted before thehigher verb but cannot occur before kom."Can Able To" Modality Test(23) Kuv qhia tau kom nws mus.I instruct can comp. him go"I can instruct him to go."Tau is grammatically inserted between the higher verb andkom, as expected. It can also follow kom; the resultingpattern, i.e., main verb plus kom tau, changes the meaningof the sentence and will be discussed in the nextsection.The facts that tsis and puas are inserted before thehigher verb and that tau is inserted between the higherverb and kom show that kom in these examples functions asa complementizer. This justifies Mottin's classificationof kom as a subordinate conjunction (p. 130). Mottinlists a few verbs which can be followed by kom (p. 131):haisqhiatxibthouxavkheevuatxhibyuam'say, speak''inform, instruct, teach, tell''tell somebody to do something''ask, beg''want''be willing, consent to''make, do on purpose''excite, entice''compel'There were numerous sentences with kom in the data I collected.The verbs preceding kom are listed below:ntuasntiabnyiam253'advise''drive away, drive out''like'

THE MORPHEME KOM 253xovthawbchebtospabtxwvnojtxiavfoomntaustsootuasaukawm ntawvpiavnyeemqwtsoyuamua'enclose with a fence''push''sweep''wait''help''forbid''eat''decide''curse''strike, hit, beat''hit, crush''fire, weapons''write''study''explain''read''shout, yell''send''force''make, do'This is a great variety of verbs. What do they have incommon? They can all be followed by an embedded sentenceintroduced by kom and conveying a result, an intention oran implicit command. It is difficult to narrow down thesemantics of kom to a single word translation when it isused as a complementizer. Mottin and Yang Dao came closestto doing so when they translated kom by "so that, inorder that." In all the sentences in which kom functionsas a complementizer, the higher subject NP affects thebehavior of the lower subject NP via a great variety ofverbs. It is my contention that any verb whose semanticsallows the subject to affect the behavior of the lowersubject can be followed by kom.Having discussed the semantics of kom as a complementizer,I will now turn to structural patterns. Thebasic pattern is the following:(NP) main verb [-kom NP VP] -This pattern applies to all the verbs I listed earlierand is illustrated below:254

254 THE HMONG IN TRANSITION(24) Hais kom nws khiav mus.say comp. her depart go"Tell her to go away."(From The Orphan and Ngao Zhua Pa, Level 1:20)(25) Peb tos kom hnub tuaj siab siab tso puastau?we wait comp. sun rise high high first Q can"Can we wait until the sun is high up?"(From The Monkeys and the Grasshoppers, Level2:9)(26) Kuv qw kom kwc ceevfaj.I shout comp. he be careful"I shout so that he's careful."/"I shout forhim to be careful."(From Tong yang, informant)When the higher and lower subjects are coreferential,the lower subject is deleted, as seen below:(27) Peb kawm ntawv kom tsawj ntawv.we study comp. be knowledgeable"We study in order to be knowledgeable."(Bertrais 1979: 104)(28) Peb tua kom yeej cobtsibwe fire weapons comp. win Vietnamesekom tshaj.communist"We fire weapons in order to defeat the Vietnamesecommunists."(From Tong Vang, informant)According to my informants, it is possible but not necessentencein (27) and (28).sary to keep peb in the lowerIn that case, the translations would be: "We study sothat we are knowledgeable" for (27) and "We fire weaponsso that we defeat the Vietnamese communists" for (28).The lower subject is also deleted when it has beenmentioned previously, as seen in the following excerptfrom a folk tale:2Jf"'"J

(29) Kuv nyiam kom sawv daws nyobI like comp. everybody liveTHE MORPHEME KOM 255mus li. Tab sis kuv nyiamthis forever. But I likeua suab sab...make noise loudua ketogetherli nolikekom (sawv daws)comp."I'd like for all of us to live together like thisforever. But I also like for us to make loudnoises..."(From Shao and His Fire, Level 2:8)In sentences where the main verb is followed by anindirect object NP, which is also the subject of the lowersentence, the lower subject is retained (30), pronominalized(31) or deleted (32,33):(30) Hais rau nws kom nws mus cheb tsev zoo zoo.say to her comp. she go sweep house well well"Tell her to go and sweep the house carefully."(Yao the Orphan, Level 2:13)(31) Haissay"Tell(Therau koj poj niam kom nwsto your wife comp. sheyour wife to go away."Orphan and Ngao Zhua Pa, L_(32) Kuv qw rau nws kom ceevfaj.I shout at him comp. be careful"I shout at him so that he's careful."(From Tong Vang, informant)khiav mus.depart go-1 1:19)(33) Nws foom rau yeej kom txhob tau zoo.he curse at man comp. NEG have good fortune"He cursed the man so that he would not have goodfortune."(From Tong Vang, informant)The lower subject could also be deleted in (30) and (31).A syntactically interesting phenomenon came up with regardto one of the examples Mottin gives in his section on kom(p. 130).256

256 THE HMONG IN TRANSITION(34) Kuv ntuas Lis kom musI advise Lee comp. go"I advised Lee to go."= (35) Kuv ntuas kom Lis mus.I advise comp. Lee goAccording to Mottin, there is no difference between (34)and (35). This was confirmed by my informants. The correspondinginformation question is:(36) Koj ntuas Lis li cas?You advise Lee what"What did you advise Lee to do?"Example (36) shows that ntuas takes a main clause objectNP. It functions like a force-type verb in English, butinstead of obligatorily deleting the lower subject NPusing Equi-NP deletion, Hmong offers a choice of deletingeither the lower subject NP (34) or the higher object NP(33). Pab "help" functions in the same way in the followingexample:(37) Pab nws kom hlob zuj zus.help him comp. grow little by little= (38) Pab kom nws hlob zuj zus.help comp. he grow little by little"Help him grow little by little."have found other verbs in the data which offer the sameIchoice when they are followed by kom:Intiabqhiatxwvxovthaw b'drive out, drive away''inform, tell, instruct, teach''forbid''enclose with a fence''push'Further subcategorization of verbs that can take thisstructure and a study to determine what accounts for thebe interesting topics to pursue.deletion choice wouldTo conclude this section I will mention an interestingfact concerning negative formation in embedded sen-257

THE MORPHEME KOM 257tences introduced by kom. In all the sentences I collectedin which the lower verb is negated, this verb ispreceded by txhob rather than tsis. This is illustratedin the following example from MottiiT86):(39) Nws pab kom tus tsov txhob tom kuv.he help comp. clf. tiger NEG bite me"He helped me so that the tiger wouldn't bite me."Mottin (1978) and Heimbach (1979) have described txhob asa particle used before the verb to express a strong negativecommand. The negative imperative thus formed canoptionally be preceded by the usual negative marker tsis,as seen in:(40) Tsis txhob txhawj.NEG NEG IMP worry"Don't worry"= (41) Txhob txhawj.NEG IMP worry(From How We Got Grain and Meat, Level 1:11,24)I formulated the hypothesis that tsis could also beoptionally inserted before txhob in embedded sentencesstarting with kom. This was confirmed by my informants,with no difference between the kom (NP) tsis txhob V andkom (NP) txhob V patterns.I further hypothesized that (tsis) txhob was theobligatory negative marker in lower sentences introducedby kom and that tsis alone could not be used in that case.My informants, however, accepted the kom (NP) tsis_ V pattern,stating that the sentences were grammatical but thatthey did not convey as strong a negative as the kom (NP)(tsis) txhob V pattern. Considering the facts that allthe sentences I found in the data followed the kom (NP)(tsis) txhob V pattern and that my informants preferredthis phttern to the kom (NP) tsis V one, I will suggestthat the former is the unmarked (preferred) pattern andthe latter the marked pattern for negative formation inembedded sentences introduced by kom.25d

258 THE HMONG IN TRANSITIONTHE IDIOMATIC PHRASE KOM TAUAs seen in (14), the modal tau "can/able to" cannotfollow kom when Isom is a main verb, presumably because ofsemantic incompatibility between the modality and thestrength of the command in kom "order, tell (somebody todo something)." Kom can, however, appear between a mainverb and tau. In that case, the sequence kom tau seems tofunction as an idiomatic phase. This point is illustratedbelow:(42) Kuv mus tau.I go can"I can go."(43) Kuv mus kom tau.I go"I will make myself go." (no matter what)When tau alone follows the main verb (13), it simplyexpresses the "can/able to" modality, as seen earlier.When kom is inserted between the verb and tau (43), itchanges the modality and implies a command, something tothe effect "not only can I do it, but I have to carry outthe action, no matter what the circumstances are." Taugoes with kom in that case, according to my informants,and the sequence functions as a separate semantic unit.Kom tau can follow a variety of verbs, as seen in thefollowing examples from Tong Vang, one of my informants:(44) Kuv nqa kom tau lub pob zeb ntawd.I lift clf. rock that"I have to be able to lift that rock."(45) Peb caum kom tau lub tsheb daj tod.we chase clf. car yellow that"We have to be able to catch up with that yellowcar."CONCLUSIONUsing the results of preverbal particle and tagformationtests, I have shown that kom can function as a

THE MORPHEME KOM 259higher verb meaning "tell, order (somebody to do something)."I have also shown that kom is used as a complementizerand discussed structural patterns. Finally, Ihave shown that, when preceded by a main verb, kom taufunctions as an idiomatic phrase meaning "can andno matter what."NOTE'This paper is drawn from my M.A. thesis (San Diego StateUniversity), in progress, on complementation in Hmong.2NEG=Negative. Other abbreviations used in the glosses ofthe example sentences are as follows: Q = question, comp.= complementizer, fut. = future, NP = noun phrase, VP =verb phrase, IMP = imperative, V = verb, clf. = classifier.3Equi-NP deletion refers to the deletion of one of twoidentical noun phrases in a sentence.REFERENCESBertrais, Yves. 1979. Dictionnaire Hmong-Francais. Bangkok:Assumption Press, Sangwan Surasarang Publisher.Heimbach, Ernest E. 1979. White Hmong-English Dictionary,revised edition. Linguistics Series IV. SoutheastAsia Program Data Paper No. 75. Ithaca: CornellUniversity, Department of Asian Studies.Johnson, Charles. 1981. Hmong Folk Tales. Saint Paul:Linguistics Department, Macalester College.. 1984. Dab Neeg Hmoob: Myths, Legends and FolkTales from the Hmong of Laos, Saint Paul: LinguisticsDepartment, Macalester College.Mottin, Jean. 1978, Elements de grammaire HmongBangkok: Don Bosco Press.20u

260 THE HMONG IN TRANSITIONYang, Dao. 1980. Dictionnaire francais-Hmong blanc.Paris: Comite National d'Entraide.YIng Dao, Se Yang and Tong Vang, personal communication.2 G i

ZERO ANAPHORA AND TOPIC PROMINENCE IN HMONGJudith Wheaton FullerMottin, in his grammar of Hmong, noted that "personalpronouns are often omitted when the sense is clear"(Mottin 1979:44; original in French). In this paper Iwill examine what this statement means in a variety ofpronominal contexts, and consider the implications ofthis absence of pronouns (called zero anaphora) in Hmongfor typological statements about the Hmong language. Iwill first outline some facts of zero anaphora in Hmong,then relate these facts to the concept of topic-prominence.SENTENCE-LEVEL ZERO ANAPHORALet us look first At some sentence-level ellipsisphenomena. Sentences with zero noun phrases occur inseveral situations; I will note these structures and dividethem into two groups. The first pronoun omissiontype, EQUI NP DELETION,' is illustrated in examples (1)and (2), where zeros are used to indicate the position ofthe missing noun phrase.2(1) Kuv xav 0 mus.fps want go"I want to gu."(2) Kuv yuav pab koj 0 cog nplej.1 ps will help 2ps plant rice"I will help you (to) plant rice."A sentence like (1) has traditionally been analyzed ashaving the underlying structure [I want [I go]], wherethe second instance of the first person pronoun is deletedby the rule of EQUI NP DELETION. The omission hereof the subject of mus 'go' is controlled by the identicalsubject of the higher verb xav 'want.' In (2), the omissionof the subject of cog 'rAant' is controlled by the261262

262 THE HMONG IN TRANSITIONsubject of the higher verb pab 'help'. In these casesthe "sense is clear" because the verbs xav 'want' and .01'help' are understood to require the deletion of the subjectof their complements under identity with the subjectand object, respectively, of the higher verb; recoverabilityof the complement subject is thus assured.IMPERATIVE SUBJECT OMISSION is also common (thoughnot mandatory) in Hmong, as shown in (3-4).(3) 0 Muab rau kuv.give to I ps"Give it to me."(4) 0 Cia nws ua nws li. (Heimbach:14)allow 3ps do 3ps like"Let him do as he likes."(3-4) the "sense is clear" because the omitted pronounInis predictably the addressee, again assuring recoverability.If nws or z1,3 for example, were 'leant, those pronounswould have to appear on the surfack.Zero noun phrases also occur in VERB SERIALIZATION,as shown in (5).(5) Thaum ntawd lawv mam thiaj li yuav 0 ncetime that 3pp then consequently want ascend0 tuaj 0 mus.come go"At that time, they will want to better themselves."In (5), if each verb is interpreted separately, its subjectis deleted under identity with the subject of theinitial verb.DISCOURSE-LEVEL ZERO ANAPHORAAnother type of zero noun phrase which occurs inHmong is main clause DIRECT OBJECT OMISSION. Examplesare given in (6-9).203

ZERO ANAPHORA AND TOPIC PROMINENCE IN HMONG 263(6) Kuv tsis nyiam 0. (Heimbach:219)1ps NEG like"I don't like (it)."(7) Kuv tsis tau ua 0. (Heimbach:307)1ps NEG ASP do"I didn't do (it)."(8) Nws mob. Mas koj puas mus xyuas 0? (Heimbach:417)3ps pain Prt 2ps Q go visit"He's ill. Will you go to visit (him) ?"(9) Kuv tsis tau hnov b dua. (Heimbach:70)1ps NEG TM hear before"I have never heard (it) before."In (6-7), pronominal object omission with the meaninggiven is obligatory. In (8), the object omission is completelyoptional. In (9), the missing object refers to anonhuman sound; if nws is inserted in the missing objectposition, this pronoun must refer to a person. While theanimacy restrictions governing optional vs. obligatoryomission are not entirely clear, object omission in generalis governed by its reference to a topic in the precedinglinguistic or nonlinguistic context. Thus the referentcannot be recovered by looking at sentence levelfacts alone.Main clause SUBJECT OMISSION, independent of IMPERA-TIVE SUBJECT OMISSION, is also a frequent phenomenon inHmong, as shown in (10-13).(10) Q Tsis tau sib fim. (Heimbach:45)NEG ASP RECIP meet"(We) haven't met yet."(11) 0 Tabkaum koj haujlwm. (Heimbach:77)hinder 2ps work"(I) have hindered you."(12) 0 txujkum dim oj? (Heimbach:4)how escape Q PRT"How will (he) escape?"264

264 THE HMONG IN TRANSITION(13) Niaj hnub 0 mus laij teb. (Lyman:47)every day go plow field"Every day (they) went to plow the field."In (10-11), the omitted subject is presumed to be thespeaker (or speaker and hearer). In the cases of thirdperson subject omission in (12-13), interpretation of thezero subject pronoun is not possible from the sentencealone, but requires additional linguistic or extralinguisticcontext.In (14), both object and subject are omitted.(14) Kuv thawb 0 0 ntog. (Heimbach:190)I ps push roll over and over"I pushed (it) and set (it) rolling."Again, the omitted word or words do not refer to anoun phrase in the same sentence; the referent is in theprevious linguistic or extralinguistic context.While all of these processes delete noun phrases,main clause SUBJECT and OBJECT OMISSION function differentlyfrom EQUI NP DELETION, IMPERATIVE SUBJECT OMISSIONand VERB SERIALIZATION.The latter three principles arecontrolled by easily specifiable syntactic conditions referringto grammatical relations and operating within thesentence. The conditions for SUBJECT and OBJECT OMISSION,on the other hand, are not easily specified, their domainis extrasentential, and they are controlled by the topic.Since a topic can be and often is mentioned in one sentenceand referred to in subsequent sen-tences, the domainof topic-controlled deletion is necessarily beyondthe sentence. In fact, SUBJECT and OBJECT OMISSION aredeletion, since thenot very good names for this type offact that the deleted NPs are subjects or objects is irrelevant.A better designation encompassing both wouldbe TOPIC-NP DELETION, as we will show.Tsao (1979), in his description of Chinese, has referredto TOPIC-NP DELETION as a process operating withina topic chain. When the domain of a topic extends overseveral sentences, Tsao refers to the sentences in thatdomain as a topic chain. Within a topic chain, the topicis mentioned once, then may be omitted in subsequent mention.It is this type of deletion process, TOPIC-NP DE-LETION (rather than the EQUI type of deletion that is265

ZERO ANAPHORA AND TOPIC PROMINENCE IN HMONG 265controlled by the sentential subject or object) that I amparticularly concerned with in the discussion of zero anaphorain Hmong.A topic chain with TOPIC-NP DELETION is illustratedin (15) (from Lyman 1979:4748).(15) Muaj ob tug txiv neeg mas yog ob tug twv tij.there 2 CLF people PRT BE 2 CLF brothersNiag hnub 0 mus laij teb.every day go plow fieldHnub laij tas,day plow finishMuaj ib hmosthere was 1 nightua li cas.howib hmos tagkis 0 rov mus saib.one night tomorrow back go lookthiaj li 0 mus zov saib yogthan go watch see was"(Once) there were two people who were brothers.Every day (they) went to plow the field. Aftereach day's plowing, when night-time was over anddawn had arrived, (they) would go back to look atthe field.... Then one night (they) went andwatched to see what happened."Within a topic chain, the question arises as to thepredictability of omission of noun phrases coreferentialto the topic. In a prototypical topic chain model, thetopic is initially expressed as a full noun phrase, andsubsequent references to the topic are deleted under i-dentity with (or reference to) the topic noun phrase.However, under certain circumstances, perhaps if the topicchain becomes long, or competing noun phrases appear,a pronoun or a repeated or substitute noun phrase may beused to refer to the topic. This raises the questionwhether there are definite rules governing TOPIC -NP DELE-TION, or whether deletion is always optional.Li and Thompson (1979), in their study of zero anaphorain Chinese, took the view that zero anaphora wasthe norm, and it was the presence of a pronoun (or fullyreferring noun phrase) in a discourse that required ex-2 6 ti

266 THE HMONG IN TRANSITIONplanation. They conducted an experiment in which they excerptedshort narratives from a Mandarin text, took outall pronoun occurrences, and gave them to native Mandarinspeakers to read, asking them to insert pronouns wherethey felt they were needed. The results confirmed Li andThompson's claims that zero anaphora is the norm, andthat native speakers vary in their judgments about wherepronouns are needed (1979:388). They found that one predictablefactor in pronoun insertion was degree of conjoinability:"the lower the degree of conjoinability betweentwo clauses, the higher isthe likelihood of a pronounoccurring in the second clause" (1979:334). Theyconcluded that since conjoinability depends on individualjudgments, predictions of pronoun insertion are only possiblestatistically, for groups of speakers, rather thanfor individual speakers.Following Li and Thompsons's procedure, I removedpronoun occurrences from a Hmong narrative written in RPAand submitted the narrative to a small number of nativeHmong speakers, asking them to insert pronouns where theythought pronouns were needed. The results indicated thatnative speaker judgments about the appropriateness ofpronoun placement were indeed variable. While no pronounswere inserted in EQUI type positions, which aresentence-level and subject-controlled, pronoun insertionoccurred in TOPIC-NP DELETION positions, w;lich are discourse-leveland topic-controlled. There was no unanimityof judgment about where the pronoun insertion occurred.For example, in the segment referred to in (15),which is part of the text used in the experiment, one respondentfelt a pronoun was required in the last sentence,following some omitted material, while another respondentand the original text used no pronoun in thatposition. Use of a pronoun in such positions may dependon speaker judgments of intelligibility, which may varywith speakers. Constraints on zero pronouns in thesesituations appear to be primarily topic-oriented.I also submitted another Hmong narrative written inRPA (a Hmong folk tale edited by Charles Johnson) to anative Hmong speaker, asking him to remove pronouns inthe narrative that he felt could be omitted. Using thisprocedure it seemed clear that intelligibility was notthe oni, criterion used to determine acceptability ofsentences without coreferential pronouns; even where co-2 CV

ZERO ANAPHORA AND TOPIC PROMINENCE IN HMONG 267ref eren tiality was clear, pronouns were sometimes required,apparently for syntactic reasons.3 This situationis shown in (16), where nws in the second sentenceis obligatory, according to this native speaker.(16) Nws hie nws cev ris tsho zoo cia.3ps take off 3ps CLFclothes good put asideNws hnav nws cev ris tsho tawv nas tawv3ps put on 3ps CLF clothes skin rodent skinnoog.bird"He took off his good clothes.skins."He put on animalZero pronouns are more acceptable when the sentences withcoreferential NPs are not merely conjoinable but when atransition word is used, as in (17), where Sib Nab may beomitted in the second sentence if ces, 'then', is inserted.(17) Sis Nab ntsia saum Nkauj Ntsuab lub qhov raiSis Na gaze up Ngao Njua CLF windowSis Nab 0 ces txawm tshuab tshuab qeej.Shee Na then PRT blow kheng"Shee Na looked up at Ngao Njua's window.played his kheng."He thenConstraints c-, omission here appear to be syntactic aswell as topic-centered. However, much more study isneeded to uncover and interpret the facts of these deletionconstraints.Languages differ in the restrictions they place onzero anaphora. English, for example, restricts zero anaphoraquite severely. While (18-19), showing a type ofdeletion that Tsao calls SPEAKER AND HEARER DELETION, areacceptable in informal speech, (20-24) are either unacceptable,(marked by an asterisk) or marginally acceptable,(marked by a question mark).268

268 THE HMONG IN TRANSITION(18) 0 Went to the store while you were gone. 0 Boughtsome blueberries.(19) 0 Go to the store while I was gone? 0 Buy anyblueberries?(20) *When I saw John yesterday, 0 realized that he wassick.(21) That table is ugly. *I don't like 0.?6 Doesn't go with the buffet.(22) *Regarding that table, 0 isn't big enough to servethis many people.(23) *Regarding that table, I want to refinish 0.(24) ?Before you set the table, put a tablecloth on 0.On the other hand, in a language like Chinese, as wehave noted, zero anaphora is much less restricted. Accordingto our investigations so far, Hmong falls somewherebetween Chinese and English in having zero anaphorain more contexts than in English, but with more restrictionsthan Chinese.TOPIC-PROMINENCE AND ZERO ANAPHORAGundel (1980) has classified a number of languagesaccording to the degree to which they restrict zero anaphora.This classification is shown in (25).(25)I II IIILisu Chinese Numbami Hebrew Spanish EnglishLahu Japanese Kekchi Arabic Italian FrenchThai Korean Yukatan Turkish Baltic GermanVietnamese Fijian Latin Slavic26

ZERO ANAPHORA AND TOPIC PROMINENCE IN HMONG 269Languages with least restricted zero anaphora areshown at the left in I, those with most restricted zeroanaphora at the right in III. Those in the middle, inII, are intermediate, allowing zero anaphora only in somesituations. As Gundel points out, an interesting featureof this classification is the degree to which it correspondsto the classification of subject-prominent andtopic-prominent languages developed by Li and Thompson(1976): the languages that have the least restrictedzero anaphora are the same languages that are the mosttopic-prominent, and the languages which have the mostrestricted zero anaphora are the ones that are the leasttopic-prominent. Gundel's generalization stating thispoint is given in (26).(26) The more topic-prominent a language, the less restrictedits use of zero anaphora (Gundel 1980).Li and Thompson have proposed the typology of topicprominentversus subject-prominent languages as a toolfor examining the sentence structure of languages to determinewhether the language is more fruitfully describedin terms of topic-comment structure or subject-predicatestructure (1976:460). Li and Thompson characterizedtopic-prominent languages as shown in (27).(27) Characteristics of topic-prominent languages:a) surface coding of topicb) passive construction nonexistent ormarginalc) no "dummy" subjectsd) existence of "double subjects"e) topic controls coreferential constituentdeletionf) verb-finalg) no constraints on what may be topich) topic-comment sentences are part of therepertoire of basic sentence types.Subject-prominent languages, by contrast, typicallydo not code topics; they have processes such as pasIvization,which refer to subject, and they require "dummy"subjects to fill subject position. Subject in subject-270

270 THE HMONG IN TRANSITIONprominent languages is important in controlling coreferentialdeletion, subject-predicate division rather thantopic-comment division is characteristic of sentence patterns,and constraints may be placed on the expression oftopic (for example, the topic may have to be a subject).I would now like to examine Hmong sentences accordingto Li and Thompson's characteristics of topic-prominentlanguages, and then use Tsao's criticisms of Li andThompson's statements about Chinese to show that zero anaphorais the most relevant of these characteristics inassessing the "topic prominence" of a language.(a) Surface coding of the topic is an indicationthat the language requires the topic to be salient. Codingmay be a morphological marker, as in Lisu, Japaneseand Korean, or position (invariably initial sentence position),as in Chinese. Hmong has no invariant morphologicaltopic marker. Topics are often made explicit bythe introductory expressions hais txog or hais tics, aswill be illustrated below.(b) Hmong has a passive construction containing theword aug which is sometimes translated into English aspassive and appears to be regarded by Hmong speakers as apassive. Examples of its use are shown in (28-31).(28) Pov tau raug nees tuam nws.Pao TM horse kick 3ps"Pao got kicked by a horse."(29) Naghmo kuv lub qhws ntsej raug luagyesterday I ps CLF earring somebodynyiag lawm.steal TM"Yesterday somebody stole my earring."(30) Nws raug tua tuag tawm.3ps kill die TM"He was killed."(31) Tooj raug luag ntes.Tong someone catchTong was caught by someone."271

ZERO ANAPHORA AND TOPIC PROMINENCE IN HMONG 271However, the passive word raug cannot be used intranslating English sentences such as "The room had beencleaned by Ntxhais" or "A new kind of rice was studied bythe scientist." These sentences must be rendered inHmong in active form, as in (32-33).(32) Ntxhais tau cheb chav pw.Ntxhais TM clean room-bed(33) Tus kws txuj tau kawm ib horn nplej tshiab.CLF scientist TM study one kind rice newThe types of verbs in use with raug suggest that PASSIVEin Hmong is limited to an "adversity" situation. In addition,the word mutt is not specifically a passive marker,but a verb with the meanings "to encounter," "to meetup with," "to incur" or "to suffer," according to theHeimbach dictionary. Other facts bolstering the identificationof raug as a verb are its placement in verb positionin the sentence, and its use with the aspect markertau, as in (28). Passive in Hmong is thus either nonexistentor quite restricted, to the extent that one canread many pages of Hmong texts of various types withoutencountering a single instance of raug. The passive constructionin Hmong thus conforms with Li and Thompson'scharacterization of passive in topic-prominent languages.One use of passive constructions is to make nonsubjectnoun citruses into subjects so that they can serve astopics. Li and Thompson state that the reason passive isnot needed in topic-prominent languages is that "any nounphrase can be the topic of a sentence without registeringanything on the verb" (1976:467). However, the existenceof the topic chain suggests another mode for registeringa noun phrase as a topic in Hmong. If, for example, chav2w 'bedroom' in (32) were to be a topic, it could appearin topic position in a previous sentence, then becomezero in the following clause, as in (34).(34) Chav pw sw heev (mas) Ntxhai yuav cheb 0.room-bed disorderly very (PR Ntxhais will clean"The tedroom is very disorderly, (so) Ntxhais willclean (it)."Notice, however, that this expression of j.9p)cl requiresconsideration of a domain beyond the simple set' lee.

272 THE HMONG IN TRANSITION(c) "Dummy" subjects occur in English in sentenceswith nonreferential it and existential there. Hmongappears to have no dummy subjects.4 The verb muaj 'tohave' is used clause-initially in an existential sense;(15) in the narrative given previously is an example ofthis.Expressions such as "It might rain today" are typicallygiven in Hmong as (35).(35) Tej zaug hnub no yuav los nag.maybe today will come rain(co The occurrence of double subjects, typical in atopic-prominent language like Chinese, appears to be marginalin formal and written Hmong, although it may bemore acceptable in conversational spoken language. Forexample, (36) is an acceptable Hmong sentence, while(37) is better re-phrased as (38).(36) Kuv lub tsev mas kuv tau yuav tsaib no.1 ps CLF house PRT, 1ps TM buy last year this"I bought my house last year."(37) Tsheb mas coy Toyota thiab Datsun zoo heev.car PRT CLF Toyota and Datsun good very"Cars, Toyota and Datsun are very good."(38) Hais txog tsheb mas, coo Toyota thiab Datsun zoospeak about car PRT CLF Toyota and Datsun goodheev.very(e) Regarding coreferential deletion, it is clearfrom the preceding discussion that both subject and topiccan control coreferential deletion in Hmong. The crucialcases here involve sentences with distinct top: and subject,and a subsequent zero pronoun which could potentiallyrefer to either, as in (39).(39) (Hais txog) daim teb / thav av ntawd (mas), tsob(speak of) CLF field / CLF land that (PRT) CLFnplej tuaj loj loj yog li no mas 10 muaj nqi heevrice grow big consequently PRT valuable very273

ZERO ANAPHORA AND TOPIC PROMINENCE IN HMONG 273"That field/land, rice grows very big, so (it) isvery valuable."Li and Thompson's claim is that in these cases,prominentin topic-languages, topic overrideslersubject asof control-deletion, and where subject must controlthe sentence isdeletion,unacceptable. In Hmong, thein zero pronoun(39) can refer to either the topic- or theThus Hmong is equivocalsubject.about topic or subject-prominencewith regard to coreferential deletion.(f) Li and Thompson characterizelanguagestopic-prominentasverb-final. Although it is true thatlanguagesthethey identify asal,topic-prominentitare verb -fin-is not clear what relevance this fact has to topic-prominence.Hmong is verb-medial, although itwe have seen, some topic-prominenthas, as(g) Hmong characteristics.can have objects as well as subjectstopics, as shown in (40).as(40) Hais txog tsheb mas, kuv tsis nyiam tsav 0.speak about car PRT 1 ps NEG like drive"Speaking of cars, I don't like to drive (them)."With respect to the range of phenomena Li and Thompsonconsider, then, Hmong does not constrain heavily whatbe topics.can(h) Regarding sentences which are basic in Hmong,sentences (36-38) show thatof theovert topic-comment sentencesJapanese/Chinese type are possible but somewhatmarginal in Hmong.What Li and Thompson have shown in their typologyan interesting iscorrelation between certain properties ofsentences and a predilection for a languageits to organizestructure according to topic-commentof theserelations. Somesentence-level properties, such asing for topic and surface cod-topic-commert sentencesmanifestation of as basic, are atopic-prominence. Other case;, such asno dummy subject and verb-final, show an interestingrelation,cor-but no apparent intrinsic relati.Inship toic-prominence,top-although a dummy-subjectintrinsicallyrequirement isrelated to subject-prominence.sionTheof inclu-some properties, such as no passive,subjects,no dummyand topic-comment sentences as basicdesigned to showseemedthat in these languages the subject is274

274 THE HMONG IN TRANSITIONof little importance. In fact, Li and Thompson takepains to show that in Lisu and in Mandarin it is sometimesimpossible to identify a sentence subject.However, there is no reason to assume that a languagewhich gives a prominent role to topic must give alesser role to subject. Japanese and Korean mark bothsubiects and topics, and Li and Thompson themselves considerthese languages to be both topic-prominent and subject-prominent.It is also clear that a language likeHmong, which has some topic-prominent characteristics,also gives the subject a well-defined role: subject hasa characteristic sentence-initial position, and controlscoreferential deletion in sentence-level processes, as wehave seen.Tsao (1979:91) makes a crucial point in comparingsubject and topic influence when he states that subjectand topic "belong to different levels of grammatical organization."Subject is relevant at the sentence level,while topic is relevant at the discourse level. Thus,regarding sentence-level structural facts, while they mayshow interesting correlation with topic organization andmay indeed be related to it, the relationship is indirect.For example, as I have suggested, a language likeHmong may have a marginal passive not because the topicrather than the subject is important in Hmong, but becauseorganization at the discourse level, where subjectis not relevant, obviates the necessity for the use ofthe passive construction, or because constraints on thePASSIVE morpheme force the use of alternative constructions.The grammaticality of topic-comment sentences assuch is clearly a sentence-level matter; the rarity ofsuch sentences, even if they are "basic", can only be explainedon the discourse level, where the notion of thetopic chain shows clearly that explicit topic-comment expressiononly occurs when the topic is introduced.Li and Thompson's characteristics of topic-prominentlanguages are all sentence-level characteristics, withone exception. Control of coreferential deletion in coordinateclauses is, although Li and Thompson do not presentit this way, clearly related to discourse processes.This type of coreferential deletion is similarto that occurring in topic chains in that deletion occursunder coreference with the topic.27

ZERO ANAPHORA AND TOPIC PROMINENCE IN HMONG 275Because of its direct connection with the organizationof topic in discourse, then, zero anaphora is themost relevant of Li and Thompson's characteristics forassessing topic-prominence. The definition of "topicprominent"can thus be extended to mean that such a languagehas certain ways of organizing discourse topics,including use of zero anaphora, and that certain sententialstructures provide signals of this organizationalpattern.While many of these matters require further study,it is clear that the term "topic-prominence" is separableinto sentence-level and discourse-level processes.Since, as Tsao points out, the proper domain for topic isdiscourse, it is appropriate to look to discourse-levelprocesses such as zero anaphora for defining features ofa topic-prominent language.CONCLUSIONIn conclusion, I have shown that zero anaphora inHmong is less restricted than in E.1glish, since subjectand object NPs may be zero when the referent is the discoursetopic. Zero anaphora is, however, more restrictedthan in Chinese, since a connection between sentencesrather than mere conjoinability is sometimes required foracceptable sentences with zero NPs. While topic clearlycontrols coreference in topic chains, in the crucial situationwhert a sentence has both subject and topic as in(39), the results are equivocal: control by either subjector topic is possible. Hmong thus occupies an intermediateposition with regard to coreferential deletion:between English, a prototypically subject-prominent laDguage,and Chinese, a prototypically topic-prominent language.It is interesting that this intermediate positionis also reflected in characteristics: Hmong has a marginalpassive and does not use "dummy" subjects, characteristicsof topic-prominent languages; 1)11 is not verb -final, and it is not entirely clear that double-subjectsentences are basic. Sentence-level and extrasententialprocesses thus reinforce each other in elucidating therole of topic in Hmong.276

276 THE HMONG IN TRANSITIONACKNOWLEDGEMENTSwould like to thank Vang Vang and Yang Dao forItheir help with the Hmong sentences and texts, and BruceDowning, Jeanette Gundel, Martha Ratliff and CynthiaPerry for their helpful comments on earlier drafts ofthis paper. Any remaining problems are, of course, myown.NOTESIEQUI NP DELETION is a term used to indicate the deletionof one of two identical noun phrases in a sentence.2Other abbreviations used in the glosses of the examplesentences are as follows: la refers to the first personsingular pronoun, 221 to second person singular,221 to third person singular, 3pp to third personplural, NEG = negative, ASP = aspect or tense marker,PRT = particle, Q = yes-no question word, RECIP =reciprocal, CLF = classifier, BE = any form of the verbbe.3Charles Johnson has pointed out to me that narrativessuch as these may require more pronoun use than othertypes of speech.4Some speakers accept sentences with dummy subjects,but say they are seldom, if ever, used. Other speakersreject non referential nws completely.%Chinese, a Subject-lerhCLject bnguage, is the exception.Li and Thompson argue that Chinese is becomingverb-final.REFERENCESGundel, Jeanette K. 1980. "Zero NP-Anaphora in Russian:A Case of Topic-Prominence." Papers from the Parasessionon Pronouns and Anaphora. Chicago: ChicagoLinguistics Society.2 7'r

ZERO ANAPHORA AND TOPIC PROMINENCE IN HMONG 277Heimbach, Ernest E. 1979. White Hmong-English Dictionary,revised edition. Linguistics Series IV. SoutheastAsia Program Data Paper No. 75. Ithaca: CornellUniversity, Department of Asian Studies.Johnson, Charles, ed. 1981. Nkauj Ntsuab thiab Sis Nab:Dab Neel; Tiaj Rhawv Zeb. Ngao Njua and Shee Na: TheStory of the Plain of Jars. St. Paul: Lingustics Department,Macalester College.Li, Charles N., and Sandra A. Thompson. 1976. "Subjectand Topic: A New Typology of Language." Si,bject andTopic, ed. by Charles N. Li, pp. 457-489. New York:Academic Press Inc.. 1979 "Third-Person Pronouns and Zero-Anaphora inChinese Discourse." Syntax and Semantics 12: Discourseand Syntax) edited by Talmy Givon, pp. 311-335. New York: Academic Press.Lyman, Thomas Amis. 1979. Grammar of Mong Njua (GreenMiao): A Descriptive Lynguistics Study. Published bythe author.Mottin, Jean. 1978. Elements de grammaire Hmong blanc.Bangkok: Don Bosco Press.Tsao, Feng-Fu. 1979. A Functional Study of Topic inChinese: The First Step Towards Discourse Analysis.Taipei, Taiwan: Student Book Co., Ltd.27

INVESTIGATING LITERACY: APPROACHES, TOOLS,AND THEIR CONSEQUENCES FOR INQUIRYGail WeinsteinBACKGROUNDChou, a young man of 25, sat patiently beside thepastor until it was time to give a Bible reading inHmong. During his two-minute reading, the Hmong in thecongregation looked up to listen, and the Americansseemed to relax and let the music of the language washover them. After the service, someone introduced me toChou as an English teacher who was interested in the refugeesin the neighborhood. He invited me to visit at hishome.Ithroat.took up the invitation despite a cold and a soreAfter finding Chou's rowhouse, the only one intactin a row of gutted and abandoned structures, I rangthe bell and waited. I was welcomed into a second-floorapartment. Two old beds, where two babies slept in heapsof blankets, had been pushed together against a wall.Plants crowded a table next to a TV against another wall.On the bathroom door, there was a magazine picture of avery muscular wrestler gritting his teeth. As soon as mycold became apparent, Chou's wife, Sal, cut some gingerroot and rubbed it on my neck and arms. This marked thebeginning of a lasting friendship. (The names of Chou,Sai and all other Hmong people in this paper have beenchanged to protect anoiymity.)Chou is among the better educated Hmong who fledfrom Laos. The son of a town merchant, he was able toattend school for four years before becoming a soldier.He guesses, with no aid from birth documents, that he wasabout twelve or thirteen when he left school to become asoldier. By then, he had already had what would prove tobe some precious experience with reading. Sai, on theother hand, did not encounter printed materials until shewas an adult in a Thai refugee camp. Unlike Chou, whosefirst written language encounters were with Lao, Sai readand wrote her first words in English shortly before herjourney to the United States.27927

280 THE HMONG IN TRANSITIONThe Hmong in Philadelphia, like many other recentHmong refugees, are faced with the transition from farmingin the hilltops of Asia to coping with the people andinstitutions of a Western urban center. If we are to un-Howderstand their situation, what are we to watch for?do we make sense of what we see? What is the role of literacyin adaptation? How do we interpret the processesby which Chou and Sai and their kin grapple with theirnew lives?PERSPECTIVES ON THE ROLE OF LITERACYAnthropologists have long contended that it is notonly difficult to shed our cultural biases and observe"objectively"it is impossible. As effective social scientists,the best we can do is to recognize what thosebiases are, and to take into account what effect theywill have on what we find.The first section of this paper very briefly reviewsthe work of some researchers and scholars who have ponderedthe role of literacy in human cognitive and socialdevelopment. By looking at thcir burning questions wecan detect their underlying assumptions. By scrutinizingtheir research tools, we can better evaluate the scopeand limits of the tools available to us.Literacy and Evolution: The Great TransformationEarliest discussions about literacy centered on itsconsequences on human cognition. Havelock (1963) arguesthat alphabetic writing, which originated in Greece,changed the structure of human thinking and logic. Goody(1968) builds on Havelock's assertion that a phonetic alphabetis linked with the eventuality of abstraction, andits corresponding abilities. In their work among theWolof, Bruner (1973) and Greenfield (1972) found that illiteratesscored much lower on sorting and labeling tasksthan their literate counterparts, and were more "contextdependent"in communication. They conclude from theirresearch that writing promotes cognitive development, andsuggest further that "technologies" available in a given2G u

INVESTIGATING LITERACY 281culture determine the "range and abilities" of its members."Symbolic technologies," they argue, "push cognitivegrowth better, earlier, and longer than others."English teachers will be among the first to concurthat Hmong students have considerable difficulty in theclassroom. In view of these difficulties, some of thesesuggestions are compelling. Olson (1977) examines the"shift from utterance to text" as a phenomenon characterizingindividual psychological as well as cultural processes.He argues that as individuals develop, spokenlanguage becomes less ambiguous, more explicit, and moreautonomous as a representation of meaning. Of individualsand social groups, Olson writes:The bias of written language toward providingdefinitions, making assumptions and premises explicit,and observing formal rules of logic producesan instrument of considerable power forof real-building an abstract and coherent theoryity. [P. 273]This argument suggests that there exists a continuumalong which one could map the stage of development of anyindividual or culture (whose concept of reality is correspondinglycoherent or incoherent). Is this model accurate?Does it explain the difficulties that Hmong peopleface in the classroom? More important, is it useful?Scribner and Cole (1981) address themselves to the"writing crisis" in the United States in a manner thatprovides a new framework that is useful for examiningliteracy. In a discussion of the Vai of northwest Liberia,Scribner and Cole demonstrate that the effects ofliteracy are often confused with effects of formal si_aooling.In an extensive study (1981) they question traditionalevaluative measures for testing consequences ofliteracy. They find that tnose Vai who are literate withoutschooling do better in some cognitive tasks than nonliterates,but only in tasks involving closely relatedskills. Some literacy skills are transferred to othertasks, Lut within a very specific range. "Logic," theyfind, is little affected by nonschooling literacy in theVal people.Scribner and Cole examine uses of literacy in English,Arabic and Vai scripts in a functional framework,281

282 THE HMONG IN TRANSITIONlooking to see which skills arr: used by whom for whichparticular ends. Their aim i: to learn about cognitiveconsequences of literacy, specifically in areas whereprevious scholars have made assertions without empiricalevidence. By examining the uses and functions of skillsin context, Scribner and Cole are able to demonstrate howsocial organization creates the conditions for a varietyof literary activities. With these kinds of questions inmind then, it becomes possible to examine the social organizationthat encourages or restrains uses of literacyas well as the social and intellectual significance ofthese skills.Trouble at School: Approaches For Problem SolvingScribner and Cole examine uses of oral and writtenlanguage as well as schooled and nonschooled literacy todiscover intellectual effects. Other scholars have studiedthese distinctions to understand and explain the difficultiesfaced by children as they navigate their waythrough urban schools. Cook-Gumperz and Gumperz (1977)argue that the move from home to school demands a changein communicative tasks in which children are required toprocess "decontextualized" information in new social andlinguistic codes. That is to say that a formal schoolsetting involves introduction of information that is notof immediate relevance to the social situation. Whilelanguage in the home might center around immediate concernssuch as preparation for mealtime or hours of availablesunlight for grandma's sewing, the language ofschooling is removed in content from immediate experience,and is structured in different ways.Sco lion and Scollon (1981) contrast their two -year.old daughter with other children at Fort Chipewyan inAlberta, Canada. They discover that she has skills in"decontextualization of the information structure," includingfictionalizing her own role in narration. Shetells a story about an experience she had, referring toherself in the third person. In a fascinating discussion,Scollon and Sco llon demonstrate how a preschoolerhas an orientation that is characteristically literatecompared with a ten-year-old Chipewyan child who canread, write and even type! That is to say, literacy isnot a set of mechanical coding and c:ecoding skills, but282

INVESTIGATING LITERACY 283rather a way of processing information which will affectways of interacting.Scollon and Scollon conclude, as do others, thatthese skills are the result of specific socializationpatterns in the forms of conscious parental instruction.They suggest that prose, or "essayist literacy" is underlyingour (middle class) caretaker talk, and language andactivity structures of Western schooling. Michaels (1981)examines language in classrooms with black and whitechildren. She finds that white teachers interpret blackchildren's differences in narrative style during "showand tell" time as behavioral or cognitive deficits.While the work of Scollon and Scollon and Cook-Gumperzand Gumperz make it clear that there are a variety oforientations that are socially and situationally determined,Michaels raises the dilemma that only one style islegitimized in certain settings.Researchers who have approached literacy with thepurpose of understanding and solving problems of childrenin schools have raised new questions and challenged oldways of viewing literacy. They suggest that literacy isnot only a set of mechanical skills residing in individuals.Rather, new insight can be gained by viewing literacyalso as a mode of communication that takes on meaningwithin specific social contexts.Literacy and the Social OrderScholars who have investigated literacy practices intraditional societiesGoody (1968) among themhavefound that writing is associated with a wide variety ofpolitical, social and economic activities. Reder andGreen (1983), in a thorough study of literacy activitiesin an Eskimo fishing village, follow two major historicalstrands in the development of literacy on the island:Russian and English. By observing the practice of literacythatis, who writes what, where and under what conditionstheydiscover that there are distinct domains ofliteracy activity associated with each strand. In thisstudy, Reder and Green present an analysis in which theydemonstrate that social activity both shapes and isshaped by the distribution of literacy activities. There-283

284 THE HMONG IN TRANSITIONfore, literacy provides a window into the structure ofthe social order, and the social arrangement furthers theunderstanding of literacy practices in cultural context.Heath (1980) brings the framework closer to home.Examining the functions and uses of literacy in a blackworking class community in the southeastern UnitedStates, she argues that there are discontinuities betweenthe meanings of reading and writing for families in thecommunity as opposed to in the schools where their childrenare enrolled.Ogbu (1980) makes a very strong case that culturaldiscontinuity alone cannot explain the "failure" of blackchildren in American -.:hools. He argues that blacks, andother "caste-like" minorities find themselves in a societywith such strong gatekeeping mechanisms (e.g., jobceilings) that no amount of mastery of literacy skills in"essayist" or any other style is likely to affect or altertheir position in the economic order. By looking atthe black example, it becomes apparent that observing thefunctions and uses of literacy within a minority group isnot sufficient for understanding the role of literacy inadaptation. It it; also necessary to examine the socialand economic context within which the minority must operate.Literacy and Adaptation: Preparation for InquiryMuch of the literature on literacy and "oral/literate"cultures makes generalizations that are broad andtar-reaching. These generalizations, while appealing,become problematic in light of the bothersome detailsthat :.,re yielded when researchers study the use of languagein everyday life. For these reasons, I argue thatresearchers should bring several assumptions to the task.1. The meaning of reading and writing at e problematic.Studies of people in communities show that definitionsof reading, writing and literacy are problematicboth within and between groups. As Gilmore (1982)listens to claszroom teachers lament that the kidsdon't know how to write, she observes the childrenpassing a se. ies of elaborate notes to one anotherunder their desks. The social context will determinewhat kinds of reading or writing "count" or not.2-84

INVESTIGATING LITERACY 2852. The relationship of speech and text are also problematic.While Olson's distinctions between the natureof utterance and text are stimulating, researcherswho study actual language use find that even thesedistinctions become fuzzy. Tannen (1982) finds thatfeatures used earlier to characterize text versusspeech could be found to differing degrees in thespeech of two different groups. Lakoff (1982) reflectson ways in which writers use both oral andliterate strategies in written communication toachieve specific effects. Observations of some Hmongfamilies reveal that oral and written modes of communicationare being used in interesting ways. Lettersthat relatives in other states or countries sendto these families are usually read out loud to severallisteners. Letters often include cassette tapeswith oral messages. Speech and text are not distinctentities with clear boundaries. They are modes ofcommunication within a communicative economy, inwhich changes in technology and/or social organizationchange the balance of how these modes are usedand how they interact.3. The third assumption is twofold: first, that individualsare rational beings whose behavior reflects rationalchoices within their perceived options, andsecond, that inquiry into human behavior must takeinto account the individual, the group(s) to whichs/he belongs, ariI the wider social context in whichthe group(s) operate. The "orality" or "literacy" ofcultures is meaningless if we know nothing of thosewho participate in them. Accounts of individuals aremeaningless without recognition of the wider contextin which they operate.4. Investigation must have an empirical base. Suspicionsabout the nature of literacy are interesting,but investigation of real people using language forspecific ends holds the most promise for discovery,and for interesting surprises.We are fortunate, indeed, to have the opportunity toexplore the nature of literacy in our own backyard. Whatis most fortunate is that we have the chance to collabo-285

286 THE HMONG IN TRANSITIONrate with our rugged, resourceful new neighborstheHmong.SOME SOCIAL CONSEQUENCES OF LITERACY:REFLECTIONS OF A BACKYARD ETHNOGRAPHERTo begin to explore the role of literacy in everydaylife, I arranged to spend four weeks with Chou and Saiand their two children. On weekdays, after our respectivedays of studying, childcare and teaching, we sharedthe evening meal. Weekends afforded more time for visiting,sewing, and excursions (i.e., to Reading to visita sick brother or to the Spectrum for an evening of wrestling).I discovered that four weeks was time enough to dolittle more than lay the foundations of a friendshipiffriendship grows from sharing not only great adventures,but also daily routines. Also, as I looked on while Chouand Sai navigated their way through welfare and educationalbureaucracies, community activities and a varietyof interactions with both Americans and Hmong, I mademany discoveries.I will suggest in these pages that literacy, as oneamong many modes of language, can be viewed as (1) a toolused for negotiating with new institutions, (2) a toolpossessed by those who mediate between culture groups,and (3) a tool associated with articulating new socialstatus in a changing social order.By describing some glimpses I have caught through mytiny keyhole, I hope to create a framework in which literacycan be examined as an addition as well as an agentof change in a whole communicative economy, with profoundeffects on the relationships of the participants.Literacy As a Tool for Negotiating with New InstitutionsOne Tuesday afternoon, Chou received a letter fromhis caseworker saying that he had to show up at her officeFriday morning, or his "case" would be "closed."The letter instructed him to bring four documents withhim, including one which had to be stamped by an agencyseveral miles away (and not easily accessible by publictransportation), and others which required picking upvarious forms with signatures from different places.286

INVESTIGATING LITERACY 287Chou had already planned to take off from school onThursday afternoon to pay phone and electric bills, Hesays he has to pay them in cash because if he puts morethan a very small amount of money in a checking account,his benefits will be cut. This left only Thursday morning,since he had a test in school on Wednesday that hefelt he could not miss.That evening, Chou spent time usually set aside forhomework to fix his bicycle. He knew that he could notcover all necessary bases on foot in one day. He calledhis English teacher to tell her he would be absent onThursday and Friday.On Friday, Chou reported to idc caseworker with allof the forms that he had gathered. Etnth his signatureand Sai's, were required, and both 'nut to be made in thepresence of the caseworker. So when Chou got home atabout midday, he took over the child care so that Saicould repeat the journey. We took the Broad Street subway.I showed Sai how to distinguish the northbound fromthe southbound train, and how to find the right stop forfuture ref erence.At the welfare office, we waited in bolted-downwooden seats for Sai's name to be called. On the wallacross from where we were sitting were two written signs:"EFFECTIVE 11-8-81 Foodstamps will be given out only between2:30 and 3:30 p.m." and also, "PLEASE NOTE: Anyfootistamps not kicked up within five days will be returnedto Harrisburg:. Sai's name was called. Shesigned the form and we left.Chou comments that he likes his present caseworkerbecause "she help us everything." Just lucky, he reflects.In his previous apartment in another neighborhood,the old caseworker "window broken, he don't care.No heat, he not do anything. New baby born, he don'thelp extra money." Some neighbors have difficulty evenwith the "good" caseworker. An older woman living onefloor below does not know how to decipher forms, letterscr bills. Preparing the required documents is an insurmountabletask for her. One evening Chou disappearedwith a pan full of meat. Returning to the apartment, heexplained that he had intervened with the caseworker onthe woman's behalf, getting $125 and $60 for gas on twooccasions. Therefore, the woman lets Chou and Sai use287

288 THE HMONG IN TRANSITIONher oven from time to time to help them save money ontheir own gas bill.When the old woman cannot solve a problem, she sometimesbrings it to Chou. When Chou is stuck, he turns tome or to another American in his church. He had done thison many occasions with puzzling bills or difficult homework.Sometimes a favor is returned in kind, as illustratedabove with the gas money. Other requests for helpcause conflict.Chou had been complaining for a couple of days abouta man who wanted him to write a fraudulent letter to welfare.The man, who lived six blocks away, had asked Chouto claim that he lived with him so that he could pick upa welfare check at Chou's address. I arrived one afternoonto find the two in frank discussion. When the manleft, Chou was angry. "He want me to cheat for him....He not my relative, he never help me anything. He justmake trouble for me." Chou felt that the request was inappropriate,and that it jeopardized his own standingwith welfare. I advised him not to do it, as had anotherAmerican.I'm not sure why Chou was called upon for this favor.He is neither a neighbor nor a clan member. I suspectit may be related to the fact that Chou is among thefew literate household heads who would be able to performthe task.Chou later received a call from a man who he explainshas a leadership position in Philadelphia. As inmany other cities, social service agencies hired the earlierHmong arrivals who became bilingual to work in theresettlement process. These people have become the primemediators between the caseworkers and the Hmong populationin the area. Not coincidentally, Chou tells me, themembers of some clans seem to fare best in matters thatrequire attention from the social service system. Forexample, Chou said that when he first arrived, the leadercalled the police when Sai was in labor. He did not takethe time to accompany the couple to the hospital as hedid for certain other families. Because he spoke verylittle English at the time, Chou felt helpless when thedoctors decided to perform a Caesarean section on Sai.He is sure that it could have been avoided with the rightadvocacy. In Laos, this leader would appropriately beexpected to aid members of his own clan. As a caseworker288

INVESTIGATING LITERACY 289in Philadelphia, however, he is seen by Americans as arepresentative of "the Hmong" and is asked to be an advocatefor members of many clans. The old and new expectationsdo not always harmonize.The leader, then, asked Chou to do this welfare letterfor the young man. While listening to the Americanson one hand, Chou felt, I suspect, trapped by powerfulsocial and political pressures from within the Hmong community.The dilemma had no clear solution. As the phonecontinued to ring, and Chou's homework lay neglected, heshouted in exasperation, "Too many calls! Too many questions!"With that, he pulled the phone wires out of thewall and sat down to do his homework, complaining, "Everybodywant something from me!"From these ex.imples, it becomes clear that decodingand composing documents play an important role in economicsurvival. Those who have literacy skills must usethem to gain and maintain benefits, and those who don'tmust often rely on those who do. Who may rely on whom isproblematic. Whereas dependence relationships once residedwithin families and clans, new categories such ascaseworker/client, cotenant/client or urban neighbor/clientpose new possibilities and pressures for differentkinds of interaction in this new setting.Literacy As a Tool Used by Those Who Mediate BetweenCulture GroupsMembers of Chou's church feel their way as they accommodatethe hefty new population of refugees. Incorporatingthe Hmong into religious life was a priority.On any given Sunday, one or several Hmong couples arebaptized, becoming full members of the "church family" (aterm used by one of the Sunday school teachers).Chou used to spend several hours each Saturdaystudying the Bible with the pastor. Thus prepared, heled Sunday school Bible lessons in Hmong for those whocould not understand English. The group consists mostlyof new arrivals, women and older men. With the Lao Bibleand English bible close at ;land, Chou is able to synthesize,each lesson to convey it to the group.Until their language mastery improves, newcomers dependon interpreters like Chou to open a sort of lifelinewith the church, permitting exchange of information and289

290 THE HMONG IN TRANSITIONparticipation that would otherwise be impossible. In aconversation with me over a year ago, the pastor confidedthat he was having difficulties at that time because theHmong leader/translator had just moved with many in hisclan to resettle in Rhode Island. To his relief, Chouhas moved into the vacant position, bringing "wisdom"from the American religious leader into the world of theHmong, and providing a way for the Hmong to become partof the church community.When Chou does his short reading for the whole congregation,he is, in another sense, bringing a bit ofHmong culture over for consumption by Americans. Hisseat next to the pastor during the service, and his nameprinted in the service program legitimize both the taskand Chou himself as mediator in the exchange.Because his knowledge of English makes him accessible,I suspect, several congregation members have calledChou to offer him small wages for doing work in theirhomes. When he is unable to comply with all offers onhis one precious Saturday, he is usually free to supplythe name of an alternate.Chou's role as mediator extends past the boundariesof the church. During my stay at Chou's home, a socialworker from a neighboring church called, asking him totranslate a sign that requested users of a tree clothingroom to limit the amount that they took, from English toHmong. Chou had been identified to her as an able writer,and she contacted him immediately. That evening heput aside his homework and wrote and rewrote the messageuntil he was satisfied that it was well done. The socialworker picked it up that same evening, thanking him forthe effort.This woman, a community worker, depended on Chou toconvey information to Hmong people who were using theservices of her own church. Chou is therefore becoming amediator for the neighborhood as well as for his ownchurch.Chou spends time deliberately cultivating relationshipswith Americans. Before I moved in with him and hisfamily, he would often call me on tne telephone for noapparent reason except to chat. One afternoon during myfour-week stay, when I returned to his apartment, I foundhim going systematically through his memo-pad phone directorycalling each American on his list, one by one.

INVESTIGATING LITERACY 291The content of each conversation was more or less thesame, as he asked each person what was "up" and spoke ofhis own state of affairs. My impression was that he enjoystrying to keep his American "friends," includingteachers, church membersand oddballs like ethnographersengagedas long as possible. The decision he andSal made to let me live with them for a month was anotherdaring move to allow an American into his world.Incorporation of Americans into Chou's life surelyhas its benefits, but it must also have its price. AfterI moved in with Chou and Sai, a young Hmong man began"accidentally" kicking Chou repeatedly during soccergames. Sai found herself the object of suspicious questionsat church. Chou and Sai asked which Hmong people Ihad told about the stay, and asked that I keep the arrangementprivate. I didn't know whether the conflictarose out of jealousy, suspicion or some other factor.At one point, I asked Chou and Sai if they wanted meto leave. They assured me that I could stay, but theirrequest for my discretion made it clear what a sensitiveissue this was. There is surely ground for much investigationto better understand the gains and sacrifices involvedin becoming a bridge between two culture groups.Literacy As a Tool Associated with Articulating NewSocial Status In a Changing Social OrderThe New Year: Tradition and Change. One November,my oldest student, fifty-eight-year-old Jou, invited meto join her in New Year's preparations. She led me to ahome on Powelton Avenue where thirty Hmong women werecrowded into one roont, wrapping colorful sashes overtheir black velvet garb. Coin- and silver-bedecked childrenran in and out from the street, playing with eachother. Jou's daughter helped the other women bind theirhair in dark turbans with striped bands criss-crossedover them. One lovely teenager, who was heavily ladenwith coins and beads and had black velvet wrapped aroundher calves, threw a shiny green silk jacket on over herdress. Across the back, in large letters, was printed"ROLLER DISCO".A couple of months before this New Year celebration,a meeting of the Hmong Association was called. A coupleof hundred Hmong went by bus to gather at the Indochinese231

292 THE HMONG IN TRANSITIONCenter. The association was then headed by a young manwho is in his thirties. Jou's son showed me a flyer, inHmong, with instructions and news of committee responsibilitiesfor different aspects of the affair. It outlinedfinancial arrangements, job responsibilities andsubsequent meeting times.In the hills of Laos, New Year was an event anticipatedall year. Heads of clans in each village, alwaysthe eldest men, met to arrange the event. Apparently NewYear's festivals were planned at different times for differentplaces, in order that villages might reciprocallyinvite one another to their activities. Young men andwomen woulo line up in rows facing each other, throwingback and forth a cloth ball that the girls had made fromtheir sewing scraps. Boys would sing, serenading, beginninga process of courtship that would become manifestwith a rash of marriages following the New Year.How has literacy as a new communicative mode playeda part in changing the social order? Previously, a householdhead would call the meetings by word of mouth. Responsibilitieswere delegated and carried out accordingly.As Hmong people find themselves spread throughout acity, written materials become useful to organize meetingsand arrange for bus transportation. As informationsharingbegins to depend on a literate mode, what happensto the role of elders in leadership? How will decisionsbe made when young people are the only ones who have directaccess to information and resources?On the second day of the New Year festivities, costumedyoung men and women line up facing one another on astrip of park grass in Philadephia's Germantown. Theythrew green tennis balls back and forth. Jou complainedthat the boys didn't know the words to the traditionalsongs anymore. But can a Hmong boy still impress a girlwith blessings and prayers he has memorized? Theseskills will no longer be crucial to prove himself a goodhusband. In an environment where information is recordedon paper, what traits will be sought or desired in amate?The Craft Sale. It takes time for an outside observerto discover the talents of Hmong women. Returningto the apartment in the afternoons, I would often findSai sewing decorative squares while she rocked her youngestchild on her back in a brightly embroidered baby car-292

INVESTIGATING LITERACY 293rier. Characterized by tiny precise stitches, reverseapplique and stunning colors, the handiwork is awesome.To my delight, Sai often works with her sister and mother,as well as other passers-through. All are happy tochatter or sit quietly as I join them with my crocheting.From such gatherings, I got the idea of inviting myfriends to a party at which the Hmong women could displaytheir pieces for sale. So I planned one. The party succeededto the tune of $700 of collective sales. Themoney was distributed to the individual artists whosepieces were sold.I suggested to Chou and Sai that each woman contributeone or two dollars for a kitty to cover refreshmentsfor the party, and to start some collective savings forrenting craft tables or even eventually saving for astorefront. Chou spoke out: "Hmong people can pay forfood at the party, but not to save money together. Myname (clan name), I help (members of samedan). We don't trust money together."There were the divisions, starkly clear. Familyhelps within family, clan helps within dan. Cooperationwas only possible to repay the American authority. Aninteresting conflict then arose. By miscommunication betweenmyself and another American who was keeping books,money was not collected for reimbursing me for the refreshments.I brought it to Chou's attention. "Don'tworry," he comforted me. He would collect the money fromthe people in Sunday school. He could not collect fromthe other sale participants. "They not obey me," hesaid. "But people in my church, I think we can helpyou." A new unit of cooperation is being formedmembersof a congregation who can be influenced to act by a youngman who may or may not share thei - name.Chou once commented to mit, "I don't have familyhereI have my church. People help me, give me a desk,dresser, chair...." Indeed, the church has been a sourceof furniture and clothing for Chou and Sai. During myshort stay at their home, individual congregation memberslent a space heater when the heat was broken andmoved to investigate the delay in bringing Chou's orotherover from Thailand, among other things. They have providedthose things for which, in the past, Chou couldonly turn to "family" or dan.293

294 THE HMONG IN TRANSMONIt remains to be seen how, as Hmong refugees adaptto their new urban environment, literacy will take itsplace within the communicative economy. Likewise, as thenew mode of communication enters the social fabric, itwill be interesting to note what impact literacy has onthe process by which relationships between people evolveand change.NOTE: On Being an EthnographerIt would be easiest to say that I wanted to understandmy students in order to be more effective as a literacyteacher, that I wanted to know how they processedthings, what they could use, and what was important, so Icould make the classroom work. The truth is, however,that I became an English teacher because I wanted to findout how these people made sense of the worldnot theother way around! Peasant farmers, who once organizedtheir lives around the rhythms of subsistence survival,find themselves in a foreign, literate city. How do theymanage?To many of the Hmong I have known, I am an Englishteacher. I help them unravel the language, tell themabout their new city and decipher bills and other strangedocuments. To Chou and Sai I have been a frequent visitor,an unsuccessful language learner, an appreciativeconsumer of Hmong food, a source of rides, homework helpand tickets to the Spectrum wrestling matches. I havealso been a hostess to parties where Hmong women can selltheir crafts. And I am one who does not understand therelationships between actors in the "Hmong Community"- -community being my own choice of word and perhaps not onethat the Hmong would use to descrLm themselves. I havealso been a source of conflict for Chou and Sai, whosehospitality has caused various tensions for them withothers in the community. I don't know what other thingsI am to them. A source of favors? A social worker? Anenigmatic curiosity? A lone person to take in and nurture?A source of language input and help? A friend?To me, Chou and Sai have been many things. They aremodel survivors. From Laos, across the Mekong, to boringdirty refugee camps, to Powelton where they were robbedand their neighbors' children beaten by angry black kids,2J4

INVESTIGATING LITERACY 295they somehow managed to continue with gentleness, courtesyand good humor. They are a keyhole into a complexworld that I have barely glimpsed. They are young people,making decisions about how to balance the old andthe new. They have been hosts who have fed 1;:e, housedme, invited me to sleep next to them when it was cold.They are fun companions to watch wrestling with, and theyare family to share good meals with. They are newcomerswho need tips about getting around, and they are new languagelearners who can use my help in some areas.I suspect that they have given me far more than Ican give them. The "homestay" is over, but the feelingof home remains. In response cc their obvious hurt/disappointmentwhen I prepared to move back home, I left mynightgown for occasional overnight visits. That seemedto appease their suspicion that I was pulling out forever.Indeed, if luck is with me, I will know Chou andSai for a long timeand have the good fortune to continuea process with them of having fun together, sharingthings, figuring each other out, and learning from eachother more about the world and how it can look from differenteyes.REFERENCESBruner, J.S. 1973. "From Communication to Language: APsychological Perspective." Cognition 3:255-287.Cook-Gumperz, Jenny, and John Gumperz. 1977. "From Oralto Written Culture: the Transition to Literacy". InWriting: The Nature, Development, and Teaching ofWritten Communication. Vol I: Variation in Writing:Functional and Linguistic-Cultural Differences, editedby Marcia Farr Whiteman, pp. 89-109. Hillsdale,N.J.: Lawrence Er Ibis= Associates.Gilmore, Perry. 1982. "Gimme Room: A Cultural Approachto the Study of Attitudes and Admission to Literacy."Ph.D. dissertation, University of Pennsylvania.296

296 THE HMONG IN TRANSITIONGoody, Jack. 1968. "The Consequences of Literacy." InLiteracy in Traditional Societies, edited by JackGoody. New York: Cambridge University Press.Greenfield, P.M. 1972. "Oral or Written Language: TheConsequences for Cognitive Development in Africa,the United States and England." Language and Speech15:199-77.Havelock, Eric A. 1963. Preface to Plato. Cambridge,Mass.: Belknop Press of Harvard University Press.Heath, Shirley. 1980. "The Functions and Uses of Literacy."Journal of Communication 30(1): 123-134.Lakoff, Robin. 1982. "Some of My Favorite Writers areLiterate: The Mingling of Oral and Literate Strategiesin Written Communication." In Spoken and WrittenLanguage, edited by D. Tannen, Norwood, N.J.:Ablex Publishing Corporation.Michaels, Sarah. 1981. "Sharing Time: An Oral Preparationfor Literacy." Presented at Ethnography in EducationResearch Forum, University of Pennsylvania.Ogbu, John. 1980. "Literacy in Subordinate Cultures:The Case of Black Americans." Paper presented at theLiteracy Conference, Library of Congress, Washington,D.C., July 14-15.Olson, David R. 1977. "From Utterance to Text: The Biasof Language in Speech and Writing." Harvard EducationalReview 47(3): 257-281.Reder, Steven, and Karen Green. 1983. "Contrasting Patternsof Literacy in an Alaska Fishing Village."International Journal of the Sociolo: of Lan ua e42: 9-13.Scollon, Ron, and Suzanne Scollon. 1981. "The LiterateTwo Year Old: The Fictionalization of Self." InNarrative, Literacy and Face in Interethnic Communication.Norwood, N.J.: Ablex Publishing Corporation.2 96

INVESTIGATING LITERACY 297Scribner, Sylvia, and Michael Cole. 1981. "UnpackagingLiteracy." In Writing: The Nature, Development andTeaching of Written Communication. Vol. I: Variationin Writing: Functional and Linguistic-CulturalDifferences, edited by Marcia Farr Whiteman, pp.71-87. Hillsdale, N.J.: Lawrence Erlbaum Associates.Tannen, Deborah. 1982. "The Myth of Orality and Literacy."In Linguistics and Literacy, edited by WilliamFrawley. New York: Plenum Press.291

FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISH:A LONGITUDINAL STUDY OF HMONG ADULTSKaren Reed Green and Stephen RederWithin the past ten to fifteen years, adult secondlanguage acquisition has become the subject of a numberof studies. Second language acquisition research in generalhas followed in the footsteps of studies of firstlanguage acquisition, primarily focusing on childrenlearning a language and attempting to answer the question,"Is second language learning like the first?"(e.g., Ervin-Tripp 1974). Adults' experience with languagelearning has been compared with that of children.Studies have pointed out the similarities in the processesof first and second language learning (Chamot1981; Krahnke and Christison 1983); they have also shownthat children and adults tend to follow similar patternsin the order in which they acquire certain grammaticalaspects of the second language (Du lay and Burt 1974;Bailey, Madden and Krashen 1974). Although comparisonsof rate of acquisition and proficiency attained by childrenand adults appear to vary in their findings, in generalsuch studies concur that adults and older childrenacquire the second language faster in the early stages,but that children will usually attain greater proficiencythan adults in the long run (Krashen, Long and Scarcella1979).Considerable discussion has evolved regarding factorswhich may explain these differences in rate of acquisitionand proficiency between children and adults, aswell as individual differences observed among adults.Suggestions include characteristics such as language aptitude,verbal intelligence, sense of self-esteem, willingnessto take risks, attitude toward the language orculture in which it is used (social and psychologicaldistance), and motivation to acquire the language. Otherfactors beyond the individual also considered to be influentialinclude social aspects of the interaction betweenthe two language groups (relative size of thegroups, attitudes toward each other, etc.). Researchers299298

299300 THE HMONG IN TRANSITIONgenerally agree on the importance of social and affectivefactors in second language acquisition (Gardner andLambert 1972; Krahnke and Christison 1983; Lambert 1981;Schumann 1975, 1976; Staub le 1980). However, they havefound such factors difficult to measure.Despite the growing interest in adult second languageacquisition, most of the research to date has beenof only minimal utility to individuals who teach Englishto recent immigrants and refugees in this country. Inpart, this is due to the still limited number of studieswhich focus on adults. However, the major barrier toapplying the research is the disparity between the characteristicsof the adult subjects of most studies and thebackgrounds of the many adults who have arrived in thiscountry in the last five years. In general, researchershave examined the acquisition of a host country languageby foreign students or educated foreigners (such as Americansliving abroad). Such individuals have been highlyeducated in their own countries and enter at a high levelinto the educational system in the country in which theyare studying. The differences are very clear between theeducational attainment and other background characteristicsof university-level foreign students and adult refugeeEnglish as a Second Language (ESL) students fromthird-world rural agricultural backgrounds.The present paper contributes to the growing body ofresearch on factors which influence adult second language.quisition by clarifying the important role of backgroundcharacteristics in the acquisition of English.The backgrounds of the adults in this study contrastsharply with those of participants in other studies: thesubjects here are Hmong refugeesslash-and-burn farmersfrom the mountains of Laos, individuals with little or noformal schooling and often no previous literacy skills.Through a combination of qualificative and quantitativemethodologies, the study examines these individuals' acquisitionof English in light of the differing backgroundexperiences each individual brings to the acquisitionprocess and also considers the influence of each person'scurrent activities.Our research with the Hmong has been conducted inseveral phases:1) ethnographic fieldwork and informal Englishclasses with a cohort of 10 adults,

FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISH 3012) a community survey of 332 households and3) a longitudinal study of the acquisition ofEnglish by the adults in 20 randomly selectedfamilies over a year's time.We began our work with the Hmong by conducting informalEnglish classes with five men and five women recentlyarrived in this country and settled in a large urbancenter on the West Coast. The classes were a richsource of informat i about English language and literacyacquisition and evolving perceptions about literacy, aswell as a vehicle through which to network out into thestudents' extended families. During a year and a half,we participated in the adjustment of these people to lifein their new environment and observed their growing useof English language and literacy skills in their dailylives. This first phase of the research revealed widevariation among Hmong adults in their acquisition ofspoken and written English. To investigate these differencesfurther, information was needed about the distributionof background variables and formative experiencesamong community members.A community-wide household survey (reported in Reder1982) was conducted to obtain a quantitative picture ofthe variation in factors that the previous ethnographicwork had suggested might be critical in language and literacydevelopment. Interviews were conducted with 332household headsapproximately 90 percent of the community.Topics included the educational and linguistic backgroundsof family members over the age of twelve, familymigration history, employment and English language training,economic and religious status, and any additionalconcerns voiced by the interviewees. Also included was aself-report measure of English proficiency by Hmong adulthousehold members. The results of that survey indicatedthat the process of acquisition is heavily influenced bythe background characteristics (e.g., age, previous education,literacy) which individuals bring with them tolearning a second language. In other words, people aredifferentially ready to acquire English when they enterthis country.To investigate the apparent causal links suggestedby the household survey and to measure language prof i-300

302 THE HMONG IN TRANSITIONciency objectively as it develops over time, a longitudinalstudy was designed. This study complements both priorphases of the research. On the one hand, by following asmaller number of individuals over time, the longitudinalwork could achieve greater depth than the household survey;on the other hand, by focusing on objective assessmentand documentation of individuals' language and literacyskills and knowledge at specific points of time, itprovides quantitative data which complement that obtainedin the ethnographic phase of the work.This paper will describe the methodology used totrace this group of Hmong adults' acquisition of Englishlanguage and literacy over a year's time and then focuson some of the study's preliminary results. Once we havediscussed the acquisition documented during this timeperiod, we shall examine the factors which seem to accountfor individual differences in acquisition ofEnglish.Sampling ProcedureMETHODSFrom the comprehensive community survey population,a random sample of twenty families was selected. Sincefindings from the initial phase of the research had suggestedthat most Hmong adults acquire only minimal Englishlanguage and literacy proficiency during their firstyear in this country, the sample of families was stratifiedby length of time they had been in the UnitedStates. The households in the community survey were thusdivided into four groups, based on the length of time thehead of the household had been in the United States: zeroto five months, six to eleven months, twelve to seventeenmonths and eighteen to twenty-three months. These stratawere then filled with families selected randomlyfivefrom each of the four time groups. This purposeful stratificationallowed observation of various periods of languageand literacy acquisition: the observations duringthe longitudinal study represented one year in four differentsegments of a three-year period of acquisition andadjustment. That is, at the end of the study the families3

FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISH 303had been in this country from one to three years, havingarrived between early 1979 and early 1981.Characteristics of the Study PopulationThe sample of twenty families included fifty-nineadults age eighteen and over who agreed to participate inthe year of observation and assessment. Eight additionaladults live in these households, but were not included inthe study due to extreme age, physical or mental disabilities,or reluctance. The fifty-nine participants werenearly equally divided among women and men. Their agesranged from eighteen to seventy-two; nearly half of thesample were between the ages of twenty-five and thirtyfive.As indicated in Table 1, this group as a whole hadrather limited experience with literacy and education: 80percent had not attended school in Laos; 81.5 percentcould not read Lao, a skill which was acquired throughformal schooling in Laos. Literacy in Hmong, on theother hand, was acquired informally; 29.6 percent of thispopulation reported having at least some proficiency inreading Hmong.I The ability to speak Lao could also beacquired informally, through social contacts and trade.Thus, more individuals could speak Lao than could readitslightly over half the population.Maintenance of the CohortDuring the course of this research, the Hmong communityunder studylike many other Hmong communitiesaround the countrybegan to experience the flux of secondarymigration. By the spring of 1982, approximatelythree-quarters of the population had moved to California.By the time of the study's second assessment visit (Julyand August 1982), 85 percent of the study participantshad moved to one of four differksit cities in the CentralValley. Fortunately, they were all willing to continueto participate and we were able to follow them to theirnew homes. However, by the end of the year, 28.8 percentof the individuals who had originally agreed to participatein the longitudinal study had dropped out for varyingreasons, including lack of time or interest, orhealth problems. In addition to that attrition, some individualswho remained with the study until its comple-302

304 THE HMONG IN TRANSITIONTABLE 1Background Characteristics of the Longitudinal Study PopulationFull Sampan and After AttritionVockground CharacteristicsFullsIndividuals Participatingin Both First and LastAssessmentsRangeMean18 - 7235.218 - 7135.9ts Percent n PercentMOOR 28 47.5 18 48.6Nen 31 52.5 19 51.4Proficiency in Speaking LaoNome 26 48.1 17 48.6Some 2$ 51.9 18 51.4Proficiency in Reading !monoNona 18 70.4 26 74.3Some 16 20.6 9 25.7Proficiency in Reading LaoNone 44 81.5 29 82.9Some 10 18.5 6 17.1Education in LaosNone 44 80.0 30 83.3some 11 20.0 6 16.7Education in U.S.Under 1 year 27 55.1 16 47.1Over 1 Year 22 18 52.9Length of Residence in U.S.0-S mouths13 24.5 9 25.74-11 months 24 45.3 16 45.712-17 months 12 22.6 9 25.718-23 months 4 7.5 1 2.9Notes Unless otherwise indicated. thfse data were gathered during thehousehold survey conducted in April and Noy le!).*by the time of Assessment 2 (July and August 1982)393BEST COPY AVAILABLE

FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISH 305tion had been unable to participate in the first assessment.Thus, the number of people who were part of boththe first and last assessments was reduced to 63 percentof the original group.As seen in Table 1, this attrition fortunately hadvery little effect on the composition of the cohort laterms of background characteristics, length of residencein the United States or amount of schooling taken here.Background InterviewsPrior to initiating bimonthly objective assessmentsof individuals' English capabilities, a series of indepthinterviews was conducted with each of the 20 families.These interviews focused on the families' backgroundsand experiences in Laos and Thailand and documentthe nature and extent of individuals' exposure to formaleducation, literacy and other languages prior to theirarrival in the United States, as well as their knowledgeand uses of literacy in the various environments. Informationwas gathered regarding who acquired literacy, whoused it for various purposes and how they were perceivedby others, how the uses and perceptions of literacychanged with the move from villages in Laos to camps inThailand, etc.2Bimonthly Assessment VisitsOnce the background interviews were completed, webegan to visit the families every two months, starting in1?te May and early June 1982, and thereafter continuingover a year's time. Several hours each visit were spentwith each adult family member (men and women over the ageof eighteen).During an assessment visit, time was spent with eachindividual discussing activities and experiences whichhad taken place during the two months since the precedingvisit. A series of assessment tasks was then conducted.Numerous tasks using varied materials were devised to assessindividuals' oral English proficiency, their abilitiesto read and write in English, and their knowledge ofthe uses of English literacy in this country. In addition,proficiency in reading and writing Hmong and Laowere documented. Some tasks were conducted every visitto establish developmental trends, others only periodi-304

306 THE HMONG IN TRANSITIONcally, and some only during the first and last visits tocapture overall changes.A standardized ESL test (the Basic Eng list, SkillsTest, or B.E.S.T., developed by the Center for AppliedLinguistics for use with adult Southeast Asian refugees)was administered during the first assessment visit andagain during the final (seventh) session. All other taskswere specially developed for these assessments; they includethe oral identification of pictured English vocabularyitems, recognition of written English vocabularyitems, repetition of sentences of varying length andgrammatical complexity, reading and writing of materialssampled from selected genres of written English, Hmongand Lao (e.g., a social letter, an agency form letter, ajob application, a change of address form, a sample ofdriver's test questions, etc.), and demonstration ofknowledge of the uses of written materials common indaily life in the United States.In designing these assessment tasks, we drew on ourprevious experience in helping Hmong friends learn English.Care was taken to employ not only a broad range ofwritten materials, but also diverse contexts and methods,since in this phase of the research it was not feasibleto follow this group of adults (fifty-nine individualsdispersed into five different cities) out of their homesand into their interactions using new language and literacyskills in the surrounding host community. Thus, individuallexical items were presented in varied contextse.g., as isolated vocabulary items as well as in thegenre of a form letter, bill of street sign.A "literacy portfolio" of frequently encounteredwritten artifacts was compiled, consisting of actualnewspaper ads, receipts, prescriptions, registrationforms, bills, etc., and photographs of street signs, licenseplates, store windows and so forth. Items in thisportfolio were discussed with study participants in Englishor in Hmong as the individual preferred, since thethis task was not merely to determine the ex-purpose oftent of each individual's ability to read these items,but rather his or her understanding of what they representand what functions they serve. In addition to recognitionof the item and specification of the way it isused, questions were asked about each participant's personalexperience using it, frequency of use, need for

307 FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISHassistance in using it, etc. In this way, informationwas gathered on individuals' exposure to a wide range ofliteracy items and the situations in which they are usedas well as the social networks that have developed withinthe community to cope with the new demands posed by thewider literate society. The portfolio was presented atthe beginning and end of the year of observation to documentacquisition of knowledge and changes in experiencewith the items.Other purposive data collection also took placethroughout the year. Information about individual participants'educational and employment status was updatedevery two months. Twice during the year the types of activitiesoutside the home in which individuals participatedwere discussed. The nature and frequency of theactivity and the role, if any, which literacy (in English,Hmong or Lao) played in it were carefully noted.Preliminary analyses midway through the study indicatedthe importance of native language literacy as a predictorof both oral and written English proficiency. This finding,which substantiated earlier findings from the ethnographicand survey phases of the research, prompted thecollection of additional information about the phenomenonof Hmong literacy. Discussions focused on the uses ofHmong literacy, methods and materials used for acquiringit, and the reasons some Hmong are continuing to learn toread and write Hmong in this country.MEASURING ACQUISITIONTwo of the objectives of this longitudinal studywere, first, to observe and document the ongoing processof language and literacy acquisition in the Hmong community,and second, to identify factors which influence thatacquisition. The results reported here are based on preliminaryanalyses of the first and last assessments ofthree of the measures mentioned above: (1) the Basic EnglishSkills Test (B.E.S.T.), (2) the recognition of picturedEnglish vocabulary items, and (3) the recognitionof written English vocabulary items. Comparisons of eachof these measures, between the first and last visits, indicatethat acquisition took place over the year's time.3.06

308 THE HMONG IN TRANSITION(See Figures 1-3.) The mean scores on these measures improvedfrom 13.2 to 22.7 percent.The B.E.S.T.The B.E.S.T. Core Section is described as assessingfive areas of skill: listening comprehension, communication,pronunciation, fluency, and reading and writing.All but pronunciation were scored in this study. The versionadministered in this study has a maximum possiblescore of 90. The highest score among these adults at thebeginning of the year was 88, the lowest 1; by the end ofthe year the range was 0-83. Thus, both at the start andthe finish of the year, the range of abilities in thispopulation was quite wide. The group mean for the totalscore on the B.E.S.T. was 35.57 at the start of the year;by the end it had gone up 15 percent to 40.92. (See Figure1.) Fifty-four of the fifty-nine adults in the longitudinalstudy took the B.E.S.T. at the beginning of theyear, whereas thirty-eight took it at the end. Thirtysevenpeople took both tests; comparative analyses presentedhere include only these individuals.Pictured English Vocabulary RecognitionThis task used pictures to measure the acquisitionof oral English vocabulary by Hmong adults. During eachof six bimonthly visits, participating adults were askedto identify ofally pictures of a randomly assigned set ofvocabulary items in English. The population of picturesused in this task was taken from the Peabody ArticulationCards (PAC), originally designed for use with childrenand adults whose speech production requires further developmentor remediation.3 These cards were selected foruse here becluse of the clarity and realism of the pictures.Since the medium is pictures, the words are mainlynouns; some verbs and a few adjectives (colors) areincluded. The 350 pictured words in the PAC were screenedto eliminate items which were ambiguous to Hmong adults.From a total of 210 acceptable words, six sets of thirtypictures were randomly generated. Fifteen additionalwords were selected for inclusion in all six assessments:randomly selected items and five items purposely cho-tensen for their relative simplicity and frequency of use.307

9044424031363432302126V 24H 22u; 20o; 1$4 16141210music 1BASIC ENGLISH SKILLS TEST (S.E.S.T.)nr. 37First Testbob, 14Last Test:1 Year Interval)642\;10MALSCORE(MaximumPossible 90)LISTENINGCOMPREHENSION(MaximumPossible 10)COMMUNICA-TION(MaximumPossible 44)COMPONENTFLUENCY(MaximumPossible 24)SCORESREADING SWRITING(MaximumPossible 12)

io4S444240383634323028262422co20k 18co16141210B6420TOTALSCORE(MaximumPossible 4S)FIGURE 2PICTURED VOCABULARY RECOGNITIONn 34COREWORDS(MaximumPossible IS)11111 First TestLast Test(10 Months Interval)L-- COMPONENT SCORESINONCOREWORDS(MaximumPossible 30)3

FIGURE 3WRITTEN VOCABULARY RECOGNITIONn = 34353432302826242220181614121086420IFirst Testv. N Last Test(10 )'onths Interval)TOTALCORE(MaximumPossible 35)CORENONCOMWORDSWORDS(Maximum(MaximumPossible 10) Possible 25)ICOMPONENT SCORESI310

311312 THE HMONG IN TRANSITIONThe five "easy" items were included to ensure that everyonewould know some words, thus preventing individuals atlow levels from becoming overly discouraged.Vocabulary growth was measured by assessing knowledgeof the six sets of words, one set per visit. Variationin the order of these sets across individuals madeit possible to avoid confounding the potential effects oforder of testing with growth of vocabulary over time.4The fifteen "core" items were added to each set to beused as an indicator of change or acquisition from oneassessment to the next, before the full six sets of wordshad been administered.Of course, repeated assessment ofthese fifteen core words every two months might confoundthe effects of repetition with those of spontaneous acquisition.However, once the six assessments were complete,the rate of growth of core and noncore items couldbe compared to determine approximately how much influencerepetition over assessments might have had.The scores on the noncore pictured vocabulary itemsranged from 0 to 25 (out of a total of 30) on the firstassessment and 0 to 2.3 on the last; core scores rangedfrom 0 to 13 and 0 to 15, respectively. As illustratedin Figure 2, the mean number of noncore words recognizedincreased from 11.82 on the first assessment to 13.38 onthe sixth, up 13.2 percent. The mean number of core wordsrecognized increased from 8.27 to 10.18, a rise of 23.1percent. The larger increase among core items suggeststhat in addition to the overall vocabulary growth takingplace, repetition of core items over sessions also had aneffect on learning.The interval between the first and last assessmentsof this and the task described below was approximatelyten months. Thirty-four individuals participated in bothfirst and last assessments of these tasks.Written English Vocabulary RecognitionIn this task, reccSnition of isolated written lexicalitems was used as one measure of literacy. Using aprocedure similar to that used to generate materials forthe pictured vocabulary task, six partially overlappingsets of items were developed and represented to individualsin varying orders over the six visits. Twenty-fiveitems were unique to each set and ten were common to all

FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISH 313six.5 During each of six assessment visits, participantswere asked to read and paraphrase each item in the givenset of thirty-five English words, phrases and common abbreviations,written in boldface type on three-by-fiveinch index cards.Scores on the noncore written vocabulary itemsranged from 0 to 20 on the first assessment and 0 to 21on the last; core scores ranged from 0 to 8 and 0 to 9,respectively. The group mean for noncore items rose 22.7percent, from 3.88 words out of 25 to 4.77. (See Figure3.) The mean core item score rose 29.3 percent, from3.31 words out of 10, to 4.15. Clearly, this task wasmore difficult for the majority of participants than wasthe pictured vocabulary recognition: A full 50 percentof those participating in both first and sixth assessmentsrecognized no noncore written words at all in thefirst session and that figure dropped only slightly, to47 percent, during the last session. Thirty-eight percentknew no core written words the first time, 29 percentknew none the last. This contrasts greatly with thepictured vocabulary results, in which only one to two individualsknew no noncore items.Relative ProficienciesBeyond documenting the acquisition which has takenplace, this study examined individual variations in acquisition.The Hmong's acquisition of English is by nomeans uniform. Like other immigrants before them, notall Hmong adults master the English language or literacy.A wide range of proficiencies and levels of functioningexist even among individuals who have been in this countryfor several years. As noted above, the range of proficiencyamong these longitudinal study participants isconsiderable: on each of the measures reported here,scores ranged from 0 to close to the maximum possible.Such differences in learning need to be better understood.In general, individuals acquire some Englishproficiency early in their resettlement here. Accordingto the results of this study, acquisition continues at aslow pace over time. Analyses of these results furthershow that the relative levels of proficiency within thegroup remain much the same as acquisition progresses.That is individuals who scored low at the beginning of3.1.2

314 THE HMONG IN TRANSITIONthe year of observation continue at a low level relativeto the group at the end of the year. The best predictorof how a person performs an assessment task at the end ofthe year is how he or she performed at the beginning. (Onthe B.E.S.T., for example, individuals' scores on thetest at the first of the year predicted 89 percent of thevariance in scores on the test at the end of the year.)When we consider the progress made by a smaller group ofHmong adults over the past three years (that of the Hmongadults to whom we taught English at the beginning of thisresearch project), a similar pattern emerges: all havemade progress over those three years, but their relativepositions %within the group with regard to their Englishproficiency remain the same today as they were threeyears ago.FACTORS AFFECTING ACQUISITIONIf individuals tend to maintain their relative levelof proficiency (there are exceptions, of course), whataccounts for the initial differences? That is, what factorsunderlie individual differences in the acquisitionof English? Our analyses indicate there are many variablescorrelated with individual differences in acquisition.Many of these variables reflect differences in theexperiences individuals had in Southeast Asia, prior tomost Hmong's attempts to learn English;6 others involveexperiences concurrent with the learning process.? Forexample, a consistent pattern is evident between individuals'assessment scores and their personal backgroundcharacteristics, particularly formal education in Laosand proficiency in reading Hmong or Lao. For the mostpart, the educated group scores higher than the noneducatedand th,: literates higher than the nonliterates.(See Table 2.) Similarly, a consistent pattern emergesin relation to individuals' current activities, such asattending school, having American friends or using Englishat the store. (See Table 3.) All of these variablesshow a relationship to proficiency in the English languageand literacy. The difficulty lies in determiningwhich of these many correlations reflect causal relationshipswith English acquisition.3i 3

TABUS 2English Proficiency Scores by Personal Eackground CharacteristicsProficiencyMASUKIIISScaleValuesEducation in LaosProficiency inReading LaoProficiency inReading SoongMOM Some Pons Bowe Non. Some Women PenSex9.1.0.T4. n 31 6 29 6 26 9 le 190-90First Test Total 20.1 70.3 29.7 60.0 27.3 e2.0 23.9 46.6Last Test Total 34.4 74.7 35.2 73.5 32.2 69.3 27.4 53.7Pictured Vocabulary a 30 4 27 4 24 7Noncom Words - First 0-30 10.9 19.3 11.1 20.3 10.0 17.4 n.s.*moots Words - Last 0-30 12.5 20.0 12.9 21.0 12.5 19.0 n.s.Core Words - First 0-15 7.7 12.3 0.1 12.3 0.0 11.0 n.s.Core Words - Last 0-159.0 13.1 n.s.Written Vocabulary n 30 4 27 4 24 7 15 19Noncom Words - First 0-25 2.2 16.5 2.6 15.3 2.3 11.0 1.3 5.9Wormers Words - Last 0-25 3.1 17.0 3.6 16.3 2.9 13.1 n.s.Core Words - First 0-10 2.6 6.0 2.9 1.0 2.5 7.0 2.0 4.2Core Words - Last 0-10 3.5 9.0 3.9 9.9 3.5 7.7 n.s.tet Analyses of variance indicated differences in proficiency by these personal characteristics tobe significant at least at the .05 level unless otherwise noted.Difference is not significant at the .05 level.314

TABLE 3English Proficivon Scores by Current Personal ActivitiesAttending School Speak English Speak to Speak EnglishNov or Within Nave American With American Americans Speak English to theLast 2 Mouths Friends Friends Daily at the Store DoctorProficiencyScaleMeasures values No Yes No Yea No Yes No Tee No Tee Mo Tee11.11.S.T. n _13 24 26 11 2O 9 19 le 21 16 30 60-90First Test Total 15.6 44.11 25.11 51.5 29.3 55.0 24.1 47.7 23.1 51.9 211.5 70.3Last Test Total 17.5 53.6 34.6 55.9 35.0 59.3 26.5 56.2 27.2 50.9 34.2 74.7Pictured Vocabulary n 10 21 24 7 15 16 16 15 25Nonoore Words - First 0-30 0.0 14.2 n.s. n.a. 0.9 15.4 9.9 14.7 10.11 19.2Noncom Verde - Last 0-30 9.0 16.5 n.s. n.s. 10.1 17.11 11.3 17.0 13.1 20.0Core Words - First 0-15 6.3 O.@ n.s. 0.0 11.0 7.1 10.1 7.2 10.2 n.s.Core Words - Last 0-15 7.2 12.0 n.s. n.s. 0.5 12.3 0.7 12.3 9.6 14.0Written Vocabulard n 10 21 23 @ 24 7 15 16 16 15 25 5dancers Words - First 0-25 .3 6.1 2.7 5.5 2.7 9.7 1.5 G.@ 1.1 7.7 2.4 r4.2M0000[*lbas - Last 0-25 .2 7.6 n.s. B.S. 1.5 0.7 2.1 11.6 3.6 14.2Core Words - First 0-10 .7 4.9 2.S 5.5 n.s. 1.6 5.3 1.9 5.2 3.0 7.0Core Words - Last 0-10 1.0 5.9 n.s. n.s. 3.0 6.0 2.S 6.4 3.9 O.,i Ana yam variance indicated differences in proficiency __ these personal activities to s . _._ .gn can . a . leastthe .05 level unless otherwise noted.rl 1 ',. '*Difference is not significant at the .05 level.aJBEST COPY AVAILABLE

FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISH 317Interpretation of the relationship between activitiesor experiences occurring at the same time as the acquisitionbeing measured is problematic. At best, we canonly say there is a positive association, since the influencesmay flow both ways. For example, ESL traininghas a very significant correlation with all of the assessmentresults to date. The mean score at the end ofthe year for individuals with less than a year of ESL inthis country was 25.9 out of a total of 90 points possibleon the B.E.S.T., whereas the mean for those with morethan a year of ESL was 56.8. Similarly, those not takingESL within the last two months had a mean of only 17.5;for those presently attending ESL, it was 53.6. Althoughit would thus appear that ESL training helps, we cannotsay from these results whether continued attendance inESL classes improves proficiency, or whether individualswho are having success with language learning in general(outside of school) are the ones who continue to takeESL, whereas others who are less successful get discouragedand drop out. The same difficulty of interpretationholds for other kinds of ongoing experiences, such ashaving American friends. Perhaps only those individualswho have reached a certain level of proficiency are willingor able to make social contacts with Americans. Thefact that the people with American friends are also theones who exhibit high levels of proficiency does not necessarilymean that having American friends increasesone's proficiency.Interpretation of associations between proficiencyand prior characteristics or experiences is morestraightforward. In such cases, it can be assumed thatone variable is causally prior to the other (e.g., thatage may affect English acquisition but English acquisitionmay not "affect" one's age). Using multiple regressionanalyses, we have found that individual differencesin acquisition of English language and literacy arelargely determined by such prior variables, that is, byindividuals' English proficiency as measured by theassessment tasks de3cribed abovethe B.E.S.T. and thepictured and written vocabulary recognition tasks. Althoughthe degree of their influence varied from task totask (see Table 4), the following characteristics werefound to exert the greatest influence on individuals'English proficiency:316

TAILS 4Regression &justices for Predicting Inglish ProficiencyAdjustedProficiency MeasuresPredic asta Slaniff^ancaB.R.O.T.First Test.602 Proficiency in ReadingSoong 5.943 .4220 11.406 .002Alp -.730 -.3590 9.026 .004Education in Laos 3.436 .2737 4.004 .035Last Test .344 Proficiency in ReadingRoom; 5.615 .3549 0.024 .020Age -.920 -.3902 10.576 .003Sea 15.316 .2970 4.044 .039Pictured VocabularyNoncors WordsFirst Test.236 Proficiency in ReadingLao 1.051 .3442 4.303 .047Last Teat .342 Age -.364 -.6035 16.612 .000Corso WordsFirst Test .250 tin-.155 -.5319 11.443 .002Last Test .122 Age -.161 -.4979 9.559 .006Written VocabularyNoncore WordsFirst Test.590 Education in Laos 1.256 .3917 7.701 .011Proficisncy in ReadingMoon 1.565 .3010 7.323 .012Age -.150 -.2026 6.735 .040Last Teat .502 Education in Laos 1.450 .4201 0.703 .007Proficiency in ReadingBoone 1.643 .3750 6.965 .014Core WordsFirst Test.416 Proficiency in Readingamong .911 .4540 0.752 .006Proficiency in ReadingLao .952 .3425 4.904 .034Lest Test .436 Ago-.129 -.4494 10.000 .004311Proficienoy in ReadingOmni .072 .4100 0.400 .007

FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISH 3191) prof iziency in reading Hmong (a person literatein Hmong tends to score higher than someone whois not),2) at (a younger person tends to score higher thanan older person), and3) education in Laos (an individual who receivedsome formal schooling in Laos tends to scorehigher than someone who is not educated).As seen in Table 4, one of these three variables ora combination of them was found to exert a significantinfluence on the results of each of the proficiency tasksreported here. For example, these three characteristicspredict 60 percent of the variance among individuals'performances on the B.E.S.T. at the beginning of theyear. At the end of the year, proficiency in readingHmong and one's age are again important predictors. (Aminor variation occurs in the third variable to enter thepredictive equation, but this variablesexonly contributes5 percent to the amount of variance predicted.) Ageis a primary influence on individuals' performances onthe pictured vocabulary recognition, and education inLaos and proficiency in reading Hmong are the major predictorsin the written vocabulary recognition task. To alesser extent, proficiency in reading Laowhich was primarilyacquired through formal schooling and is thushighly correlated with education in Laos (.79)occasionallyexerts a significant influence.These results, based on objective measures of languageand literacy over time, confirm what was suggestedpreviously through the survey methodology and our priorethnographic experience with the Hmong: the importanceof background ciaracteristics. Thus, although this longitudinalstudy is based on a relatively small sample ofindividuals, the findings provide some insight about therelative influencethat is, the interrelationshipsoffactors which affect adult second language acquisition.That literacy, age and previous education are influentialin adult language learning is not surprising. Thesevariables have long been thought to be important. Butexploring why and how they are influelicial and why otherbackground factors are not may help us to understand318

320 THE HMONG IN TRANSITIONfurther the process of second language acquisition byadults Ṫhe finding that age, and sometimes sex, continue toplay a role in predicting English proficiencyeven afterthe effects of their correlation with other backgroundexperiences are statistically controlledsuggests thecontinuing influence of those historical, social and culturalfactors which enabled only certain members (primarilyyoung men) of current Hmong refugee communities to goto school when they lived in Laos. That is, the socialexpectation that young people (men, in particular) shouldgo to school may be combining forces with the sense ofinadequacy experienced by adults finding themselves inthe midst of a totally different, complex, often unintelligibleenvironment, and with the discouraging observationmade by most parents that their children arelearning faster and more than they are. Over and overagain people expressed the belief that they wrre too oldto learn (even though they were in their late twenties ormid-thirties). The continuing influence of age as a predictormay also be simply a result of the still preliminarystage in which most Hmong adults find themselves intheir acquisition of English. Teenagers have arrivedwith more education and literacy than their parents andreceive more concentrated schooling here. In addition,because of their rapidly increasing language and literacyskills, they are constantly being called upon by adultsto use those skills. Thus, the idea that the young learnfaster becomes a self-fulfilling prophecy.One of the factors that is puzzling for its absenceof influence is the amount of time individuals have beenin the United States. While other background characteristicsdo predict acquisition, the passage of time in theUnited States per se does not appear to be a significantpredictor. The reasons for this are not clear. The factthat time is not a predictor is not because individualshave stopped learning. The results reported here showacquisition taking place nowand these individuals havebeen here from two to four years.Time is usually considered to be a significant factorin adult language acquisition because it is assumedto reflect the amount of individuals' linguistic input.But factors other than time may actually determine theamount of English input individuals receive. It is clear312

FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISH 321from our fieldwork that Hmong adults have very differentopportunities for exposure to and use of the host language.Some Hmong have suggested, as one reason for notmoving to areas of high Hmong concentrations in California,that the more heavily concentrated the Hmong populationis in an area, the more socially isolated Hmongadults are from contacts with native English speakers.Following this theory, in areas like the Central Valleyof California or the Twin Cities of Minnesota, adultswould have less social contact with native speakers.Self-reported data on social contacts with Americans bystudy families before and after the move to Californiahave been gathered, but have yet to be analyzed. Ethnographicwork after the longitudinal study found thatindividuals in a heavily Hmong community in Californiafelt they had infrequent contacts with native Englishspeakers.Another characteristic conspicuous for the absenceof its influence is the ability to speak Lao. A commonassumption is that experience learning a second languagecan facilitate learning a third. Ervin-Tripp (1973:95),for example, speaks of different learning strategiesbased on different language acquisition experiences. Inmultivariate analyses of our data, fluency in Lao doesnot play a significant role at all. The point to ponderhere is the finding that previous literacyHmongliteracyexerts more influence on acquisition of spokenEnglish than does previous bilingualism. The relativelylimited influence of literacy in Lao is less difficult tounderstand. Once the relationship of Lao literacy toeducation in Laos is statistically controlled, previouseducation appears as a more influential factor than literacyin Lao.We are still at a preliminary stage in analyzing therelationship of literacy in Hmong to acquisition of English.The nature of this relationship is not quite asdear -cut as that of some of the other background characteristicsbecause acquisition of Hmong literacy is stilltaking place in this ountry: 47 percent of those who reportedthemselves as literate in Hmong have learned it orare still learning it here in this country. There may besome levels of mutual influence between English acquisitionand Hmong literacy. For example, it is possiblethat acquisition of English (if one learns the alphabet)3,20

322 THE HMONG IN TRANSITIONcould facilitate subsequent acquisition of Hmong literacy.However, considering these results together withthe previous survey results and our ethnographic experiencewith the Hmong, it seems likely that the primary influenceis in the other directionthat is, that Hmongliteracy is affecting English acquisition.There are several reasons why Hmong literacy mightaffect English acquisition. An obvious one is that bothEnglish and Hmong literacy (as practiced by this group ofHmong) use the Roman alphabet. To individuals literatein Hmong, the characters used to write English at leastlook familiar. But since the Romanized Popular Alphabet(RPA) orthography for Hmong does not represent the phonologyof English, it is not clear how familiarity with thealphabet itself would foster acquisition of spoken English.Several possibilities come to mind. First,adults' acquisition of oral and written English may be sointertwined that the obvious advantage which the familiarityof the Roman alphabet seems to bestow on Englishliteracy acquisition carries over to spoken English aswell. Second, many Hmong participate in formal ESLtraining, which relies heavily on the use of primers,worksheets, boardwork and other written materials, sothat those familiar with the alphabet may progress morequickly and benefit more from formal training. Third, aspreviously reported, Hmong adults use their native scriptactively as a tool in learning Englishcopying new material,transliterating its pronunciation and translatingit into Hmong. All of these applications of Hmong scriptare later studied at home. The social segregation ofHmong adults and their resulting isolation from nativeEnglish speakers may well give a significant edge tothose able to "study" English through the tool of Hmongliteracy. If this is the case, the English proficiencygap between people literate in Hmong and those who arenot might be expected to widen further in California.Another kindof factoran affective onemay be atthe apparently powerful role of Hmong literacy inwork inEnglish language acquisition. Most individuals literatein Hmong script chose to become literate on a selfselectedbasis (in contrast, for example, to most of theindividuals literate in Lao, who were sent by others toschools in Laos where Lao literacy was learned). Perhapsthere is overlap between the motivations or aptitudes321

FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISH 323underlying acquisition of Hmong literacy and English;individuals who desire to learn Hmong script tend to bepredisposed to learning English. The fact that manyHmong believe that literacy in Hmong script facilitatesacquisition of English seems consistent with this line ofexplanation. In a surprising twist, the acquisition ofnative language literacy in Hmong may be seen not as anindicator of interest in cultural maintenance, but ofopenness to change, if not assimilation.CONCLUSIONSThis longitudinal study set out to document Englishacquisition by a group of Hmong adults over a year's timeand to explore the factors which appeared to be affectingtheir progress. Based on the three measures reportedhere, we found that acquisition was indeed taking placethat people were learning slowly, but still progressingtwo to four years after their arrival in this country.Individuals' relative positions within the group withregard to proficiency level remained much the same overtime.Both the experiences individuals have prior to comingto this country and their current activities arehighly correlated with their English proficiency. rue tothe ongoing nature of current activities (such as attendingschool, having American friends, speaking English atthe store, etc.), it is difficult to isolate their mutualinfluences on one another. The causal relationship ofbackground variables with English acquisition can be analyzed,however. Multivariate analyses indicate that certainbackground characteristicsespecially proficiencyin reading Hmong, age and education in Laosare powerfulpredictors of English proficiency. Further data mayidentify the reason why some variables seem to influenceacquisition, whereas others do not. As noted, many othermeasures not considered in this paper were gathered duringthe study. As the additional data are analyzed, particularattention will be paid to the apparently strongrole played by literacy in Hmong and to the pervasiveinteraction between literacy and language acquisition.322

324 THE HMONG IN TRANSITIONACKNOWLEDGEMENTSThere are numerous people whose help made this studypossible. Our colleagues, Michael P. Sweeney and Mary L.Cohn, contributed thoughtfully to the study design andlabored tirelessly in the field. We are equally gratefulto Dan X. Mua, our research assistant, and the many Hmonginterpreters who arranged and facilitated interviews andaccompanied us throughout the fieldwork. Thanks also toour colleague, Nancy Faires Conklin, for her comments onthe manuscript. And finally, we would especially like toexpress our appreciation to the twenty Hmong families whowelcomed us into their lives and tolerated seemingly endlesshours of questions and discussions. Their good willand hospitality formed the basis for this research.This work was funded by a contract from the NationalInstitute of Education ;Ho, 400-83-0005). The opinionsexpressed in this publication do not necessarily reflectthe position of the National Institute of Education, andno official endorsement by the Institute should be inferred.NOTES'These figures were compiled for the longitudinal studypopulation from the proficiencies reported in the householdsurvey. Ensuing visits with these individuals indicatedthat more literacy in Hmong exists than was previouslyreported. This discrepancy is due in part tocontinuing acquisition of Hmong literacy in this country.These more recent data have yet to be fully analyzed.2The results of these background interviews have beenpresented in Reder, Green and Sweeney (1983) and inReder, Green, Sweeney and Cohn (1983).3Developed by Smith (1971).4A Latin Square design was used to vary the order of administrationrandomly across subjects.323

FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISH 3255Each of the six sets of noncore items consisted of tenitems randomly selected from the pictured vocabularyitem sets and fifteen items chosen from the literacyportfolio materials and other literacy artifacts inwhich these vocabulary items were more contextualized.The core words included five words from these last twosources and five easy words common in everyday life.32i

Pearson Correlation. of Boors. by Personal Background CharacteristicsProficiencyMeasures Ago SexProficiency inSpeaking LaoProficiency inReading MangProficiency inReading LaoEducationin LaosEducationin U.S.TimeinU.S.B.S.S.T.First Tam! Total -.58 .50 .32 .67 .63 .68 .56 .04Last Test Total -.5a .52 .30 .63 .56 .59 .61 .06Pictured VocabularyNoncom Words - First -.53 .04 .02 .46 .50 .44 .64 .17Nbncors Words - Last -.67 .15 -.01 .40 .44 .34 .66 .24Cora Words - First -.66 -.03 -.05 .38 .41 .41 .59 .14Cora Words - Last -.64 .10 .01 .38 .33 .34 .57 .04Written VocabularySonoocs Words - First -.41 .39 .23 .10 .70 .78 .43 .13Noncom WOrds - Last -.40 .31 .12 .65 .64 .69 .52 .28Cora Words - First -.45 .34 .08 .60 .55 .56 .64 .35Coes Words - Last -.59 .22 -.06 .54 .51 .54 .58 .20is Unless ot cm sf indicated, data on background c aractsc st es wars gathered duringthe household survey conducted in April and May 1981.of individuals who took both first and last Nis!by !ha time of Assessment 2 (July and August 1982)3° J

Pearson Correlations of Scores' by Current Personal ActivitiesProficiencyMeasuresAttending SchoolNow or WithinLast 2 MonthsINave AmericanFriends11Speak EnglishWith AmericanFriendsSpeak toAmericansDailySpeak Englishat the StoreSpeak Englishto theDoctor11.1.6.T.First Test Total .55 .46 .49 .52 .63 .6SLast Test Total .67 .36 .41 .56 .62 .56Pictured VocabularyMowers Words - First.47 .26 .34 .52 .36 .49Somme Words - Last .S1 .26 .29 .55 .41 .37Core Words - First .47 .32 .37 .45 .44 .36Core Words - Last .56 .29 .33 .46 .46 .41Written VocabularyMoncore Words - First.45 .44 .46 .44 .55 .72Moncore Words - Last .52 .29 .32 .54 .46 .60Core Words - First .63 .31 .33 .60 .52 .49Coro Words - Last .60 .32 .27 .47 .56 .47ndivldwls who took both first and last test326

328 THE HMONG TRANSITIONREFERENCESBailey, Nathalie, Carolyn Madden, and Stephen Krashen.1974. "Is There a 'Natural Sequence' in AdultSecond Language Learning?" Language Learning 24(2):235-243. Reprinted in Second Language Acquisition:A Book of Readings, edited by Evelyn MarcussenHatch, pp. 362-370. Rowley, Mass.: Newbury HousePublishers, Inc., 1978.Chamot, Anna Uhl. 1081. "Applications of Second LanguageAcquisition Research to the Bilingual Classroom."Focus, No. 8, Sept. Rosslyn, Va.: NationalClearinghouse for Bilingual Education.Du lay, Heidi, and Marina Burt. 1974. "Natural Sequencesin Child Second Language Acquisition." LanguageLearning, 24(1):37-53. Reprinted in Second LanguageAcquisition: A Book of Readings, edited by EvelynMarcussen Hatch, pp. 347-361. Rowley, Mass.:Newbury House Publishers, Inc., 1978.Ervin-Tripp, Susan M. 1973. "Structured Language Acquisition."In Language Acquisition and CommunicativeChoice: Essays by Susan M. Ervin-Tripp, edited byAnwar S. Dil, pp. 92-129. Stanford: Stanford UniversityPress.Gardner, Robert C., and Wallace Lambert. 1972. Attitudesand Motivation in Second-Language Learning,.Rowley, Mass.: Newbury House Publishers, Inc.Krahnke, Karl J., and Mary Ann Christison. 1983. "RecentLanguage Research and Some Language TeachingPrinciples." TESOL Quarterly 17(4):625-649.Krashen, .Stephen D., Michael A. Long, and Robin C.Scarce Ila. 1979. "Age, Rate and Eventual Attainmentin Second Language Acquisition." TESOL Quarterly13(4):573-582.qot.,"04

FACTORS IN INDIVIDUAL ACQUISITION OF ENGLISH 329Lambert, Wallace E. 1981. "Bilingualism and LanguageAcquisition." Paper presented at Conference onNative Language and Foreign Language Acquisition, atNew York Academy of Scier::es, January.Reder, Stephen. 1982. "A Hmong Community's Acquis;of English." In The Hmong in the West: Observaiand Reports, Papers of the 1981 Hmong Rese .,Conference, edited by Bruce A'. Downing and DouglasP. Olney, pp. 268-303. Minneapolis: Center forUrban Regional Affairs, University of Minnesota.Reder, Stephen, Karen Reed Green, and Michael P. Sweeney.1983. Acquisition and Use of Literacy in theHmong Community of Newton: 1981 Annual Report of theFunctional Literacy Project. Portland, Ore.: NorthwestRegional Educational Laboratory.323

PAP,T FOURHEALTH CARE ISSUES32J

INTRODUCTIONAmos S. DeinardIn an effort to provide high quality medicalforcareIndochinese refugees, medical clinics in the UnitedStates trained and hired bilingual interpreters. It soonbecame apparent, however, that language was not the onlybarrier to be hurdled. Health care providers must makeother modifications to win the confidence of groups of peopleunaccustomed to the Western approach to medical care.Medical personnel at first did not appreciate howoverwhelming it was to be confronted with so much thatwas new. Indochinese groups resisted pressure to acceptquickly and on faith new methods of curing. This was especiallytrue for the Hmong, who had the least experiencewith Western-influenced medicine. The health care professionalsdid not know how the different refugee groupsconceptualized causes and cures of disease. The rush todo good before adequate background information could begathered resulted in conflicts that prejudiced consumersagainst some specific medical treatments.For human service organizations to survive and toremain credible, they must continually consider how theycan best meet the needs, both objective and subjective,of those who use their services. Neither the health careproviders nor the consumers alone can define these needs.Rather, there must be a constant determination that whata health service organization offers and what the consumerdesires are matched to their mutual satisfaction. Thisrequires an ongoing dialogue between the provider and theconsumer. Without such considerations, instead of developinga harmonious relationship with the consumer, thehealth care system may be faced with angry, suspiciousand uncooperative parents and patients. A fragmentedrather than a unified health care program may emerge asparents and patients, terrified at the thought of certainprocedures, shop around for noninvasive or nonsurgicalalternatives. Both the consumer and the provider mustacknowledge the notion that good health care requirescompromises between their different cultural perspectives.333330

334 THE HMONG IN TRANSITIONThe following rapers touch on some of these considerations.Thy fast three help us to understand Hmongperception of illness. They suggest the possibility thatWestern medical practitioners might incorporate someHmong ideas or at minimum utilize them in some mannercomplementary to their own.Lemoine points out that "Hmong shamanism is not areligion but a way of healing." Illness is seen as an integralpart of the individual's whole being. Causation isascribed to the activities of the unseen or "spirit" aspectof life which only the shaman is able to access. Hesees particular benefit in utilizing shamanistic activitiesin the treatment of psychophysiological diseases.Bliatout also believes that many Hmong who have resettledin Western societies and who suffer from mentalhealth problems would benefit from using one or more ofthe wide variety of Hmong traditional healing arts. Hispaper provides a solid review of the types of traditionalHmong healers and of traditional beliefs about mentalhealth problems. He discusses which factors should beconsidered when deciding whether it is appropriate to refera Hmong client to traditional healers. He points out,quite astutely, that "since Hmong t 'tments never hurtanyone, but could possibly help a patient," it should beconsidered as part, of a patient's course of treatment, tobe offered in conjunction with Western regimens.Thao's paper considers the Hmong's perceptions ofillness and traditional ways of healing more from a medicalstandpoint than from the mental health standpoint.The paper contains a good discussion of the concept ofsoul loss, of ways of recallinz the soul, and of naturaland organic causes of illness. He explains the Hmong beliefin "a mandate of life," i.e., the idea that everyperscn is given a mandate to be a human being, with illnessocurring because the mandate of life predeterminesillness and death occurring, should if ensue, becausedeath is due. Thao points out that "the Hmong haveevolved many different methods to deal with each causalelement and to counteract illness as a whole. By learningmore about Hmong attitudes about health and illness,Western health care providers can do much to improve thequality and appropriateness of care for this specialgroup of new immigrants."331

INTRODUCTION 335Munger reports on field work conducted in the BanVinai refugee camp on the problem of sleep disturbancesand sudden unexpected nocturnal death syndrome (SUNDS) inHmong refugees. His paper reviews the work of other investigatorswho have studied this puzzling, complex phenomenon.Several reports of SUNDS cases in Ban Vinaierode the popular conception that SUNDS only affectsHmong who have resettled in the United States. Reactingto the sensitivity of the subject matter and to the attitudesof the Hmong toward medical research in general,Munger urges that a survey of sleep disorders not be attempteduntil a clinicalresponse can be offered to individualsidentified as having a sleep disorder. This positionappears to run contrary to the scientific processpracticed in the Western world, wherein research is oftenconducted even though no treatment is available. Mungeraddresses the need to involve scientists from both themedical and social science areas as this issue is studied,and he correctly points out the importance of teachingHmong refugees cardiopulmonary resuscitation and howto use emergency medical services.Oberg, Muret-Wagstaff, Moore and Cumming conducted acrc4s-cultural assessment of maternal and child interaction.The findings of their well-designed study, reportedhere, are provocative. On an assessment of parent-childinteractions, Hmong mothers were shown to obtain higherratings on parenting skills than a mixed Caucasian sampledid. The Hmong mothers were rated as more attentive,expressive, sensitive, responsive, and patient in childrearing.The quality of the mother-child relationshipthroughout infancy has clear implications for the subsequentdevelopment of the child. Contrary to observationslink!ag isolation, stress and lack of social support ineconomically disadvantaged groups to negative parentchildinteraction patterns, Hmong mothers demonstratethat the ability of a mother to be sensitive and responsiveto her toddler is not entirely dependent on the absenceof stress or on higher socioeconomic status. Theauthors concluded that "the sensitivity shown by theseHmong mothers seems related to a cultural priority forattentive child-rearing and the availability and use of astrong extended family network."Chun and Deinard, in their paper on undue lead absorptionin Hmong children, address an issue that has332

336 THE HMONG IN TRANSITIONbeen of primary importance to those providing health careto Southeast Asian refugeesparticularly Hmong children,in Minneapolis, Minnesotasince the first years of theirimmigration. Basic descriptive information of Hmong childrenshown to have undue lead absorption draws attentionto several features that distinguish them from Americanchildren who in the past were identified as having thesame problem. The Hmong tended to be younger, with severalbeing only six months of age at the time their unduelead absorption was detected. The problem, however, appearsto be fleeting, for the peak of Hmong cases occurredin 1981. Since 1983, the Hmong incieence is moreakin to that of the American population. Of specific interestin many of the Hmong cases is the failure to finda source of lead in the homes. The authors concluded thatthe etiology of undue lead absorption detected in Hmongchildren is far from delineated. However, in view of theHmong's documented tendency to use herbal medicines andother cultural kinds of medicines, a consideration mustbe given to the possibility that, at least in some cases,the etiology of undue lead absorption is related to thecovert use of such medications.The final paper, by Hurlich et al., describes theattitudes of Hmong towards medical research projects.The authors, in conjunction with other investigators,wished to pursue a study of the sudden unexpected nocturnaldeath syndrome and associated electrolyte abnormalitiesboth in Ban Vinai and in several locations in theUnited States. The authors summarize related projectsand describe in detail the frustrations they encounteredin attempting to initiate the project, which includedblood drawing, in Ban Vinai. Questions raised by theHmong in the camp and in the United States reflect thefrustrations that they have had with Western medicalpractices and suggest reasons why the Hmong ultimatelybecame unwilling to participate in the study. On thebasis of their collective experience in working withHmong communities inthe United States and Thailand, theauthors describe several important lessons that theylearned about undertaking research, particularly regardingthe responsibility of researchers toward refugee communities.333

SHAMANISM IN THE CONTEXT OF HMONG RESETTLEMENTJacques LemoineINTRODUCTIONYou may wonder why I chose the subject of shamanismin a meeting mainly concerned with the problems facingHmong refugees in a Western cultural environment. Thisvery morning, different orators have been debating Hmongculture, Hmong history and Hmong ethnicity from the perspectiveof change. To summarize my own experience, Iwould like to say that the present ethnic consciousness ofthe Hmongwhether they are in the United States or inFrance or (I suppose) in Australiais a strong feelingthat they all belong to a unique people who are not onlydifferent from other peoples but who are positively resolvedto avoid any kind of assimilation.In this respect, Hmong refugees are faithful to theirethnic past. Since the dawn of their history, as part ofthe Miao group of peoples in China, they managed to escapethis fate of assimilation. This is the key to explaintheir political attitudes in Indochina since they fled andsettled there in the mid-nineteenth century. Their involvementin the two Indochina wars, their early upsurgeagainst the French colonial system in the 1920s, as wellas their latest resistance to Vietnamese and CommunistLaotians who wanted to drag them into a bureaucratic slavery:all their choices, even their political mistakes,have been dictated by the same collective urge to preserveHmong ethnicity. We have to keep this in mind if we reallywant to help them in their resettlement problems. Forthey did not come to our countries only to save theirlives, they rather came to save their selves, that is,their Hmong ethnicity. It would be an egregious misunderstandingto advise them to forget it all and become anothersheep in the flock, even if we can allow them morefreedom than they could expect in the country they fled.I don't need to elaborate on this. Obviously, this meetingwould not take place if we were not all conscious ofthis fact.Let us now consider for a while the meaning of Hmongethnic consciousness among the refugee population. Be-337334

338 THE HMONG IN TRANSITIONsides being born a Hmong, it means using (at least) onedialect of the Hmong language. Through the language, allthe Hmong share a measure of Hmong traditional culture.As it is not the same amount for all, I avoid speaking of"one people, one culture." This would misrepresent thefacts. Hmong culture is still very alive, but in the pastsixty years, with the opening of the ethnic group to contactswith other people and other cultures, Hmong culturehas lost its overall command of Hmong thoughts. The Hmongindeed have experienced what has happened to most peopleson earth; they have been influenced by foreign traditions.Some of them have converted to Christianity (with all thetransformation of thinking that this entails). Some othershave given up all religious practices, whether Christianor traditional. Finally, part of these two groupsand a certain number of more conservative individuals havetried to pattern their social and political activities ona more sophisticated neighbor, namely the Lao plain-dwellerswho also were the rulers of the Hmong people's adoptedcountry, Laos. From the mixing of these diverse currents,new trends have emerged among the younger generations.The results range from dancing Lamvong to rockdancing, playing country music and singing with a guitarin a band, to free matching and marriage, to divorce andsexual liberation, and so forth.In time of adversity and persecution, Hmong culturehas always conjured up a messianic gathering of Hmong peoplein order to oppose what was thought of as alien oppression.This was the case with Pachai, who dared torebel against French colonial agents from 1918 to 1921.This was also the case of Ya Shong Leu, who in 1967 challengeda Hmong authority in Western guise: General VangPao, the administrator of Military Region No. 2. The reasonI mention this latest messianic reaction to the growinganxiety of the Hmong who were facing more and more destructivewarfare, is that Ya Shong Leu's ideas did notdie with him in the early seventies. They must havespread widely among the people, for they surfaced againafter the great flight to Thailand in 1975. These ideashave produced a new Hmong church, whose influence is nowswiftly spreading among the refugees of the younger generation.The establishment of a church instead of the gatheringof an army is in itself a remarkable adaptation tothe circumstances. In the seclusion of a refugee camp,

SHAMANISM IN THE CONTEXT OF HMONG RESETTLEMENT 339those who established the new fait;1 could observe (andprobably regret) the growing impact of Catholic and Christianchurches on their fellow tribesmen.Thus, when we now speak of Hmong culture, we mustkeep inmind not only a perennial Hmong tradition, but alsothat of the iollowers of Christian and Catholic Churches,the neo-traditional standards of the new faith, andfinally a materialistic current. I must apologize forhaving expended so much on the religious situation. Ifound it necessary to locate the cultural setting of therefugees before introducing Hmong shamanism. Whateverhistorical influences it may still carry, Hmong shamanismremains a central part of Hmong traditional culture. Andit is that part which is now facing the conflicting trendsof Christianity and neo-Hmong cults.SHAMANISM: A HMONG WAY OF PSYCHOTHERAPYHmong shamanism is not a religion but a way of healing.Hmong traditional beliefs form a religion of soulsand spirits, gods and devils. It also answers man's mostagonizing questions about the genesis of life, death anddisease, and the perpetuation of life through reincarnation.All these Hmong metaphysics are well explained inan initiatory song to the dead: the Qhua Ke ("Showing theWay") (see Lemoine 1983). Shamanism deals only with prophylaxisfrom disease and death. Contrary to some reports,a shaman never performs religious rites as such,like the funeral, the releasing of a dead person's soul orthe commemorative buffalo sacrifice to dead parents. If ashaman is seen taking part in a ritual it is never as ashaman but in his capac!ty as a rituali4c after he has secureda further qualification. All qualificationsfromplaying and singing musical parts in weddings and funerals,to dealing with a corpse and, later on, with his returningsoulare learned orally from a renowned specialist.An initiation fee is usucaly paid to the master.Shamanism is also a tradition transmitted from master todisciple. But it differs from all other kinds of learning.Besides being taught, it has to be inherited.The shaman's activity differs from that of the ritualist.One may say that it consists of a shaman exploringthe unseen part of reality in order to chase devils336

340 THE HMONG IN TRANSITIONand restore somebody's health. The total reality thateveryon' experiences in this world includes what you seeand what you do not see. For an ordinary Hmong person, itdoes not follow that unseen reality does not exist. Theunseennamely the souls of the people, the spirits of thedeparted, tutelary gods, wild spirits and evil influencesisstrongly perceived as existing even when ordinaryhuman eyes cannot catch a glimpse of it. The maindifference between a Hmong shaman and an ordinary fellowis that the shaman can enter the unseen part of realityand deal with it. For that he needs to put himself inwhat Michael Hamer (1980) rightly called "a shamanicstate of consciousness." It means that he must leave thisside of reality and enter the unseen. Two of his accessoriessymbolize this transition. The first is the veilwith which he covers his head, meaning that he is absentfrom this side of reality, blind to us. The other one isthe lamp that must be alight on his a!tar, shedding lighton the unseen.To enter the unseen, a Hmong shaman, as most shamansin other cultures, has to reach this state of shamanicconsciousness. He attains this goal through a trance.Trance is a distinctive feature of the Hmong shaman. Onecannot be a shaman without a predisposition to go into atrance. But trance is not only achieved through physicaltraining. It is an inherited ability that one cannot haveat will. It must be provoked by spirit helpers. Thesounds of the gong, the rattle and the finger bell certainlygive the tempo and help the shaman to alter his ownconsciousness. But they accompany rather than precede theshaman's singing. The trance always starts first, for itis, as the shamans themselves explain, rovoked by spirithelpers.SPIRIT HELPERS AND SHAMAN'S VOCATIONWhere do the spirit helpers come from? When a Hmongshaman dies, his altar is undone and thrown away in thejungle. But his tools are carefully kept by his sons. Infact when he feels himself about to die, he summons allhis sons to come to his bedside and gives them the contentsof his bowl of magic water to drink. This bowl isalso known as "the dragon pond" (lub paas zaaj). It is33'7

SHAMANISM IN THE CONTEXT OF HMONG RESETTLEMENT 341the place where the dragon who rules lightning and thundercomes to rest when invited by the shaman. To most EastAsian peoples, thunder and lightning, in their capacity tokill suddenly, are the paragon of magic arts. And severaltraditions, including Chinese Taoism, claim to derivetheir magic from thunder power. A Hmong shaman also enjoysa special relationship with the Thunder-Dragon.Properly invited, it will become another spirit helperDrinking the Dragon-Water (dtej Laag) means that the childrenwill absorb the essence of magic in their bones andflesh in order to keep it in the descent line. "I shallgive them my bowl to drink because I want them to keep myspirit helpers. If they are incapable of shamanizingthemselves, then I shall be reborn as one of their sons, Ishall be able to be a shaman again...," says Chu Yao, aproficient master.As a matter of fact, the spirit helpers of a deceasedshaman will not exactly set themselves up in the bodies ofhis sons. What will stay with them after drinking themagic water is a mark, a token of the former alliance.The spirit helpers themselves will return to their base:the Cave of Nya Yi on a high cliff in the third layer ofheaven. This is the original abode and the common base forall the Hmong shamans' spirit helpers, Nya Yi (or Shibeing the first shaman who initiated the practice. Shamansof this world are but the heirs of his talents.The dead shaman himself will join his troop as a headspirit (thawj neeb). After a while, when he thinks thatthe time has come, he sends two of his troops, the InspiringSpirit (Leej Nkaub) and the Spirit of the Trance(Tsheej Xeeb), to "importune" one of his sons. These twospirits are essetial to the vocation of the shaman. Theywill make him sick in a strange way, with fever, attacks,shivering and pain, until he calls for a shaman who willdiagnose his illness as the vocation. This shaman willsummon his patient's spirit helpers and the veracity ofhis diagnosis will be proven by the subsequent relief experiencedby his patient. So it becomes plain that hisproblem is the vocation imposed on him by these spirithelpers, who will not leave him alone until he starts callingthem regularly, following the instruction of the consultantshaman. The latter thus becomes his first master.(It should be noted here that women as well as men canbecome shamans, although female shamans are much less com-

342 THE HMONG IN TRANSITIONmon. The Hmong would explain that a male soul has beenreincarnated in a female body.)Then follows a training period which may last two orthree years, maybe more, during which the disciple learnsfrom his master not only how to call his spirits, but alsothe ways of the world beyond: the important souls, i.e.,the different parts of the self; the way to restore theintegrity of a soul once it has been captured by anotherpower or become a runaway that needs to be fetched back;and last, but not least, the different wild and evil spiritsthat he will have to fight. His spirit helpers cometo him already structured as a task force with scout spirits,engineer spirits, horsemen and foot soldiers. TheInspiring Spirit and the Spirit of the Trance, who neverpart from him, act as staff and liaison officers.Other spirits who have a specific function will bedescribed as animals or insects whose features correspondto that function. For example, the vanguard of the engineerspirit will be the spider who can stretch a thread of"copper and iron" for the troops to settle on. Other engineerswill throw a suspension bridge of "copper andiron" to let the foot soldiers and the horsemen crossspace from one point to the other. This thread and thisbridge appear in the shaman's house where one can see athread of cotton following a specific course between hisaltar and the main door, and on certain occasions, a cloth'bridge also stretched between the altar and the main door.Another insect is a type of fly, the Sphex, whose propertyis to anesthetize its prey (other insects) in order tocarry them alive to its nest. This peculiar hunting habitis used by the Sphex spirit of the Hmong shaman when itis told to get hold of a very volatile soul: the protrudingshadow. Another of these functional spirit helpers isthe woodpecker, who can pull worms and caterpillars fromthe growing bamboo soul. Other animals like the bear orthe elephant will be used for their strength) The hawkand the eagle will be sent to catch a shadow soul inflight, and so on. However, the large number of spirithelpers remain anonymous: horsemen or soldiers, or a Chinesedetachment to remind us of the Chinese sources of theshaman's tradition.3 3 d

SHAMANISM IN THE CONTEXT OF HMONG RESETTLEMENT 343THE SHAMAN AT WORKOnce a shaman is established as a full-fledged practitioner,he erects in his house a two-storied altar.His master comes to summon once more his spirit helpersand invite them to settle there as soldiers in their barracks.Training is not yet completed and each year themaster will come and help his disciple to send his spiritson holiday to Nya Yi's Cave. But the people will startcoming to consult him in his house. More likely, unableto move, they will send someone of their family to invitehim to their house. He will go there with minimal equipment:gong, rattle, finger-bell, divination blocks andveil. All these are carried in a bag by the emissary ofthe sick person. The patient's kin will erect a makeshiftaltar for summoning the shaman's spirits. A bench of thehouse or of a neighboring house will be prepared for theshaman to sit on, because the shaman in most of the tranceis riding his "dragon chargec," his "speedy steed," his"ship of wind and clouds." This fantastic Pegasus is ofcourse represented by the bench on which he is rockingbackwards and forwards, impersonating a very suggestivehorse-riding.His first intervention, as he calls it: "ua neebsal " ,2 will be an exploration of the sickness in order toidentify its causes. If his diagnosis proves right, thesick person will feel great relief within a few days.Then he will decide to call the shaman back to perform thesecond part of his intervention, namely the "healing" (uaneeb khu).3 What need is there, will you ask me, to havea healing follow when the patient is already better? Thisis indeed the great challenge of the Hmong shaman. Psychoanalystssay that a patient is cured when he does notwant to come and pay his analyst anymore. The Hmong situationis the reverse. When a patient feels better hewants to pay his healer who did not charge him anythingyet. If a patient feels no better after a few days or afew weeks, he only calls another shaman to try his talent.But if he feels relief, it means that the shaman's interventionwas appropriate. Then, in order to fulfill hisprevious agreement with wild spirits or gods, or to enforcehis analysis of the disease, the shaman must performa healing ceremony with the participation of his patients.340

344 THE HMONG IN TRANSITIONIt is a dramatic re-enactment of his findings in theunseen, a show of the various causes of the previous illness.For example, supposing that the illness has beenprovoked by the running away of the running bull soul; hewill summon his spirit helpers to go with him to the placewhere he found that soul and bring it back home. In thefirst session (ua neeb sai) when he was still searching,his patient would not participate at all. But this time,he is invited to stay in the middle of the room and assoon as the shaman has brought back his running bull soul,an assistant will sacrifice a piglet in order to build amagic corral around the patient's running bull soul. Theshaman will eventually seal the fence around the runningbull by stamping blood from the sacrificed pig on tne backof his patient. Another pig's life could be used to makean umbrella over the patient to hide him from evil spirits,and so on. In that case life is considered as a materialobject. So a pig's life may be used to repair humanlife.The sacrifice of one or several pigs is usually acentral part of this performance. And it is not (as onemay guess) in order to provide food to spirits but tomend or to strengthen a collapsing soul. For that heneeds to use a similar substance, spirit to back spirit,life to strengthen life. What he does in a pig sacrificeis to take hold of the pig's soul and use it to supportor protect his patient's soul. Both human and animalsouls are thought to be made of the same substance andman can reincarnate in a domestic animal. In respect tothe spirit, the shaman will always burn a small amount ofjoss money to provide him with money for the journey inthe world beyond. But this animal soul will be kept prisoneras an assignment given by the shaman until the end ofthe year. Meanwhile the shaman keeps the lower jaws ofthe sacrificed pigs in a basket close to his permanentaltar. He will burn all of them when sending his spiritson holiday, meaning then that they are freed and can leavetheir duty.The Hmong shaman has two main obligations. One isrestoring the self, which he divides up into five differentsouls: the protruding shadow ( ntsuj, dluab ntsujhlauv), the reindeer soul (nyuj caab nyuj kaub), the runningbull (nyuj Laag ny_ui rhi), the chicken soul (ntsujgat/ ntsuj noo), and the growing bamboo soul (ntsuj xyooj3 4 1

SHAMANISM IN THE CONTEXT OF HMONG RESETTLEMENT 345ntsu j ntoo). This division could be paralleled with theFreudian analysis of the Ego. It has the same purpo3e,namely to come closer to the complexity of man's psyche.By finding and bringing back the runaway, he helps hispatient to recover his psychic balance. But there remainsan obligation: to take over his patient's fight for life.Like antibiotics in a weak body, his fierce troops supplyresistance to the evil miasmas and spirits. Exorcism isthe most thrilling part of his performance. Cleansing orpurification by magic water blown on the patient is alsovery common. All these actions are very theatrical andcontribute to prop up the patient's morale. Certainly, agood Hrr.ong shaman has a decisive influence in this respectand I have been surprised to see that this technique wassuccessful not only with Hmong fellow tribesmen sharingthe same culture, but also with Westerners, provided thatthe shaman's analysis is explained to them. Some shamansnow in Western countries h- remarked that their treatmentof hysteria is far int. t fficient than the drugs ofthe psychiatrist.It does not mean that shamans are opposing the useof drugs or any other therapy. They are often themselvesdistinguished herbalists a-.d they always welcome the useof Western drugs in conjinction with their own magic. Itis not that they doubt their own power, but that they acknowledgetheir healing as essentially psychosomatic. Diseasesand infections may remain out of reach of theirspirit fighters. From my own observations I would con -dude that, in spite of a wild appearance, Hmong shamanismis generaily a positive therapy to counteract anxiety andother psychic and psychosomatic disorders.SHAmANISM AND THE PSYCHOMENTALCOMPLEX OF THE HMONG REFUGEESAfter this brief description, I hope that you now enjoya clearer idea of the part played by the Hmong shamansin Hmong traditional society. This part is obviously torelieve anxiety and to cure psychophysiological diseases.In a traditional context, back in Laos and far from anyWestern medicine, all the diseases will be put into theshaman's hands. But he himself will possibly use herbalmedicine instead of spirit helpers. If no local medicine

346THE HMOIYG IN TRANSITIONworks, he will then try magic and spirit helpers. As soonas a more sophisticated medicine is introduced in the village,it will be used jointly or separately for the patient'ssake, with the shaman's approval.The people who worked in refugee camps must have noticeda large number of shamans at work, making an almostcontinuous musical background all the day long. Tnis maybe indicative of a high degree of anxiety and psychic vulnerabilityfor the Hmong in the refugee camps. But I muststress that it was already the case in the context of atraditional village. I remember counting no less than 30shamans in a cluster of 90 households.In third countries, the number of shamans has decreasedfor various reasons. Some did not want to leaveThailand, where they lived in an exclusively Hmong environment,which they enjoyed very much in spite of thedifficulties of being in a refugee camp. Others were excludedfrom the resettlement list as useless religiouspractitioner'.The fact is that when Hmong refugees are free intheir religious attitudes, they keep calling in shamansto help, even when they have to cover long distances tobring them. The situation in the refugee camp and thesituation in a Western country during the adaptationstage seem to me to be broadly parallel, generating anincreased anxiety about the present and the future. Itwould take too long to enumerate here all the differentsources of stress ending up in more-or-less deep depression.I refer you here to studies by Westermeyer andVang (1983a, b).The causes for anxiety range from unemployment andthe shame of living on welfare, to the religious pressureof a fundamentalist sponsor. It is not in my capacity tojudge the different effortssuccesses and failuresthathave happened in the resettlement of the Hmong in theUnited States. Broadly speaking, they parallel what Ihave observed in France. But when thinking of helpingthe Hmong to preserve their selves, one should pay moreattention to the delicate workings of their psychic balance.I must say that I was shocked to hear in St. Paullast year that the religious sponsors of some of my Hmongfriends would allow them to practice recalling the souls,but not shamanism. This seemed to me a gross misunderstandingof the shaman's role. He is not competing with3 43

SHAMANISM IN THE CONTEXT OF HMONG RESETTLEMENT 347a medical doctor, nor with a priest nor a minister, butwith Dr. J. Westermeyer and his like: the psychiatristand the psychologist. I know Dr. Westermeyer well and Idoubt that he would advise any of his Hmong patients notto see P shaman.Last 'car, together with British anthropologist,Christine Mougne, I conducted an inquiry on the problemof sudden death during sleep among Southeast Asian refugees.To make a long story short, our findings have shedlight on the possible triggers of sudden death among subjectsphysically predisposed to cardiac arrest. Amongthe triggers we found were dreams and night terrors andthe modem substitute for dreams: TV shows. But thestrangest thing is that many of the cases we investigatedwere people who had already adapted or were adaptingthemselves. They had given up their traditional protectionand were highly motivated to competition and successin Western societies. This very motivation drew theminto "the fast lane" of American society, with all theresulting stress and anxiety. The young intellectualswho died were almost exclusively either Christian oratheist. They had no resources yet in the Western contextto ease their stress or anxiety. And they had abandonedtheir traditional protection and help. May I ventureto say that perhaps the comfort of a shaman beforethe stress had beceint: too high would have prevented theirheart failure?But I must also add that there are instances inwhich the shaman did not succeed. Whatever his insight,the shamanlike the psychoanalystnever gets total controlof his patient's mental health, and even when thetreatment has been successful, it may not necessarily bepermanent.In any case, at the time of the seizure the shamanand the psychiatrist are powerless. Only a resuscitationteam that arrives within five minutes of the seizure hasany chance to save the patient, as has been demonstratedin Seattle and Santa Ana, California.I shall conclude by a thought which came to me whilewatching a Hmong shaman curing a young, educated Germanstudent from a deep and long-lasting melancholy. Comparinghis work to psychiatric procedure, I noticed thatwhile the analyst tries to provoke self-analysis byscratching the wounded part of the self, a Hmong shaman344

348 THE HMONG IN TRANSITIONwill provide an explanation which avoids all selfinvolvementof the patient. He is always represented asa victim of an assault from outside powers or of an accidentalseparation from one part of his self. When thissituation has been identified and overcome by the shaman,health is recovered. At no point has there been a feelingof guilt associated with suffering.4 Maybe in thehealing power of the Hmong shaman's art there is a lessonwhich the psychotherapist could learn.NOTES'For instance, to raise up a fallen soul.2Va "to do, to manage," neeb "spirit helpers", sai "tosee". The shaman is himself a txiv neeb "father (ormaster) of spirit."3Va "make", neeb "spirit helpers", khu "heal(cr cure)".4Even when illness is provoked by angry ancestorsto whomthe patient did not offer proper rituals, he is not madeto feel that he is "bad," only that he made a mistakethat can be corrected.REFERENCESLemoine, Jacques. 1983. L'Initiation du mort chez lesHmong, Bangkok, Thailand.. Le chamane Hmong sur les Chemins de l'Au-dela(in presirWestermeyer, Joseph; Tou Fu Vang, and John Neider. 1983a."Migration and Mental Health Among Hmong Refugees,"Association of Pre- and Postmigration Factors withSelf-Rating Scales." The Journal of Nervous andMental Disease, vol. 171, no. 2, pp. 92-96.. 1983b. "Refugees Who Do and Do Not Seek PsychiatricCare, An Analysis of Premigratory and PostmigratoryCharacteristics." The Journal of Nervousand Mental Disease, vol. [71, no. 2, pp. 86-91.345

GUIDELINES FOR MENTAL HEALTH PROFESSIONALSTO HE' P HMONG CLIENTS SEEK TRADITIONALHEALING TREATMENTBruce Thowpaou BliatoutMany mental health professionals are beginning tounderstand and accept that some members of minoritygroups who suffer from mental health problems may behelped by using cultural support systems and/or traditionalhealing practices. It is probable that a greatmajority of the Hmong resettled in Western societies whosuffer from mental health problems would benefit fromusing one or more of the wide variety of Hmong traditionalhealing arts,Some mental health professionals may wish to facilitatetheir Hmong clients' needs to seek traditional healingarts, but lack the background knowledge about what isconsidered acceptable to Hmong individuals or what isavailable in the Hmong community for patient use. Thepurpose of this paper is first to familiarize readerswith Hmong traditional beliefs about mental health. Thesecond purpose is to introduce the various Hmong traditionalhealing arts and explain how they can help certainHmong mental health clients. Third and lastly, this paperwill suggest how to identify those Hmong clients mostlikely to benefit from dual (Western and Hmong) treatmentsversus those who may progress better by using onlyone type of treatment.HMONG TRADITIONAL BELIEFS ABOUTMENTAL HEALTH PROBLEMSThe Hmong society has long recognized that someHmong individuals suffer from mental health problems. Infact, they have their own terminology for those who sufferfrom such illnesses. The Hmong associate the liveras the organ which governs a person's emotions, in contrastto the Western association of the heart. For theHmong, it is the liver which is thought to most directly34934d

350 THE HMONG IN TRANSITIONaffect individuals' personalities and mental stability.In the Hmong language, the liver is often used to describea person's character. For example, a person whois "pure-hearted" is considered to have a "white liver"or slab dawb.Hmong Terminology of Mental Health ProblemsWhen the Hmong refer to a person as having mentalhealth problems, they usually use one of a variety of"liver" terms. These "liver" terms do not in any wayimply a physiological problem with the liver. Rather,the terms are idiomatic and refer only to emotional problems.These are some examples:"Slab Phem" or Ugly Liver. When a Hmong person issaid to have slab Ahem, it usually means the person hassuddenly become destructive and abusive."Nyuab Siab" or Difficult Liver. This usually indicatesa patient who is suffering from excessive worry.Patients often become confused and cry. Many suffer fromloss of sleep."Tu Siab" or Broken Liver. This term refers to individualssuffering feelings of guilt or grief."Slab Luv" or Short Liver. Patients with this problemoften are those with suddenly appearing aggressivebehavior, and a change from normal to extreme bad temper."Kho Siab" or Murmuring Liver. Those suffering fromkho slab usually exhibit the development of certain nervoushabits. Commonly seen are constant coughing, humming,whistling or shaking of the head. Patients oftenbegin to talk of death and suicide."Lwj Siab" or Rotten Liver. Some clients who are unableto accomplish their goals or are unhappy with theirpresent lives often develop 1%2_1/ siab. Clients such asthese usually exhibit symptoms of loss of memory and delusions.Relationship Between Mental Health Problemsand Traditional Hmong ReligionThe Western mental health practitioner should recognizethat a Hmong person's belief about his or her healthand mental health status is firmly tied in with some tra-3 4 7

GUIDELINES FOR MENTAL HEALTH PROFESSIONALS 351ditional Hmong religious beliefs. Those Hmong who havenot yet converted to a Western religion are usually ancestorworshipers. This is not to be confused withtype of theBuddhist religion found in Southeast Asia, whichmany of the Lao ethnic group follow. Unfortunately, noHmong literature on their form of ancestor worship exists.The Hmong have handed down their religious practicesand rituals from father to son over the generationswithout any written text. Although this has led to someregional, clannal and family differences in some ritualdetails, most unconverted Hmong continue to follow thesame general broad outlines of ancestor worship.Hmong families who follow traditional Hmong ancestorworship practices usually believe that the majority ofmental health cases are caused by certain religious problems.The most common situations thought to cause theseproblems to individuals are either the loss of one's soulor souls, or problems with spirit entities.Hmong Beliefs About Soul(s) and Mental Health. TheHmong believe that each person has three major souls.One soul lives in the head area, one in the torso areaand one in the legs area. It is believed that upon death,one comes back to live with the descendants, one remainsat the grave site, and one goes back to heaven and mayeventually be reborn (either as a human or somethingelse). That is how the Hmong believe in both ancestorworkship and reincarnation.All souls must be united with the body for a personto have good health --and particularly good mental health.Loss of one or more souls is thought to cause a wide varietyof mental health problems. The more souls that arelost, the more serious the symptoms of illness usuallyare. In addition, the longer one has been separated fromone's soul or souls, the more severe the symptoms of illnessbecome.Whatever the reason for the loss of soul or souls,the mental health symptoms are usually one or more of thefollowing: depression, disorientation, delusions, .loss ofappetite, nightmares, aggressive behavior, change in personalityand inability to sleep. Very often victims succumbto a "liver" disease. It is the family's responsibilityto seek a cure for the person who is sick fromloss of soul.34d

352 THE HMONG IN TRANSITIONHmong Beliefs About Spirits and Mental Health. TheHmong ancestor worshipers believe that the world is coinhabitedby a variety of spirit entities. These spiritsare thought to affect a person's health and mental healthstatus sometimes. A brief review of the Hmong types ofspirits which are thought to commonly affect the mentalhealth status of individuals follows below.1. Ancestor spirits: Ancestor spirits are the spirits ofancestors who have died. As stated earlier, one ofthe souls of each ancestor comes back to live withhis or her descendants. Ancestor spirits must bekept satisfied to insure the good health and prosperityof the family. Proper burial procedures and thecarrying out of certain rituals to strictest detailare considered necessary to keep ancestor spiritscontent after death. In addition, certain annualceremonies must be carried out correctly. If theseduties are not performed, or are performed incorrectly,the offended ancestor spirit may cause illnessor mental health problems to the erring descendantor to a member of the descendant's family. Thecausing of illness is considered a form of communicationrather than a malicious act. The family ofthe ill person then becomes responsible to find outwhich ancestor is unhappy or needs something and toperform the necessary conciliatory ceremony.2. Nature spirits: The Hmong believe that almost allthings occurring in nature are governed by a naturespirit. Ponds, streams, rivers, hills, valleys,trees, rocks and even wind currents are believed tohave a spirit. Some nature spirits are thought to bevery powerful, especially those which have a postserving "god," while others are lesser, localizeddeities. The localized nature spirits are the oneswhich sometimes cause illness or mental health problemsto some Hmong persons.Although nature spirits are usually uninterestedin human beings' affairs, if offended, they may causehealth or mental health problems to the offendingperson or to a member of that person's family. Unfortunately,it is difficult to know how to avoidoffending nature spirits.3 4 d

GUIDELINES FOR MENTAL HEALTH PROFESSIONALS 3533. Evil spirits: Many Hmong believe that in addition tonature spirits, there are innumerable evil spiritsthat also coinhabit the earth. Most evil spirits arethought to live in areas not populated by mankind.This is why most Hmong persons prefer not to travelin uninhabited spots alone. However, evil spiritscan also follow humans to their villages and homesand cause all sorts of illness, mental health problemsand even some disasters such as a accidents,financial ruin and so forth. Evil spirits have noknown reason for their choice of victim. The spiritsneed not have been offended. Being a victim of anevil spirit is just bad luck.It is a little more difficult to deal with illnessor mental health problems caused by evil spirits.This is because only sometimes can the evilspirit be bribed into accepting an animal sacrificein return for allowing the victim to regain goodhealth. Sometimes the evil spirit is not willing torelinquish the victim and a Hmong person skilled in acertain healing art must then be called in to fightthe evil spirit away.4. House spirits: The last of the spirit entities theHmong believe to cause health or mental health problemscommonly are the house spirits. Every Hmonghouse is thought to be inhabited with a set of housespirits. These include two door spirits, one for thefront door and one for the back or side door; twofireplace spirits, as the traditional Hmong home usuallyhas two fireplaces; and four corner spirits, onefor each corner of the house. Usually house spiritshelp foster the well-being of the family. However,if offended, the house spirit may then cause illnessor mental health problems in the family.Usually it is relatively easy to help patientswith house spirit problems. House spirits are alwaysbenevolent and usually minor sacrifices and the appropriateceremonies will restore the patient to goodhealth.Other Hmong Beliefs About Mental Health. There aretwo less common causes for Hmong mental health problems.Although these two causes of mental health problems may330

354 THE HMONG IN TRANSITIONappear to be less believable or acceptable than the a-forementioned reasons, for some Hmong persons, thesecauses are very real. Several cases of Hmong mentalhealth problems have been linked to these two causes anda large majority of Hmong still believe in these forces.1. Curses: The Hmong believe in the power of righteouscurses. These curses are not like voodoo or other"black magic" curses. Rather, Hmong curses only havepower if the person cursing is morally right tocurse.If one's mental health proble.as are caused by acurse, it becomes necessary to locate the cursingperson and convince the person to take back thecurse. If the cursing person is already dead, it becomesnecessary to contact the spirit of that person.Since the cursing person is always morally right, thevictim or the victim's family will usually have tomake some sort of conciliatory act and/or pay a fee,as well as to conduct a series of ceremonies.2. Tame evil spirits: Certain persons are thought toknow "magic spells" which, when spoken, take the formof an evil spirit. This is not a wild evil spiritlike the ones previously mentioned, but an evil spirittamed by a person. Most Hmong claim that the originsof these tame evil spirits lies with other ethnicgroups found in Laos. It is said that some Hmongpersons paid a fee to learn this power from theseother ethnic groups and, in turn, passed the power onto other Hmong indi-eicuals.Tame evil spirits are thougt to have one of twopossible powers. One is the power to magically implantforeign objects such as bones, nails or rocksinto victims. The victims then become ill or crazed,and death can follow quickly. The other power is thepower to symbolically suck the blood and thereforethe life of victims away.Only tame evil spirits with the ability to implantforeign objects in a person can take them out.And only tame evil spirits able to suck blood canfight other tame evil spirits of that type. Relativesof victims of tame evil spirits must seek otherowners of tame evil spirits and pay them to fight for351

GUIDELINES FOR MENTAL HEALTH PROFESSIONALS 355the life of the victim. Alternately, they may findthe owner of the tame evil spirit attacking theirrelative and bribe him or her to call off the attack.TYPES OF TRADITIONAL HMONG HEALERSJust as the Hmong believe there is a wide variety ofcauses for mental health problems, they have a wide varietyof healing arts that can help cure Hmong mentalhealth patients. It must be remembered that any one personcan know more than one healing art. Since most ofthese arts are learned, many Hmong health practitionersknow a variety of traditional healing skills.Hmong DiagnosticiansSome Hmong health practitioners are only diagnosticians.They provide patients with the probable cause fortheir mental health problem or illness, and advise themon what type or which healer they should contact. Thepatient or the patient's family can then proceed on thisadvice to obtain a cure. This is considered an importanthealth care function because each Hmong health practioner,even among those following the same discipline, isthought to have a different level of power. Sometimesonly a certain person will have the power to cure a patient.Hmong Fortuneteller. This art is thought to havebeen originally 'learned from the Chinese. The Hmong fortunetelleruses a series of charts, reads various omensand consults with certain gods, and is then able to tellclients various types of information. Clients who havemental health problems or who are ill can find out whatthe cause of their problem is because the Hmong fortunetelleris able to tell whether the cause is due to spiritentities, loss of soul, a curse or whatever. In addition,he can sometimes tell the patient what type of healingpractitioner is most likely to succeed in curing the victim.Of course, Hmong fortunetellers are often consultedfor reasons other than health.ERR Readers. Many Hmong ask an egg reader for adviceat times. Persons who have learned the skill cantalk to a freshly laid raw chicken egg. They will per-3 5 2

356 THE HMONG IN TRANSITIONform the right ceremony, then ask the egg the necessaryquestions. Most times the egg will be asked which healthpractitioner can best cure a patient.Rean Readers. Bean readers are also usually consultedwhen a patient does not know which health practitionerto approach. Three dry soy beans are placed in abucket of water and are questioned as to whether a certainhealth practitioner is the right one for a patient.If the beans form a certain configuration, then the answerto the question is affirmative.Rice Grain Readers. Rice grain readers perform thesame function as Bean Readers except that they use threegrains of rice in a cup of water as their tools.Ear Readers. Ear readers can tell clients how manyand which souls have been lost. Clients are taken outinto bright sunlight. The ear reader gently holds theclient's ear out and looks at it while sunlight streamsthrough it. The fine lines of the ear tell the ear readerwhat has happenl.td to that person's souls.Basket Talkers. Certain people are able to talk tospirits using a large basket. The spirits are sometimesable to tell why certain patients have become ill.When evening comes, a basket is dressed up to looklike a person. A "head" is attached and a long stickstuck through to represent arms. The basket is taken tothe door and a spirit is invited to inhabit the basket.The basket talker can then ask the spirit in thebasket questions about almost any subject. If there is asick person, the basket spirit can give information as towhy that person is sick.won Talkers. Some persons are thought to be ableto visit the spirit world by using a spoon spirit. Thespoon talker waves a spoon over the fire three times, andthen places it down by the fireplace. The spoon spiritthen leads the spirit of the spoon talker to the spiritworld. While in the spirit world, the spoon talker cantalk to spirits and ask them questions.This skill is looked upon skeptically in some Hmongcircles while others believe in it. In Laos, spoon talkerswere often asked to talk to spirits simply to entertainor pass time during the village evenings; the practicewas only occasionally done to diagnose a patient.In the United States and probably other Western societies,the Hmong do not practice this as much, probably353

GUIDELINES FOR MENTAL HEALTH PROFESSIONALS 357because it is not as easy to build a big fire in theevening.Person with a "Neng."A person whose body hosts a"neng," or healing spirit, is probably considered theHmong diagnostician with the greatest abilities. He orshe is also a healer. All of the other Hmong diagnostictechniques and healing arts can be learned. But a personwith a neng must inherit his skill and his neng from a-nother member of his clan.Each neng has its own level of power and ability.However, almost all of them can diagnose most medical andmental health problems. The host of the neng must have aspecial altar and certain tools which he uses to help patients.Persons with a neng use their skills exclusivelyto help ill people, in contrast to other diagnosticians,who sometimes find out information on other subjects.Hmong_Traditional HealersAfter a Hmong patient has been diagnosed, a curemust be found. Over the centuries, the Hmong have developeda variety of healing skills. Sometimes more than oneskill must be used to cure a patient. This is usuallynot a problem because there is little or no professionaljealousy among conventional Hmong healing practitioners.Each practitioner is bound by custom not to be competitiveand is secure in his or her own ability.Herbalist. Almost every Hmong family has a portionof its garden devoted to herbs. In most Hmong families,herbs are quite commonly used for a variety of healthproblems. However, there are certain persons who are especiallyskilled in using a larger variety of herbs.Masseuse. The Hmong use massage as a technique tocure a variety of body aches. The most skilled masseusesare able to bring relief to head, muscle and even someforms of stomachaches. They are able to massage withoutirritating or hurting the skin.At times, masseuses rub herbs into the skin of thepatient while massaging. At other times, hard boiledchicken eggs and silver bars are used in the massagetechnique.354

358 THE HMONG IN TRANSITIONNeedle Users. The Hmong have not developed the artof acupuncture. However, they do on occasion use needlesfor "blood pressure" problems. Most commonly used areregular sewing needles.Soul Callers. When a person's soul is lost orthought to be lost, often the family of that person willperform a soul calling ceremony. There are three variationsof the ceremony. However, whatever the variation,at least a pair of chickens (one rooster and one hen,each about two to three months old) and one egg must besacrificed.The first variation is the least complicated. Usuallymost Hmong elders and heads of household know how todo it. The positions of the claws, the jaws, the eyesand the skulls of the chickens all say whether the soulhas returned or not. If the soul has not returned in afew days, additional chickens must again be sacrificed.The second variation of soul calling is usually performedfor those who suspect they know where their soulwas lost. For example, if the patient was suddenlyfrightened at the riverside, the soul caller will go tothe riverside. He or she will then perform the necessaryceremonies and say the correct words and lead the soulback to the village and the house of the patient. Again,at least two chickens and an egg must be prepared in thesame manner aescribed earlier.The third variation involves a time period of a day.Three soul-calling ceremonies are performed in the home:one at sunrise, one at midday, and one at dusk. However,chickens need be sacrificed only at dusk.Person with a "Neng." A person with a neng isthought to be the most skilled of the Hmong healing practitioners.He or she is most commonly asked to assistonly in relatively serious medical or mental healthcases. While in a trance, the person with a neng can notonly find out if the patient is suffering from problemswith spirits, but also if the patient needs assistancefrom other health practitioners in case it is not a spiritualproblem, or if another person with a neng wouldbetter be able to help the patient.When dealing with illnesses caused by ancestor, natureor house spirits, the person with a neng usuallymakes a deal with the offended spirit.

GUIDELINES FOR MENTAL HEALTH PROFESSIONALS 359When dealing with evil spirits, the person with aneng must sometimes literally fight with the spirit. Heor she will pick up the neng sword and tambourine andjump about the room fighting the spirit.At times, when the person with the neng finds outthat the patient is simply suffering from old age andthat it is time for the patient to die, the person with aneng may even visit heaven and ask the god in charge ofhuman beings' 'visas" on earth if the patient's life canbe extended in return for a sacrifice.Persons with a neng are also skilled in finding people'slost souls. While in a trance, friendly spiritscan help the neng locate a :ost soul and entice it backto the patient.Persons with a neng are thought always to be ethicallymoral people. This is because the neng which selectsthem are supposed to select goodhearted people andkeep their hosts from doing any medical evil. They aregreatly respected in the Hmong community and a very rigidprotocol must be followed when requesting a person with aneng to take a medical or mental health case.FACTORS TO CONSIDER WHEN REFERRINGA HMONG CLIENT TO TRADITIONAL HEALERSHaving reviewed Hmong attitudes on mental healthproblems and how the Hmong community traditionally treatsmental health problems, I will now consider which Hmongclients of Western mental health professionals would mostlikely benefit from referral to traditional healers. TheWestern practitioner should make several considerationsbefore arriving at his or her decision.Consideration of Cross-cultural Differencesin Initial Assessment of ClientSome Hmong clients may give unsuspecting Westernmental health practitioners the firm impression that theclient is suffering from delusions and/or a wanderingmind. This is because a Hmong person's description ofhis or her symptoms may sound very unusual to the Westernear. What Hmong society often feels is a straightforwarddescription of a disease may cause a Westerner to be con-356

360 THE HMONG IN TRANSITIONvinced that the client needs mental health care. For example,clients who complain about their blood being unableto flow, or bugs eating away their flesh, or rocks intheir stomach are often thought to be suffering from delusion.In addition to complaints about strange symptoms,the Hmong client may give an unbalanced appearance becausehe or she may not seem to be able to answer questionsin an appropriate manner. It is normal for a Hmongclient, especially an elder one, to ramble as he or sherecounts his or her health or mental health history.Accounts of dreams or encounters with spirit entities mayeven be included.The Western mental health practitioner must considerthat these mannerisms and accounts are within the normsof Hmong society in relation to mental health problems.While deciding the course of Western treatment, professionalsshould also consider referring the client to traditionalhealers. Perhaps some clients may not even needWestern mental nealth counseling, while others may besuited for Western health care and not traditional care.Hmong clients who complain of being depressed, beingforgetful, having loss of appetite and so forth have inthe past often been treated successfully by traditionalmeans. It is almost safe to say that most Hmong clientsabove the age of twenty-five years would receive someform of emotional support if he or she received a form oftraditional treatment, either in conjunction with Westerntreatment, or alone. It must be remembered that Hmonghealing treatment does not prohibit use of other medicalsystems, so dual (Western /Hmong) treatments can readilybe accomplished if the client desires. For those youngerthan twenty-five } , s of age, particularly teenagers,referral to traditional treatments may not be as necessaryunless the teenager belongs to an extended familywhich places great value on Hmong healing arts.35/

GUIDELINES FOR MENTAL HEALTH PROFESSIONALS 361Availability of Hmong Health Practitionersin the CommunityIt is the custom that a Hmong health practitionernot advertise his or her skill. Thus, it is often hardfor Westerners to obtain a list of who and what skillsare available in the local Hmong community. Usually theHmong grapevine is quite adequate in providing this information,but on occasion, certain clients may not haveaccess to this information either.Most large Hmong communities have quite a. large varietyof healing persons with different skills available.Unfortunately, some of the smaller Hmong communities maynot. However, when traditional treatment is desired butnot available, Hmong patients sometimes are willing totravel to another town to seek cures. At times, familymembers in another city can ask a person with a neng toperform a ceremony without the patient being physicallypresent.Western mental r:ealth practitioners who deal largelywith Hmong clients should be encouraged to become familiarwith the social organization of their local Hmongcommunity. Usually there will be one or more communityleaders who can supply information regarding availabilityof traditional healers. It must be remembered, though,when discussing traditional healing arts, that youngHmong community leaders may not know as much as Hmongelders and clan leaders.Clan or Family PreferencesIt isthe custom of Hmong families to seek cures forsick family members, rather than the sick person seekinga cure for himself or herself. Thus it is difficult foryounger Hmong persons without an extended family to seektraditional health treatments. It must also be rememberedthat when utilizing traditional health practitioners,some fees may be involved. Depending on what is involved,fees can he nothing, part of the animal sacrificeor as high as in the hundreds of dollars. In addition,it must be remembered that if one utilizes certain typesof Hmong health practitioners, one must be prepared tosacrifice chickens and/or a pig. Families who cannotafford, or do not know how to obt&in, or do not have the358

362 THE HMONG IN TRANSITIONtransportation to obtain these items, may hesitate toseek traditional treatments. In Laos, each Hmong familykept at least a few live chickens and pigs around theirfarm. So obtaining live sacrifice was not the problem ithas become in Western societies.Some Hmong persons may prefer to utilize only Hmonghealth practitioners who are of the same clan, or who arerelatives. This may be in the hope that the fees will beless, or the sacrifice for the spirits less expensive.It may also be due to less friendly relations betweensome clans. Therefore, it is sometimes difficult for anoutsider to recommend with confidence any particularHrnong health practitioner without considering interclannalas well as personal relations. It may be best to advisea client to seek preliminary counseling from his orher clan leader.Religious Background of ClientThere are sizable numbers of Hmong in Western communitieswho have converted to some form if Christianity.A large number of these belong to the Christian do MissionaryAlliance group, while some are Catholic and othersbelong t.' miscellaneous Protestant groups. Almost allHmong Christians are reluctant to participate in soulcallingceremonies or to accept treatment from a personwith a neng. Some even go so far as to charge thatchickens or pigs sacrificed for these health practicesare "devil's meat." Thus, if the Western health practitionerencounters a Hmong Christian, it would be safestto suggest that if the patient wishes traditional treatment,he or she should seek only the herbalists, masseuses,and/or needle users.Most Hmong ancestor worshipers are probably more receptiveto using any variety of Hmong traditional healingarts. It might be mentioned here that many Hmong clientsmay not admit to being ancestor worshipers for fear thatthe Western health practitioner may not understand thisreligion. One must tactfully question a client as to hisor her true beliefs before deciding whether to recommendtraditional healing treatments.353

GUIDELINES FOR MENTAL HEALTH PROFESSIONALS 363CONCLUSIONThere are many factors to consider before a Westernmental health practitioner may decide that a client wouldbenefit from traditional healing treatments. However,since Hmong health treatments never hurt anyone, butcould possibly help a patient, it should be seriouslyconsidered as part of a client's course of treatment.Even if the traditional treatments may not appear to doany clinical good to the Western observer, the supportthe patient feels from his family as the family goesabout performing the traditional ceremonies provides thepatient with positive feelings. The sacrificed animalsalmost always lead to a small party of sorts to consumethe meat, and this too can engender positive feelings.The Hmong have been treating their own mental healthproblems for centuries and although many of their methodologiesmay sound very unusual to Westerners, many Hmongmental health patients have been successfully cured orhelped using these methods. There is enough of a varietyof different Hmong health disciplines that almost anyIlmong client may find something that suits his or hercase. Even if a Western health practitioner may not actuallywant to refer a client to a Hmong health practitioner,it is the intent of this paper to encourage himor her to foster an understanding environment in thecounseling sessions and to accept a Hmong client's terminologyof mental health problems, and perhaps even tosuggest that the client may, on his own, seek traditionalhealing treatment without fear of ridicule.360

HMONG PERCEPTION OF ILLNESSAND TRADITIONAL WAYS OF HEALINGXoua ThaoThe Hmong have a traditional system of dealing withillness. This traditional system consists of shamanism,herbal medicines or a combination of the two. In leavingLaos, the Hmong knew that their traditional ways of healingwould be challenged by Western medical practices intheir countries of resettlement (Catanzaro and Moser1982). In my experience as an interpreter at a largemetropolitan hospital, I have observed that few Westernhealth care providers have prior experience in caring forpatients who do not share their scientific rationalism.This paper attempts to provide Western health careworkers with an introduction to the Hmong perception ofillness and traditional methods of healing. Most of thesetraditional .practices are still carried on by the Hmongin their new countries. It remains for future researchto discover what and how modifications have been instituted.The draft of this paper was discussed with manyHmong experts who have knowledge of shamanism and herbalmedicines. Many Hmong bilingual workers who have experiencein Hmong medical practices also have read it.The Hmong healing system shares many common elementsof traditional folk medicines found in numerous ruralvillage settings (Snow 1974). Incantations, rituals andherbs have been the traditional medicines for most ofhistory (Balikci 1967). In the Hmong world, shamanismand herbal medicines come in a variety of forms and procedures.I will discuss some elements of Hmong shamanismand herbal medicines, the concept of soul loss, ways ofcalling back the soul, the mandate of life, the differentcauses of illness, the traditional use of herbal medicinesand, finally, the highest and most complex ritualin Hmong shamanism, called Ua Neeb.365361

366 THE HMONG IN TRANSITIONHMONG SHAMANISMHmong shamanism is a repository of many esotericrituals and incantations. They constitute the main processesthe Hmong use to cure and are substantially basedupon oral texts which only a few individuals know. Thesefew individuals could be males or females. They becomeexperts through a period of training, and they are theonly ones to practice. Hmong shamans keep an altar athome to convince the community of their skills.The altar is the shaman's "diploma," announcing hisor her expertise to individuals who seek help, much likethe framed certificates that hang in a Western physician'soffice. The altar also symbolizes the shaman'sdevotion to the sick and suffering.Shamans perform rituals at the altar and elsewhere,depending upon the type of ritual. However, for the ritualsand incantations to work, every step in the proceduremust be followed and everything done as prescribed.The individual must believe in the process and have confidencein the shaman. One often offers the shaman food,money or some words of thanks as a sign of trust and appreciation.There are many rituals and incantations. Like drugs,each is good for a symptom or a specific illness. Thereare rituals and incantations for headache, vomiting,rash, eye and ear complications, soul loss, etc.HERBAL MEDICINESIn the Hmong community, herbal medicines rally a-gainst illness along with shamanism. Herbal medicinesare part of the women's world. As the women claim, "Weare given the gift and knowledge. Only we can make herbalmedicines effective. Men are much too careless."Medicinal herbs are used on a trial-and-error basis.Not every woman has the knowledge of and access to theseherbs. Gifted women pass on their knowledge informallyto their daughters or grandaughters. In fact, experts inthis area of healing rarely teach anyone unless the noviceis a relative or a trusted friend.3G2

HMONG PERCEPTION OF ILLNESS 367The effectiveness of the herbs depends extensivelyon the attentiveness given to their preparations. Withoutcare and proper use, they are futile. If the herbsfail to satisfy people's expectations, as a female herbalistsays, "they curse the experts and condemn the medidnes."Unlike shamanistic rituals, however, one herbalmedicine can be used for more than one illness.The Hmong perceive illness as a condition that rendersan individual incapable of eating, drinking or gettingout of bed. An individual who can still force himselfor herself out of bed is hardly considered ill.Thus, a person is sick only when he or she cannot performor fulfill his or her social tasks and responsibilities.illE CONCEPT OF SOUL LOSSHmong believe there are many sources of illness.But the primary cause is the loss of soul. Illness canalso be of natural causes, supernatural or spirit causes,magical causes or organic causes. Each of these causeshas certain characteristics and is determined by the conditionand history of the illness. Usually a mild illnessis believed to be of natural or organic causes whilethe serious ones are ascribed to the supernatural, spiritor magical causes.Though there are various beliefs, soul loss deservesfirst mention. To the Hmong, every person has a soulwhich lives in and governs the body. However, the soulmay leave a person's body when the person is frightenedor alone in a dark place, or when he or she feels depressedand lonely due to separation from family or lossof a loved one. In any highly emotional circumstances,there is a possibility that the soul will go away.When a baby cries constantly at night, the Hmongwould say that the baby might have lost its soul. Customarily,an adult family member takes a burning stick tothe door and swings it back and forth. This shines theway for the lost soul to return. If that fails to pacifythe baby, then the family member places some corn or ricein a tray and waves it in circular motion above the baby'shead. While doing so, he or she pleads to the soulto come back.363

368 THE HMONG IN TRANSITIONChildren or adults who have lost their souls ordinarilybecome thin, tired or pale. They sleep heavilyand lose appetite. A shaman claims that behind the earlobesof these people, a tiny blister exists on top of asmall capillary. When these symptoms appear, they shouldnot be ignored because they indicate soul loss. If leftunattended, the person will become sick. The soul, then,will be difficult to call back.Normally, the wandering soul returns to its host.However the vagrant soul can also change into other formsWhen the soul has transformed, as the Hmong be-of life.lieve, it does not recognize its owner. It cannot return.The Hmong say, "The soul cannot find its way home." Ifthe transformed soul is accepted into a new form of being,then the owner becomes seriously ill. Death can result.If a ritual is not performed to call the soul back,the sick person will not get well.However, if the soul transforms and is not acceptedinto a new form of being, it then wanders until it ishungry and exhausted. In this state, the soul would notsurvive long. It is willing to come back to its owner ifthe necessary measure is taken to get it back.Because the Hmong believe the soul leaves the bodyeasily, wherever they go, before returning home, they usuallyutter words that say it is time to go home, and thesoul, wherever it is at that moment, must return, too.When the Hmong are at an unfamiliar place or even whenthey are out at a picnic, they rarely return quietly.Parents would call loudly to their children's souls. Individualscan also call their own souls. If the parentsfail to summon the souls back, their imprudence doomstheir children to eventual sickness. Out of this beliefthat the soul needs to know its owner and to be calledinto the body springs the practice of giving a newbornbaby a name in a welcoming ceremony on the third day.WAYS OF CALLING THE SOUL BACKTraditionally there are many ways to call the soulback. If the person is sick because the soul cannot findits way home, then the family of the stricken individualprepares two small rectangular boards. The parents takethem to the intersection of two pathways near the house364

HMONG PERCEPTION OF ILLNESS 369and place them in a criss-cross pattern. This placementsymbolically serves to guide the soul back home. If anystranger happens to pass by at that occasion, the parentswould ask for a few words of blessing.If the person is seriously ill because the soul either,has gone too far or was transformed into anotherform of life, then an animal has to be sacrificed. Beforethe animal is slaughtered, the shaman performs theritual by standing inside the door, chanting a lengthyincantation. When the incantation is complete, the animalis killed. "As head is cooked and brought back tothe door. At this time, the shaman performs the secondand last round of the calling ritual. The Hmong believethe animal's soul will replace the person's soul. Thetype of animal depends on the shaman as well as the situation.For instance, if the child falls into a pond, thena duck is needed. However, sometimes a dog is used; atother times, a pig, chicken or cow is acceptable.Although there are many methods, each with a differentprocedure, the use of an egg is most popular. Initiallythe shaman massages the sick with some ginger,starting from the head, then the fingers and toes. Themassage gradually moves toward the center of the body,and the shaman continually says he is gathering and callingthe soul back. As he finishes, the ritual begins.Normally, three poles and one of the sick person'sshirts are required. The first pole is staked outsidethe door, the second at the threshold, and the third afew feet into the house. A string is tied to them. Theshirt is placed on the ground at the third pole with itsfront facing inward. As the shaman stands at the thresholdcalling, another person must sit at the third pole,holding an egg upright on the palms of both hands whichare laid on top of the shirt. Whenever the egg fallsbackward, as the shaman chants, that symbolizes the returnof the soul. Immediately, the shaman cuts thestring. The egg and shirt are steamed together. Thenthe egg is burnt to destroy the path which led the soulastray. The shirt can be worn again. To further stabilizethe soul, white strings are tied to the wrist of thesick person.3C

370 THE HMONG IN TRANSITIONTHE MANDATE OF LIFEIn addition to soul loss, there is the belief of a"mandate of life." The Hmong believe that every personis given a mandate to be a human being. On it, the SuperiorBeing foretells the length of the person's existenceand his or her fate. An individual, therefore, becomesill because the mandate of life predetermines illness anddies because death is due.NATURAL CAUSESBesides the soul and the mandate of life, nature cancause illness to men. Historically, the Hmong settle inthe highlands and mountainous areas. They grow up adaptingto nature, hills and jungles. This ecological relationshipinfluences their perception of illness, which isstrongly based upon the equilibrium of men and nature.Also, the locations of Hmong villages expose them to aconstant range of weather. They are sensitive to itsfluctuation and the seasonal changes.Because men and nature must coexist in equilibrium,the Hmong consider illness a distortion of this relationship.Rheumatism is a good example. The pain of jointsand the aching of the body are symptoms the Hmong associatewith the coming of either warm or rainy weather. Fever,cold, runny nose and coughing are ascribed to theshifting of the man-nature relationship.The Hmong are conscious of the impact of naturalfactors upon the health of an individual. They exercisepreventive measures. A mother who has jirt given birthto a baby is discouraged from participating in the family'sdaily work for a month. They believe that her bodyis "unclean" and already in "disequilibrium" with nature.Forcing herself to work in such a condition will make herill in later life. When the weather changes, she will becomesick. Furthermore, such a mother should be verycareful in using water and food. She must eat right.Hot is always preferred because, as a Hmong woman says,"cold water and food will make our blood congeal and giveus sickness. Our skin will become wrinkled and 'cryitchy in old age."366

HMONG PERCEPTION OF ILLNESS 371ORGANIC CAUSESAlthough the Hmong place great credence in naturalcauses of illness, they also find organic explanationsappropriate. Illnesses which exhibit physical evidencesuch as stomachache, chicken pox, nose bleeding, orleprosyare always accepted to have some organic bases.A weak individual is thought to lack something in thebody.When this person eats well but gains no weight, theHmong claim that there is not enough fat and blood tofeed his or her body. To get more fat and blood (andthus strength), a chicken is killed and cooked with someherbal medicines which give the healing power. The chickenis needed because the herbs are not usually eaten directly.This procedure is repeated three or four times.TRADITIONAL USE OF HERBSThe Hmong use herbal medicines to deal with some ofthe organic illnesses. Massage is often the initial stepto treat a stomachache. Then certain herbal medicinesare used. Sometimes some dry roots are sliced into tinypieces and put in a glass of hot water for the person todrink. Other times the person eats the pieces. The proceduredepends on the types of herbs and must be followedcorrectly. For example, if the herbal roots are given toa male, there should be nine slices. If given to a female,seven are required. The number has to be exact.In addition to the utilization of herbs, the Hmongalso believe that sometimes a person is sick becausethere is too much pressure inside the body due to stress.To relieve the internal tension, the yolk of a hardboiledegg and a silver bar are wrapped together in apiece of cloth. The person either sits still or liesdown, and someone rubs the pack over the person's back,arms and forehead.Although herbal medicines help many symptoms, theyare not used for every illness. The Hmong still turn tothe rituals and incantations for many conditions whichappear to have organic bases. When a baby or an adulthas a rash in the mouth, a specific incantation remediesit. A hoe and some cotton are used. While the shaman' is367

372 THE HMONG IN TRANSITIONoutside, another person sits inside the door with the individualwho has the rash. This other person points thehoe's handle toward the sick person's mouth. The shamanlights the cotton and chants a three-verse incantation.The person holding the handle responds to each line respectively.Most Hmong prefer thin method over herbalmedicines or foreign medications.There are also very specific lcantations for earand eye problems. The shaman points he socket of a hoeto the inflamed ear. A small paddle is used to strikethe hoe gently while the shaman chants the incantation.The trouble is relieved by its words. For a person withrepeated eye pain, the shaman points two 'r three burningincenses to the troubled eye and chants the incantation.When he finishes, he stakes those incenses beside thedoor. For conditions that lack appropriate incantationsor herbal medicines, the sick person will be brought tomedical attendants using Western practices as a lastresort.MAGICAL CAUSESMagical, sources can also cause illness. The Hmongbelieve words can cause illness when they are cursed uponothers. Furthermore, there are magical people who cancast spells on someone to cause sickness. They can sendan "egg stone" to the person.! When the person is hitby this inanimate object, unless it is discarded, theindividual can die.When someone is inflicted with this magical thing,it can be removed by an incantation. The shaman asks thevictim for the place of pain and begins to suck out thatarea. He then spits the magical stone into a bowl ofwater. The stone is thrown away, and the person shouldfeel fine. A Hmong shaman asserts that it is bad tolearn to cast magical spells because one has to use theacquired skills. If not, the magical spells can turnharmful to the expert or members of the expert's family.368

HMONG PERCEPTION OF ILLNESS 373SUPERNATURAL OR SPIRIT CAUSESAside from natural, organic and magical causes, thebelief in spirits makes up a large system encompassingvirtually every serious illness. In addition to severityof the condition, an illness caused by the spirits mustshow certain consistent patterns. The person feels morepain or is disturbed more often at a certain period oftime. When that time arrives the next day, the personexperiences the same plight. Such a characteristic ofthe illness is readily attributed to the doings of spirits.Furthermore, if the sick person shows some kind ofpeculiar physical symptoms such as mumbling, trembling,convulsion or the deformation of facial expressions, theHmong most often describe that as being hit by the spirits.Spirits dislike being disturbed. If people are notcareful and do not keep away, they can be hit and becomesick. The individual catches illness by touching and destroyingthe spirits' places or by cursing and ignoringthem. When a person sees whirlwind, for example, theperson should stay away at a distance. If such an individualignores the phenomenon and is hit, the spirit ofthe whirlwind can make the person dizzy. A pregnant womandisturbing the spirit of a stream when she crosses itcan expect a miscarriage.The spirits and the people must live in harmony.Otherwise, the spirits can afflict them with illness,plague or natural catastrophe. An unhappy spirit mightabduct the soul and threaten it to make the host severelyill. In that case, an animal is needed to get the soulback. Even if the spirit does not kidnap the soul, itcan come to the individual and disturb the soul in thebody. In such cases, the individual experiences excruciatingpain when the spirit is present and alleviationwhen it goes away. The illness is expected to persist ifnothing is done to chase away the spirit, to hide thesoul of the sick person or to negotiate with the spirit.WAYS OF DEALING WITH SPIRITSThere are several ways to encounter the spirits.All involve rituals and incantations. A person who has369

3 0HMONG PERCEPTION OF ILLNESS 374been hit by the spirit of the sun, for example, wouldhave pain in the head in late morning and again in lateafternoon. To get rid of this spirit, a small crossbowis made. An arrow, with an egg shell inserted into it, isshot over the person's head in the directions of north,south, east and west. The arrow, however, is shot onlyafter the incantation has been chanted by the shaman.This ritual is done early in the morning when the sunjust starts to rise and is repeated for several mornings.The Hmong believe that the arrow transfixes thesun and, as it sets, the spirit go-.s away as well. Thearrow and the egg are posted outside the house.There is also an incantation to drive away the spiritthat causes vomiting. The sick person puts five coinsinto a glass of water. As the shaman begins chanting, hepokes the water slightly with the handle of a hammer.After the three verses, the person is given the glass ofwater for three sips. If the sick person is in bed, thenthe shaman turns the glass of waterhowever much of itis leftupside down beneath the bed. But if the sickperson is brought to the shaman's house, then the shamanplaces the glass of water upside down where the sick personsits. A few days afterwards, the glass is removed,and the shaman keeps the coins. The water serves as theantidote for this condition.Sickness can be caused by all kinds of spirits, andeach incantation only works against one type. So, to beeffective, the right one must be used. A certain ritualrequires the use of knives. After it is performed, theknives are placed at the door to scare away the spirit.Another incantation is accompanied by the spatting into afire around the sick person. The shaman holds an oillamp and takes in some water into the mouth. The wateris spit back over the oil lamp to cause a flash of fire.These are extreme cases for incantations, and the shamanmust be paid some money. After these rituals, a treebranch is always put up in front of the house to keepoutsiders from entering. It is feared that outsidersmight make way for the spirit to return.In addition to chasing the spirit away, a person'ssoul may be hidden so that the spirit cannot find it.Thus, it will go away. The shaman takes a few hairs fromthe person, puts them in a glass of water and chants theincantation. The glass is sealed very carefully with a

MONO PERCEPTION OF ILLNESS 375piece of white cloth. It is placed underneath the bedwhere the sick person is.Another method is to use charcoal instead of hairs.Three pieces of charcoal are put in the glass of water.In addition, five bamboo meshes are made. When the shamancompletes the incantation, four bamboo meshes are puton poles in four different corners of the house. Thefifth one is staked near where the sick person sleeps.The glass is also carefully sealed and placed underneaththe bed. This ritual is traditionally called "sealing ofwater" and serves two functions. First, it camouflagesthe soul so that the spirit does not recognize and cannotbother it. Second, the ritual confines the soul so thatit does not wander off to disturb the spirit.THE RITUAL CALLED UA NEEBWhen a person is sick and various incantations failto help, the person resorts to a particular shaman whohas the ability to travel to the spirit world to searchfor the lost or kidnapped soul. This ritual is called UaNeeb. Through it the shaman comes in contact with thespirits and negotiates the price for getting the soulback to the owner. This process is the ultimate expressionof Hmong shamanism.These shamans must have some decent characteristics'IA appropriate qualities to be called or chosen to thisvocation. The recurring sickness they experience convincesthem that they have been called, even againsttheirwill, to learn this craft of healing. A Hmong shaman proclaims,"If I refused to obey the sign and did not startthe craft, I would always be sick."Some shamans perform Ua Neeb in half an hour whileothers may begin in the morning and end in the afternoon.Others may begin in the afternoon and end late in theStill others may begin early in the morning andevening.end the ritual the following morning. The length of timefor the ritual varies from one individual shaman to anotherand certainly depends on the practice of the teachershaman from whom one learned the procedure. Althoughthe time length varies, a shaman who finishes the ritualin less than half an hour is suspected and might evenlose credibility in the eyes of the sick person.371

376 THE HMONG IN TRANSITIONBefore Ua _Neeb, the shaman needs certain objects.Incenses and an egg in a bowl of rice are placed on thealtar, which is found against the front wall of thehouse. The shaman covers his face with a black piece ofcloth and begins. A gong is also played along by theshaman's helper. When the shaman is on the way to thespirit world, he communicates this fact to his helper whothen silences the gong. As the shaman searches for thecauses of the illness, he constantly sends back messages.The helper relays the messages to the family of the sick.When the ritual stops, the shaman explains to thefamily what he has found. Sometimes the shaman discoveredthat the soul was lost ar4 could not find its wayhome, but he has gathered it back to the body. In thatcase, no future ritual is needed. However, if the shamandiscovered that the soul of the sick person has been kidnappedor transformed, then he asks the family to get ananimal and sets another time for the next set of rituals.The type of animalbe it chicken, pig, dog or cowiswhat the shaman negotiated with the spirit.In the second ceremony, the family of the sick personprepares the animal for sacrifice. When the shamangets to the spirit world, he starts negotiating with thespirits. When the concession is accepted, the shamansends back the message. The helper lets the family know,and immediately the animal is slaughtered. Some ritualpapers will be stained with the animal's blood and beburnt as a rayment to the spirits.This is the most exhaustinghealing task.A TURNING POINTIf illness persists after Ua Neeb, the family turnsto the "position of an egg." In this traditional practice,an egg is stood up on a bottle that lies flat on atray of rice. The positioning of an egg serves as anoutlet to search for another shaman or node of healing.When the egg suddenly stands up at the mention of a shaman'sname or a new set of rituals the family goes tothat shaman for help, or prepares the new set of rituals.A new process then begins.37 )

HMONG PERCEPTION OF ILLNESS 377CONCLUSIONShamanism and herbal medicines make up the Hmong'straditional world of dealing with illness. Shamanism iselaborate, esoteric, and based on rituals and incantations.Herbal medicines, however, are solely in the women'sworld and also come in a variety of forms and procedures.Though the concept of soul loss is most stronglybelieved, illness can be caused by many other sourcessuch as nature, physiology, mankind, magic and spirits.The Hmong have evolved many different methods to dealwith each causal element and to counteract illness as awhole. By learning more about Hmong attitudes towardh-ealth and illness, Western health care providers can domuch to improve the quality and appropriateness of carefor this special group of new immigrants.ACKNOWLEDGEMENTSI would sincerely like to thank my puents, Mr.Chadang and Mrs. Xang Yang Thao, for their patience andsincerity in sharing wit me the Hmong's traditional artsof healing. I extend my appreciation to Mrs. Youa ChatMr. Cher Ker Moua, Mr. Pakoua Vang, and Mr. Chong LeeChue for their enthusiasm in teaching me our ways of doctoring.I wish to say ua tsaug to John Finck and DonnaDryer for their help in writing this paper.NOTElAn egg stone is a tiny piece of stone that the shamansucks out from an area where the victim feels a shockingpain.REFERENCESBalikci, Asen. 1967. "Shamanistic Behavior among theNetsilik Eskimo." In Magic Witchcraft and Curing,edited by J. Middleton. Austin: University of TexasPress.373

378 THE HMONG IN TRANSITIONCatanzaro, A. and R.J. Moser. 1982.Refugees from Vietnam, Laos, and247(9): 1303-1307.Snow, L. 1974. "Folk Medical Beliefstions for Care of Patients."Medicine 81: 82-96."Health Status ofCambodia." JAMAand Their Implica-Annals of Internal374

SLEEP DISTURBANCES AND SUDDEN DEATH OF HMONGREFUGEES: A REPORT ON FIELDWORK CONDUCTEDIN THE BAN VINAI REFUGEE CAMPRonald G. MungerSudden unexpected deaths during sleep continue to bea significant public health problem of Southeast Asianrefugees in the United States. Sudden deaths in sleep ofHmong refugees in Thailand were reported by Hmong ininterviews in the United States in 1981, shortly aftersensational newspaper reports of sudden deaths appearedin this country (Munger 1982). These refugee accountsand an interest in studying the problem of sudden deathas a phenomenon of migrant populations led the author toconduct field studies in Southeast Asia. A study of suddendeath in sleep of Hmong refugees was conducted in theBan Vinai refugee camp, it., northeastern Thailand, fromOctober 1982 to June 1983.One goal of this fieldwork was to document whetheror not sudden unexpected deaths during sleep occurredamong Hmong refugees in Thailand. The next step was tostudy the characteristics of sudden death victims (cases)as compared to randomly selected individuals in Ban Vtnaiof the same ethnicity, age and sex (controls). The studydesign thus used a "case-control" method (Schlesselman1982). The purpose of this report is toofdescribe casessleep disorders which may be related to the suddendeaths in sleep. Presented here are case reports of foursudden death victims with a history of previous nonfatalsleep disturbances and reports of similar nonfatal sleepdisturbances occurring to individuals who areing at the time ofstill liv-this writing. Such cases of nonfatalsleep disturbances are consistent with thethathypothesisdisorders of the control of respiration duringsleepthe sleep apnea syndromesmay be related to theSoutheast Asian sudden deaths in sleep (Munger 1982).Nonfatal sleep disturbances may identify a group of individualswho are at a highsleep.risk of dying suddenly during379375

380 THE HMONG IN TRANSITIONSudden death of adult Southeast Asian refugees 'nthe United States during sleep has been recognized as rsyndrome. The distinction must be made between asyndrome, a collection of signs and symptoms, and adisease, defined by specific abnormalities of bodilystructure or function. The disease process that is theunderlying cause of the Southeast Asian sudden deathsyndrome is unknown; further clinical studies arenecessary. The aistinction between disease and illness isalso important. Illness refers to an individual's perceptionof disease and disability, which may be quitedifferent from the physician's perception of the diseaseinvolved. Interview studies of Hmong refugees cannotprecisely define the underlying disease process of thesudden death syndrome, but the observations and explanationsoffered by Hmong refugees can provide clues to thechain of events responsible for this malady.The pattern of sudden deaths in sleep amongSoutheast Asian refugees has stimulated research inseveral disciplines, including the social sciences, medicineand epidemiology. The history and methods of someof these studies and some reactions of the Hmong communityto these studies are reviewed in this volume byHurlich, Holtan and Munger (p. 427). The initial resultsof studies summarized below serve to guide the course offuture research and public health programs.SUDDEN DEATHS OF SOUTHEAST ASIANREFUGEES IN THE UNITED STATESThe United States Centers for Disease Control (CDC)has reviewed at least seventy-nine cases of sudden, unexpecteddeath of Southeast Asian refugees in the UnitedStates occurring between 15 July 1977 and 30 April 1983(Baron et. al. 1983). The first cases of these suddendeaths were recognized by Larry Lewman, M.D., of theMultnomah County, Oregon, Medical Examiner's Off iCe. Ofthe seventy-nine cases reported to the CDC, thirty-ninewere Laotian Hmong, twenty-one were ethnic Lao, ten wereKampuchean and nine were Vietnamese. Only one of thevictims was female. All of the deaths occurred duringsleep and none of the victims had a signifi^.ant illnessprior to death. Witnesses to the deaths often described376

SLEEP DISTURBANCES AND SUDDEN DEATH 381moaning, groaning or chokingseveral victims had seizure-likedeath. Routine autopsies andnot indicate a cause of deathControl 1981; Baron et al. 1983).The suddenness of the deaths and therespiratory sounds, andactivity at the time oftoxicological analyses did(U.S. Centers for Diseaselack of specificpathologic changes are consistent with death due tocardiac dysrhythmia. The CDC received three reports ofdocumented abnormal cardiac electrical activity in thevictims; ventricular fibrillation was confirmed in eachcase. Three other Asian men have been successfully resuscitatedfrom episodes of ventricular fibrillation insleep (Tauxe et al. 1982; Tauxe, personal communication).All three patients recovered from the initial episode andshowed no evidence of coronary artery disease. One patientwas found to have self-limiting polymorphic ventriculartachycardia which was easily induced by electrophysiologicstimulation. Quinidine therapy successfullyabolished this response; procainamide had no effect.Tauxe and his colleagues concluded that the sudden deathsyndrome is the result of a primary cardiac abnormalitypredisposing certain individuals to lethal cardiacarrhythmias.Detailed studies of the cardiac conduction system ofSoutheast Asian sudden death victims are being conductedby Robert Kirschner, M.D., Deputy Medical Examiner, CookCounty, Illinois, and Friedrich A. 0. Eckner of theUniversity of Illinois School of Medicine. Conductionsystem abnormalities which may predispose individuals tosudden death have been observed by these investigators(Edcner et al. 1983).The review of pathology reports and the case-controlstudy conducted by the CDC have ruled out infectic sagents, unusual physical illness, pharmacologically activesubstances and agents of chemical or biological warfareas causes of the Southeast Asian sudden deaths(Baron et al. 1983). The CDC study also did not find anysignificant emotional experiences or stressful eventsthat characterized cases as compared to controls; thisinvestigation was not designed to be an in-depth study ofcultural or psychological factors contributing to thedeaths. The CDC investigation found no evidence that thesudden death victims under study were related, although3,77

33,382 THE HMONG IN TRANISITIONit was reported that one relative of a sudden death victimdied under similar circumstances in Laos.A study of cultural, socioeconomic and familial factorspossibly related to the sudden deaths of Hmong inthe United States was conducted by Dr. Bruce ThowpaouBliatout (1983). Bliatout conducted case investigationsfrom his unique position as a native-Hmong researcher.Of the forty-five cases investigated by Bliatout, eightof the victims had at least one other relative who diedunder similar circumstances. Bliatout presents this evidenceto support the hypothesis that a genetic factor maypredispose some individuals to sudden death. Bliatoutsuggested two triggering mechanisms that he thought mightact upon a background of genetic susceptibility to suddendeath. The first triggering factor he suggested isstress due to the inability to continue certain religiouspractices in Western countries, inability to find traditionalhealers or difficulty in adapting to a new lifestyle.The second triggering mechanism suggested byBliatout is an acquired weakness due to exposure toagents of chemical warfare in Laos.A clustering of sudden deaths within families doesresponsiblenot necessarily mean that a genetic factor isfor the deaths. Family clusters may also result f rom anenvironment the family shares from psychological or socialfactors they share, or from chance alone.Bliatout's study, conducted at his own expense, would bestrengthened by further support which would allow theaddition of a control group with which the cases could becompared.Professor Jacques Lemoine and Ms. Christine Mougneinterviewed twenty-nine families of Hmong, Lao and Mien-Yao sudden death victims in the United States (Lemoineand Mougne 1983). In four of the twenty-nine families ofsudden death victims, evidence was found of similardeaths occurring among close male relatives. Lemoine andMougne speculated that night terror may lead to suddendeath in sleep. The sudden death cases were classifiedinto two categories: victims of "culture shock" whosedeaths occurred within days or weeks after arrival in theUnited States and victims who died after a longer periodof time in this country. Members of this latter groupwere thought not to exhibit acute psychological trauma,but perhaps were exhausted, had family quarrels or were

SLEEP DISTURBANCES AND SUDDEN DEATH 383provoked by terrifying images from television. Lemoineand Mougne concluded that the Southeast Asian suddendeaths are a result of the interaction of psychological,cultural and physical factors. A case-control comparisonwould be more convincing than a presentation of selectedcases and should be incorporated in future research onthe sudden death syndrome. Lemoine and Mougne also donot provide statistical data to support their hypothesisthat two distinct categories of sudden death exist.Based on their experiences in Southeast Asia and on thereaction of refugees interviewed in the United States,Lemoine and Mougne responded with a "qualified no" whenasked whether sudden deaths in sleep occur in the refugees'own homeland, noting that "mainland Southeast Asianpopulations have never had a sophisticated medical surveillancesystem such as found in the United States(Lemoine and Mougne 1983, p. 10).The sudden deaths of Southeast Asian refugees arestrikingly similar to sudden deaths of Filipino men,known in Tagolo as bangungut. On the basis of the circumstancesof the deaths, the distribution of the time ofdeath, and the distribution of the rates of death by age,Munger (1982) concluded that the Southeast Asian andFilipino sudden deaths are part of the same syndrome,which is perhaps a regional phenomenon in Southeast Asia.The studies of the Southeast Asian sudden deathshave not yet identified a factor that is alone responsiblefor the deaths. The cause of the syndrome of suddendeaths is most likely a complex interaction of genetic,environmental and behavioral factors; the author'sresearch on sudden deaths of Hmong in the Ban Vinai refugeecamp supports the hypothesis that each of these threefactors is involved. The evaluation of genetic, environmentaland behavioral hypotheses is in progress and willbe presented elsewhere. Evidence linking the suddendeaths of Hmong with sleep disorders is presented in thisreport. Sleep disorders may be more prevalent in So!itheastAsian populations than currently recognized.THE BAN VINAI REFUGEE CAMPThe Ban Vinai refugee camp is located in northeasternThailand, near the Mekong river bordering Laos.37J

384 THE HMONG IN TRANSITIONThe camp population in November 1982 was 33,117, including32,933 Hmong, 143 ethnic Lao, 20 Khamu, 10 Mien-Yaoand 9 Thai Dam (UNHCR 1983). The case-control study waslimited to the Hmong refugee population. The problem ofsudden death in sleep could not be studied in Ban Vinaias it had been in the United States. Death certificateswere rarely completed in Ban Vinai; most deaths occurredoutside of the hospital. Even in the relatively rareevent of a death in the hospital, a death certificate wasnot likely to be completed. Reverence for the spirits ofthe deceased was not overpowered by laws requiring postmortemexaminations; autopsies were unthinkable in BanVinai.A United States State Department physician, upon areturn trip to Ban Vinai, remarked that he previously wasunable to turn up any evidence of sudden death in sleep.The expatriate health workers were aware of only one victimof sudden death in sleep, a twenty-six-year-old Hmongman who worked in the public health program. These initialreports were less than encouraging; where were thesudden death victims, if they existed at all? It soonbecame apparent that most relief workers had little contactwith what was now the largest Hmong community in theworld. The largest number of expatriate volunteers wereworkers who arrived each morning, worked within the hospitalcompound, then left the camp in the evening. Thehealth workers had contact with a small proportion of theHino.ig refugees who were ill. The bamboo fence around thehospital was symholic of a wide cultural gap between theWestern medical workers and the Hmong community. Even ifthis gep did not exist, sudden death cases would still bemore likely to occur in residences than in the hospitaland go unreported. Thus the hospital was not a usefulsetting for locating sudden death cases, although thestaff was very supportive of the author's research project.Soon after the author's arrival in Ban Vinai a meetingwas arranged with Mr. Vang Neng, the chief leader ofthe Hmong in Ban Vinai. The purpose of this meeting wasto outline what was known of the sudden death syndrome ofSoutheast Asian refugees in the United States and to seekVang Neng's advice Ind assistance for a study of similardeaths in Ban Vinai. Vang N .ng was well aware of thesudden deaths of Hmong in the United States and thought330

SLEEP DISTURBANCES AND SUDDEN DEATH 385similar deaths occurred in Ban Vinai; he said that aletter was sent to the United Nations High Commissionerfor Refugees (UNHCR) field office in Ban Vinai which describedseveral unexpected sudden deaths that hadoccurred recently,RESEARCH METHODSAn effective surveillance network that would reportca:-..cs of sudden death had to be organized. Informationwas sought on any death that was sudden, unexpected orunexplained. Sudden death cases were defined as anydeath occurring withintwenty-four hours of the onset ofacute illness; thus this study was not limited to suddendeaths in sleep. Discussions with Vang Neng included howto use best the existing social structure in Ban Vinai tolocate cases. Vang Neng arranged a meeting with theHmong Advisory Council at which the study proposal waspresented. The Hmong Advisory Council includes centerchiefs, clan leaders and elders. The center chiefs,leaders of the eight districts in Ban Vinai, mediatedisputes and oversee the distribution of food, suppliesand mail within each district. Seventeen clans are representedin Ban Vinai, each having an appointed leader.The clan leaders, usually older and more influential thanthe center chiefs, are the most important negotiators onmatters concerning clan members. After the introductorymeeting, individual interviews were held at the officesof center chiefs and in the homes of clan leaders. Theleaders were overlapping components of a surveillancenetwork; the center chiefs searched for cases withintheir districts, which included many clans, and clanleaders searched for cases within their clan, includingcases of members who may have lived in different centers.Once cases were reported, interviews with the victims'relatives were sought. The interviews were conductedwith the assistance of Mr. Vang Lia, a :4mong Interpreterappointed by Vang Neng. Vang Lia was selected on thebasis of his exceptional linguistic talents and staturein the community.Interviews were conducted with relatives or spouseswho lived with the victim and either witnessed the deathor discovered the body of the deceased. Notes were taken381

386 THE HMONG IN TRANSITIONon second-hand reports of sudden deaths, but formal interviewswere conducted only with actual witnesses.Interviews were conducted at the home of the informantafter permission had been granted bythe center chief orclan leader. The interview was initially "open-ended"with the informant describing how the death occurred andany events considered to be related to the death. Aninterview form was then used to record demographic data,the circumstance of death, migration history, medicalhistory, sleep habits, occupational history, militaryexperience, personal habits and dietary history. A genealogywas constructed of all relatives who could be rememberedby household members; it included information onresidence and age of living relatives and the age, date,location and cause of death for deceased relatives. Casereports of sudden death in sleep were crosschecked by Ms.Elizabeth A. Booton, C.R.N., and a female Hmong interpreter,Ms. Mai Lee Lor, both unaware of the results ofthe initial interview. Comparison interviews of individualsrandomly selected from the UNHCR's list of campresidents (controls) were conducted and were matched tocases by age, sex and ethnicity. The case interviewsincluded information on the characteristics of the deceasedas reported by a relative. Rather than interviewingthe control subject directly, a relative of thecontrol was interviewed, so that the information on casesand controls was more comparable. Further, the relationshipof the informant to the control subject was matchedto that of the informant and the deceased in the caseinterview with which the control was compared.CASE REPORTSTwenty-seven formal interviews were conducted in BanVinai with relatives of sudden death victims. Sixteen ofthese were cases of sudden unexpected death during sleepin which the victim was not previously ill. Data onthese cases and controls are being analyzed at the timeof this writing and will be reported elsewhere. Fourcase reports were collected in which victims of suddendeath in sleep had experienced previous nonfatal sleepdisturbances, including abnormal respiratory sounds,breathing difficulties or transient loss of conscious-382

SLEEP DISTURBANCES AND SUDDEN DEATH 387ness. None of tile controls experienced such sleep disturbances.Several personal accounts of similar sleepdisturbances were also obtained from Hmong refugees inBan Vinai who were still alive at the time of thiswriting.Case 1In October, 1981, a thirty-nine-year-old Hmong maledied during sleep in the Nong Khai refugee camp, also innortheastern Thailand. The wife of this man, interviewedin Ban Vinai, reported that at midnight her husband madea noise during sleep that sounded like an obstruction inhis throat. The abnormal respiratory sounds were madefor several minutes. The wife cared to her husband; hedid not respond and his body was flaccid. The subject'sbreathing while unconscious was reported to be very slow.The wife massaged the husband and he regained consciousnessin four to five minutes. The wife asked her ,-,usbandwhat was wrong and he replied that something had caughtin his throat. No physical object was known to havecaused the obstruction. The subject went back to sleepuntil 7:00 A.M.At 7:00 A.M. the subject held his infant whileseated (-41 the bed and his wife went into the kitchen tocook. A relative came to the kitchen soon after andreported that the woman's husband was regaining consciousnesoand the infant was on the floor, having eitherbeen dropped or set down. Her husband complained of apain in his chest; he had never complained of this before.The wife gave her husband a massage, and whiledoing so noticed that his eyes had rolled back. Shecalled to him, buL he was unconscious again. The husband'sarms and legs flexed and his entire body becamerigid. The woman noticed that her husband was notbreathing and tried to administer mouth-to-mouth resuscitation.The man was taken to the camp hospital, but wasdead on arrival.Case 2A forty-year-old Hmong man was found dead in his bedin Ban Vinai on 6 Jule 1981. At 6:00 that morning, thewife of thy; man got up to make a fire for cooking. The3&3

388 THE HMONG IN TRANSITIONwoman heard the children crying and called to her husbandto take care of them. Her husband did not answer and shereturned to the bedroom to see what was wrong. The womanfound her husband dead on their sleeping platfoirn; hisbody was lying undisturbed on the bed with his head stillon the pillow. The wife remarked that her husbandappeared to be sleeping, with no signs of a struggle ordistress.This victim's wife described dreams her husband hadin Laos in which a ghost, called a poi ntxoog,' wouldappear. The zi ntxoog is a female spirit, which is believedto live in the jungle and is greatly feared.Heimbach reported (1979, p. 28) that poj ntxoog refers toa spirit that rides a tiger and whose father is a tiger.Interviews with Hmong in Ban Vinai generally confirmedHeimbach's characterization; the P1 ntxoog may take awaythe spirits of the living, thus causing illness anddeath, and shamanistic ceremonies may be performed toprotect against mi. ntxoog. The subject told his wifethat the mj ntxoog would sometimes visit during sleepand sit on his chest. Breathing during these episodeswas very difficult; the subject would make a sound ineither a moan or a cry, and then his wife would beawakened. At other times, the subject would awakenfirst, then wake his wife to tell her what had happened.Such sleep disturbances were first noted by the subject'swife in 1979 in the Laotian highlands, and were estimatedto occur once every two or three months. The subject hadbeen in Thailand less than a year at the time of hisdeath. The sleep disturbances had not yet occurred inThailand; the last one was in 1980, when the family wasstill in Laos.Case 3Near midnight on 12 June 1981, a Hmong man heard histwenty-five-year-old son moaning and groaning in hissleep in the Nong Khai refugee camp. The man called tohis son and asked what was wrong. The subject was awakenedwithout great difficulty and reported that his legshurt. He described the pain as a tingling sensation, asif needles were sticking him in both thighs, and saidthat his legs felt very weak. The subject returned tosleep. In the morning the subject's legs were numb, so3

SLEEP DISTURBANCES AND SUDDEN DEATH 389his wife gave him a massage. Concerned about the sleepdisturbance that occurred the night before and the numbnessof his son's legs, the father then took him to thecamp hospital which was just across the road.At 7:00 A.M. the camp physician examined the subjectbut could not find any abnormalities. The physician requestedthat the patient return later in the day so thata chest x-ray could be obtained. The family returnedhome shortly after 8:00 A.M. and family members stayedhome all day. At 2:00 P.M. the subject said that he felttired, asked his wife for a blanket, and went to bed.Shortly after the subject fell asleep, the family noticedthat his right arm jerked and rested perpendicular to hisbody. The parents called to their son to ask him whatwas wrong, but theie was no answer. The subject's bodywas motionless on the bed and he could not be awakened.The relatives carried the unconscious man across the roadto the camp hospital. The physician ordered that thepatient be rushed to the Thai hospital in Nong Khai. Thesubject was dead upon arrival at the Thai hospital and adeath certificate was issued in English and in Thai. Thephysician gave a copy to the family and instructed themto present the certificate if anyone inquired about thedeath. The cause of death was listed as "unknown."Case 4At about 10:00 P.M. on 23 August ;978, in Ban Vinai,the mother and father of an eighteen-year-old Hmong girlheard their daughter making abnormal respiratory soundsduring sleep. Unable to awaken their daughter, the parentsimmediately responded by giving her a massage andsticking needles into the inside of her arm. The motherof the subject knew of similar cases of sleep disturbancesand had learned the massage and acupuncture treatmentfrom other Hmong refugees in Ban Vinai. At the onsetof this treatment the subject was breathing with difficulty,as if something was obstructing her throat; theabnormal breathing sounds lasted several minutes. Thesubject then awakened and asked why her parents were massagingher and sticking her with needles. The parentsreplied that she was making noises in her sleep and couldnot be awakened. The subject could not explain what happened,and went back to bed, sleeping normally for the3836

390 THE HMONG IN TRANSITIONrest of the night. The next morning, the subject feltnormal.On 11 April 1981, the subject was sleeping in thesame bed as her eighteen-year-old female cousin; the subject'smother and father were also sleeping in this room.At 3:30 A.M., the subject's arm moved while she wassleeping, hitting and awakening her cousin. The cousinand the parents then heard abnormal respiratory soundsthat were said to be the same as in the episode of 23August 1978. The parents again responded by massagingtheir daughter immediately after hearing the sounds shemade during sleep. The sounds lasted two to three minutes,then gradually subsided. The subject was unresponsiveand having seizures, with her body moving side toside and her legs kicking. The family tried without successto detect a pulse on the subject's chest. When therespiratory sounds ceased, the subject's body was flaccid.Perspiration was noted over the entire surface ofher body. The girl was dead within minutes after thesounds ceased, although the family continued the massagein vain for one hour.Other Sleep Disturbances Noted in Ban VinaiFor the purposes of the case-control study, a casehad to be narrowly defined; sudden death during sleep /asa prerequisite. Soon after the fieldwork in Ban Vi.iaibegan, several Hmong people reported that they had sleepdisturbances similar to those already presented. Leaderswere asked about problems in sleep, but because a nonfatalsleep disturbance is much less dramatic than adeath, it is more likely to remain unnoticed outside ofthe household. A systematic survey of households forpeople with sleep problems was not feasible due to ashortage of time and funds. We located five Hmongpeople--three women, twenty, twenty-five and thirty-twoyears of age, and two men, thirty-two and fifty-five yearsoldwith histories of sleep disturbances in which therelatives were alerted by moaning, choking or groaningsounds and seizure-like activity, and the subjects wereunresponsive to attempts to awaken them for some time.Hmong refugees who were well aquainted with the study ofsudden death in progress brought these five cases to theauthor's attention, thinking that these nonfatal episodes3 Sri

SLEEP DISTURBANCES AND SUDDEN DEATH 391might be caused by the same factors leading to death inothers. Four of these cases had multiple episodes, someas often e.s two to three times a month. All of thesepeople were anxious about their condition and feareddying in sleep.A sixth individual, a thirty-five-year-old Hmongwoman, was observed having sleep disturbances by themedical staff in the Ban Vinai hospital, and is the bestdocumented case of a nonfatal sleep disturbance in BanVinai. This patient was admitted to the Ban Vinai hospitalon 10 May 1983, with complaints of diarrhea, numbnessof the extremities for one month and tightness of thechest. The admitting physician suspected that vitamin Bdeficiency and intestinal parasites were the basis of herillness. Over the next three days, while receiving vitaminsupplements and antiparasite medication, the numbnessand pain in her extremities gradually decreased; generalweakness and chest pain persisted.At 12:30 A.M. on 17 May 1983, an elderly Hmong manstaying in the hospital noticed that the woman was havinga seizure in her sleep, with her fists clenched and armsshaking. A Hmong nurse on duty was notified immediately;he could not detect any heart sounds with a stethoscope.The patient was not breathing and artificial respirationwas given. Breathing was restored in five minutes. Theattending physician, Dr. Bibba Holland, arrived soonafter and noted that no reason could be found for thecardiorespiratory arrest and seizure in sleep. The nextday the patient complained of numbness in the hands andlower legs, as well as general weakness. Headache andchest pains, described as a squeezing sensation, wereintermittent. Walking did not Increase the chest painbut rather eased it somewhat.On 18 May 1983, at noon, the patient experiencedanother episode of cardiopulmonary arrest accompaniedwith a seizure. A Hmong nurse heard a brief groan andnoticed that the patient was unconscious on her bed. Thepatient's face was described as blue; the hands were saidto be white. The patient began having a seizure and becameincontinent of urine; cardiopulmonary resuscitation(CPR) was initiated by the Hmong nurse. Dr. Hollandarrived and confirmed that heart sounds were not audiblewith a stethoscope and that the patient was not breathingand cyanotic; he then continued CPR. The seizure stopped381

392 THE HMONG IN TRANSITIONin less than five minutes and the patient regaineo regularcardiac rhythm and breathing. Within a half-hour thepatient, while very drowsy, was able to answer some questionsand recalled having a severe headache before theepisode. The patient requested a massage, with acupuncture.The hospital statf did not respond to thisrequest, but the patient's family did. Bruises were lefton the patient's arm by the vigorous massaging andpinching.Over the next three days, the headaches, chest painand numbness subsided, although the patient continued tofeel tired and weak, with a poor appetite and abdominalpain. A chest x-ray showed no evidence of tuberculosisor cardiac enlargement. No abnormalities were noted inthe physical examination or in the routine blood, urineand stool tests. The patient was discharged on 25 May1983 with a tentative diagnosis of hypovitaminosis andseizures of unknown origin.The patient and her family were interviewed severaltimes in the hospital and at home after discharge. Thepatient had experienced a seizure during daytime sleep on8 May 1983, two days before her admission to the BanVinai hospital. After the subject was found to be havinga seizure and to be nonresponsive, a vigorous massage wasgiven by her family, during which she regained conciousness.Further interviews revealed that a similar episodehad occurred six months earlier; shortly after midnightthe woman made abnormal respiratory sounds during sleep,alerting her husband. The woman's limbs were rigid andshe was incontinent of urine. Her husband called otherrelatives to assist in giving a massage. The woman didnot fully regain consciousness until about three hourslater.One more episode that occurred in early 1982 wasrecalled. The woman collapsed one morning in the marketwhere she made and sold textile goods, which was next tothe camp hospital. Her teenage daughter witnessed thisepisode and reported that two American nurses were nearbyand began CPR because the woman was not breathing. Afterthe woman's breathing was restored, she was taken, stillunconscious, to the camp hospital. The woman remained asa patient for three or four days. Hospital records arenot kept longer than three months, so no documentationwas available.3 8 d

SLEEP DISTURBANCES AND SUDDEN DEATH 393DISCUSSION AND CONCLUSIONSOf the sixteen cases of sudden, unexplained deathduring sleep studied in Ban Vinai, four included a historyof nonfatal sleep disturbances. Two of the fourcases had sleep disturbances in which relatives werealerted by abnormal respiratory sounds, described asmoaning, choking or groaning, accompanied by seizure-likeactivity, and the subjects were unresponsive to attemptsto awaken them. The third case experienced repeated episodesof difficult breathing during sleep. The fourthcase was found groaning in sleep with later complaints ofweakness and numbness in his legs; the next day he diedsuddenly during daytime sleep. None of the sixteen controlsubjects experienced any of these sleep disturbvices.Six Hmong refugees in Ban Vinai, all alive at thetime of this writing, also reported episodes of seizurelikeactivity and unresponsiveness in sleep. The relativesof each of the six subjects, all alerted byabnormal respiratory sounds, responded with a vigorousmassage, pinching and acupuncture treatment.The purposeof this treatment was to release the "bad blood" from thebody, which is perceived as a small amount of dark, thickblood, believed to be "cool." If the "bad" or "cool"blood is allowed to circulate, it is thought to "cool"the limbs and tventually the heart, causing death.Massage is a common treatment used when a Hmong personbecomes ill. Vigorous massage, which may includeslapping, hard pinching, even hitting with a shoe, andacupuncture in special sites are a specific response toloss of consciousness. The knowledge and practice ofthis treatment seemed to be spreading quickly in BanVinai.The reports of sleep disturbances presented here arean indication that sleep disorders may be related to thesudden deaths of Hmong refugees during sleep, and may bemore common than previously recognized. Lemoine andMougne (1983, p. 18) described the case of a Hmong manwho had been successfully resuscitated from a nearlyfatal episode of ventricular fibrillation. The subjectdescribed a sleep disturbance that had occurred in Laosfive years before. The subject described being under theattack of a spirit called a dab col which rests on thechest of a sleeping person38d

394 THE HMONG IN TRANSITIONYou want to listen, you can't hear; you want tospeak, you are dumb; you want to call out, youcannot; you feel you are dying; you want to runaway. You piss with fear in your sleep.Tobin and Friedman (1983) also described a case ofrepeated sleep disturbances of a twenty-two-year-oldHmong man in the United States. After three nights ofdisturbed sleep, the subject described his problem to aHmong resettlement worker. Tobin and Friedman wrote:The first night he woke suddenly, short ofbreath, from a dream in which a cat was sittingon his chest. The second night, the room grewdarker, and a figure like a black dog came to hisbed and sat on his chest. He could not push thedog off and grew quickly and dangerously short ofbreath. The third night, a tall, white-skinnedfemale spirit came into his bedroom from thekitchen and lay on top of him. Her weight made itincreasingly difficult for him to breathe, and ashe grew frantic and tried to call out he couldmanage but a whisper. He attempted to turn ontohis side, but found he was pinned down. After 15minutes, the spirit left him, and he awoke,screaming (Marshall Hurlich, this volume, p. 440).The breathing difficulties were interpreted by the subjectand other Hmong people as a spirit problem and ashamanistic ceremony was performed to protect the subjectfrom further spirit visits.The accounts described in this paper and publishedpreviously support the hypothesis that the broad group ofdisorders of respiration during sleep, known as the sleepapnea syndromes, may be related to the Southeast Asiansudden death syndrome (Munger 1982). This is not an explanationof the deaths, but rather a plausible mechanismin which genetic, environmental and behavioral factorsmay increase the risk of sudden death. Sleep apnea wasdiscounted as a factor contributing to these suddendeaths in a CDC report which stated (1981, p. 582),"According to their families, none of the victims manifestedclinical signs of the sleep apnea syndrome, suchas obesity, snoring, frequent nocturnal awakening or3'30

SLEEP DISTURBANCES AND SUDDEN DEATH 395hypersomnolence." A later report by Baron et al. (1983,p. 2950) stated that "refugee victims did not manifestsymptoms of sleep apnea." As stated previously, sleepapnea is a broad group of disorders of respiration duringsleep, not a single syndrome. Obesity, snoring, frequentnocturnal awakening and hypersomnolence characterize oneof the earliest described sleep apnea syndromes, thePickwickian syndrome (Jung and Kuhlo 1965), but do notrepresent necessary symptoms for all sleep apnea problems.Sleep apnea has been found to be more prevalent inmales, may be asymptomatic and may give rise to fatalcardiac arrhythmias (Block et al. 1979; Cherniak 1981;Phillipson 1978). Only sleep studies will determine withcertainty whether or not one or more syndromes of sleepapnea are prevalent in Southeast Asian refugee populations,and whether or not these disorders are related tosudden death in sleep. Sleep disorders do not aloneeasily explain why the deaths occur during sleep. Suddendeaths in sleep of Southeast Asian refugees may be relatedto both sleep apnea and cardiac abnormalities.Clinical and cultural studies of sleep disorders ofSoutheast Asian refugees are warranted. A survey ofsleep disorders should not be attempted until a clinicalresponse can be offered to individuals identified ashaving a sleep disorder. Investigators from the clinicaland social sciences must work closely with SoutheastAsian community leaders, family leaders and researchers.Although the mechanism of sudden death' and factors whichincrease the risk of sudden death are unclear, it isknown that quick intervention has already prevented somedeaths. Training in CPR and the use of emergency medicalservices should be a high priority for health educationprograms serving Southeast Asian refugees.ACKNOWLEDGEMENTSThe support and assistance of my wife, Elizabeth A.Booton, through often difficult situations, has contributedgreatly to all pnases of my dissertation research.The Royal Thai Government, especially the NationalResearch Council and the Ministry of the Interior, kindlygranted approval for the extended research in Ban Vinai.Dr. A.G. Rangaraj, Ms. Ailsa Holloway, Mr. Daniel Bellamy3Oi

396 THE HMONG IN TRANSITIONand Ms. Wadtanee Phiktakpao of the United Nations HighCommission for Refugees provided generous assistance inBan Vinai and Bangkok. I am also grateful for theassistance of The Christian Medical Team-Christian andMissionary Alliance of Thailand, The International RescueCommittee, The American Refugee Committee, World Visionof Thailand, and Mr. Hang Sao, Dr. Bruce ThowpaouBliatout, Mr. Vang Neng, Mr. Vue Mai, Mr. Chong Moua Lee,Mr. Chong Koua Vang, Mr. Hang Doua, Mr. Vang Lia, Ms.Mai Lee Lor, Dr. Amara Pongsapich, Dr. Bibba Holland, Dr.Marshall G. Hurlich, Dr. Neal Holtan, Dr. Robert Tauxe,Dr. Tom Prendergast, Ms. Nampet Panichpant-M., Dr. SteveHelgerson, Mr. Bob Johnson, Mr. Doug Hulcher and Ms.Andrea Crossland. All errors of fact or interpretationare my own. Financial support was provided in part bythe Intergovernmental Committee for Migration and theGraduate Student Research and Travel Fund of theUniversity of Washington, Seattle, Washington. Finally,I offer my deepest sympathy to the Hmong families whohave suffered great losses, and my appreciation for theircooperation and patience during this research.NOTES1The orthography used for Hmong terms In this paper isthat of G. Linwood Barney and William A. Smalley, asused by Ernest Heimbach in his White Hmong-EnglishDictionary (1979).REFERENCESBaron, R., S. Thacker, L. Gorelkin, A. Vernon, W. Taylor,and K. Choi. 1983. "Sudden Death Among SoutheastAsian Refugees." Journal of the American MedicalAssociation 250 (21):2947 -51.Bliatout, B.T. 1983. Hmong Sudden Unexpected NocturnalDeath Syndrome: A Cultural Study. Portland, Oregon:Sparkle Publishing Enterprises, Inc.394

SLEEP DISTURBANCES AND SUDDEN DEATH 397Block, A.J., et al. 1979. "Sleep Apnea, Hypopnea andOxygen Desaturation in Normal Subjects: A StrongMale Predominance." New England Journal cf Medicine300(10):513-517.Cherniak, N.S. 1981. "Respiratory Dysrhythmias DuringSleep." New Eng4nd Journal of Medicine 305(6):325-330.Eckner, F., R.H. Kirschner, and K. Baron. 1983. "SuddenNocturnal Death in Southeast Asian Refugees: A ProgressReport." Proceedings of the 1983 Meeting ofthe American Academy of Forensic Science. AbstractG-32.Heimbach, Ernest E. 1979. White Hmong-English Diction-Hi. Ithaca, New York: Linguistic Series IV, SoutheastAsia Program, Cornell University.Jung, R., and W. Kuhlo. 1965. "Neurophysiological Studiesof Abnormal Night Sleep and the Pickwickian Syndrome."Progress in Brain Research: Sleep Mechanisms18, p. 140. Amsterdam: Elsevier.Lemoine, J., and C. Mougne. 1983. "Why Has Death Stalkedthe Refugees?" Natural History, November: 6-19.Munger, R.G. 1982. "Sudden Death in Asian Populations:The Case of the Hmong." Hmong in the West: Observationsand Reports. Papers of the 1981 Hmong ResearchConference, edited by Bruce T. Downing andDouglas P. Olney. Minneapolis: Center for Urban andRegional Affairs, University of Minnesota.Phillipson, E.A. 1978.Sleep." American118:909-939.Schlesselman, J.J. 1982.Conduct and Analysis."Control of Breathing DuringReview of Respiratory DiseaseCase-Control Studies: Design,New York: Oxford Press.3 Ji

398 THE HMONG IN TRANSITIONTauxe, R.V., et al. 1983. "Sudden Nocturnal Death inLaotian Refugees." First International Symposium onPublic Health in Asia and the Pacific Basin. Honolulu:School of Public Health, University of Hawaii.Tobin, 3.3., and J. Friedman. 1983. "Spirits, Shamans,and Nightmare Death: Survivor Stress in a HmongRefugee." American Journal of Orthopsychiatry 53(3):439-448.Urited Nations High Commissioner for Refugees (UNHCR)."Population Statistics of the Ban Vinai RefugeeCamp." Unpublished. Bangkok, Thailand.United States Centers for Disease Control. 1981. "Sudden,Unexpected Nocturnal Deaths Among Southeast AsianRefugees." Morbidity and Mortality Weekly Reports30:581-89.3J4

A CROSS-CULTURAL ASSESSMENTOF MATERNAL-CHILD INTERACTION:LINKS TO HEALTH AND DEVELOPMENTCharles N. Oberg, Sharon Muret-Wagstaff,Shirley G. Moore and Brenda CummingOver two hundred years ago the Hmong people migratedfrom China to the mountaintops of Laos (Le Bar and Suddard1960). In the last decade, as a result of war, thousandshave come to the United States and to the Twin Citiesmetropolitan area. Their arrival here has given us theopportunity to explore from a cross-cultural perspectivethe differences in the perception of health and diseaseand the utilization of the health care system. In addition,we have attempted to compare and contrast Hmong andCaucasian toddler development, parenting techniques andmaternal-child interaction, all of which are so importantfor subsequent growth and development.METHODThe present study was conducted in the PediatricClinic at Hennepin County Medical Center from November198? through March 1983. The subjects consisted oftwe.ty-four mother -child dyads, twelve each from theCaucasian and Hmong communities of the Twin Cities. Theaverage ages at the time of the study for Hmong andCaucasian children were 20 months (range: 18.0 to 21.5)and 19 months (range: 18.0 to 20.3), respectively. Allfamilies were of low socioeconomic status. In addition,all children were born at Hennepin County Medical Centerand were part of a larger longitudinal study (Muret-Wagstaff and Moore 1983) which was beguo in the newbornperiod.During a one-hour pediatric clinic visit for eachchild and parent, a pediatrician (Charles N. Oberg) whowas unaware of the results of the earlier newborn studyand was trained in the use of the cited instruments, collecteddata in the areas of health, development and par-399395

396400 THE HMONG IN TRANSITIONent-child interaction. First, sociodemographic data anda medical history were obtained by inter view. Next, adevelopmental assessment was conducted, using the DenverDevelopmental Screening Test (Frankenburg, Goldstein andCamp 1971), followed by a complete physical examination.Throughout the clinic visit, mother-child dyads were observedand rated on several scales of mother-child interaction.These were (1) items from the Mother-Child RatingScales (Egeland, Deinard, Brunnquell and Taraldson1%5); (2) Ainsworth's System for Rating Maternal-CareBehavior (1976); and (3) interaction items from the HomeObservation for Measurement of the Environment (HOME)(Caldwell and Bradley 1978). The maternal-child interactioninstruments are described below.Mother-Child Rating ScalesFourteen items from the Mother-Child Rating Scaleswere chosen from those used by Egeland, Deinard, Brunnquelland Taraldson when they observed feeding, play andclinic waiting room behavior of mothers and infants fromthe same urban population from which the present Caucasiansample was drawn. Eight of the fourteen items wereamong ten which were weighted significantly toward Mother'sCaretaking Skills and Mother's Affective Behaviorfactors in the original authors' analysisfactors whichstrongly contributed to differentiation between subsequentadequate and inadequate parental care groups(Egeland, Deinard, Brunnquell, Phipps-Yonas and Crichton1979; Vaughn, Taraldson, Crichton and Egeland 1980). Selectionof the fourteen items was directed by their relevancyto the toddler period and the likelihood of theirexpression during a typical pediatric visit, confirmed bypilot testing. The items focus on the specific qualityand variability of the mother's handling of the child andresponsiveness to fussing and crying, as well as her generalpatience, attentiveness and expressiveness towardthe child. Each item is scored on a nine-point scale,with scale points defined for individual items, highscores indicating optimal performance.Ainsworth's System for Rating Maternal-Care BehaviorAinsworth's nine-point rating scales represent amore global assessment of the parent-child relationship

LINKS TO HEALTH AND DEVELOPMENT 401in four areas: Sensitivity vs. Insensitivity to theBaby's Communications, Cooperation vs. Interference, Acceptancevs. Rejection, and Accessibility vs. Ignoringand Neglecting. The Sensitivity vs. Insensitivity to theBaby's Communication scale, which distinguished betweenthe consistently high-scoring Hmong and more variableCaucasian mothers in the newborn period, has been mostwidely used in other research (Vaughn, Taraldson,Crichton and Egeland 1980; Whitt and Casey 1982) and isde-cribed here in greater detail.1. Sensitivity vs. Insensitivity to Baby's Communication.This scale reflects the mother's ability to perceiveaccurately the signals and cues implicit in her infant'sbehavior. A mother rated as highly sensitive,or ), is one who is exquisitely attuned to her child'ssignals. She responds promptly and appropriately tothe child's subtle cues. She is empathic and respectsthe child's point of view. Her interactions with thechild seem complete and well-rounded, and she monitorsthe child and the environment closely, anticipatingproblems. A rating of 7 is considered a sensitive response.The mother generally interprets the child'ssignals accurately and is never seriously out-of-tunewith the baby's state. However, she may be less sensitiveto more subtle cues. For example, she may feedhim when he fusses and sucks his fist, but she mightmiss the message in his uncomfortable squirming. AA mother rated 5, or inconsistently sensitive, is visiblymore ambivalent. She may be sensitive on occasion,but is inconsistent and at times may be unawareof her baby's whereabouts or needs. An insensitivemother is scored 3. She fails to respond promptly orappropriately. However, the capacity for sensitivitymay be seen on some occasions. Fi.ally, a highly insensitiveresponse, or 1, is characterized by a motherwho is geared almost exclusively to her own wishes,moods and activities. She responds to the baby's signalsonly if repeated, prolonged and intense, and onlyif they are in 'seeping with her own needs.2. Cooperative vs. Interference. The central issue addressedby this scale is whether the mother's initiationsof interaction are geared to the baby's state or3 9?

402 THE HMONG IN TRANSITIONwhether she interrupts and fragments the child's ongoingactivity. A highly cooperative mother capitalizeson spontaneity and sees her baby as a separateperson whose activities have a validity of their own.A highly interfering mother fails to acknowledge thisautonomy and is arbitrary in her interference.3. Acceptance vs. Rejection. This scale deals with themother's behavior that indicates the extent to whichshe integrates the positive and negative feelings thatcome with parenthood. The highly accepting mother respectsthe fact that the baby has a will of his own,even when it opposes hers. Her periods of irritationand frustration are brief, and she accepts responsibilityfor his care without resentment. At the otherend of the continuum, highly rejecting behavior includespervasive irritation and scolding, rough handlingand ill-concealed anger, and maternal escalationof conflict in power struggles.4. Accessibility vs. Ignoring and Neglecting. This scaledeals with the mother's awareness of and responsivenessto child. A highly accessible mother is aware ofthe baby consistently; she actively acknowledges andresponds to him or her. An inaccessible mother ignoresher baby and in this sense she neglects herbaby. It is the simple acknowledgment of the baby'sreal presence that is important to this scalenot thequality of her response to him tapped by the Sensitivity-Insensitivityscale.HOMEFinally, a subset of ten parent-child interactionitems from the HOME was used to assess (1) the emotionaland verbal responsiveness of the mother, (2) the avoidanceof restriction and punishment, and (3) maternal involvementwith the child.This instrument uses a Yes/No dichotomous scoring.The rating is such that the higher the total score, themore optimal the environment for the child. Such scoresderived from the use of the complete instrument have beenrelated to later evidence of positive cognitive development(Elarao, Bradley and Caldwell 1975).39d

LINKS TO HEALTH AND DEVELOPMENT 403The examiner was trained in the use of the observationalinstruments, and pilot testing was conducted overa three-week period luring medical visits in the pediatricclinic in which this study was to take place. In pilottesting it was noted that both parent and child behaviorbecame more stilted with two observers presentrather than one. Therefore, in an attempt to maintainhigh validity of study observations, we decided to conductfive reliability observations on Hmong and Caucasianchildren who were from the same clinic population but whowere not participants in the project sample. Interscorerreliability between the study examiner and a second simultaneousindependent observer, defined as within 1-pointagreements for 9-point scales and exact agreement foryes/no dichotomous ratings, was .83 and 1.0 for the twoparent-child dyads tested immediately prior to the study,1.0 for the two tested while the study was in progress,and .87 for a mother-child pair tested at the conclusionof the study. Interobserver reliabilities for theMother-Child Rating Scales, Ainsworth's System for RatingMaternal-Care Behavior, and the HOME, respectively, were.94, .95, and .93.Statistical analyses of the sociodemographic andmedical data and scores from the mother-child interactionscales were carried out with two-tailed t tests and Fisher'sexact test. Denver Development Screening Tests werescored and compared according to the manual (Frankenburg,Dodds and Fandal 1973). Thorough physical examinationsfollowed a standard format (McKay 1975) and are reporteddescriptively.Sociodemographic DataRESULTSWhile in early infancy each child in the study livedin a household with the mother and at least one otheradult, interviews with parents during the toddler periodrevealed that the make-up of the family network and theintegration of family roles was much different for thetwo groups, as might be expected. The family size waslarger for Hmong than for Caucasian families with a significantlygreater number of adults (t = 2.06, 2

404 THE HMONG IN TRANSITIONand children (t = 3.73, 2

108-98-00-70-60-percent 58 -483828 -18 -Figure 1MEDICAL EVENTSII HmongCaucasian8AccidentsEmergency Hospitaliza IngestionsVisits tions40i

406 THE HMONG IN TRANSITIONzation of emergency room services, and a tendency forCaucasian children to experience more burns, trauma, ingestionsand hospitalizations. The Caucasians as a groupreported over three times as many accidents as the Hmong.Development AssessmentThe Denver Developmental Screening Test showed nochild who would have been classified as developmentallydelayed. However, the Hmong parents did tend to reportfewer items passed in the language domain.Physical ExaminationGrowth patterns of all children were normal exceptfor one obese female in each group. The physical examinationsshowed an abundance of cold symptoms typical forthe season. There were no significant medical findingsthat differentiated the two groups.Mother-Child Interaction MeasuresTable 1 presents the results from the Mother-ChildRating Scales and demonstrates markedly different patternsbetween the two cultural groups. The Hmong mothersscored significantly higher (2

LINKS TO HEALTH AND DEVELOPMENT 407Table IComparison of Hmong and Caucasian Dyadson Mother-Child Rating Scales*HmongCaucasianX SD X SD1. Attentiveness 8.58 0.52 6.17 1.642. Quality of PhysicalContact 8.92 0.29 5.42 1.313. Facility in Caretaking 8.58 0.79 6.12 1.754. Speed of Responsivenessto fussing and crying 8.67 0.65 6.42 1.685. Effectiveness of mother'sresponse to crying 8.58 0.90 5.83 1.806. Variability of mother'sbehavior 8.08 0.90 4.92 1.677. General expressivenessof mother towards child 8.50 0.52 5.17 1.708. Frequency of expressionof positive regard 8.17 0.58 4.92 1.889. Frequency of expressionof negative regard 9.00 0.00 7.67 1.1610. Mother's alertness 8.08 0.79 5.25 1.3611. Appropriateness ofmother's interaction 8.50 0.80 5.33 0.7812. Delight 8.67 0.65 4.42 1.56403

408 THE HMONG IN TRANSITIONHmongCaucasianX SD X SD13. Quality of mother'sverbalizations 8.92 0.29 6.08 1.4414. Mother's patience 8.92 0.29 5.83 1.27(X = mean; SD = Standard Deviation)*Every Item shows significant group difference, 2-tailedt test, p < .001Table 2Comparison of Hmong and Caucasian Dyads on Ainsworth'sScales of Maternal-Care Behavior and the HOMEHmongCaucasianX SD X SDCooperation vs. Interference* 8.58 0.52 5.67 1.30Acceptance vs. Rejection* 9.00 0.00 6.50 1.17Accessibility vs. Ignoring* 8.83 0.39 5.50 1.68Sensitivity vs. Insensitivity* 8.58 0.67 5.17 1.27HOME t 10.00 0.00 8.40 1.73(X = mean; SD = standard Deviation)t2-tailed t test, 2

Figure 2Cooperation11Acceptance3Hippiirg-AriilJAN If:III HmongAccessiblity11q2 CaucasianSensitivityIII1111 ' 111 2 3 4 5 6 7 0 9Ainsworth's Scale forMaternal Care Behavior4 0 5

410 THE HMONG IN TRANSITIONaveraged three times as many emergency room visits aschildren of moderately sensitive to highly sensitivemothers.It should be noted that items on the mother-childinteraction measures in all probability are highly correlatedwith each other. Consequently, t-tests on individualitems tend to inflate the magnitude of group differencesfound. Nevertheless, the data comparing Hmong andCaucasian mothers' parenting behaviors are so strikingoverall that they clearly appear to be reliable and replicable.The decision to apply t-tests to individualitems of the Mother-Child Rating Scales and Ainsworth'sScales of Maternal-Care Behavior did present the possibilityof detecting items (or clusters of items) thatmight fail to differentiate the two groups of mothers, orthat might favor Caucasian mothers. In fact, however,there appear to be no areas in which Caucasian mothers asa group scored better than Hmong mothers.DISCUSSIONThe Hmong sample in this study obtained higher ratingson parenting skills on each of the four Ainsworthscales and on the HOME assessment instrument, indicatingthat these mothers were rated as more attentive, expressive,sensitive, responsive and patient in child-rearingthan Caucasian mothers. Hmong mothers also were more effectivein protecting their children from harm, evidencedby fewer emergency room visits for accidents and ingestions,and they reported a more nutritious diet for theirtoddlers than did Caucasian mothers, despite similar incomes.In contrast, the Caucasian children were morelikely to be adequately immunized than the Hmong children.This difference strikes at a major dissimilarityin the perception of health and disease. The AmericanCaucasian families have come to accept the preventiveapproach to health as routine, wh-le Hmong families generallyhave had no experience with this particular medicalapproach. The conflict of having the child receivean injection when he or she appears well, coupled withreports of significant nerve injury secondary to injectionsreceived in Laotian villages, makes some Hmong parentsreticent to accept immunization as an important as-4 0 6 .

LINKS TO HEALTH AND DEVELOPMENT 411pect of health care.Although the ratio of adults to children in the twopopulations was approximately the same (with Hmong householdshaving larger numbers of both), Hmong families appearedto support the mother in her responsibilities to agreater extent than Caucasian families. There were significantlyfewer disruptions from separation and divorcein the Hmong population and more sharing of child careamong the adults in the familiesparticularly by thegrandmothers present in the home.Anecdotally, during clinic visits with the Hmong,the observer was particularly impressed with the effectivenessof the infant's use of the mother as a securebase for exploration, the fluidity of interaction, subtleadjustments made by the mother specifically to accommodatethe child, and obvious parental pride in the child'ssmallest accomplishments. While some Caucasian dyadsdemonstrated similar patterns, the group differed in thata number of parents responded to the child in rigid,stereotyped ways, were sometimes unaware of the child'swhereabouts in the exam room, demanded performance, spokein irritated tones and showed clear disappointment in thetoddler's failure to meet parental expectations.Despite the high degree of reliability Oberg attainedin the administration of all instruments, a singleobserver conducting all assessments and making allmother-child ratings is an important potential source ofbias. Three kinds of findings from this and other reportssupport the validity of the current results.First, a high degree of consistency is apparent acrossother measures, including such indicators cf maternalcare as nutrition, accidents and emergency room visitsfor toddlers. Differences in these factors were observedbetween cultural groups and, in the last case, betweenhigh- and low-sensitivity Caucasian mothers. Second, thepatterns of associations between variables which distinguishcultural groups in this study have been noted inother research. For example, strong social networks seemto support positive mother-baby interaction patterns(Crockenburg and Smith 1982), and family stress, disruption,and tense or distant interaction patterns have beenlinked to childhood ingestions and emergency room visits(Sibert 1975; Sobel and Margolis, 1965). Third, althoughr.ports of Hmong mother-infant relations are sparse, they407

412 THE HMONG IN TRANSITIONare quite consistent with the present findings.Bernatzik (1963) notes that, in the traditional Hmongcommunity, the child is regarded as "the most treasuredpossession a person can have" (p. 77) and describes thebaby's life as rich in physical contact and social interactionwith the active participation of mothers, grandmothersand older siblings. Newlin-Haus (1982) comparedten-to-twelve-month-old infants of immigrated Hmong inPortland with those of white middle class American familiesand found that Hmong mothers held and touched theirinfants more frequently and were more extensively comfortingand consoling, while American Caucasian mothers,although attentive, playful and vocal, were often inclinedto interrupt the baby's activities and lived inobject-filled rather than people-filled environments.The quality of the mother-child relationshipthroughout infancy has clear implications for the subsequentdevelopment of the child. Ainsworth, Blehar, Watersand Wall (1978) demonstrated that early patterns of careare significantly related to subsequent patterns of attachment.The child raised in a caring, consistent andsensitive manner is likely to develop a secure attachmentwith the care-giver and to utilize the care-giver as abase from which to investigate the environment. Thechild cared for in a fragmented and insensitive manner ismore likely to develop an attachment pattern characterizedas anxious and insecure. Later, the two-year-oldchild previously assessed as securely attached is morelikely than the insecurely attached child to show greatercooperation, persistence at task, and enthusiasm duringproblem-solving (Matas, Arend and Sroufe 1978). In addition,Sroufe (1983) found that children who were securelyas opposed to insecurely attached as toddlersshowed greater ego resiliency at four years of age.Isolation, stress and lack of social support in economicallydisadvantaged groups have been implicated asfactors in negative parent-child interaction patterns(Vaughn, Egeland and Sroufe 1979; Yarrow, Rubenstein andPedersen 1975; Zegiob and Forehand 1975). Within a lowsocioeconomic group, mothers of maltreated children aremore likely to have shown insensitivity to the child'sneeds in the infancy period than those of children whosubsequently are not maltreated. Additionally, anxiousand/or avoidant rather than secure relationship patterns4 od

LINKS TO HEALTH AND DEVELOPMENT 413are more common in the early toddler period for the previouslyinsensitive group (Egeland and Sroufe 1981).In the present study, the children in both of thesepopulations appear to be healthy and within the normalrange for their ages developmentally. Nevertheless, themarked differences between the two groups in the adequacyof parenting place the Caucasian children at greater riskfor subsequent adaptational problems if there is not improvementin the quality of mothering many are receiving,in the general state of the parent-child relations, andin the circumstances which perpetuate these difficulties.Hmong mothers, on the other hand, demonstrate that theability of a bnother to be sensitive and responsive to hertoddler is not entirely dependent upon the absence ofstress or on socioeconomic status, as is commonly construed.The factors related to their particular strengthsin providing uniformly responsive caregiving deserve furtherstudy. The sensitivity shown by these Hmong mothersseems related to a cultural priority for attentive childrearingand the availability and use of a strong extendedfamily network. The payoffs are not necessarily seen ingrowth parameters or in general developmental testing,but possibly in fewer home accidents and in positiveparent-child relationships which bode well for thechild's future. The data invite our attention to the importanceof the maternal-child relationship and how it isinfluenced by cultural factors. Such understanding andawareness is essential when providing pediatric care andcounseling.REFERENCESAinsworth, M. 1976. System for Rating Maternal-care Behavior(Test #008053). Princeton, New Jersey: EducationalTesting Service.Ainsworth, M., M. Blehar, E. Waters, and S. Wall. 1978.Patterns of Attachment. Hillside, New Jersey:Erlbaum.Bernatzik, H.A. 1963. "Akha and Miao: Problems of AppliedEthnography in Farther India." Translated byA. Nagler. Human Relations Area Files.409

414 THE HMONG IN TRANSITIONCaldwell, B.M., and R.H. Bradley. 1978. Home Observationfor Measurement of the Environment. LittleRock: University of Arkansas.Cr'ckenberg, S.B., and P. Smith. 1982. "Antecedents ofMother-infant Interaction and Infant Irritability inthe First Three Months of Life." Infant Behavior andDevelopment 5: 105-119.Egeland, B., A. Deinard, D. Brunnquell, S. Phipps-Yonas,and L. Crichton. 1979. "Final report: A prospectivestudy of the antecedents of child abuse." Unpublishedmanuscript. (Available from Byron Egeland,University of Minnesota, Minneapolis).Egeland, B., A. Deinard, D. Brunnquell, and B. Taraldson.1975. "Three-month Observations of Feeding Situations,Six-month Observations of Feeding and PlaySituations, waiting room observation scales." Unpublishedmanuscripts. (Available from Byron Egeland,University of Minnesota, Minneapolis).Egeland, B., and L.A. Sroufe. 1981. "Attachment and EarlyMaltreatment." Child Development 46:71-76.Elarao, R., R. Bradley, and B. Caldwell. 1975. "The Relationshipof Infants' Home Environment to MentalTest Performance from Six- to Thirty-six Months: ALongitudinal Analysis." Child Development 46:71-76.Frankenburg, W.K., J.B. Dodds, and A.W. Fandal. 1973.Denver Developmental Screening Test Manual/Workbookfor Nursing and Paramedical Personnel. Boulder:University of Colorado Medical Center.Frankenburg, W.K., A.D. Goldstein, and B.W. Camp. 1971."Me Revised Denver Developmental Screening Test:Its Accuracy as a Screening Instrument." The Journalof Pediatrics 79:988-995.Le Bar, F.M., and A. Suddard. 1960. Laos: Its People, ItsSociety, Its Culture. New Haven, Connecticut: NewHaven Human Relations Area Files Press.4i u

LINKS TO HEALTH AND DEVELOPMENT 415Matas, L., R.A. Arend, and L.A. Sroufe. 1973. "Continuityof Adaptation in the Second Year: The RelationshipBetween Quality of Attachment and Later Competence."Child Development 49:547-556.McKay, R.J. 1975. "General Considerations in the Care ofSick Children: Clinical Evaluation of Infants andChildren." Nelson Textbook of Pediatrics (10th ed.),edited by V.C. Vaughn, R.J. McKay and W.E. NelsonPhiladelphia: W.B. Saunders.Muret-Wagstaff, S., and S.G. Moore. 1983. "Hmong vs.American Caucasian Newborn Behavior and RearingPractices." Paper presented at the biennial meetingof the Society for Research in Child Development,Detroit, April 1983.Newlin-Haus, E.M. 1983. "A Comparison or Proxemic andSelected Communication Behavior of Anglo-Americanand Hmong Refugee Mother-Infant Pairs." Doctoraldissertation, Indiana University, 1982. DissertatationAbstracts International 43(8):2724B-2725B.(University Microfilms No. DA8300862.)Sibert, J.R. 1975.Have Ingested641-651.Sobel, R., and J.A.in Children: A641-651."Stress in Families of Children WhoPoisons." British Medical Journal 3:Margolis. 1965. "Repetitive PoisoningPsychosocial Study." Pediatrics 35:Sroufe, L.A. 1983. "Infant-Caregiver Attachment andPatterns of Attachment in Preschool: The Roots ofMaladaptation and Competence." In Development andPolicy Concerning Children With Special Needs. ThtMinnesota Symposium on Child Psychology (Vol. TO;edited by M. Perlmutter. Hillsdale, New Jersey:Erlbaum.Vaughn, B.E., B. Egeland, and L.A. Sroufe. 1979. "IndividualDifferences in Infant-Mother Attachment at 12and 18 Months: Stability and Change in Families UnderStress." Child Development 50:971-975.411

416 THE HMONG IN TRANSITIONVaughn, B.E., B. Taraldson, L. Crichton, and B. Egeland.1980. "Relationships Between Neonatal BehavioralOrganization and Infant Behavior During the FirstYear of Life." Infant Behavior and Development 3:47-66.Whitt, J.K., and P.H. Casey. 1982. "The Mother-Infant Relationshipand Infant Development: The Effect of PediatricIntervention." Child Development 53:948-956.Yarrow, L., J. Rubenstein, and F. Pedersen. 1975. Infantand Environment: Early Cognitive and MotivationalDevelopment." New York: Wiley.Zegiob, L., and R. Forehand. 1975. "Maternal InteractiveBehavior as a Function of Race, Socioeconomic Status,and Sex of the Child." Child Development 46:564-568.

UNDUE LEAD ABSORPTION IN HMONG CHILDRENKarl Chun and Amos S. DeinardINTRODUCTIONIn the spring of 1981, the Bureau of Maternal andChild Health at the Minneapolis Health Department, alongwith other institutions in Minneapolis and St. Paul, becameaware of an extraordinary number of Hmong childrenidentified with undue lead absorption. In retrospect,the screening data for lead were present as far back asthe fall of 1980, but had been lost in the mass of otherrefugee health problems at the time. As time becameavailable to Investigate this "epidemic" of lead poisoning,however, the incidence melted away so that it is nogreater today than the incidence of undue lead absorptionnormally found in Minneapolis. The finding of lead in afolk remedy during the summer of 1983 in St. Paul offeredsome insight into the seemingly unexplainable high leadlevels found in some very young children, but leaves unansweredthe question of why Hmong children in generalhave had such high lead levels.This report describes the experience of the MinneapolisHealth Department, one of the major health care providersfor the Hmong in Minneapolis, and does not includethe experience in St. Paul. lo put the problem in perspective,the Minneapolis Bureau of Housing Inspectionsis the sole agency that receives reports of undue leadabsorption from the entire city. In 1980 they receivedthe usual number of case reports, twenty to thirty. In1981 the Hmong alone accounted for twenty-four cases,The number of case reports has de-over half the total.dined since that year.BACKGROUNDLead poisoning, which has been recognized at leastsince the Roman Empire, is a clinical entity with numerousmanifestations. Lead colic in children is generallyrecognized by irritability, lessening of play activity,417413

418 THE HMONG IN TRANSITIONheadaches, poor appetite, abdominal pain, vomiting andconstipation. In its most severe form, lead encephalopathyis characterized by vomiting, stumbling, fallinglevel of consciousness, coma and seizures, with subsequentmental retardation, blindness and paralysis.Fortunately, these frank features of lead poisoninghave become less common since efforts to reduce lead inthe environment first began in this country in the 1950s.In Minneapolis, clinical lead poisoning is rarely encountered.The subject of this paper is an entity known asundue lead absorption which refers to excessive absorptionand retention of lead without overt clinical symptoms.At these levels children are primarily at risk fordeveloping behaviors that result in impaired learningabilities: more distractable, not persistent at tasks,dependent, impulsive, easily frustrated, tendency to daydream,inability to follow simple directions or sequencesof directions, and lower overall function (Needleman etal. 1979). In population studies IQ scores proved to bemeasurably lower, in the range of four to ten points, dependingon the level of lead exposure (Rutter 1980). Unfortunately,these findings have been demonstrated inchildren whose blood lead levels have been well belowthe thirty mg/di (micrograms per deciliter) level currentlydefined by the Centers for Disease Control inAtlanta, Georgia, as representing undue lead absorption.Biochemical evidence of lead toxicity (including some ofthe lead screening tests) has also been shown to occur inthe range of ten to twenty mcg/dl.Exposure to lead is a by-product of industrial society.Industries that use lead represent an occupationalsource. In the urban environment leaded paint once accountedfor a large proportion of identifiable sources;however, since the ban on leaded indoor paint in 1957,paint has become :.ss of a problem. Leaded gasoline probablyproduces much of today's urban lead, which is a-vailable to children to children in high concentrationsin dust and especially in inner city neighborhoods.Children are at a greater risk of lead's toxic effr'ctsfor a number of reasons: (1) Children absorb agreater proportion of a given load of ingested lead thanadults. Adults absorb approximately five to ten percentof ingested lead, whereas children absorb in the range ofthirty to fifty percent (Ziegler et al. 1978). (2) The414

UNDUE LEAD ABSORPTION IN HMONG CHILDREN 419brain is rapidly growing and acquiring skills at an earlierage, so that the greater amount of lead that crossesinto the brain has a devastating effect on development(Rutter 1980). (3) diets deficient in protein or calciumenhance the absorption of lead. Because of the rapidgrowth in a young child, nutritional deficiencies arequite common.METHODSThe data to follow come from two sources: the SoutheastAsians who have been through the screening programfor iron deficiency and undue lead absorption at the MinneapolisHealth Department (hereafter referred to as theHealth Department) and a few additional cases detectedthrough Hennepin County Medical Center. Comparisons willbe made with two earlier reports on the same screeningprogram at the Health Department from 1978 to 1980 (Yip,Schwartz and Deinard 1983) and a similar program at MinneapolisChildren's Health Center, a private hospital andclinic, from 1979 to 1980 (Yip, Norris and Anderson1981).The Health Department clinics routinely screen allchildren for lead with a screening test called the erythrocyteprotoporphyrin (EP) test. Children from one tosix years of age are tested yearly. Those children whoseEP exceeds fifty mcg /dl are further tested to determinetheir lead level. In accordance with the Centers forDisease Control's guidelines, an elevated blood lead levelis defined as thirty or more mcg/dl. Children withelevated blood leads are classified into risk groups bythe degree of elevation of the EP and blood lead level.Risk class II is low risk, class III moderate risk andclass IV high risk. All children with blood lead levelsof thirty or more are reported to the Minneapolis HousingInspections Division, which inspects the housing forsources (., lead and makes and enforces recommendationsfor removing identifiable sources. Children in class IIare followed without medical intervention other than removalof the source, and education on av,iding lead.Those in risk class III are generally further analyzed bya lead mobilization test to quantitate the total bodylead burden and in this way to determine the necessity of415

420 THE HMONG IN TRANSITIONtherapy. Those in class IV are hospitalized for a threetofive-day course of chelation therapy to remove leadfrom the body. Follow-up continues until lead levelsfall into and remain in a normal range.RESULTSOf 870 screenings of Hmong children up to six yearsof age at the Health Department from 1980 to March 1983,the incidence of undue lead absorption was 4.3 percent.If only the highest at-risk age group is considered(children up to thirty-six months of age), the incidenceof undue lead absorption increases to 5.5 percent. Incomparison to a survey of the same clinic's populationbefore the arrival of the Southest Asians, the incidencewas 1.3 percent, demonstrating a three-fold increase inthe Hmong (2-tail t-test: p < .0001) (Yip, Schwartz andDeinard 1983). The Minneapolis Children's Health Center,from 1979 to 1980, reported an incidence of 2 percent,a two-fold increase in the Hmong (2-tail t-test:p

UNDUE LEAD ABSORPTION IN HMONG CHILDREN 421Table 1YEARLY INCIDENCE OF UNDUE LEAD ABSORPTION INHMONG CHILDREN, RECORDED BY DATE OF FIRSTELEVATED ERYTHROCYTE PROTOPORPHYRIN.1980 1981 1982 1983 TotalChildren under 6 yearsCases 5 20 9 2 36Screenings 127 258 366 95 846Proportion .039 .078 .025 .021 .043Children under 3 yearsCases 5 18 8 2 33Screenings 84 179 259 76 598Proportion .060 .101 .031 .026 .055Of the forty-six Southeast Asian children identifiedwith undue lead absorption by the Health Department andHennepin County Medical Center through August 1983,forty-two were Hmong and four were Laotian. The mean agewas twenty-four months at the time of discovery; however,a more accurate reflection of the age groups was the medianage of nineteen months. All but four children wereless than four years old, and the older children wereidentified by blood-testing the entire family for leadafter the discovery of a younger sibling index case. Byadjusting this group to include the Minneapolis Children'sHealth Center, the mean age was twenty months,eleven months less than the mean of thirty-one months atthe Children's Hospital (Yip 1981). Fifty-nine percentwere male and 41 percent were female, in keeping with thegeneral notion that males are affected somewhat more oftenthan females. The majority (65 percent) fell intothe lowest risk classification, which is no differentfrom that observed in the Minneapolis Children's HealthCenter group (Yip, Norris and Anderson 1981).The medical work-up of children with undue lead absorptionusually includes studies of radiographs of theabdomen to determine recent ingestion of paint chips, and417

422 THE HMONG IN TRANSITIONradiographs of the knees, where deposition of lead in thegrowing portion of bones will indicate long-term exposure.No child has had radiographic evidence either ofrecent paint chip ingestion or of long-term exposure. Afurther measure of total body lead burden is obtained bya mobilization test. The child is given an injection ofEDTA (ethylenediaminetetraacetic acid), which then formsa soluble complex with lead in the soft tissues of thebody. This soluble complex is carried to the kidneys andexcreted in the urine, where the excreted lead can bemeasured. None of the moderate risk children in classIII has shown a high total body lead burden by this test.The high risk children in class IV are usually hospitalizedand treated immediately, so that a mobilizationtest is not routinely performed.Housing inspections are routinely conducted to identifyleaded paint on interior surfaces. For ten instancesin which a completed inspection report is available,eight houses had interior leaded paint and two didnot. However, most of the housing units with leadedpaint had no easily ingestible peeling paint; thus acausal link between paint and body lead burden is difficultto establish. Anecdotally, one apartment complexhas produced six of the cases. The only apartment withlead in the interior had no children in it; the otherapartments were all free of lead. Mielke (in press; hasestablished that the Hmong live in areas of high soillead, as do all inner city residents, which may serve asanother source of ingestible lead, but this does notexplain why Hmong children in particular absorb lead andother residents in the same area do not.SUMMARYThe basic descriptive information on Hmong childrenwith undue lead absorption shows several features thatdistinguish them from other America; children with unduelead absorption.They tend to be of a lesser age, withthree only six months of age at the time of detection. Anoticeable trend in the incidence of undue lead absorptionhas occurred between 1980 and 1983, with the peakincidence in 1981 and an incidence more akin to that ofthe general American population in 1983.418

UNDUE LEAD ABSORPTION IN HMONG CHILDREN 423One of the more intriguing aspects is the failure tofind a source of lead in the homes. Certainly leadedindoor paint can be found, but most of the paint has beenconsidered a poor source since it was not peeling orpeelable. The etiology of undue lead absorption in Hmongchildren is far from clear, both for the general populationand in individual cases.DISCUSSIONThe usual investigation of children with undue leadabsorption has failed to uncover the common sources:peeling leaded paint, plaster, lead in jewelry or otherhousehold items, lead in pottery, industrial exposure orpica (the excessive mouthing of objects). We have routinelyasked about the use of herbal medicines, knowingthat Chinese herbal medicines and Mexican folk medicineshave been identified as a cause of lead poisoning.Indeed, in St. Paul doctors identified one six-month-oldchild who had been given a lead-contaminated folk medicine,my-loo-ah, an earthen mixture whose source is notclear to us. We have been unsuccessful in obtaining ahistory of pay-loo-ah use in all subsequent new cases.Radiographic studies have failed to implicate paintchips as a possible source. Mobilization tests haveshown low total body lead burdens, and radiographic evidenceof long-term exposure by lead lines in the longbones is lacking. One would not expect long-term exposurebecause of the intermittent screening for lead inchildren since their arrival in the United States and thelow likelihood of lead exposure in the Thai camps.Although the refugees in the Thai camps have never beentested for lead, the EP screening tests done on arrivalin Minneapolis do not indicate that lead exposure couldhave occurred before arrival in this country.Do the Hmong absorb, retain and excrete lead in adifferent manner from others? Age and diet determine theamount of lead absorbed through the gut. Younger childrenabsorb more of a given load of leadapproximately30 to 50 percent in contrast to 5 to 10 percent byadults. Diets deficient in calcium or protein enhancelead absorption. Conceivably, the high incidence in 1981could have been from a poorer nutritional status in415

424 THE HMONG IN TRANSITIONregard to calcium and protein, and perhaps an improvedintake of calcium and protein over time has correctedthese deficiencies.Further investigation of hypotheses on the uniquenessof the Hmong experience with lead has become lesspossible in Minneapolis, as the "epidemic" proportionsreached in 1981 have disappeared. St. Paul, with alarger Hmong population, does not report a diminishingnumber of cases. The Health Department's approach to allchildren with undue lead absorption concentrates onattention to the usual environmental sources, mouthingactivities, nutrition and childhood development. Additionally,we can offer to test for the possibility oflead contamination in household folk medicines, a phenomenonalso seen in some Chinese and Mexican folk medicines.Because undue lead absorption is a silent affliction,efforts to communicate its harmful effects on achild's development have been difficult, especially sincelead is one of many newly encountered problems inthe United States. Until society develops a policydirected at further efforts to eliminate lead and itssources from the urban environment, public education onlead and identification of cases must suffice to avoidlead's deleterious effects.ACKNOWLEDGEMENTSThe authors wish to acknowledge Ms. Brenda Grabensteinfor the analysis of the blood samples and Ms.Virginia Miller for tabulating the data. The computerizedstorage and analysis of the data was possiblethrough a grant from the University of Minnesota ComputerCente7. Mr. Gerald Takumi and Dr. Lydia Caros providedthe data on housing inspections.REFERENCESMielke, Howard. In press. On soil lead levels inMinneapolis-St. Paul and with specific reference toHmong neighborhoods.

UNDUE LEAD ABSORPTION IN HMONG CHILDREN 425Needleman, H.L., C.E. Gunnoe, A. Levition, R. Reed,H. Peresie, C. Maher, and P. Barrett. 1979. "Deficitsin Psycho logic and Classroom Performance ofChildren with Elevated Dentine Lead Levels." NewEngland Journal of Medicine 300: 689-695.Rutter, Michael. 1980. "Raised Lead Levels and ImpairedCognitive/Behavioural Functioning: A Review of theEvidence." Supplement to Developmental Medicine andChild Neurology 22:1-26.Yip, R., T.N. Norris, and A.S. Anderson. 1981. "IronStatus of Children with Elevated Blood Lead Concentrations."Journal of Pediatrics 98:922-925.Yip, Ray, Samuel Schwartz, and Amos S. Deinard. 1983."Screening for Iron Deficiency with the ErythrocyteProtoporphyrin Test." Pediatrics 72:214-219.Ziegler, E.E., B.B. Edwards, R.L. Jensen, K.R. Mahaffey,and S.J. Fomon. 1978. "Absorption and retention oflead by infants." Pediatric Research 12:29-34.42i

ATTITUDES OF HMONG TOWARDA MEDICAL RESEARCH PROJECTMarshall Hurlich, Neal R. Holtanand Ronald G. MungerINTRODUCTIONMigrant populations are always of interest to epidemiologists(Kasl and Berkman 1981) and biological anthropologists(Baker 1981). Besides carrying certain infectiousdiseases from place to place, migrating groupsprovide observers with large-scale "natural experiments"by removing themselves from one set of environmental factorsand substituting a new set. Migrant studies haveaided in the understanding of health conditions such ascancer, in which diet and environmental agents arethought to play a role (Haenszel and Kurihara 1968; Modan1986; Knudson 1977).However, the response which the migrant populationitself has to a biomedical research effort usually goesunreported. This report deals with the responses which arefugee migrant population from Laos has had to the variousattempts made to investigate an unexplained, suddenunexpected nocturnal death (SUND) syndrome reported sinceits arrival in the United States (United States Centersfor Diesease Control 1981; Munger 1982; Baron et al.1983). Attention is given to how these studies have beenreceived by a specific refugee community in the UnitedStates and in a refugee camp in Thailand, and how thestudies have created problems preventing the successfulcompletion of physiological research in a refugee camp inThailand. It is not our intention to review biologicaland medical results of the research efforts made; rather,we wish to focus on certain research and data-gatheringprocedures and on responses of the refugee community.BACKGROUNDOver one hundred thousand Hmong people have moved aspolitical refugees from their native country of Laos toThailand, and then to the United States and other coun-427422

428 THE HMONG IN TRANSITIONtries for permanent resettlement (Office of Refugee Resettlement1934). The complex health system in theUnited States has allowed close scrutiny of the Hmong'shealth status, leading to the first reports of SUND syndromeamong them (CDC 1981). One of us, Ronald Munger,spent nine months in the Ban Vinai Refugee Camp in LoeiProvince of northeastern Thailand in 1982 and 1983 collectingdata on sudden deaths there (Ronald G. Munger,this volume, p. 379).Meanwhile, in Minnesota, several physicians at St.Paul-Ramsey Medical Center were noticing occasional Hmongpatients with unexplained abnormalities in their serumelectrolytes, including low levels of potassium and magnesium(Holtan et al. 1983). Considering the role thatserum electrolytes play in the regulation of the heartbeat,they began to speculate on the possibility, thusfar unproven, that these abnormalities were related tothe SUND syndrome.In December 1982, a party of five health workersdealing full-time with refugee health at St. Paul-Ramseytook a pleasure and fact-finding trip to Thailand. Twoof them, a nurse and a doctor, had been medical volunteersat Ban Vinai in 1980, and a third was a Hmong-American whose family still resided in the camp. Aftermeeting by chance at Ban Vinai, the St. Paul team andMunger shared information about the SUND syndrome and theelectrolyte abnormalities which had been observed in theHmong patients in the United States. It was thought thatblood and urine samples collected at Ban Vinai could besent to St. Paul for analysis in order to determinewhether or not there were electrolyte abnormalities inany part of the Hmong population in Ban Vinai.Shurtly thereafter, several researchers at St. Paul-Ramsey Medical Center began communicating with other researchersat the University of Washington about hypotheses,study design, logistics and cooperative arrangements.A comprehensive proposal to study Hmong in BanVinai and in several locations in the United States waswritten and submitted for funding consideration. Twosmall research proposals, pilot studies for the largerproposal, were approved for funding by the St. Paul-Ramsey Medical Education and Research Foundation.One of the proposals, entitled "Electrolyte and hormonalregulation in clinically healthy Hmong subjects,"423

424HMONG ATTITUDES TOWARD A MEDICAL RESEARCH PROJECT 429entails the investigation of fifty apparently healthyHmong, twenty-five of each sex. Both sexes a.,: includedin order to determine base-line data for a wide segmentof the Hmong population whit:I': is at apparent risk forSUND (CDC 1981; Baron et al. 1983). Each of the subject .would have blood tests and urine tests over a twer,four-hour span of time. Subjects ,vould also have amal diet history, a brief medical history, a short ncal examination, and a genealogical listing of directancestors. In addition, they would wear a Holter heartmonitor for a day to look for abnormal cardiac rhythms.Criteria for subject selection included apparent goodhealth, absence of thyroid disease, age between twentyfiveand forty-four years, and absence of vomiting ordiarrhea. This study is now in progress in St. Paul(January 1984), but is not sufficiently advanced to judgeits eventual success.In the same proposal, funding was also approved toanalyze blood and urine samples from a comparable grouprf. Hmong subjects in Ban Vinai. For the St. Paul pilotstudy, the fifty subjects are adequate to define the normalcircadian pattern of hormone and electrolyte levelsfor the Hmong population. The St. Paul subjects couldtherefore serve as the standard against which to comparethe Ban Vidal subjects. Therefore, the subjects in Thailandneeded to be sampled only twice each, once at theeight A.M. cizcadian peak of serum electrolytes and onceat the midnight circadian trough. Subject selection criteriawere: (a) residence in the camp of at least twoweeks, (b) male, and (c) the same medical criteria appliedto the subjects in St. Paul. For this Hmong population,we decided to test only ..iales, because most SUNDcases among refugees in the United States are males, andbecause we didn't want to complicate what we anticipatedwould already be a delicate investigation. The Ban Vinaisubjects were also to have medical histories, physicalexaminations, diet histories and genealogical interviewstaken. Holter monitors were not proposed for use in BanVinai. Research funds for the project in Thailand wereprovided by the St. Paul-Ramsey Medical Center.Tht, coordinator of the research project in theUnited States. Dr. Neal Holtan, and Munger, the coordinatorof the work in Ban Vinai, spent time with manypersons and organizations to obtain cooperation for com-

430 THE HMONG IN TRANSITIONpletion of the Thailand phase of the project. In spite ofall efforts, the researchers did not succeed in obtainingthe blood and urine samples in Ban Vinai. They did, however,gain experience in conducting research in Thailandand in interacting with the Hmong people to achieve agoal. We believe it is useful to share in detail theproblems they encountered so others can avoid them in thefuture, as well as learn from them.First lorts to the Centers for Disease ControlLarry Lewman, chief medical examiner in Portland,Oregon, was the first to report cases of sudden unexpectednocturnal deaths among Southeast Asian refugeesto the Centers for Disease Control (CDC) in Atlanta,Georgia, in February, 1981. i3etween February and December,1981, the CDC had been notified of thirty-eight SUNDcases throughout the United States. Of these, twenty-livewere Hmong, eight were among other Laotian ethnic groups,four were Vietnamese and one was Kampuchean. The deathsreported had occurred between 15 July 1977 and 28 October1981 (CDC 1981). In response to these deaths, the CDCinitially planned to establish a SUND surveillance network,to review the various causes of deaths among Indochineserefugee populations in the United States between1975 and 1980, to investigate individual SUND cases, toconduct genealogical investigations of cases, to reviewrefugee electrocardiographic status and to investigatethe question of prior exposure to toxic chemicals inSoutheast Asia (Vernon and Baron 1981). It was assumedat the time that ethemical warfare had been conductedagainst Hmong people in Laos, and that such exposurecould constitute a risk factor for the SUND syndrome.To achieve these goals, an investigator from theCDC, using a case-control study design, interviewed membersfrom the families of twenty-six Hmong and ethnic LaoSUND victims and interviewed three groups of matched controisusing the same questionnaire (Schlesseiman 1982).not questions from the prepared form covered birthplace,migratory history, occupational history, military history,the reported exposure to toxic chemical weapons,foods eaten, marital and reproductive history, healthstatus, drug use and other topics. No significant differenceswere reported between cases and controls (Baron425

HMONG ATTITUDES TOWARD A MEDICAL RESEARCH PROJECT 431et al. 1983).Many Hmong leaders were not initially aware of theCDC investigation. Unfortunately, some individuals inthe Hmong community said that they believed that the CDCstudy would determine the biological cause of the SUNDsyndrome. Some said they believed that CDC investigatorsthought the SUND cases were related to episodes ofpresumed chemical warfare in Laos. The lack of findingsdid nothing to clarify this point for the Hmong or to putsuspicions to rest.Cultural Studies of Sudden DeathThe first completed investigation into the culturalaspects of the SUND syndrome among the timong was conductedby Bliatout (1982). He initially presented hisstudy to the Hmong community as a research effort whichmight reduce the psychological fear associated with thesudden death syndrome, and also as an investigation offive hypotheses which had been proposed as potentiallyexplanatory. Bliatout, himself a Hmong, emphasized inhis interviews that his study could not guarantee an answer,but was instead part of a long-term effort neededto uncover the causes of sudden deaths among Hmong refugees.'Bliatout interviewed surviving family members offorty-five Hmong SUND victims who died between 1973 and1982. His sample, however, does not exactly overlap thatobtained by the CDC. Of his forty-five cases, thirtyeightoccurred in the United States, three in Laos, twoin Thailand, and one each in Canada and France. Twentysixoccurred between 1977 and 1981 (compared to twentyfiveHmong cases reported to the CDC during approximatelythe same period), and thirty-seven occurred since 1978.A second investigation of the cultural associationsof the SUND syndrome was conducted by Lemoine, who interviewedrelatives of SUND victims in the United States andFrance, gaining access to the Hmong, Mien and Lao communitiesthrough the network of his former students fromLaos (Lemoine and Mougne 1983).Neither Bliatout nor Lemoine limited their interviewsto prepared questionnaires, and, as both are intimatelyfamiliar with Hmong culture, they were able todesign their studies to be sensitive to the culturalattitudes of the individuals they wished to interview.426

432 THE HMONG IN TRANSITIONReactions to Studies DescribedDuring and since these studies, we have interviewedHmong in a variety of leadership positions. During thecourse of these interviews, it became apparent that manyHmong felt that the CDC was withholding information fromthem.1 They asked the following questions: Several investigationsof the ,FUND cases had been made. Why wasn'tthe answer available? Did CDC "know" the cause, but refuseto reveal its knowledge? Was it because the primary"cause" was in fact chemical warfare, and the CDC did notwish to contribute to the developing debate about the useof toxic chemicals in Southeast Asia? Was it because CDCdid not wish to provide fuel for arguments made by someindividuals that Hmong should be eligible for U.S. warveterans' benefits? Was it because the CDC felt thatHmong were not sophisticated enough to understand thecause of the sudden deaths? Or was it because the CDCdid not care about refugees?Additional Biological StudiesA fourth get of studies of the ''D cases were undertakenby pathologists, quite removeo from the view ofthe refugee community. Post-mortem studies were performedon tissues from the victims to determine if there weretraces of drugs or poisons, or if the victims had anatomicalabnormalities which were the cause of death (Eckner,Kirschner and Baron 1983).The law in every state provides that in cases of unexpectedand unexplained deaths, autopsies must be performed.Consequently, tissue samplesincluding in manycases the entire heartwere removed from a number ofSUND victims under the direction of local medical examiners.Studies of cardiac pathology are in progress, focusingon the conduction system and the aortic root.Although some anomalous structures have ben suggested,the physiological significance of the findings is notclear (Eckner and Kirschner 1983).In most cases where autopsies were performed on SUNDvictims, the Hmong victims' families did not understandthe procedures and results. The families' permission wasnot routinely requested; the autopsy protocols were notuniformly described; the need for tissue samples was not42'1

HMONG ATTITUDES TOWARD A MEDICAL RESEARCH PROJECT 433always discussed. Some Hmong families, upon receivingthe corpse c.,f their dead family member, noticed that thebody had been cut open. The relatives then proceeded toopen the body and found, to their horror, that some organshad been removed and that sometimes the organs hadnot been properly repositioned in the body.EFFECTS OF PREVIOUS RESEARCH EXPERIENCEON CURRENT PROJECTSThe background of sudden-death studies is importantbecause it very likely conditioned the responses of theHmong in Ban Vinai during the recent attempt to obtainblood and urine samples.Let us start with the fact-finding trip made by fivehealth workers from St. Paul to refugee camps in Thailandin November and December, 1982. During their stay in BanVinai, Mao Thao, a Hmong-Americar woman living in St.Paul, and the first Hmong-American to visit Ban Vinai,agreed to answer questions about life in the UnitedStates at a public meeting. Over ten thousand peoplefor:.,ed the unexpectedly large crowd. These were amongthe questions Mao Thao was asked: (St. Paul PioneerPress, 1983)."When Hmong people die in the United States, is ittrue that they are cut into pieces and put in tin cansand sold as food?""After you die, why do American doctors try to openup your head and take out your brains?""Why do some Hmong in Minnesota go to sleep and neverwake up again?""Do American doctors eat the liver, kidney, andbrain of Hmong patients?""Why do American doctors draw so much blood from patients?"These questions suggest some of the concerns, misunderstandingsand fears in the Hmong community (not only42a

434 THE HMONG IN TRANSITIONin Ban Vinai but a!:,o in the United States) concerningthe health care practice in this country. These problemsstem in part from the following factors:1. Leadership. Misunderstandings have arisen from failureof American medical researchers to approach theHmong community through the structure of its trueleadership. Part of this problem results from thefact that most Americans do not know who the "trueleaders" are in the Hmong community and why it is importantto consult with them. The Hmong communitystructure includes exogamous clans and their formermilitary leaders. The lack of communication betweenresearchers and community leaders allows for considerablespeculation on the part of Hmong leaders andcommunity members about the true intentions of medicaland social science researchers.A related problem is that, although clan membersrely on clan leaders for advice, clan leaders frequentlyare not in a position to give advice in anarea where they themselves are uncertain about thelong-range consequences of advice they might give,such as the participation of clan members in research.If clan leaders fear loss of esteem withinthe clan when their advice "fails," they might refuseto accept the responsibility of potential failure.,2. Explanations of medical procedures. Misunderstandingshave arisen from the failure to explain the justificationfor normal medical laboratory procedures. Therelationship between a person feeling ill, his goingto a medical clinic, having a sample of tissue (usuallyblood) removed, laboratory analysis of theblood, diagnosis of the finding, determination ofmedication and administrative of medication is unclearto many Hmong. The ri tionale behind this chainof events has not in general been explained to Hmongpatients in research ' projects or in health clinics ina manner which the Hmong understand. This lack ofsuccessful communication leaves much room for speculationby Hmong about medical researchers' and practitioners'motives. For example, it raises the typesof questions asked at Ban Vinai: Why do doctors takeso much blood? Why do they remove organs?42:3

HMONG ATTITUDES TOWARD A MEDICAL RESEARCH PROJECT 4353. Are researchers hiding something? Some difficultiesstem from the perception by some Hmong that, sinceseveral studies have already been done into the problemof sudden deaths, American doctors already knowwhat causes the deaths, but they are not telling. Wespeculate that at least three factors contribute tothis difficulty. First, most Hmong who are subjectsin research projects do not understand the conceptualbasis for undertaking research, nor the relationshipbetween research efforts and the treatment of medicalconditions. Consequently, when a Hmong subject isseen by a doctor during the course of a research effort,the Hmong individual may expect either to bechecked for a medical condition or to receive medicationfor it at the end of the "visit." While thisdoes not apply to all Hmong individuals, it does seemto characterize the majority of Hmong, who only sincetheir arrival in the United States are beginning toreceive Western education and literacy skills (Officeof Refugee Resettlement 1984).Second, problems result when researchers, in effortsto "sell" research effort to the subject community,promise greater returns for the research thancan in fact be expected. A researcher may be temptedto tell the Hmong interpreter that this research"will" contribute to an effort that "might" determinethe causes of sudden death which can then "help lead"to a cure or at least preventive action. The Hmonginterpreter most likely does not or even cannottranslate the subtleties of the English words "may"and "might" to the Hmong subject, who is then leftwith expectations that participation will lead to acure. Thus, difficulties arise when Hmong subjects,perceiving American medicine as powerful, believethat the answer can be quickly learned.4. Methods. A fourth set of problems derives from themethods of investigation used. The unexplained removalof organs at the autopsy, if not adequately explained,generates a response which is based on Hmongbeliefs about the relationship between organs of thebody and the body's souls. For example, some Hmongsay the body has seven souls, one of which resides inthe heart. As Chindarsi notes:4 30

436 THE HMONG IN TRANSITIONWhen a man dies the souls of the ears, nose andmouth remain in the grave, but the souls of theeyes and heart go to the afterworld, from whichthey will return in due course to be reincarnated.Some Hmong say that after the body rotsaway, the souls of the ears, nose and mouth leavethe grave to rejoin the souls of eyes and heartin the afterworld. It is the souls of the eyesand heart which wander at night and cause theirowners to dream (The Religion of the Hmong Njua,1976).The practice of removing organs at autopsy possiblyled to the sort of questions raised publicly inBan Vinai, when some individuals asked, "Do Americandoctors eat the organs of dead Hmong people?"5. Additional problems exist which are beyond the scopeof this report. For example, the popular media havetended to portray the problem of the sudden deaths ina light which makes the deaths seem bizarre, and byimplication, the Hmong are made to seem bizarre or atleast unintelligible (Monagan 1982).This background affected Munger's attempt to secureblood and urine samples in Ban Vinai, and influenced thedesign of the long-term study proposed for St. Paul andSeattle.THE BAN VINAI EXPERIENCEMunger and Elizabeth Booton, a nurse practitioner,spent ten months in Ban Vinai Refugee Camp, Thailand,during 1982 and 1983. Between 5 February and 3 Jule,1983, they attempted to organize a collection of urineand blood specimens from a sample of Hmong in the camp.Aided by project directors in the United States (Holtanand others), this effort required the cooperation of severalThai ministries, including the National ResearchCouncil, the Ministry of the Interior, the Secretary Generalof the Thai Medical Council, and the Thai MilitaryCommander of Ban Vinai. Due to the political volatilitycreated by refugee popu...ations in Thailand, securing ap-431

HMONG ATTITUDES TOWARD A MEDICAL RESEARCH PROJECT 437proval proved difficult and time-consuming. Our experiencesuggests that negotiations are best facilitated bypersonal meetings with the proper official and that importantcorrespondence should, when possible, be handdelivered.In one case several months were spent waitingfor approval because key officials did not receive importantcorrespondence.Logistical and personnel support were requested fromseveral American agencies, such as the American RefugeeCommittee in Minneapolis and in Bangkok, and the InternationalRescue Committee in the United States and Bangkok.The Christian Medical Team and World Vision Internationalalso provided support. Assistance was also requestedfrom the office of the United Nations High Commissionerfor Refugees, Bangkok.During these four months, many letters and telephonecalls crossed between the United States and Thailand;many meetings were held; many expected and unexpectedobstacles were overcome. (Among the obstacles were theneed to figure out how to ship dry ice from Bangkok toBan Vinai, and how to prepare and preserve and ship bloodspecimens). Some individuals in the American agenciesmentioned in the preceding paragraph acted as couriersfor letters and for telephone messages, and some were onsite in Ban Vinai and offered their services during thestudy. The usefulness of their support proved that goodrelations with voluntary agencies holding influence inBan Vinai are essential.The most important meetings, however, were those betweenAmerican investigators and the Hmong leadership inthe United States and in Thailand. The relationship betweenthese parties is the single mo: important factorin carrying out this kind of research project. A greatdeal of time was spent in negotiations before the projectand during the sampling procedures. The following discussionfocuses mainly on this aspect of the project, dueto the importance of these interactions.After clearance with the Thai ministries, Munger metwith the Hmong Advisory Council (HAC) in Ban Vinai on 7May 1983, because the HAC had previously requested thatthe Thai government approve all work done. The councilconsists of clan leaders, center chiefs and Hmong elders.The purpose of the meeting was to consider the bloodstudy project, at the same time discussing sudden deaths.1j

438 THE HMONG IN TRANSITIONof limong in the United States, Thailand and Laos, and thelaborator; work completed in St. Paul on serum electrolytelevels of Hmong. The leaders said they approved ofa blood study being done in Ban Vinai as part of the investigationinto the cause of sudden death, but that theywanted to pick the sample of individuals who donatedblood for the study. The Hmong leaders suggested thatthey pick twenty-five subjects whom they classified as"rich" and twenty-five they classified as "poor," becausethey thought that living conditions and marginal nutritionof the poor might be contributing to the problems ofsudden deaths. Further, they approved the idea that thesubjects could be compensated with food and drink.A second meeting with the HAC, the Thai MilitaryCommander of Ban Vinai and Munger was held on 30 May todiscuss the specifics of the protocol for obtaining bloodsamples. An immediate topic of discussion was the amountof blood to be drawn from each subject. Munger initiallyrequested a sample of 20 cc at two separate times foreach subject. The council, while it had initially agreed,was now not so willing. Members made a counter-proposalthat 2 cc of blood be drawn at the midnight sampling, and3 cc of blood drawn at the morning sampling. Their mainconcern seemed to be about the amount of blood drawn, andthey consequently negotiated for the smallest amount possible.With less blood, fewer diagnostic tests could beperformed; Munger, however, agreed to the reduction inblood volume drawn, since one major goal of the effortwas to work out all procedures for obtaining specimens.Munger met next with Vue Mai, president of the HAC.Vue Mai related the reasons that Hmong were reluctant togive blood, including fears that giving blood would makean individual weak and susceptible to illness, and fearover what would be done with the blood. For example, heindicated that many Hmong believe that if blood is transfusedinto another person and if the recipient becomesill, then the donor may become ill as well.On 2 June 1983, a third meeting with the HAC includedvigorous debate between Hmong leaders about whether ornot to allow the blood sampling project. The central issueswere fear of illness because of giving blood and theamount of blood requested. Not all leaders agreed to thestudy being done.43J

HMONG ATTITUDES TOWARD A MEDICAL RESEARCH PROJECT 439The leaders did reach a consensus acknowledging theimportance of studies of blood samples as part of the researchefforts on the SUND syndrome. However, most peoplein the Hmong community were unwilling to give bloodsamples, stating that they already felt ill or weak, andthat giving blood would make them feel worse. Anotherconcern expressed was that medical assistance should bemade available after the blood sampling, to treat anycomplications that might arise. Many of the leaders werewilling to donate their own blood, but only if there wasa health assessment beforehand and if the total amount ofblood collected was 5 cc or less. The leaders agreed toreconvene the next day with volunteers from their ownBans to have their blood pressure, pulse and generalhealth conditions assessed. Those found in good healthwould then give blood samples that night.Unfortunately, time was running out. Booton examinedthirty-one males on 3 June. Ten volunteers did notmeet the screening criteria because of high blood pressure,report of illness, enlarged thyroid gland or obviousfear of the procedures. Fourteen of the seven teenclans in Ban Vinai were represented.The effort to collect blood samples was canceledsoon after this initial screening. The main reason forthe cancellation was that a medically qualified personwho was part of the research team could not stay in BanVinai for a significant time after the sampling. An attemptto obtain blood samples over a short period of timemight have been successful, but the fact that the researchteam had to leave soon after would have endangeredthe prospect for a more complete study at a later time.DISCUSSIONThe authors of this paper by now have had a fairamount of collective experience in working with the Hmongcommunities in the United States and in Thailand. We feelwe have learned several important lessons from the reviewof studies of the SUND phenomenon, and our participationin some of the investigations.4 These lessons, discussedbelow, deal with undertaking research in Thailand, withconducting research within the Hmong community and withthe responsibility of researchers toward refugee communities.434

440 THE HMONG IN TRANSITIONA. Undertaking research in Thailand.1. Frequent and personal communication between researchersand Thai officials is essential. Similarly, goodrelations with voluntary agencies in Bangkok, inrefugee camps and in the United States are also essential.All agencies have limited resources andcannot be expected to provide much assistance or personnel.2. Thai researchers and assistants should be included instudies conducted in Thailand, whenever possible. Wefeel that this is an obligation of foreign researchersfor several reasons. First, foreign researchersare guests in Thailand and should act to bolster andutilize Thai research capabilities. Second, engagingThai researchers may make both them and the foreignresearchers more cognizant of research opportunitiesand public health problems. As Hmong individuals becomeincreasingly educated and sophisticated in conductingresearch, and as some become interested ininvestigations into SUND, they should also be invitedto participate in investigations.B. Conducting biomedical and biosocial research with theHmong community.1. Hmong interpreters must be carefully selected. Theselection should be based not only on language ability,but also on the status this individual has inthe community. After discussing our research and thelanguage requirements needed for interpretation, itproved most productive for us to follow the recommendationsof Hmong leaders in our choice of interpreters.We suspect research procedures and rationalecould be better explained if interpreters first wentthrough the research protocol as subjects.2. The most crucial aspect, as already indicated, is therelationship between the research group and the Hmongcommunity. Research efforts must proceed initiallythrough the network of Hmong leaders, requiring thatresearchers go to the effort needed to identify theappropriate leaders. Time must be provided for full13 6-

HMONG ATTITUDES TOWARD A MEDICAL RESEARCH PROJECT 441discussion of all research goals, methods and rationales.Careful attention must be paid to communityconcerns, such as (in this case) fears attendant uponthe removal of samples of blood. Suggestions by Hmongof additional factors which the research group oughtto considersuch as the possible role of nutritionand poverty as risk factors for SUNDshould be fullyevaluated and discussed.3. Researchers should schedule sufficient time for negotiationsand remain available for a period of timeafter completion of data collection to respond to anyphysical conditions claimed to be the result of samplingor testing procedures.C. General considerations.1. Some of the problems discussed above may be avoidedby careful negotiations between researchers and thesubject group. This exchange gives subjects additionalopportunity to provide clues and suggestionsof potential interest to researchers, and it enablesresearchers to determine if their approach to thecommunity is effective or possibly if their researchis threatened by local political developments whichdo not deal with the perceived quality of the research,but with such issues as differential accessto resources resulting from, for example, medicalscreening and payment of subjects.2. It is necessary to describe exactly what a specificresearch effort is designed to learn, as well as how.Often, this proves to b.:. a difficult task, especiallywhen it is necessary to minimize use of culturallyspecificcode words such as "research," "experiment,""test" and "risk factor." Explanations of the rationaleunderlying a research activity in layman's terminologycan lead to increased clarity for the researcheras well.436

442 THE HMONG IN TRANSITIONFinally, as should be obvious from this case study,one research effort is related to other research effortsdealing with the same problem, especially in the minds ofthe subject populations. We now approach a Hmong communitythat is already sensitized by its experiences withprevious research efforts. For the continued success ofstudies attempting to uncover the cause and prevention ofthe SUND phenomenon, researchers must develop methods todeal with the numerous complicating factors discussedhere.ACKNOWLEDGEMENTSWe wish to thank those members of the Hmong communitywho gave their time to discuss research dealing withSUND. Valuable comments on an earlier draft of thisreport were made by Dr. Robert Tauxe, CDC; Dr. AmosDeinard, University of Minnesota Medical School; Mr. JohnFinck, Office of Refugee Resettlement, State of RhodeIsland; Ms. Toyo Biddle, Office of Refugee Resettlement,Social Security Administration; Ms. Sherry Muret-Wagstaf, f, Pediatrics, Hennepin County Medical Center,Minneapolis. The authors are solely responsible for theviews expressed in this report.NOTES'Personal communication to M.G. Hurlich. Bliatout didnot interview any controls in his study, so his resultsare not comparable to those from the CDC. The five hypotheseshe examined to explain SUND syndrome in theHmong were: (!) The possibility that maintenance of traditionalanimistic religious beliefs was associated withthe likelihood of suffering SUNDnot supported. (2)The possibility that use of traditional health practiceswas associated with SUND. (3) The possibility that SUNDcases clustered in certain clans or lineages. (4) Thepossibility that the regions of geographic origins werenot randomly distributed among the SUND victims. (5)The possibility that SUND victims differed from the averageHmong individual with respect to socioeconomicfactors.4 3'r

HMONG ATTITUDES TOWARD A MEDICAL RESEARCH PROJECT 4432We wish to emphasize that these statements reflect opinionsof Hmong people interviewed about work conducted bythe CDC, and that their opinions are no doubt influencedby many of the factors outlined in this report.3Dr. Amos Deinard alerted us to this problem.4For example, both Hurlich and Munger assisted the CDCinvestigator, Dr. Roy Baron, in his epidemiological investigationin part of Washington State; Hurlich assistedLemoine in his interviews in western WashingtonState; Holtan is involved in studies of serum electrolytesof Hmong in St. Paul; Munger has conducted an epidemiologicaland cultural investigation of the RIND phenomenonas it occurs in Ban Varai, Thailand; all threeauthors have interviewed various Hmong individuals aboutinvestigations into the problem of SUND.REFERtNCESBaker, P.T. 198l. "Migration and Human Adaptation." InMi ration Ada taticn and Health in the Pacific editedby C. Fleming and I. Prior. Wellington, NewZealand: Wellington Hospital.Baron, R., S. Thacker, L. Gorelkin, A. Vernon, W. Taylor,and K. Choi. 1983. "Sudden Death Among SoutheastAsian Refugees." Journal of the American Medical Association250(21):2947-51.Blicicout. B.T. 1982. Hmong Sudden Unexpected NocturnalDeath Syrorome: A Cultural Study. Portland, Oregon:Sparkle Publi3hing Enterprises, Inc.Brataas, Anne. 1983. "Hmong Medical Interpreter FieldsQuestions from Curious." St. Paul Pioneer Press, 20March 1983, p. 10A.Chindarsi, N. 1976. The Religion of the Hmong Nt.a.Bangkok: The Siam Society.Eckner, F., R.H. Kirschner, and R. Baron. 1983. "SuddenNocturnal Death in Southeast Asian Refugees: A Pro-.438

444 THE HMONG IN TRANSITIONgress Report." Proceedings of the 1983 Meeting ofAmerican Academy of Forensic Science. Abstract G-32.Eckner, F., and R. Kirschner. 1983. Letter. See also:Baron, R., and Kirschner, R. 1983. "Sudden Night-timeDeath Among Southeast Asians too." Letter. TheLancet 8327:764.Haenszel, W., and M. Kurihara. 1968. "Studies of JapaneseMigrants. I. Mortality from Cancer and Other DiseasesAmong Japanese in the United States." Journal of theNational Cancer Institute 40:43-69.Holtan, N.R., F.E. Kaiser, T. Davin, and E. Haus. 1983.Unpublished Observations. St. Paul-Ramsey MedicalCenter, Departments of Medici:le and Pathology.Kasl, S.V., and L. Berkman. 1981. "Health Consequencesof the Experience of Migration." Annual Review ofPublic Health 4:69-90.Knudson, A.G. 1977. "Genetic and Environmental Interactionsin the Origin of Human Cancer." In Geneticsof Human Cancer, edited by J.J. Mulvihill, R.W.Miller and J.F. Fraumeni, Jr. New York: Raven Press.Lemoine, J., and C. Mougne. 1983. "Why has death stackedthe Reiugees?" Natural History, November 1983:6-19.Modan, B. 1980. "Role of Migrant Studies in Understandingthe Etiology of Cancer." American Joumal ofEpidemiology 112(2):289-95.Monagan, D. 1982. "Curse of the Sleeping Death." ScienceDigest, April 1982: 36-7.Munger, R.G. 1982. "Sudden Adult Death in Asian Populations:The Case of the Hmong. In The Hmong in theWest: Observations and Reports, Papers from the 1981Hmong Research Conference, edited by Bruce T.Downing and Douglas P. Olney. Minneapolis: Centerfor Urban and Regional Affairs, University ofMinnesota.43j

HMONG ATTITUDES TOWARD A MEDICAL RESEARCH PROJECT 445Office of Refugee Resettlement, Social Security Administration.1984. "The Hmong Resettlement Study."Prepared by Literacy and Language Program, NorthwestRegional Educational Laboratory. Portland, Oregon.Schles J.J. 1982. Case-Control Studies: Design,Conduct, Analysis. New York: Oxford Press.United States Centers for Disease Control. 1981. "Sudden,Unexpected, Nocturnal Deaths among SoutheastAsian Refugees." Morbidity and Mortality Weekly Report30:581-89.Vernon, A., and R.C. Baron. 1981. Discussion on Hmongsudden deaths. Held at the Centers for DiseaseControl, 20 April 1981. Review notes and cassettetape.440

CONTRIBUTORSBruce T. Bliatout: Refugee Coordinator, City of Portland;B.A. (Sociology), University of Hawaii; M.P.H. (InternationalHealth & Health Services Admin.stration),School of Public Health, University of Hawaii; M.S. Hygiene(Tropical Diseases), School of Public Health dcTropical Medicine, Tulane University; Dr. P.H.C.-A.B.D.(Tropical Diseases), School of Public Health dc TropicalMedicine, Tulane University; Ph.D. (Health ServicesManagement), Century University.Karl H. Chun: Fellow, Ambulatory Pediatrics, Universityof Minnesota; M.D., University of Hawaii; Two years experiencein medical clinics in Minneapolis with Hmongpediatric patients.Mary L. Cohn: Research Specialist, Northwest RegionalEducational Laboratory, Portland, Oregon; Instructor,English as a Second Language, Mt. Hood Community College,Portland, Oregon.Robert Cooper: Resettlement Officer (Indochinese) UNHCR,Geneva; Deputy Representative, UNHCR, Lao Peoples DemocraticRepublic; Lecturer in Southeast Asian Anthropology,University of Singapore; Ph.D. Social Anthropology,University of Hull; many publications on Asiansinclude the book Resource Scarcity and the Hmong Response.Brenda J. Cumming: School Psychologist; Doctoral candidate(Educational Psychology) University of Minnesota;B.A. (Psychology) San Diego State University.Amos S. Deinard: Associate Professor of Pediatrics, Universityof Minnesota; M.D., University of Minnesota;Five years experience providing primary health care toSoutheast Asian Refugees; Bureau of Maternal and ChildHealth, Minneapolis Health Department, Minneapolis,Minnesota.447441

448 THE HMONG IN TRANSITIONNancy D. Donr111y: Ph.D. Candidate, University of Washington(Anthropology); B.A., University of Wisconsin;M.A. (English Literature), New Mexico State University;currently Principal Investigator, "Research on HmongWomen", funded by Social Science Research Council, Departmentof Anthropology, University of Washington,Seattle, Washington.Bruce T. Downing: Associate Professor of Linguistics,University of Minnesota; Ph.D., University of Texas;Research and publications on language contact, bilingualismand language learning in Vietnamese and Hmongrefugee communities.Timothy Dunnigan: Associate Professor of Anthropology,University of Minnesota; Ph.D., Scholarly interests includesocioeconomic change, linguistic acculturation,and ethnicity in Native American and Southeast Asiancultures. Publications on the Pima, Ojibwe, DakotaSioux, and Hmong.Simon M. Foss: Assistant Professor of Planning and PublicAffairs, Hubert. H. Humphrey Institute of PublicAffairs, University of Minnesota; Ph.D. (Planning),University of California at Los Angeles; Ten years ofeconomic development experience in Asia, Africa, Carribbeanand North America.Judith W. Fuller: Ph.D. candidate, University of Minnesota(Linguistics); B.A. (Psychology) DePauw University;M.A. (Linguistics) University of Minnesota.Beth L. Goldstein: Ph.D. candidate, University of Wisconsin-Madison(Educational Policy Studies); ThailandRefugee Instruction Project, Ban Vinai, Thailand; Instructor,Chinese University of Hong Kong.Karen R. Green: Research Associate, Literacy and LanguageProgram, Northwest Regional Educational Laboratory;B.A. (Spanish), Lewis and Clark College; M.A. (LatinAmerican Studies), The University of California, LosAngeles; specialist in cultural change efforts and literacyand language development among selected ethnic442

CONTRIBUTORS 449groups; four years of field research with Hmong refugeesin the United States.Catherine S. Gross: Coordinator and Trainer, InterculturalRelations Institute, Palo Alto; Special CurriculumDirector, Language Pacifica Institute; B.A. (Anthropologyand Sociology), University of California SantaBarbara; TESL Certificate, University of California;ESL Instructor, Indochinese Language and Literacy Project;eleven years overseas experience with Departmentof State.Glenn L. Hendricks: Associate Professor of Anthropology;Coordinator, Southeast Asian Refugee Studies Project;University of Minnesota International Student Adviser;research and publications in area of migration and resettlementfrom Caribbean and Southeast Asia.Neal R. Holtan: Staff Physician, Department of Medicine,St. Paul-Ramsey Medical Center; Assistant Professor ofMedicine, University of Minnesota Medical School; B.A.(Biology), Saint Olaf College; M.D., The University ofIowa; M.P.H. (Epidemiology), The University of Minnesota;American Refugee Committee, Thailand.Marshall G. Hurlich: Assistant Professor of Anthropology,Department of Anthropology, University of Washingtoa,Seattle; Ph.D. (Physical Anthropology), SUNY-Buffalo; Extensive research with Hmong refugees in theUnited States, dealing with nutrition, growth and development,demography, resettlement history, secondarymigration, sudden unexpected night-time deaths.Annie C. 7aisser: Graduate student in Linguistics, SanDiego State University; B.A. (English and Linguistics),University of Strasbourg, France; M.A. Thesis on HmongSyntax, San Diego State University, San Diego, California;E.S.L. Instructor, San Diego Community CollegeDistrict.Gary Y. Lee: Coordinator, Lao Community Advancement Conperative,Sydney, Australia; Public Relations Officer,Hmong-Australia Society; Youth Development Worker forIndochinese Refugees, Ethnic Communities Council of New443

450 THE HMONG IN TRANSITIONSouth Wales; Teacher in Social Welfare, Sydney TechnicalCollege; M.A. (Social Work), University of NewSouth Wales; Ph.D. (social anthropology), University ofSydney; specialist in Hmong and refugee resettlement.Jacques Lemoine: Charge de recherche, Centre nationalede recherche scientifique, Paris; Coordinator of theThai and Miao-Yao Projects of CEDRASEMI (School of AdvancedStudies in the Social Sciences, Paris).Sarah R. Mason: Historian, Researcher, Southeast AsianRefugee Studies Project, University of Minnesota; B.A.(History) College of Wooster; M.A. and Ph.D. (AsianHistory, United States History, United States-Asian Relations)Northern Illinois University; Research Associate,Minnesota Historical Society; specialist in historyof Asian and Asian American women; extensive experiencein Asia.William H. Meredith: Assistant Professor, University ofNebraska-Lincoln; Ph.D. (Community and Human Resources),University of Nebraska-Lincoln.Shirley G. Moore: Professor of Child Psychology, Instituteof Child Development, University of Minnesota; Co-Director of the Center for Early Education and Development;M.A. (Child Development and Early Childhood Education),University of Iowa. Iowa City; Ph.D. (ChildPsychology), University of Iowa, Iowa City.Ronald G. Munger: Ph.D. candidate, Department of Anthropology;M.P.H. candidate, Department of Epidemiology,University of Washington. Fieldwork on sudden death insleep in the U.S., the Philippines, and Thailand.Sherry Muret-Wagstaff: Director, Southeast Asian Children'sClinic, Hennepin County Medical Center; M.S.(Public Health), University of Minnesota; Ph.D. student,Institute of Child Development, University ofMinnesota.Charles N. Oberg: Congressional Science Fellow in ChildDevelopment, Society for Research in Child Development,Washington Liaison Office, Washington, D.C.; B.A.444

CONTRIBUTORS 451(Child Psychology) University of Minnesota; M.D. (Pediatrics)University of Minnesota Medical School and Hospitals;M.P.H. (Maternal and Child Health) Universityof Minnesota School of Public Health.Douglas P. Olney: Ph.D. Candidate (Anthropology), Universityof Minnesota.Laurel A. Owensby: Graduate student (Linguistics), SanDiego State University; Instructor, ESL, San Diego CommunityCollege District.Martha S. Ratliff: Ph.D. candidate (Linguistics), TheUniversity of Chicago; Senior Research Associate, TheLinguistic Atlas of the Middle and South AtlanticStates; B.A. (English), Carleton College; M.A.T. (EnglishEducation), The University of Chicago; five yearsexperience teaching ESL to Indochinese.Stephen Reder: Director, Literacy and Language Program,Northwest Regional Educational Laboratory; B.A. (Psychology),Stanford University; Ph.D. (Cross-culturalPsychology), The Rockefeller University; specialist inliteracy and language development among selected ethnicgroups; director of research on adult literacy developmentSoutheast Asian refugee English language training,Hmong resettlement and American Samoan adaptation inthe United States.George P. Rowe: Associate Professor, Department of HumanDevelopment and the Family, University of Nebraska,Lincoln; Ph.D., Florida State University; has taught atthe University of Missouri and North Dakota State University;current teaching and research interests indude:families in crises, long-term marriages, and kinrelationships in the retirement years.Louisa Schein: Graduate student (Anthropology), Universityof California at Berkeley; B.A., Brown University;1982, Arnold Fellow, Research in Thailand and the People'sRepublic of China; 1980-81, refugee resettlementwork in Providence, Rhode Island; production of documentary"The Best Place to Live" on Hmong resettlement.445

452 THE HMONG IN TRANSITIONWilliam A. Smalley: Professor of Linguistics, Bethel College,St. Paul, Minnesota; Ph.D. (Anthropological Linguistics),Columbia University; researcher and translationconsultant in Laos, Thailand and Vietnam; participantin the development of the current romanized Hmongwriting system.Nicholas C. Tapp: Research student, School of Orientaland African Studies, London University; M.A. Hons.Cantab. (English Literature); M.A. (Southeast AsianStudies), London; four years in Thailand, and eighteenmonths fieldwork with White Hmong for Ph.D. in SocialAnthropology.Xoua Thoa: Student, Dartmouth Medical School, Hanover,New Hampshire; B.A. (Biology), Brown University.Shur Vang Vangyi: Indochinese Refugee Economic DevelopmentConsultant, LTG Associates, San Francisco, California;MBA, United States International University;currently working toward a DBA degree in General Managementat the United States International University;four years of experience as Trade Director, Internaland External, for the Laotian Government, Ministry ofNational Economy; six years experience as ExecutiveDirector of Lao Family Community, a nonprofit organizationassisting the Indochinese refugees adjusting totheir new life in America.Gail Weinstein: Adult Educator; B.A. (Anthropology); M.S.(TESOL); Ph.D. Candidate (Sociolinguistics), Universityof Pennsylvania; teaching/travels in China and Thailand;admirer of Hmong people and culture.

INDEX"Absorption and Retention of Lead by Infante,425Acquisition and Use of Literacy in the HmongCommunity of Newton. 1981 Annual Reportof the Functional Literacy Project, 329"Aesthetic Language in White Hmong", 235AFDC, 199African Language Structures, 243African, 219"Age, Rate and Eventual Attainment in SecondLanguage Acquisition", 328Agncultural Extension Service, 117Ainsworth, M., 412-413Ainsworth's System for Rating Maternal-CareBehavior, 400-401Ajchenbaum, Yves, 10, 21"Akha and Miao: Problems of AppliedEthnography in Farther India", 413Alberta, Canada, 282AFL-CIO, 108American Refugee Committee, 396, 437Anderson, A.S., 419, 420, 421, 425Anderson, Pamela, xi,Andrews, Frank M., 125, 130An h u1,78Animism, 47, 48, 52n, 55"Antecedents of Mother-infant Interactionand Infant Irritability In the First ThreeMonths of Life", 414Anthropological Quarterly, mu, 52"Applications of Second Language AcquisitionResearch to the Bilingual Classroom",328Arend, R A., 412, 415Arkansas, 177Asian Needlecrafters,160-162, 165, 166, 167,168, 169, 171Atlanta, Georgia, 418, 430"Attachment and Early Maltreatment", 414Attitudes and Motivation in Second-LanguageLearning, 328Attopeu, 31Australia, xi, xii, 5, 55, 59, 60, 6L 63, 64, 65,69, 337Australian Capital Territory, 62Austronesian, 219Bahnar, 219Bailey, Nathahe, 299, 328Baker, P.T , 427, 443Balika, Asen, 365, 377Ban Vinai, 13, 14, 15, 20n, 31, 38, 335, 336,379-395, 428-429, 434, 436-438, 443nBangkok, 396, 437, 440Banton, M 70Bantu, 219Bar-Yosef, Rivka Weiss, 159, 172Barney, G Linwood, xm, 17, 21, 121, 122,123, 124, 130, 396nBaron, R.C., 380-381 395-397, 427, 429-430,432, 443-445Barrett, P , 425Barth, Freclank, 78, 84, 138, 142Basic Educational Opportunity Grants, 200Becker, Sandra, 119Beecham, May, 119Beijing, 74, 78Bellamy, Daniel, 395Berkman, L., 427, 444Bernatzik, H.A., 412-413Bertrais, Yves, 245, 254, 259Bibliography of the Hmong (Miao) of SoutheastAsia and the Hmong Refugees in theUnited States (A), 53Biddle, Toyo, 442Bilingual Instruction, 116Bilingual Program, 108-109Bilingualism, 47"Bilingualism and Language Acquisition", 329Bini, 221, 224"Black" Miao, 78Blehar, M., 412-413Bliatout, Bruce Thowpaou, 21,177, 334, 382,396, 431, 442n, 443Block, A.J., 395, 397"Blue Hmong", 46, 79Bolinger, Dwight, 224, 234Booton, Elizabeth A., 386, 395, 436Bradley, R.H., 400, 402, 414Brataas, Anne, 443"Bnef Descr ption of the Miao Language" (A),85"Brief Introduction to China's Minority Education"(A), 84"Brokerage, Economic Opportunity, and theGrowth of Ethnic Movements", 173Brown University, 83nBruner, J.S , 280, 295Brunnquell, D., 400, 414Buddhist (Buddhism), 47-48, 351Bureau of Maternal and Child Health at theMinneapolis Health Department, 417Bureau of Research and Training, 152, 156Burma, 88Burt, Manna, 299, 328Buyi, 764534 4 ?

454 THE HMONG IN TRANSITIONCCSDPT Handbook: Refugee Services inThailand (The), 21Calderon, Eddie, 152, 156Caldwell, B.M., 400, 402, 414California, 55,100, 177, 182, 184, 195, 198,199, 303, 321, 322Cambodia, 101Camp, B.W., 400, 414Canada, 431Caros, Lydia, 424Case-Control Studies Design, Conduct andAnalysis, 397, 445Casey, P.H., 401, 416Castile, George Pierre, 81, 84Castro, Fidel, 103Catanzaro, A., 365, 378Catholic, 41, 47, 339Center for Applied Linguistics, mil, 306Center for Migration Studies, xccCenters for Disease Control, 418, 445Central Institute in Belong, 80Central Nationalities Institute, 74Central Valley, 182, 184, 185, 186, 303, 321Chadang, 377Chamot, Anna Uhl, 299, 328Chao Fa, 17, 47Cha, Youa, 377Chen, P.S.J., 70Chenan, Father, 119Chemiak, N.S., 395, 397Children's Hospital, 421Chin, 88China (Chinese), 5, 7, 11, 15, 27, 28, 33, 36,42, 43, 53, 55, 67, 73, 74, 87, 88, 89, 90,148, 224, 275, 337, 342, 355, 399China. Land of Many Nationalities, 85China's Geographic Foundations, 84China's Minority Nationalities, 85"China's National Minorities", 85"China's Policy on National Minorities", 85Chindarsi, N., 44.,Chinese Academy of Sciences, 80Chinese Geomancy, An AnthropologicalAnalysts of, 94Choi, K., 396, 443Chomsky, Noam, 245Chong Koua Vang, 396Chong Lee Chue, 377Chong Moua Lee, 396Chnstian (Chnstianiky), 16-17, 48, 61, 149,162, 166,338-339, 347, 362Chnstian & Missionary Alliance Group, 362Christian Medical Team, 437Chnstian Medical Tea:n-Chnstian, 396"Chnstianity. Innovation cn Meo Culture", 21Christison, Mary Ann, 299, 300, 328Chu Yao, 341Chuan Qian Dian, 79Chun, 335Clark, Margaret, 135, 143Clarke, R., 121, 122, 131"Classifier in the Wei Ning Dialect of the MiaoLanguage in Kweichow" (The), 236Cohen, Y , 70Cohen, Mary, 100, 177, 324, 324nCole, Michael, 281, 282, 297Collected Papers, 235Committee for Coordination of Services toDisplaced Persons in Thailand, 21Commonwealth Schools Commission, 63Communist Laotians, 337"Comparison of Proxemic and SelectedCommunication Behavior of Anglo-American and Hmong Refugee Mother -infant Pairs" (A), 415Comprehensive Employment and Training Actof 1973 (CETA), 111Congdon, Glona, 119Conklin, Nancy Faires, 324"Consequences of Literacy" (The), 296"Continuity of Adaptation in the Second Year:The Relationship Between Quality ofAttachment and Later Competence", 415"Contrasting Patterns of Literacy in an AlaskaFishing Village", 296"Control of Breathing During Sleep", 397Cook County, 381Cook-Gumperz, Jenny, 282, 283, 295Cooper, Robert, xii, 4, 66, 70Copenhagen, 101"Corporation and the Entourage. A Comparisonof Thai and Amencan SocialOrganization" (The), 172Corum, Clam 243Cressey, George c,abcock, 75, 84Cnchton, L , 400-401, 414, 416Crockenberg, S.B., 411, 414Crossland, Andrea, 396Cuban Refugee Program, 103Cumming, B.J., 335Current Trends in Linguistics and AdjacentArts and Sciences, 52"Curse of the Sleeping Death", 444Dab Neeg Hmoob: Myths, Legends and FolkTales from the Hmong of Laos, 259Dallas, 190, 194Dallas-Fort Worth, 15, 177, 178, 182Dalton, G., 56, 70Davin, T., 444De Vos, George, 75, 84, 136, 143"Deficits in Psychologic and Classroom Performanceof Children with Elevated DentineLead Levels", 425d

INDEX455Deinard, Amos S., 335, 400, 414, 419-420,425, 442, 443nDenver Development Screening Tests, 400,403, 406Denver 7-, ^lopmental Screening Tzst ManualWorkbook for Nursing and ParamedicalPersonnel, 414Description and Comparison in CulturalAnthropology, 70"Designs from Another World", 172"Desocialization and Resocialization TheAdjustment Process of Immigrants", 172"Development Dynamics: Hmong Refugees inAmerica", 120Development and Policy Concerning ChildrenWith Special Needs. The Minnesota Symposiumon Child Psychology, 415Dewey, Joyce, 119Than Dong Bei, 79, 80Dictionary ofMong Njua, 235Dictionnaire francais-Hmong blanc, 260Dictionnaire iimong-Francais, 259Diffloth, Gerard, 219, 221, 222, 225, 233,234, 234nDil, Anwar S., 328District 916 Vo-Tech, 108, 112Divorce, 128-129Dodds, J.B., 403, 414Don& 76, 79Donnelly, Nancy, 100Dooley, Tom, 17, 21Downing, Bruce T., 15, 19, 21, 22, 71, 100,156, 177, 187, 209, 235, 276, 397, 444Dravidian, 219Dray, Nancy L., 233Dryer, Donna, 377Dulay, Heidi, 299, 328Duluth, 104Dunnigan, Timothy, mil, 46, 47, 52Durrenberger, E.P , 88, 94Duvander, Sandra, 119East and West Metro Refugee Health Projects,111Eckner, Friedrich A 0 , 381,397,432,443-444Edge of Tomorrow (The), 21Edwards, B B., 425Egeland, B., 400-401, 412-416Eisenstadt, S.M., 162, 170, 172Elarao, R., 402, 414Elements de grammaire Hmong blanc, 235,259, 277English as a Second Language (ESL), 106, 111,118, 146, 153, 186, 190, 192, 213,300Ervin-Tripp, Susan M., 299, 321, 328Eskimo, 283Essays on the Problem of Tribe, 94Ethnic Adaptation and Identity: The Karen onthe Thai Frontier with Burma, 95Ethnic Groups and Boundaries, 84Ethnic Groups of Mainland Southeast Asia,52"Ethnic Identification in China", 84Ethnic Identity: Cultural Continuities andChange, 84, 143Ethnic School, 63Ethnicity, 3, 50-51, 55, 63, 67, 73, 75, 76,89,92, 337Euro-American, 162Europe, 29European Colonialism, 89Evers, H.D., 70"Explorations of Acculturation: Toward aModel of Ethnic Identity", 143"Expressive Phonology and Prosaic Phonology",234"Expresswes in Semai", 234Family Day Care Training Program, 116Family Service of St. Paul, 116Fandal, A.W., 403, 414Fass, Simon M., 100, 102, 120, 177Faydang Lobliayao, 16Federal Food Program for Women, Infants andChildren, 115Fei Hsiao-tung, 74, 75, 76, 84Feuchtwang, S., 87, 92, 94Filipino, 383"Final Report: A Prospective Study of theAntecedents of Child Abuse", 414Finrk, John, 100. 177 377, 442Fishman, Joshua A., 47, 52Fleming, C., 443"Folk Medical Beliefs and Their Implicationsfor Care of Patients", 378Fomon, S.J., 425Forehand, R., 412, 416Fort Chipewyan, 282Fort Smith, Arkansas, 177, 178, 182, 190,194Fort Worth, 15"Four Syllable Coordinative Constructions inThe Miao Language of Eastern Kweichow",235France (French), xi, 28, 36, 66, 73, 148, 337,338, 346, 431Frankenburg, W.K., 400, 403, 414Fraumeni, Jr, J.F., 444Frawley, William, 297Fresno, California, 177, 178, 182, 184, 185,186, 194Freudian, 345Friedman, J., 394, 398449

456 THE HMONG IN TRANSITIONFriends, Followers, and Factions, 172From Communication to Language A PsychologicalPerspective", 295"From Oral to Written Culture. The Transitionto Literacy", 295"From Utterance to Text. The Bias of Languagein Speech and Writing", 296Fu Mao-Chi, 80, 84Fujian, 78Fuller, Judith, 213, 215Functional Study of Topic in Chinese TheFirst Step Towards Discourse Analysis (A),277"Functions and Uses of Literacy" (The), 296Galbally Report, 63, 71Gall, Allen, 177Galt School, 107Gardner, Robert C., 300, 328Geddes, William R., mu, 18, 21, 66, 70Geertz, C., 58-59, 70"General Considerations in the Care of SickChildren: Clinical Evaluation of Infants andChildren", 415"Genetic and Environmental Interactions inthe Origin of Human Cancer", 444Genetics of Human Cancer, 444Geneva, xii, 101Gensmer, Karen, 119George, Isaac, 237, 243Gersmehl, Carol, ionGilmore, Perry, 284, 295"Gimme Room. A Cultural Approach to theStudy of Attitudes and Admission to Literacy",295Givon, Talmy, 277Goldstein, A.D , 400, 414Goldstein, Beth, 99Goodenough, W., 68, 70Goody, Jack, 280, 283, 296Gorelkin, L., 396, 443Grabenstein, Brenda, 424Graduate Student Research and Travel Fund,396Graham, D.C., 94Grammar of Mong lshua (Green Miao) ADescriptive Linguistics Study, 277Greece, 280Green Hmong, 66Green, Karen Reed, 177, 215, 283, 29o, 324n,329Greenfield, P.M , 280, 296Gross, Catherine Staumpos, 99, 156Guangdong, 78Guangxi, 76, 78, 79Guizhou, 224, 76, 77, 79, 80, 81, 82Gumperz, John, 282, 283, 295Gundel, Jeanette, 268, 269, 276Gunnoe, C.E., 425Haenszel, W , 427, 444Haley, Pamela, 160, 172Halpern, Joel, 17, 21Hammond, Ruth, xiiHan Chinese, 74, 76, 79, 80, 81, 89Hang Doua, 396Hang Sao, 396Ham, 76Hanks, Lucian M., 162, 165, 172Harrier, Michael, 340Hams, Jane, 172Hartwell, Lucy, 119Hassoun, Jean-Pierre, 10, 21Hatch, Evelyn Marcussen, 328Hattiberg, Patricia, 119, 119nHaus, E., 444Havelock, Eric A., 280, 296Haw, 15"Health Consequences of the Experience ofMigration", 444Health Interpreters Training Program, 110,111, 115"Health Status of Refugees from Vietnam, Laos,and Cambodia", 378Heath, Shirley, 284, 296Heimbach, Ernest E., 84n, 214, 219, 221,222,234, 245, 246, 259, 277, 388, 396n, 397Helgerson, Steve, 396Helm, J., 89, 94Hendricks, Glenn, 177Hendriksen, Georg, 163, 172Hennepin County Medical Center, 399, 419,421, 442Hickey, Gerald C., 52Hinton, P., 89, 94Hispanics, 108Histoire des Mao, 53Histoires d'insertion de groupes familiauxHmong refugees en France, 21Hmong Advisory Council (HAC), 385, 437,438Hmong Artwork Association, 168, 170, 171Hmong Association, 46"Hmong Attitudes Toward Language Maintenance:A Community Sponsored TranslationProject", 52Hmong Community Survey, 120n, 191"Hmong Community's Acquisition of English"(A), 209, 329"Hmong Families in Long Beach 1978-1982:Between Incorporation and Rejection", 156Hmong Farm Project (Minnesota Agriculture4 3- 0

INDEX 457Enterpnse for New Americans), 117Hmong Folk Tales, 243, 259"Hmong Literacy, Formal Education, and TheirEffects on Performance in an ESL Class",209"Hmong Medical Interpreter Fields Questionsfrom Curious", 443"Hmong Refugee Community in San DiegoTheoretical and Practical Implications of itsContinuing Ethnic Solidarity" (The), 71Hmong Research Conference, 185, 190"Hmong Resettlement Study Site Report:Dallas-Fort Worth, Texas" (The), 21-22,191Hmong Resettlement Study, 175, 179, 184,187, 191, 192, 445"Hmong student Association of Dallas andFort Worth" (The), 21Hmong Sudden Unexpected Nocturnal DeathSyndrome: A Cultural Study, 396, 443"Hmong and Mien Textile Art. Motif andMeaning", 173' Hmong in Laos" (The), 131Hmong in the West: Observatio "s and Reports(The), 71, 156, 209, 209, 235, 329, 397,444"Hmong vs. Amencan Caucasian NewbornBehavior and Reanng Practices", 415Hmong-Australia Society, 62Hmoob Dawb, 79, 87Hmoob Ntsuab, 79Hoa Young, 119Hobsbaum, E J., 71Holland, Bibba, 391, 396Holloway, Ailsa, 395Holtan, N.R., 380, 396, 428, 429, 443n, 444Holter, 429Holverson, Colleen, 119Home Observation for Measurement of theEnvironment (HOME), 400, 402-403, 414Hsu, Francis L.K., 89, 94Huang, 80Huang-ying, 84Hubei, 78Hulcher, Doug, 396Hume, Shirley, 119Hunan, 76, 77, 79Hurlich, Marshall G., 171, 177, 336, 380, 394,396, 442n, 443nIllinois, 381Impact Aid, 168"Impact of Indochinese Resettlement on thePhillips and Elliot Park Neighborhoods inSoutn Minneapolis" (The), 156Indian, 11"Individual Differences in Infant-MotherAttachment at 12 and 18 Months. Stabilityand Change in Families Under Stress", 415Indochina Migration and Refugee AssistanceAct of 1975, 103Indochina (Indochinese), 152, 333, 337Indochinese Center, 291-292Indochinese Community Project, 148, 155nIndochinese Refugee Education Guides,General Information Series M16, xiiiIndustrial Revolution, 29Infant and Environment: Early Cognitive andMotivational Development, 416"Infant-Caregiver Attachment and Patterns ofAttachment in Preschool:The Roots ofMaladaptation and Competence", 415Institute of Linguistics and Philology, 80Institutes for Nationalities, 80Intergovernmental Committee for MigrationSeminar on Adaptation and Integration ofPermanent Immigrants, 101Intergovernmental Committee on Migration,148, 396International Rescue Committee, 396, 437Introduction to Indochinese History, Culture,Language and Life, 131"Introduction: The Dyadic Basis of Clientelism",172Iowa, 204"Iron Status of Children with Elevated BloodLead Concentrations", 425Is There a 'Natural Sequence' in Adult SecondLanguage Learning?", 328Isla Vista Youth Project, 152Isla Vista, 99, 145-158Israel, 159Jaisser, Annie, 213, 214Jakobson, Roman, 224, 234n, 235Janssen, Laurel, 119japan (Japanese), 219, 274Japenga, A., 149, 156Jensen, R.L., 425Jiangxi, 78Jinuo, 75Johns, Brenda, 220, 235Johnson, Bob, 396Johnson, Charles, 237, 243, 245, 259, 266,276n, 277Johnson, Kathleen, 119Jung, R., 395, 397Kachin, 88, 89, 92Kaiser, F.E., 444Kampuchea, 380, 430451

458 THE HMONG IM TRANSITIONKasl, S V., 427, 444Kaufman, S., 135, 143Keesmg, Roger M., 49, 52Kershaw, Greet, 152, 155, 156Keyes, C., 95Khamu, 384Khasi, 219Khmu', 10Kim, Kong-On, 221, 235Kin Groups and Social Structure, 52Kinship, 49-50Kirschner, R.H., 381, 397, 432, 443, 444Knudson, A.G., 427, 444Korea (Koreans), 221, 225, 270, 274Krahnke, Karl J., 299, 300, 328Krashen, Stephen D., 299, 328Kuhlo, W., 395, 397Kunstadter, P., 130Kurihara, M., 427, 444Kushner, Gilbert, 84Kweichow, 42L'initiation du mort chez k: Hmong, 348La fabuleuse aventure du peuple de l'opium,52Lahu, 88Lakoff, Robin, 285, 296Lambert, Wallace E., 300, 328, 329Lande, Carl H., 163, 172Language (language issues), 4, 46-47, 63-64,78-80, 187-193, 197, 211-330Language Acquisition and CommunicativeChoice: Essays by Susan M. Ervin-Tripp,328Lao Community Advancement Co-operative,65-66, 69Lao Family Community, 175, 177Lao Mumg Plai, 19nLaos Civil War, 29, 37Laos Constitutional Monarchy, 43Laos (Lao, Laotian), xi, 4-5, 7-9, 10, 11-13,16-18, 20n, 24-25, 28, 30, 32-33, 34, 38,39, 43-45, 47,63, 66, 67,69, 88, 101,113,121, 124, 127, 129, 136-137, 139, 145,148, 149,165, 179, 183, 292, 294, 303,305, 314, 319, 320-322, 338, 380, 382,384, 410, 431Laos: Its People, Its Society, Its Culture, 414"Laotian Health Statistics", 21Larteguy, J.an, 42, 52Latin American, 162Le Bar, Frank M., 43, 52, 399, 414Le chamane Hmong sur les Chemins de l'Audela,348Leach, Edmund R., 78, 84, 88, 89, 95Lee, Gary Yia, xnLehman, F.K , 88, 95Lemome, Jacques, xis, 17-18, 21, 78, 84, 87,95,165,173, 334, 339, 348, 382, 383, 397,431, 443n, 444Les Hmong du Laos face au develloppement,22"Les ecritures des Hmong", 21"I et Isla Vista Eat" (LIVE), 153Levition, A., 425Lewman, Larry, 380, 430Li, Charles N , 237, 238, 243, 26E 269, 270,271, 273, 274, 275, 276n, 277Liberia, 281Lin Yueh-Hwa, 77, 84Linguistics and Literacy, 297Linnekin, Jocelyn S., 51n, 53Lsu Project: A Sono-Medical Study of theL,su of North Thailand, 94Lau, 88, 92, 270, 274Literacy, 17, 18, 190"Literacy in Suboi dinate Cultures. The Caseof Black Americans", 296Literacy in Traditional Societies, 296"Literate Two Year Old: The Fictionalizationof Self", (The), 296Liu Chun, 76, 85Loei Province, 428Lobo, 91, 92"Lobo Story: 'The Great God of 0-LI-Bi-Zih'by Lin Kuang-Tien" (A), 94Long Beach, 152Long Cheng, 11, 20n, 29Long, Michael A , 299, 328Loojmem (Geomancy), 87-93"Lord of the Sky", 47Lor, Mai Lee, 386, 396Los Angeles Times, 149Luang Prabang, 9, 11, 31, 32Lushan, 224Lutheran Immigration and Refugee Service,101Lyfoung, Lysao, 19Lyfoung, Touby, 16Lykke, Miriam, 233Lyman, Thomas Amis, 219, 235, 265, 277Ma Touan-Lin d'Hervey de Saint-Denys, 76,85Madden, Carolyn, 299, 328Magic Witchcraft and Curing, 377Mahaffey, K.R., 425Maher, C., 425Man in Adaptation The Cultural Present,70Manchu Government, 27Mandarin Chinese, 78, 274452

INDEX 459Mantel-Haenszel, 420Margolis, J A., 411, 415Marie Lee, Sister, 119Martin, Samuel E., 221, 235Marx, K., 56, 71M.6xism, 56Mason, Sarah, 15, 21, 99, 177Matas, L., 412, 415"Maternal Interactive Behavior as a Functionof Race, Socioeconomic Status, and Se--; ofthe Child", 416McCarthy, John J., 224, 235McDonough Homes, 115McGrath, Debreana, 155McKay, R.J., 403, 415Medical Examiner's Office, 380"Meeting of Young Minds in 24 Languages"(A), 156Mekong, 11, 294Mekong River, 35, 383Melville, Margarita B., 135, 143"Meo of Xieng Khouang Province" (The),130"Meo-An Incipient Church" (The), 21Merced, CA, 184, 185Meredith, William, 99Mexican-American, 170Mexico (Mexican), 41, 423, 424Mhu, 221, 224Miao Language Team, Chinese Academy ofSocial Sciences, 85"Miao Nationality Village" (A), 85Miao and Yao Lii 7uisvc Studies, 235, 236"Miao-Man Peoples of Kweichow" (The), 84"Miao Their Origins and Southwards Migrations"(The),95Michaels, Sarah, 283, 296Middleton , 377Mielke, Howard, 422, 424Mien, 10, 160, 162, 166, 168, 169, 431Mien-Yao, 382, 384Migrants of the Mountains: The I.. al Ecologyof the Blue Mtao (Hmong 1404a) ofThailand, xiii, 21, 70"Migration and Human Adaptation", 443"Migration and Mental Health AmongHmong Refugees", 348Migration and Refugee Assistance Act of1962, 103Migration, Adaptation and Health in thePacific, 443Millar, Susan Bolyard, 142Miller, R.W., 444Miller, Virginia, 424Minneapolis, 15, 111, 112, 114, 115, 152,336, 417, 423, 424, 437Bureau of Housing Inspections, 417,Children's Health Center, 419, 420-421;Commun'ty College, 108; Health Department,'119, 420, 421, 424, HousingInspections Division, 419; Programs,113; Refugee H,memaker Program, 105,113, 115; Technical Institute, 107, 112Minneapolis-St Paul, (cf. St. Paul-Minn.)Minnesota Battery Test, 107Minnesota Department of Public Welfare, 116Minnesota Papers in Linguistics and the Philosophyof 1.. guage, 52Minnesota, 15, 104-119, 177, 207, 336, 428"Minority People in Kweichow" (The), 84Missionary Alliance of Thailand, 396Missoula, Montana, 207Modan, B , 427, 444Modern China Series, 85Mom Phuv, 10-11, 15Mon-Khmer, 219, 221, 225Monagan, D., 436, 444Mongolians, 74Monkeys and the Grasshoppers (The), 254Moore, Ruth L., 155, 335Moore, S.G., 415Moos Plais, 9-15, 19nMoser, R.J., 365, 379"Mother-Infant Relationship and InfantDevelopment. The Effect of Pediatric latervention"(The), 416Mottin, Jean, 220, 235, 245-253, 255, 256,259, 261, 277Moua, Cher Ker, 377Mougne, Christine, 347, 382-383, 393, 397,431, 440Mua, Dan X., 324Multnomah County, Oregon, 380Mulvihill, J.J., 444Munger, R.G., 335, 379, 380, 383, 394, 397,427-429, 436-438, 443n, 414Muong Ngoi, 19nMuong Sai, 19nMuret-Wagstaff, S , 335, 415, 442Murphey, Herbert, 119Musgrave, John K., 52Mutual assistance associations, 44-45"Myth of Orality and Literacy" (The), 297Nagler, A., 413Naida Reda, 119Nam Ou, 11Nam Phong, 15, 18Nam Poui, 11-12, 13, 15Nampet Panichpant-M, 396Narrative, Literacy and Face in lnterethnicCommunication, 296National Alliance for Polish People, 209

460 THE HMONG IN TRANSITIONNational Endowment for the Humanities, xiiNational Mental Health Needs Assessment ofIndochinese Refugee Populations, 156"National Minorities Enjoy Regional Autonomy",85Nationalities Affairs Commission, EditingCommittee of the Publishing Series for FiveKinds of Nationalities Problems, 85Nationalities Institute, 78, 81"Natural Sequences in Child Second LanguageAcquisition", 328Needleman, H.L., 425"Negotiating Marriage in Hmong Society, AnExample of the Effect of Social Ritual onLanguage Maintenance", 52Neider, John, 348Nelson Textbook of Pediatrics, 415N^Ison, W.E., 415"Neurophysiological Studies of AbnormalNight Sleep and the Pickwickian Syndrome",397New South Wales, 62"New Year in a New Land Religious ChangeAmong the Lao Hmong Refugees in SanDiego" (A), 71Newlin-Haus, E M., 412, 415Nigerian Kwa, 224Ninh Phan, 119Nkauj Ntsuab Thiab Sis Nab: Dab Neeg TunRhawv Zeb. Ngao Njua and Shee Na: TheStory of the Plain of Jars, 277Nong Khai, 387Norris, T.N , 419-421, 425Norsu, 89North Vietnam, 7Northern Thai, 33Northwest Reg'onal Educational Laboratory,175, 177"Notes on Expressive Meaning", 234Oberg, Charles N , 335, 394, 411Ogbu, John, 284, 296Olney, Douglas P., mu, 52n, 53, 71, 100,120n, 156, 177, 209, 235, 329Olson, David R , 281, 285, 296One Year After Arrival: The Adjustment ofIndochinese Women in the United States,1979-1980, 120Opium, 32, 34, 35, 36, 39, 57, 66, 68"Opium and the Miao", /0Opium Trade 16, 27Opium War, 27"Oral or Written Language: The Consequencesfor Cognitive Development in Africa,the United States and England", 296Orange County, California, 177, 178, 182,186-187, 190, 194Oregon, 177, 198ORR Hmong/Highland Lao Workshop, 184Orphan and Ngao Zhua Pa (The), 255Orthography, 17-18, 46-47Owensby, Laurel, 213, 214P'an Yuan-en, 220, 235Pachai, 338Paine, Robert, 165, 166, 172,173Pakse, 31Panghai, 79Pathet Lao, xi, 9,13, 17, 29, 30, 33"Patron-Client Relations as a Model of StructuringSocial Exchange", 172Patrons and Brokers in the East Arctic, 172,173Patterns of Attachment, 413Pedersen, F., 412, 416Pederson, Janice, 119Peirce, Charles Sanders, 234n, 235Peng Jianqun, 85People's Daily, 85People's Republic of China, 73, 82, 83Peoples of the Earth, 131Peresie, H., 425Perlmutter, M., 415l'erry, Cynthia, 276"Persistent Cultural Systems: A ComparativeStudy of Identity Systems that Can Adaptto Contrasting Environmrnts", 53, 85Persistent Peoples: Cultural Enclaves in Perspective,84Peuples Etrangeres c la Chine, 85Phiktakpao, Wadtanee, 396Philadelphia, 280, 289, 292Phillipson, EA., 395, 397Phipps-Yonas, S., 400, 414Phone Savane, 32Phonemes and O- thography: Language Planningin Ten Minority Languages of Thailand,22"Phonetic Symbolism in Korean", 235"Phonological Features and MorphologicalStructure", 235Pickwickian Syndrome, 395Pierce, R , 135, 143Political Systems of Highland Burma, 84, 95Pollard, Samuel, 80Polygyny, 123, 127, 129, 130Pongsapich, Amara, 396"Population Statistics of the Ban Vinai RefugeeCamp". 398Portland, Oregon, 175, 177, 178, 182, 184,186, 412, 430Poweltrm, 294454

INDEX 461Precapitalist Economic Formations, 71Preface to Plato, 296"Preliminary Study of the Descriptive Wordsin the Miao Language of Eastern Kweichow"(A), 235Prendergast, Tom 396Prior, 1., 443"Problems of Consonants and Tones Hmong(Meo, Miao)" (The), 22Proceedings of the International Associationof Historians of Asia,95Protestant, 16-17, 47, 362Providence, 177, 178, 182, 194Purnell, Herbert C., 85, 235, 236Qhua Ke, 339Qian Dong, 79, 80Qingshui River, 78"Quest for the Essence of Language", 235Radcliffe-Brown, A.R , 58, 71"Raised Lead Levels and Impaired Cognitive/BehaviouralFunctioning. A Review ofthe Evidence'", 425Ramsey County Child Health Consortium, 111Ramsey County Nursing Service, 115Rangaraj, A.G., 395Ratliff, Martha, 213, 214, 276"Recent Language Rescarch and Some LanguageTeaching Principles", 328Reder, Stephen, 100, 177, 190, 209, 215, 283,296, 301, 324n, 329Reed, R , 425Refugee Act of 1980, 101, 103Refugee Program Office, 111"Refugees Who Do and Do Not Seek PsychiatricCare, An Analysis of Premigratoryand Postmigratory Characteristics", 348Regina Project, 113-115"Relationship of Infants' Home Environmentto Mental Test Performance from Six- toThirty-six Months: A Longitudinal Analysis"(The), 414"Relationships Between Neonatal BehavioralOrganization and Infant Behavior Duringthe First Year of Life", 416"Religion as Cultural System", 70Religion of the Hmong Njua (The), 443Religion, 47-48, 57-59, 149Renmin Ribao, 77, 85Repatriation, 23-40"Repetitive Poisoning in Children A PsychosocialStudy", 415Report of the Review of Post-arrival Programmesand Services for Migrants, 71Report on the Ethnic Schools Programme, 70"Respiratory Dysrhythtruas During Sleep", 397"Revised Denver Developmental ScreeningTest: Its Accuracy as a Screening Instrument"(The), 414Rhode Island, 177, 178, 290, 442"Rime Assonance and Morphemic Analysis",234Robson, Barbara, 190, 209"Role of Migrant Studies in Understanding theEtiology of Cancer", 444Roman Catholic, 16-17Romanized Popular Alphabet, 215, 217, 218Romanucci-Ross, Lola, 84, 136, 143Roniger, Louis, 162, 172Ross, Darlene, 119Rowe, George, 99Royal Thal Government, 395Rubenstein, J., 412, 416Ruey, `1111-Fu, 88, 95Rutter, Michael, 419, 425Saint Paul-Minneapolis, 15, 46, 102, 111-113,115, 119, 177, 178,182,184,417, 423-424,429, 433, 436, 443nSaint Paul, Public Heal, Department, 115;Ramsey Medical Ce ter, 428; TechnicalVocational Institute (TVI), 107, 108, 111,113-115Sam Neua, 52nSan Diego, 55, 67San Diego State University, 259:,Santa Ana, California, 175, 347Santa Barbara, 152, 155Santa Barbara's Summer Solstice Parade,153-154Savannakhet, 31Savina, F.M., 42, 53Sayaboury, 11, 31, 32Scarcella, Robin C., 328Schein, Louisa, 5Schlesselman, J.J., 379, 397, 430, 445Schmidt, Steffen W., 172,173School of Advanced Studies in the SocialSciences, xiiSchumann, 300Schwartz, Samuel, 419-420, 425Scollon, Ron, 282, 283, 296Scollon, Suzanne, 282, 233, 296Scott, G.M., 55, 65, 67, 71"Screening for Iron Deficiency with the ErythrocyteProtoporphynn Test", 425Scribner, Sylvia, 281, 282, 297Se Yang, 260Seattle, 186, 396, 436455

462 THE HMONG IN TRANSITIONSebeok, Thomas A., 52Second Language Acquisition. A Book ofReadings, 328"Segmentary Kinship in an Urban Society.The Hmong of St. Paul-Minneapolis",52"Selective Acculturation of Female MexicanMigrants", 143Semai, 219, 221, 224"Serial Verb Constructions in Mandarin ChineseSubordination or Co-ordination?", 243Shamanic trances, 64"Shamanistic Behavior among the NetsilikEskimo", 377Shan, 88, 92Shanghai, 78Shao and His Fire, 255"Sharing Time ekn Oral Preparation for Literacy",296Sibert, J R., 411, 415Sichuan, 78, 79, 82"Sleep Apnea, Hypopnea and Oxygen Desaturationin Normal Subjects: A StrongMale Predominance", 397Small Business Administration, 206Smalley, William A., 4, 17, 22, 396nSmith, P., 324n, 411, 414Smith-Stark, T. Cedric, 243Snow, L., 365, 378Sobel, R., 411, 415Social Indicators of Well-being, 130Social Security Administration, 442"Sociology of Language. An InterdisciplinarySocial Science Approach to Language inSociety" (The), 52Some of My Favorite Writers are Literate:The Mingling of Oral and Literate Strategiesin Written Communication", 296Sop Tuang, 12-15, 20n"Soul-calling", 64"Sound Symbolism in Korean", 235"Southeast Asian Textiles", 172Southeast Asian Tribes, Minorities, andNations, xiii, 130Southeast Asia (Asians), 42, 83, 101, 102-106,112, 116, 162, 165, 306, 336, 347, 351,379, 381, 383-384, 395, 419, 421, 432Special Intermediate District 916 VocationalTechnical Institute, 107Spicer, Edward H., 41, 42, 45, 53, 81, 85"Spirits, Shamans, and Nightmare DeathSurvivor Stress in a Hmong RefLgee", 398Spoken and Written Language, 296Sre, 219Sronfe, L.A., 412-415Stockton, 184Storlie, Rebecca, 119Strecker, David, 220, 233, 235"Stress in Families of Children Who Have IngestedPoisons", 415"Structured Language Acquisition'", 328Structured Tests in English Language, 107,109Studies in Tai and Mon-Khmer Phonetics andPhonology, 234"Studies of Japanese Migrants. 1. Mortalityfrom Cancer and Other Diseases AmongJapanese in the United States", 444"Study of Kinship Systems" (The), 71"Subject and Topic. A New Typology ofLanguage", 277Suddard, A., 399, 414"Sudden Adult Death in Asian Populations.The Case of the Hmong", 397, 444"Sudden Death Among Southeast Asian Refugees",396, 443"Sudden Night-time Death Among SoutheastAsians Too", 444"Sudden Nocturnal Death in Laotian Refugees",398"Sudden Nocturnal Death in Southeast AsianRefugees: A Progress Report", 397Sudden Unexpected Nocturnal DeathSyndrome (SUNDS), 335, 336, 347,379-395, 427, 428-429, 431-442,443n"Sudden, Unexpected Nocturnal DeathsAmong Southeast Asian Refugees", 398, 445Sweeney, Michael P., 177, 324, 324n, 329Sydney, Australia, 5, 69System for Rating Maternal-care Behavior, 413Tagolo, 383Takumi, Gerald, 424Tannen, Deborah, 285, 296, 297Taoism, 341Tapp, Nicholas, 5Taraldson, B., 400, 401, 414, 416Tasmania, 62Tauxe, R.V., 381, 396, 398, 442Taylor, W., 396, 443Teachout, Richard, 119Texas, 15, 177Thacker, S., 396, 443Thai, 13, 18, 36, 66, 136, 165, 279, 436Thai Dam, 10, 384Thai Medical Council, 436Thai Military Commander, 436, 438Thailand, xi, xiii, 4-5, 7-8, 12, 14, 28-31, 35,37, 47, 61, 66, 68, 73, 88, 179, 202, 293,305, 338, 379, 383, 387, 427-428, 431,433, 437-439, 443nMinistry of the Interior, 395, 436;4 1.Jo,

INDEX463National Research Council, 395, 436,Secretary General, 436Thao, 334Theo, Chue, 185Thao, Mao, 433Thao, Xang Yang, 377The Hmong in the West, xiiiThomas, Elaine, 160, 173Thompson, Sandra A., 237-238, 243, 265-266,269-271, 273-277Thongkum, 234"Three-month Observations of Feeding Situations,Six-month Observations of Feedingand Play Situations, Waiting Room ObservationScales", 414Tibetans, 74Tobin, J J., 394, 398"Tonal Icons in Bini", 236Toos Tub, 11, 15Traditional Tribal and Peasant Economies, 70Training for Hmong Liaisons, 113"Transactional Basis of Influence; White MenAmong Naskapi Indians" (The), 172"Transition in Perspective A Study of theHmong of Isla Vista", 156Trejo, Francisco, 119Tribesmen and Peasants in North Thailand.Proceedings of the First Symposium of theTribal Research Centre, Chiangmai 1967,94Trillin, Calvin, 152, 157Tsao, Feng-Fu, 264, 267, 270, 274, 277Ts'ao Ts'tn-yun, 220, 224, 235Tsung Yun, 75, 81 85Tswb Tchoj, 91Tupa, 75, 76Tumy, Bud, 177Twice a Minority, 143Twin Cities Hmong Community Survey, 187Twin Cities, Minnesota, 7, 9-10, 152, 183-184, 194, 321, 399"U S Journal Fairfield, Iowa Resettling theYangs", 157UNHCR Monthly Statistics, 184Un Village Hmong Vert du Haut Laos, 21, 84,95, 173Under the Ancestors' Shadow. Chinese Cultureand Personality, 94United and Equal: The Progress of China'sMinority Nationalities, 85United Nations Crop Replacement Project, 68United Nations Decade for Women, 101United Nations Development Program, 39, 40United Nations High Commissioner for Refugees(UNHCR), xii, 13, 23, 39, 40, 179,384-386, 396, 398Bangkok, 437, Monthly Statistics, 184;Workshop on Integration of Refugeesfrom Indochina in Countries of Resettlement,101United States Centers for Disease Control, 380,381, 398,427,428-432,442, 442n, 443n,445United States Department of Education, RefugeeMaterials Center, 177United States Office of Refugee Resettlement,175, 177, 435, 442, 445United States State Department, 384"Unity and Division in Hmong Social Categoriesin Thailand", 70University of California, 155University of Chicago, 233University of Illinois School of Medicine, 381University of Long Beach, 155University of Minnesota, xi, xn, xiii, 105,106,117, 175, 187Agricultural Extension Service, 204;College of Liberal Arts, xii, ComputerCenter, 424, Center for Urban and RegionalAffairs, xii, xiii, 177; MedicalSchool, 442; College of Education, 107;Southeast Asian Refugee Studies Protect,xii, SARS Newsletter, 152, 156University of Paris, xi)University of Washington, 171, 177, 396"Unpackaging Literacy", 297Van Ntxhais Xyooj, 19nVang, Blang, 119 ,Vang, Chia, 4i, 52, 124, 346Vang, Dang, 177Vang Lia, 385, 396Vang Neng, 384-385, 396Vang, Pakoua, 377Vang Pao, 9, 12, 16, 179, 338Vang, Thongsai, 177Vang, Tong, 242, 254, 255, 258, 260Vang, Tou Fu, 52, 131, 348Vang, Vang, 19, 20n, 276Vang Viang, 31Vangyi, Shur Vang, 100, 177Vaughn, B.E., 400, 401, 412, 416Vaughn, V C , 415"Verb Serialization in Kwa-Type Languages",243Vernon, A , 396, 430, 443, 445Victoria, 62Vientiane, 14, 15, 31, 32Vietnam (Vietnamese), 5, 27, 28, 33, 88, 101,380, 430Vue Mai, 396, 4384 5 v

R464 THE HMONG IN TRANSITIONWalker, Constance, 119Wall, S , 412-413Walter, Ingrid, 101, 120Wang, Fu-shih, 236Wang Shu-tang, 75, 85Washburn-Dreyer, Sharon, 119Washington, 204, 207Washington State, 443nWaters, E., 412-413Weins, Herold J , 42, 53Weinstein, Gail, 215Weiser, Ann, 243Wells, Miriam J , 170, 173Welmers, William E., 237, 243Weng Tu-chien, 74, 85Wescott, R., 221, 224, 236Westermeyer, Joseph, 346-348White Bear Lake, 107"White Hmong", 46, 79, 219-233White Hmong English Dictionary, 84n, 234,259, 277, 396n, 397Whitelock, Dons, 20nWhiteman, Marcia Farr, 295, 297Whitmore, J. 131Whitt, J K , 401, 416"Who Are The Karen, and If So, Why? KarenEthr ohistory and a Formal Theory of Ethnicity",95"Why Has Death Stalked the Refugees?", 397,444Wisconsin, 207Withey, S.B , 125, 130Wolof, 280World Vision International, 437World Vision of Thailand, 396"Wrist-stringing", 64Writing: Th2 Nature, Development, andTeaching of Written Communication, 295,297"Written Languages for China's Minorities",84Xiang Xi, 79Xiang Xi Miaozu-Tunazu Autonomous Prefecture,76Xieng Khouang Province, 16, 20n, 31-32,36-37, 39, 145Xyool, Qhov, 79Xyoot, Xab, 233Ya Shong Leu, 338Yai, Kia Mcua, 119Yai, Kuam, 233Yang, 16, 20nYang Dao, 8, 18, 19, 20n, 22, 52, 179, 245,253, 260, 276Yang, Gaoly, 119Yao, 33, 39, 76, 79, 92Yao the Orphan, 255Yaqui, 42, 51Yaqui Indians, 41Yaqws: A Cultural History (The), 53Yarrow, L., 412, 416Yi, 76Yin, 76, 80Yin Ming, 85Yip, R., 419-421, 425You Take the High Node and I'll Take the LowNode: Papers from the Comparative SyntaxFestival, 243Yunnan, 76, 77, 79, 82Yunnan-Guizhou Plateau, 75Zegiob, L , 412, 416"Zero NP-Anaphora in Russian A Case ofTopic-Prominence", 276Zhaotong-Weining, 80Zhuang, 76Ziegler, E.E., 418, 425Zos Key Xov Hlau, 19n45dKS