Art Un ticle I.1 ited Sta In the ates News - Woodring College of ...

"Those who arrive by age 12 or 13 make a quick transition to English—that's the dividingline," says Rumbaut, who has studied language assimilation in the United States for threedecades. "It's a piece of cake for those who arrive much earlier on, because of the dominanceof English in every medium in the United States, from video to the Internet. English wins."In addition, Spanish isn't the only fast-growing language in the United States: Chinese,Tagalog, Vietnamese, and Arabic have also seen impressive gains since 1990, whileEuropean languages that were once common in the United States (such as French, German,and Polish) are becoming less prevalent. But Rumbaut says that the children of newimmigrants speaking other languages will almost invariably turn to English as their primarytongue."The fate of all these languages is to succumb to rapid assimilation," says Rumbaut. "Theidea that the United States will devolve into riots and become Quebec unless everybodyspeaks English and English only, is absolutely not true. Demography will take care of theproblem itself—it is not really a policy issue."More Than One-Half of All Immigrants 'Very Proficient' in EnglishThe number of Americans speaking a language at home other than English has more thandoubled since 1980, reflecting the influx of millions of immigrants to the United States inrecent decades, particularly Spanish-speaking immigrants from Latin America. About 31million U.S. residents speak Spanish at home—easily making it the second-most spokenlanguage in the country.The idea that speaking languages other than English hinders full participation of U.S.citizenship has substantial public support. For example, in a Los Angeles Times poll ofCalifornia voters after their 1998 vote to end bilingual education in that state, three out ofevery four agreed with the statement: "If you live in America, you need to speak English." 4 Amajority of Hispanics share this attitude, according to a new Pew Hispanic Center survey: 57percent of Hispanics say that "immigrants have to speak English to say that they are part ofAmerican society." 5But a majority of those who speak languages other than English at home report themselvesalready very proficient in English, according to 2004 data from the American CommunitySurvey (see Table 1). Fewer than 50 percent of people who use Spanish or another non-English language at home speak English less than "very well"—including 48 percent of thosewho speak Spanish. Almost 70 percent of U.S. adults ages 18 to 64 who spoke Spanish in thehome said they also spoke English either "well" or "very well." 6Table 1Language Spoken at Home by U.S. Residents Ages 5 and Older, 2004Primary language© 2008 Dr. Catherine CollierAll Rights ReservedEstimate(millions)Speaks English lessthan "very well"In millionsPercentageLanguage other than English 49.63 22.31 44.94%Spanish 30.52 14.64 47.96%Other Indo-European language 9.63 3.32 34.42%2

Asian/Pacific-Islander language 7.61 3.10 49.50%Other languages 1.86 0.54 29.25%Source: U.S. Census Bureau, United States: Selected Social Characteristics: 2004 (2004)."The last people you have to tell that English is important are immigrants," says Rumbaut."English is already the de facto language of the country and of the world."The Transition to English AcceleratesU.S. immigrants are making the transition to speaking English much more quickly than didpast immigrants. Historically, this transition took three generations, with adult immigrantswho often did not learn English, children who were bilingual in English and their parents'language, and a third generation that spoke English almost exclusively.Today, however, more first- and second-generation Americans are becoming fluent inEnglish. In a study that followed more than 5,200 second-generation immigrant children inthe Miami and San Diego school systems, Rumbaut and Princeton University professor ofsociology Alejandro Portes found that 99 percent spoke fluent English and less than one-thirdmaintained fluency in their parents' tongues by age 17. 7Indeed, Census Bureau data for 2000 show that more than 67 percent of the 6.5 million U.S.children ages 5 to 17 who spoke Spanish in their homes also spoke English "very well,"while 86 percent spoke English either "very well" or "well" (see Table 2).Table 2Ability to Speak English for U.S. Residents Ages 5-17, 2000Primary language© 2008 Dr. Catherine CollierAll Rights ReservedEstimate(millions)English speaking levelVery well(millions)%Well(millions)Spanish 6.53 4.40 67.34 1.24 32.66Other Indo-European language 1.54 1.19 77.67 0.23 22.33Asian/Pacific-Islander language 1.15 0.80 69.56 0.24 30.44Other languages 0.31 0.25 80.60 0.42 19.40Source: U.S. Census Bureau, 2000 Supplementary Survey Summary Tables, table P035 (2006).Similar percentages of children who spoke a language other than English or Spanish at homealso spoke English "well" or "very well." The Census data also show that these children arelearning English at higher rates than either their parents or grandparents.Another study conducted by Rumbaut shows that more than 73 percent of second-generationimmigrants in Southern California who have two foreign-born parents prefer to speakEnglish at home instead of their native tongue (see Figure 1). 8 By the third generation, morethan 97 percent of these immigrants—Mexican, Salvadoran, Guatemalan, Filipino, Chinese,Korean, and Vietnamese—prefer to speak only English at home."The single most important indicator of how readily any speaker will switch to English andbecome fluent is the age of arrival, followed by level of education and length of stay," says%3

Article I.2 Are Signed Languages "Real"Languages?By Dr. Laura Ann PetittoEvidence from American Sign Language and Langue des SignesQuébécoiseReprinted from Reprinted from: Signpost (InternationalQuarterly of the Sign Linguistics Association), vol. 7, No. 3.1-10. French and Spanish translations available on request.In the fall of 1993, a Mr. Gilles Read sent me a letter. Mr.Read is the General Manager of the Montreal MetropolitanDeaf Community Center. He wrote to me seeking adocument that addressed the question of whether signedlanguages were "real" languages. At the time, I was surprisedto discover that there did not exist a document that simply and directly asked and answeredhis question, especially in one single source. Therefore, I wrote Mr. Read a letter, of whichan expanded version appears below. To be sure, I could not possibly have summarized forMr. Read all studies of signed languages to date, as fortunately thousands now exist. Nor wasit possible to summarize the many important studies of signed languages that have beenundertaken outside of North America. Indeed, Mr. Read was preparing to attend meetings inthe Provincial government of Québec and he needed a document that specifically addressedthe linguistic status of the two main signed languages used in these regions, in particularAmerican Sign Language (ASL) and Langue des Signes Québécoise (LSQ). Thus, my goalwas to draw together the disparate studies on signed languages that have been conducted in avery clear way, and to show how they bear on the critical question, "are signed languages'real' languages?" Although the examples are drawn from studies of ASL and LSQ, thegeneral structure of the arguments presented here are applicable to arguments for the "reallanguage" status of other natural signed languages around the world.IntroductionThis paper summarizes over thirty years of scientific research, which, in one form or another,addresses the following question: Are the natural signed languages that are used by manydeaf persons throughout the world "real" languages? Below I demonstrate that signedlanguages are indeed "real" languages. I do so by drawing evidence from three categories ofscientific research, including (i) Linguistic analyses of natural signed languages, (ii)Sociolinguistic analyses of natural signed languages, and, crucially, (iii) Biological analysesof the status of natural signed languages in the human brain.© 2008 Dr. Catherine CollierAll Rights Reserved5

identical linguistic properties found in both ASL and in spoken language. For example,studies of Langue des Signes Québécoise (LSQ), the signed language used in Québec andelsewhere in Canada by culturally French Deaf persons, have revealed that it is a fullyautonomous natural language with a unique etymology (history of the derivation of its signs):It is a grammatically distinct language from ASL and it is a grammatically distinct languagefrom the signed language used in France. Indeed, LSQ is a complete and richly complexlanguage. It has "phonological," morphological, syntactic, discourse, pragmatic, and semanticstructures that are entirely equal in complexity and richness to that which is found in anyspoken (or signed) language (e.g., Petitto, 1987b&c; Petitto & Charron, 1988; Petitto,Charron, & Briére, 1990; see also Brentari, 1991; Lacerte, 1991; Miller, 1991). Takentogether, the scientific study of the linguistic status of signed languages has demonstratedthat complete human languages are not restricted to the speech channel. Signed languagespossess all of the linguistic features that have been identified as being the essential, universalfeatures of the world's spoken languages.Sociolinguistic studies have been conducted that examine the social and cultural conditionsunder which natural signed languages are used. These studies have revealed that signedlanguages exhibit the identical sociolinguistic patterns observed in spoken languages. Likespoken languages, signed languages undergo change over time, and they demonstrate thesame types of historical change that are seen in spoken languages. For example, signedlanguages exhibit the same processes of expanding their lexicons (the set of words or signs ina language) through sign borrowings, loan signs, and compounding (e.g., see Battison, 1978;Klima & Bellugi, 1979; Woodward, 1976; Woodward & Erting, 1975). As is commonamong users of a particular spoken language (e.g., English, French), signed language userswithin distinct signed language communities (e.g., ASL or LSQ communities) demonstrateregional accents in their signing, lexical (=sign) variation depending on socio-economicstatus, and lexical variation depending on the language user's age, sex, and educationalbackground (e.g., Battison, 1978). Further, users of distinct signed languages abide bylanguage-specific (tacit) rules of politeness, turn-taking, and other discourse (conversational)patterns found in spoken languages (e.g., Hall, 1983; Wilbur & Petitto, 1981, 1983). As canbe seen with users of particular spoken languages, users of particular signed languages oftenshare beliefs, attitudes, and customs with others who use the same language, binding theminto a distinct cultural group--one that is not simply the signed version (or Deaf translation)of the majority spoken culture within which they reside (e.g., Hall, 1989; Lane, 1989; Padden& Humphries, 1988; Rutherford, 1988). In Canada, for example, the ASL and the LSQsigned language communities, respectively, are bound by a distinct collection of beliefs andattitudes that are expressed in a variety of ways. These include, for example, (a) the existenceof poetry in ASL or in LSQ, (b) humor and jokes in ASL or in LSQ, (c) indigenous artisticexpression through theatre and dance in ASL or in LSQ, (d) indigenous meeting customs andtraditions (e.g., witness the many Deaf social clubs in the ASL or the LSQ community, aswell as in the other signed language communities throughout the world), (e) Deaf religiousorganizations in ASL or in LSQ, (f) Deaf sports events in ASL or in LSQ, (g) Deafnewspapers and other publications for the ASL or the LSQ communities, and so forth. Insummary, sociolinguistic studies of natural signed languages have determined that theydemonstrate strikingly similar patterns of change, variation, and social and cultural use thatare common to the world's spoken languages.7© 2008 Dr. Catherine CollierAll Rights Reserved

Biological analyses of the status of natural signed languages in the humanbrainThough there is now wider acceptance that signed languages are "real" languages based onlinguistic and socio-cultural grounds, a persistent and powerful misconception remains. Themisconception can be summarized as follows: "Spoken language is fundamentally better thansigned language; sign is 'inferior' ('secondary') to speech. The notion that signed languagesare real languages, but somehow "inferior" or lower than the higher status spoken languages,is reminiscent of the late 19th century division of spoken languages into "high" and "low"languages. (Here, languages deemed "high" were those used in Western Europe, and "low or"primitive" languages were those used elsewhere by aboriginal peoples.) Though subsequentscientific studies have shown that the high-low classification of spoken languages is whollyfallacious, similar attitudes regarding spoken versus signed languages have not been subjectto the same sorts of scientific scrutiny. The above view about signed languages runsespecially deep because it invokes biology. At the heart of the misconception is the notionthat signed languages are biologically inferior to spoken languages. Why is this so? Theanswer involves a three-tiered set of related assumptions: First, a common quip is "mostpeople speak, so speaking must be better." I call this the "more is better" assumption. Second,drawing from the observation that "most people speak," people have further assumed that thismust "prove" that speech, alone, has been selected for over the development or evolution ofthe species (or, in "phylogeny"). Third, the assumption that speech has been selected for overhuman evolution, has implicitly been used to support the core, critical assumption about thebiological foundations of human language: the brain must be neurologically set for speechearly in the developmental history of individual human organisms (or, in "ontogeny"). Thisthird assumption has generally been regarded as being true because of the remarkableregularities observed in very early spoken language acquisition. Noting such universalregularities in, for example, the timing of the onset of speaking children's early vocalbabbling and first words, researchers have concluded that the brain and its' maturation mustbe attuned to perceiving and producing spoken language input (per se) in early life. To besure, a typical answer to the question "how does early human language acquisition begin?" isthat it is the result of, and wholly determined by, the development of the anatomy of thevocal tract and the neuroanatomical and neurophysiological mechanisms involved in themotor control of speech production (e.g., Locke, 1983; MacNeilage & Davis, 1990;MacNeilage, Studdert-Kennedy, & Lindblom, 1985; van der Stelt & Koopmans-van Beinum,1986). An implicit assumption that underlies such views is that spoken languages are bettersuited to the brain's maturational needs in development. Put another way, the view of humanbiology that underlies the prevailing third assumption is that the human brain is "hardwired"for speech and that speech is "special" or "privileged." On this view of the brain, then, signedlanguages can only be regarded as being "biologically" inferior to (or "lower" than) spokenlanguages. By extension, many educators and researchers, alike, have assumed that speech isbetter in order to achieve "normal" language acquisition.© 2008 Dr. Catherine CollierAll Rights Reserved8

Is there any evidence in support of the alleged "inferior" biological statusof signed languages?Surprisingly, with the exception of the studies reported below, the critical studies required toevaluate the above assumptions have not been conducted. As noted above, most allcontemporary answers to questions about the biological foundation of language have beenbased on the core assumption that very early language acquisition is tied to speech. There is,however, a fatal flaw with this assumption: Given that only languages utilizing the speechmodality are studied, it is in principle, a priori, impossible to find data that would doanything but support this hypothesis. Only when a modality other than speech is analyzedcan any generalization about the brain's predisposition for speech be evaluated, and,therefore, whether signed languages have the same or different status in the human brain.The critical ontogenetic evidence regarding the biological status of naturalsigned languageOver the past 12 years, research in my own laboratory has been directed at understanding thebiological foundations of human language. My central aim has been to discover the specificbiological and environmental factors that together permit early language acquisition to beginin our species. Studies of very early signed language acquisition offer an especially clearwindow into the biological foundations of all of human language (be it spoken or signed), aswell as its biological status in the brain. Spoken and signed languages utilize differentperceptual modalities (sound versus sight), and the motor control of the tongue and hands aresubserved by different neural substrates in the brain. Comparative analyses of theselanguages, then, provide key insights into the specific neural architecture that determinesearly human language acquisition in our species. If, as has been argued, very early humanlanguage acquisition is under the exclusive control of the maturation of the mechanisms forspeech production and/or speech perception, then spoken and signed languages should beacquired in radically different ways. At the very least, fundamental differences in the timecourse and nature of spoken versus signed language acquisition would suggest that each maybe processed and represented in different ways, presumably due to their differing biologicalstatus in the human brain. To investigate these issues, I have conducted numerouscomparative studies of children acquiring spoken languages (English or French) and childrenacquiring signed languages (American Sign Language or Langue des Signes Québécoise),ages birth through 36 months.The empirical findings from my cross-linguistic and cross-modal studiesare clear(i) Deaf children who are exposed to signed languages from birth, acquire these languages onan identical maturational time course as hearing children acquire spoken languages. Deafchildren acquiring signed languages from birth do so without any modification, loss, or delayto the timing, content, and maturational course associated with reaching all linguisticmilestones observed in spoken language. Beginning at birth, and continuing through age 3© 2008 Dr. Catherine CollierAll Rights Reserved9

and beyond, speaking and signing children exhibit the identical stages of languageacquisition. These include the (a) "syllabic babbling stage" (7-10 months, approx.) as well asother developments in babbling (e.g., "variegated babbling," ages 10-12 months, and "jargonbabbling," ages 12 months and beyond; Petitto, 1984, 1987a&b; Petitto & Marentette,1991a), (b) "first word stage" (11-14 months, approx.; e.g., Petitto, 1985, 1986, 1988, 1992,1993b; Petitto & Marentette, 1991b; Petitto, Costopoulos, & Stevens, in preparation), and(c)"first two-word stage" (16-22 months, approx.; Petitto, 1987a; Petitto & Marentette,1991b). Though some researchers have claimed that "first signs" are acquired earlier than"first words," subsequent analyses have revealed that the claim is wholly unfounded1.Surprising similarities are also observed in deaf and hearing children's timing onset and useof gestures. Signing and speaking children produce strikingly similar pre-linguistic (9-12months) and post-linguistic communicative gestures (12-48 months; e.g., Petitto, 1984,1987a, 1992). They do not produce more (or more elaborate) gestures, even though linguistic"signs" (identical to the "word") and communicative gestures reside in the same modality,and even though some signs and gestures are formationally and referentially similar. Instead,deaf children consistently differentiate linguistic signs from communicative gesturesthroughout development, acquiring and using each in the same ways observed in hearingchildren (see Petitto, 1992). Signing children exhibit highly similar patterns of latergrammatical development as well (ages 22-36 months, approx., and beyond), includingsystematic morphological and syntactic developments (e.g., "over-regularizations," negation,question formation, and so forth; e.g., Petitto, 1984, 1987a; see also Newport & Meier,1985). Throughout development, signing and speaking children exhibit remarkably similarcomplexity in their utterances. For example, analyses of young ASL and LSQ children'ssocial and conversational patterns of language use over time, as well as the types of thingsthat they "talk" about over time (its' semantic and conceptual content, categories, andreferential scope), have demonstrated unequivocally that their language acquisition followsthe identical path as is observed in age-matched hearing children acquiring spoken language(Charron & Petitto, 1991; Petitto, 1992; Petitto & Charron, 1988). (ii) Hearing childrenexposed to both signed and spoken languages from birth (e.g., one parent signs and the otherparent speaks) demonstrate no preference for speech whatsoever, even though they can hear.Instead, they acquire both the signed and the spoken language to which they are beingexposed on an identical maturational timetable (the timing of the onset of all linguisticmilestones occurs at the same time in both the signed and spoken modalities). In addition,such children acquire the signed and spoken languages to which they are being exposed (e.g.,ASL and English, or, LSQ and French) in the same manner that other children acquire twodifferent spoken languages from birth in a "bilingual" home, for example, one with spokenFrench and spoken English (Petitto, 1985, 1986, 1993b; see especially, Petitto, 1993a, andPetitto, Costopoulos, & Stevens, in preparation). (iii) Hearing children who are exposedexclusively to signed languages from birth through early childhood (i.e., they receive little orno systematic spoken language input whatsoever), achieve each and every linguisticmilestone (manual babbling, "first signs," "first two-signs," and so forth) in signed languageon the identical time course as has been observed for hearing children acquiring spokenlanguage and deaf children acquiring signed language. Thus, entirely normal languageacquisition occurred in these hearing children (a) without the use of auditory and speechperception mechanisms, and (b) without the use of the motoric mechanisms for theproduction of speech (Petitto, a & b; Petitto, Costopoulos, & Stevens, in preparation).10© 2008 Dr. Catherine CollierAll Rights Reserved

Significance of biological studies of early signed and spoken languageacquisitionDespite the modality differences, signed and spoken languages are acquired in virtuallyidentical ways. The differences that were observed between children acquiring a signedlanguage versus children acquiring a spoken language were no greater than the differencesobserved between hearing children learning one spoken language, say, French, versusanother, say, Finnish. Such findings cast serious doubt on the core hypothesis in very earlyspoken language acquisition: that the maturation of the mechanisms for the production and/orperception of speech, exclusively determines the time course and content of early humanlanguage acquisition. These findings further challenge the hypothesis that speech (and sound)is critical to normal language acquisition, and they challenge the related hypothesis thatspeech is uniquely suited to the brain's maturational needs in language ontogeny. If speech,alone, were neurologically set or "privileged" in early brain development, then, for example,the hearing infants exposed to both speech and sign from birth might be expected to attemptto glean every morsel of speech that they could get from their environment. Faced implicitlywith a "choice" between speech and sign, the very young hearing infant in this context mightbe expected to turn away from the sign input, favoring instead the speech input, and therebyacquire signs differently (e.g., later). Similarly, deaf and hearing infants exposed only tosigned languages from birth should have demonstrated grossly abnormal patterns of languageacquisition. None of this happened. What is most interesting about these research findings isthat the modality "switch" can be "thrown" after birth regarding whether a child acquireslanguage on the hands or the language on the tongue. Such findings have led me to propose anew way to construe human language ontogeny (see especially Petitto, 1993a&b). Speechand sound are not critical to human language acquisition. Instead, there appears to be astunning, biologically-based equipotentiality of the modalities (spoken and signed) to receiveand produce natural language in ontogeny (Petitto, 1994). The only way that signed andspoken languages could be acquired with such startling similarity, is if the brain's of allnewborns possess a mechanism that is sensitive to aspects of the structural regularities ofnatural language, irrespective of the input modality. Rather than being exclusively"hardwired" for speech or sound, our species appears to "hardwired" to detect aspects of thepatterning of language (specifically, aspects of its structural and prosodic regularities; seePetitto, 1993a&b). If the environmental input contains the requisite patterns unique to naturallanguage, human infants will attempt to produce and to acquire those patterns, irrespective ofwhether the input is signed or spoken. (For a discussion of the specific neural substrates thatunderlie this capacity in ontogeny, as well as their possible roots in phylogeny, see especiallyPetitto, 1993a&b.) In summary, the present findings prove wholly false assumptions aboutthe "biological inferiority" of signed languages relative to spoken languages. Signed andspoken languages are acquired in the same ways, and on the same maturational time course.With regard to the brain and human biology, this indicates that signed and spoken languagesengage the same brain-based mechanisms in very early language acquisition.© 2008 Dr. Catherine CollierAll Rights Reserved11

Conclusion: Are Signed Languages Real Languages?Results from studies of early language acquisition provide especially strong evidencerelevant to assessing whether signed languages are real languages. Here we see clearly thatthe prevailing assumption about the biological foundations of human language--indeed, thevery assumption upon which notions of the alleged biological superiority of speech over signrests--is not supported when the relevant studies are conducted. Specifically, no evidence wasfound that the newborn brain is neurologically set exclusively for speech in early languageontogeny. No evidence was found that speech is biologically more "special," more"privileged," or "higher" in status than sign in early language ontogeny. Instead, the key,persistent research finding to emerge is this: The biological mechanisms in the brain thatunderlie early human language acquisition do not appear to differentiate between spokenversus signed language input. Both types of input appear to be processed equally in the brain.This provides powerful evidence that signed and spoken languages occupy identical and,crucially, equal biological status in the human brain. In summary, I have outlined three linesof scientific research on the status of natural signed languages relative to spoken languages,including Linguistic, Sociolinguistic, and Biological research. All three types of researchprovide powerful converging evidence that natural signed languages are "real languages,"demonstrating all of the features characteristic of language in our species. Thus, there is noscientific reason to exclude ASL or LSQ, or any other natural signed language, from thefamily of languages used by human beings. That signed languages are real languages cannow be considered to be an unequivocal scientific fact.AcknowledgementsI thank Kevin Dunbar, Jamie MacDougall and the researchers (students and staff) whoassisted in the studies discussed within, and, crucially, the Deaf and hearing families who solovingly gave their time and support to these studies. I also thank the following for fundingthis research: Natural Science and Engineering Council of Canada, the MacDonnell-PewCentre Grant in Cognitive Neuroscience, and the McGill-IBM Cooperative Project.Footnotes1 Most all of the claims regarding the earlier onset of first signs over first words stem fromone group of researchers (e.g., Bonvillian, Orlansky, Novack, & Folven, 1983; Folven &Bonvillian, 1991). Recently, a second group of researchers (Meier & Newport, 1991) hasbased their theoretical arguments in support of a possible "sign advantage" largely onBonvillian et al's claims. The subjects in Bonvillian et al.'s studies are reported to haveproduced their "first sign" at a mean age of 8.2 months (a date that differs from hearingchildren’s' first words, which occurs at approximately 11 months). However, in their studies,the infants' "first signs" were not required to be used in a meaningful or referential way.Instead, infant manual productions containing "recognizable adult phonetic forms" withoutany "referential" content, were attributed "sign" status (=lexical or word status). What theywere actually measuring, however, is clear from the spoken language acquisition literatureand from my own work: In spoken language, we see that hearing infants around ages 7-10months begin production of "syllabic babbling," whereupon they produce vocal productionscontaining recognizable adult phonetic forms without any referential content. Similarly, in© 2008 Dr. Catherine CollierAll Rights Reserved12

signed language, we see that sign-exposed infants around ages 7-10 months also beginproduction of "syllabic babbling," albeit on their hands (Petitto & Marentette, 1991a). Thus,it would appear that Bonvillian et al., have mislabeled genuine instances of manual babblingin sign-exposed infants as being "first signs." Recall that their date for the age of first signs is8.2 months--which is smack in the middle of infants' manual (and/or vocal) babbling stage(see also Petitto, 1988, for a discussion of other methodological considerations associatedwith this research, including the over attribution of linguistic sign status to these infants nonlinguisticcommunicative gestures).© 2008 Dr. Catherine CollierAll Rights Reserved13

Article I.3 The Interpreter: Has a remote Amazoniantribe upended our understanding of Language?By John ColapintoOne morning last July, in the rain forest of northwestern Brazil, Dan Everett, an Americanlinguistics professor, and I stepped from the pontoon of a Cessna floatplane onto the beachbordering the Maici River, a narrow, sharply meandering tributary of the Amazon. On the bankabove us were some thirty people--short, dark skinned men, women, and children--someclutching bows and arrows, others with infants on their hips. The people, members of a huntergatherertribe called the Pirahã, responded to the sight of Everett--a solidly built man of fifty-fivewith a red beard and the booming voice of a former evangelical minister--with a greeting thatsounded like a profusion exotic songbirds, a melodic chattering scarcely discernible, to theuninitiated, human speech. Unrelated to any other extant tongue, and based on just eightconsonants and three vowels, Pirahã as one of the simplest sound systems known. Yet itpossesses such a complex array of tones, stresses, and syllable lengths that its speakers candispense with their vowels and consonants altogether and sing, hum, or whistle conversations. Itis a language so confounding to non-natives that until Everett and his wife, Keren, arrived amongthe Pirahã, as Christian missionaries, in the nineteen-seventies, no outsider had succeeded inmastering it. Everett eventually abandoned Christianity, but he and Keren have spent the pastthirty years, on and off, living with the tribe, and in that time they have learned Pirahã as noother Westerners have.“Xaoi hi gaisai xigiaihiabisaoaxai ti xabiihai hiatiihi xigio hoihi,” Everett said in the tongue'schoppy staccato, introducing me as someone who would be "staying for a short time" in thevillage. The men and women answered in an echoing chorus, "Xa6i hi go6 kaisigiaihlxapagaiso."Everett turned to me. "They want to know what you’re called in 'crooked head’.'" "Crookedhead" is the tribe's term for any language that is not Pirahã, and it is a clear pejorative. ThePirahã consider all forms of human discourse other than their own to be laughably inferior, andthey are unique among Amazonian peoples in remaining monolingual. They playfully tossed myname back and forth among themselves, altering it slightly with each reiteration, until it becamean unrecognizable syllable. They never uttered it again, but instead gave me a little Pirahã name:Kaaxáoi, that of a Pirahã man, from a village downriver, whom they thought I resembled. "That'scompletely consistent with my main thesis about the tribe," Everett told me later. "They rejecteverything from outside their world. They just don't want it, and it's been that way since the daythe Brazilians first found them in this jungle in the seventeen-hundreds."Everett, who this past fall became the chairman of the Department of Languages, Literature, andCultures at Illinois State University, has been publishing academic books and papers on thePirahã: (pronounced pee-da-HAN) for more than twenty-five years. But his work remainedrelatively obscure until early in 2005, when he posted on his Web site an article titled "CulturalConstraints on Grammar and Cognition in Pirahã," which was published that fall in the journalCultural Anthropology. The article described the extreme simplicity of the tribe's livingconditions and culture. The Pirahã, Everett wrote, have no numbers, no fixed color terms, noperfect tense, no deep memory, no tradition of art or drawing, and no words for "all," "each,""every," "most," or "few" --terms of quantification believed by some linguists to be among the© 2008 Dr. Catherine CollierAll Rights Reserved14

common building blocks of human cognition. Everett's most explosive claim, however, was thatPirahã displays no evidence of recursion, a linguistic operation that consists of inserting onephrase inside another of the same type, as when a speaker combines discrete thoughts ("the manis walking down the street," "the man is wearing a top hat") into a single sentence ("The manwho is wearing a top hat is walking down the street"). Noam Chomsky, the influential linguistictheorist, has recently revised his theory of universal grammar, arguing that recursion is thecornerstone of all languages, and is possible because of a uniquely human cognitive ability.Steven Pinker, the Harvard cognitive scientist, calls Everett's paper "a bomb thrown into theparty." For months, it was the subject of passionate debate on social-science blogs and Listservs.Everett, once a devotee of Chomskyan linguistics, insists not only that Pirahã is a "severecounterexample" to the theory of universal grammar but also that it is not an isolated case. "Ithink one of the reasons that we haven't found other groups like this," Everett said, "is becausewe've been told, basically, that it's not possible." Some scholars were taken aback by Everett'sdepiction of the Pirahã as a people of seemingly unparalleled linguistic and cultural primitivism."I have to wonder whether he's some Borgesian fantasist, or some Margaret Mead being stitchedup by the locals," one reader wrote in an e-mail to the editors of a popular linguistics blog.I had my own doubts about Everett's portrayal of the Pirahã shortly after I arrived in the village.We were still unpacking when a Pirahã boy, who appeared to be about eleven years old, ran outfrom the trees beside the river. Grinning, he showed off a surprisingly accurate replica of thefloatplane we had just landed in. Carved from balsa wood, the model was four feet long and hada tapering fuselage, wings, and pontoons, as well as propellers, which were affixed with smallpieces of wire so that the boy could spin the blades with his finger. I asked Everett whether themodel contradicted his claim that the Pirahã do not make art. Everett barely glanced up. "Theymake them every time a plane arrives," he said. "They don't keep them around when there aren'tany planes. It’s a chain reaction, and someone else will do it, but then eventually it will peterout." Sure enough, I later saw the model lying broken and dirty in the weeds beside the river. Noone made another one during the six days I spent in the village.In the wake of the controversy that greeted his paper, Everett encouraged scholars to come to theAmazon and observe the Pirahã for themselves. The first person to take him up on the offer wasa forty-three-year-old American evolutionary biologist named Tecurnseh Fitch, who in 2002 coauthoredan important paper with Chomsky and Marc Hauser, an evolutionary psychologist andbiologist at Harvard, on recursion. Fitch and his cousin Bill, a sommelier based in Paris, weredue to arrive by floatplane in the Pirahã village a couple of hours after Everett and I did. As theplane landed on the water, the Pirahã, who had gathered at the river, began to cheer. The twomen stepped from the cockpit, Fitch toting a laptop computer into which he had programmed aweek's worth of linguistic experiments that he intended to perform on the Pirahã. They werequickly surrounded by curious tribe members. The Fitch cousins, having traveled widely togetherto remote parts of the world, believed that they knew how to establish an instant rapport withindigenous peoples. They brought their cupped hands to their mouths and blew loon calls backand forth. The Pirahã looked on stone-faced. Then Bill began to make a loud popping sound bysnapping a finger of one hand against the opposite palm. The Pirahã remained impassive. Thecousins shrugged sheepishly and abandoned their efforts."Usually you can hook people really easily by doing these funny little things," Fitch said later."But the Pirahã kids weren't buying it, and neither were their parents. Everett snorted. "It's notpart of their nature," he said. "So they're not interested."© 2008 Dr. Catherine CollierAll Rights Reserved15

A few weeks earlier, I had called Fitch in Scotland, where he is a professor at the University ofSt. Andrews. "I'm seeing this as an exploratory fact-finding trip," he told me. “I want to see withmy own eyes how much of this stuff that Dan is saying seems to check out."Everett is known among linguistics experts for orneriness and impatience with academicdecorum. He was born into a working-class family in Holtville, a town on the California-Mexicoborder, where his hard-drinking father, Leonard, worked variously as a bartender, a cowboy, anda mechanic. "I don't think we had a book in the house," Everett said. "To my Dad, people whotaught at colleges and people who wore ties were 'sissies'-all of them. I suppose some of that isstill in me." Everett’s chief exposure to intellectual life was through his mother, a waitress, whodied of a brain aneurysm when Everett was eleven. She brought home Reader's Digest condensedbooks and a set of medical encyclopedias, which Everett attempted to memorize. In high school,he saw the movie “My Fair Lady” and thought about becoming a linguist, because, he laterwrote, Henry Higgins's work "attracted me intellectually, and because it looked like phoneticianscould get rich."As a teen-ager, Everett played the guitar in rock bands (his keyboardist later became an earlymember of Iron Butterfly) and smoked pot and dropped acid, until the summer of 1968, when hemet Keren Graham, another student at El Capitan High School, in Lakeside. The daughter ofChristian missionaries, Keren was brought up among the Satere people in northeastern Brazil.She invited Everett to church and brought him home to meet her family. "They were loving andcaring and had all these groovy experiences in the Amazon," Everett said. "They supported meand told me how great I was. This was just not what I was used to." On October 4, 1968, at theage of seventeen, he became a born-again Christian. "I felt that my life had changed completely,that I had stepped from darkness into light-all the expressions you hear." He stopped using drugs,and when he and Keren were eighteen they married. A year later, the first of their three childrenwas born, and they began preparing to become missionaries.In 1976, after graduating with a degree in Foreign Missions from the Moody Bible Institute ofChicago, Everett enrolled with Keren in the Summer Institute of Linguistics, known as S.I.L., aninternational evangelical organization that seeks to spread God's Word by translating the Bibleinto the languages of pre-literate societies. They were sent to Chiapas, Mexico, where Kerenstayed in a hut in the jungle with the couple's children-by this time, there were three-- whileEverett underwent grueling field training. He endured fifty-mile hikes and survived for severaldays deep in the jungle with only matches, water, a rope, a machete, and a flashlight.The couple was given lessons in translation techniques, for which Everett proved to have a gift.His friend Peter Gordon, a linguist at Columbia University who has published a paper on theabsence of numbers in Pirahã, says that Everett regularly impresses academic audiences with ademonstration in which he picks from among the crowd a speaker of a language that he hasnever heard. 'Within about twenty minutes, he can tell you the basic structure of the languageand how its grammar works," Gordon said. "He has incredible breadth of knowledge, is really,really smart, knows stuff inside out.” Everett’s talents were obvious to the faculty at S.I.L., whofor twenty years had been trying to make progress in Pirahã, with little success. In October,1977, at S.I.L.’s invitation, Everett, Keren, and their three small children moved to Brazil, first toa city called Belem, to learn Portuguese, and then, a year later, to a Pirahã village at the mouth ofthe Maici River. “At that time, we didn’t know that Pirahã was linguistically so hard,” Kerentold me.There are about three hundred and fifty Pirahã spread out in small villages along the Maici andMarmelos Rivers. The village that I visited with Everett was typical: seven huts made by© 2008 Dr. Catherine CollierAll Rights Reserved16

propping palm-frond roofs on top of four sticks. The huts had dirt floors and no walls orfurniture, except for a raised platform of thin branches to sleep on. These fragile dwellings, inwhich a family of three or four might live, lined a path that wound through low brush and grassnear the riverbank. The people keep few possessions in their huts – pots and pans, a machete, aknife – and make no tools other than scraping implements (used for making arrowheads), looselywoven palm-leaf bags, and wood bows and arrows. Their only ornaments are simple necklacesmade from seeds, teeth, feathers, beds, and soda-can pull-tabs, which they often get from traderswho barter with the Pirahã for Brazil nuts, wood, and sorva (a rubbery sap used to make chewinggum), and which the tribe members wear to ward off evil spirits.Unlike other hunter-gatherer tribes of the Amazon, the Pirahã have resisted efforts bymissionaries and government agencies to teach them farming. They maintain tiny, weed-infestedpatches of ground a few steps into the forest, where they cultivate scraggly manioc plants. “Thestuff that’s growing in this village was either planted by somebody else or it’s what grows whenyou spit the seed out,” Everett said to me one morning as we walked through the village.Subsisting almost entirely on fish and game, which they catch and hunt daily, the Pirahã haveignored lessons in preserving meats by salting or smoking, and they produce only enoughmanioc flour to last a few days. (The Kawahiv, another Amazonian tribe that Everett hasstudied, make enough to last for months.) One of their few concessions to modernity is theirdress: the adult men wear T-shirts and shorts that they get from traders; the women wear plaincotton dresses that they sew themselves."For the first several years I was here, I was disappointed that I hadn't gone to a 'colorful' groupof people," Everett told me, "I thought of the people in the Xingu, who paint themselves and usethe lip plates and have the festivals. But then I realized that this is the most intense culture that Iwould ever have hoped to experience. This a culture that's invisible to the naked eye, but that isincredibly powerful, the most powerful culture of the Amazon, nobody has resisted change likethis in the history of the Amazon, and maybe of the world."According to the best guess of archeologists, the Pirahã arrived in the Amazon between tenthousand and forty thousand years ago, after bands of Homosapiens from Eurasia migrated to theAmericas over the Bering Strait. The Pirahã were once part of a larger Indian group called theMura, but had split from the main tribe by the time the Brazilians first encountered the Mura, in1714. The Mura went on to learn Portuguese and to adopt Brazilian ways, and their language isbelieved to be extinct. The Pirahã, however, retreated deep into the jungle.In 1921, the anthropologist Curt Nimuendajú spent time among the Pirahã and noted that theyshowed "little interest in the advantages of civilization" and displayed "almost no signs ofpermanent contact with civilized people."S.I.L. first made contact with the Pirahã nearly fifty years ago, when a missionary couple, Arloand Vi Heinrichs, joined a settlement on the Marmelos. The Heinrichses stayed for six and a halfyears, struggling to become proficient in the language. The phonemes (the sounds from whichwords are constructed) were exceedingly difficult, featuring nasal whines and sharp intakes ofbreath, and sounds made by popping or flapping the lips. Individual words were hard to learn,since the Pirahã habitually whittle nouns down to single syllables. Also confounding was thetonal nature of the language: the meanings of words depend on changes in pitch. (The words for"friend" and "enemy" differ only in the pitch of a single syllable.) The Heinrichses' task wasfurther complicated because Pirahã, like a few other Amazonian tongues, has male and femaleversions: the women use one fewer consonant than the men do.© 2008 Dr. Catherine CollierAll Rights Reserved17

“We struggled even getting to the place where we felt comfortable with the beginning of agrammar," Heinrichs told me. It was two years before he attempted to translate a Bible story; hechose the Prodigal Son from the Book of Luke. Heinrichs read his halting translation to a Pirahãmale. "He kind of nodded and said, in his way, “That's interesting,” Heinrichs recalled.”But therewas no spiritual understanding-it had no emotional impact. It was just a story." After sufferingrepeated bouts of malaria, the couple was reassigned by S.I.L. to administrative jobs in the cityof Brasilia, and in 1967 they were replaced with Steve Sheldon and his wife, Linda. Sheldonearned a master's degree in linguistics during the time he spent with the tribe, and he wasfrustrated that Pirahã refused to conform to expected patterns-as he and his wife complained inworkshops with S.I.L. consultants. “We would say, 'It just doesn't seem that there's any way thatit does X, Y, or Z,' " Sheldon recalled, "And the standard answer--since this typically doesn'thappen in languages-was 'Well, it must be there, just look a little harder.'"Sheldon's anxiety over his slow progress was acute. He began many mornings by getting sick tohis stomach. In 1977, after spending ten years with the Pirahã, he was promoted to director ofS.I.L. in Brazil and asked the Everetts to take his place in the jungle. Everett and his wife werewelcomed by the villagers, but it was months before they could conduct a simple conversation inPirahã. “There are very few places in the world where you have to learn a language with nolanguage in common," Everett told me. "It's called a monolingual field situation." He had beentrained in the technique by his teacher at S.I.L., the late Kenneth L. Pike, a legendary fieldlinguist and the chairman of the linguistics department at the University of Michigan. Pike, whocreated a method of language analysis called tagmemics, taught Everett to start with commonnouns. "You find out the word for stick,'" Everett said. "Then you try to get the expression for'two sticks,' and for 'one stick drops to the ground,' 'two sticks drop to the ground.' You have toact everything out, to get some basic notion of how the clause structure works-where the subject,verb, and object go."The process is difficult, as I learned early in my visit with the Pirahã. One morning, whileapplying bug repellent, I was watched by an older Pirahã man, who asked Everett what I wasdoing. Eager to communicate with him in sign language, I pressed together the thumb and indexfinger of my right hand and weaved them though the air while making a buzzing sound with mymouth. Then I brought my fingers to my forearm and slapped the spot where my fingers hadalighted. The man looked puzzled and said to Everett, "He hit himself." I tried again-this timemaking a more insistent buzzing.The man said to Everett, "A plane land on his arm." WhenEverett explained to him what I was doing, the man studied me with a look of pitying contemptthen turned away. Everett laughed. "You were trying to tell him something about your generalstate--that bugs bother you," he said. "They never talk that way and they could never understandit. Bugs are a part of life.""O.K.," I said. "But I’m surprised he didn't know I was imitating an insect.""Think of how cultural that is," Everett said, "The movement of your hand. The sound. Even theway we represent animals is cultural."Everett had to bridge many such cultural gaps in order to gain more than a superficial grasp ofthe language. "I went into the jungle, helped them make fields, went fishing with them," he said."You cannot become one of them, but you've got to do as much as you can to feel and absorb thelanguage.” The tribe, he maintains, has no collective memory that extends back more than one ortwo generations, and no original creation myths. Marco Antonio Gonçalves, an anthropologist atFederal University of Rio de Janeiro, spent eighteen months with the Pirahã in the nineteeneightiesand wrote a dissertation on the tribe's beliefs. Gonçalves, who spoke limited Pirahã,© 2008 Dr. Catherine CollierAll Rights Reserved18

agrees that the tribe has no creation myths but argues that few Amazonian tribes do. Whenpressed about what existed before the Pirahã and the forest, Everett says, the tribespeopleinvariably answer, "It has always been this way."Everett also learned that the Pirahã have no fixed words for colors, and instead use descriptivephrases that change from one moment to the next. "So if you show them red cup, they're likely tosay, 'This looks like blood,'" Everett said. "Or they could say, 'This is like vrvcum' a local berrythat they use to extract a red dye."By the end of their first year, Dan Everett had a working knowledge of Pirahã. Keren tutoredKaaxáoi and a child with a monkey they have hunted. Unlike other Amazon tribes, thePirahã have resisted efforts by missionaries and government agencies to teach themfarming, and subsist almost entirely on fish and game.© 2008 Dr. Catherine CollierAll Rights Reserved19

herself by strapping a cassette recorder around her waist and listening to audiotapes while sheperformed domestic tasks. (The Everetts lived in a thatch hut that was slightly larger and moresophisticated than the huts of the Pirahã; it had walls and a storage room that could be locked.)During the family’s second year in the Amazon, Keren and the Everetts' eldest child, Shannon,contracted malaria, and Keren lapsed into a coma. Everett borrowed a boat from river traders andtrekked through the jungle for days to get her to a hospital. As soon as she was discharged,Everett returned to the village. (Keren recuperated in Belem for several months before joininghim.) "Christians who believe in the Bible believe that it is their job to bring others the joy ofsalvation," Everett said. "Even if they’re murdered, beaten to death, imprisoned-that's what youdo for God.Until Everett arrived in the Amazon, his training in linguistics had been limited to fieldtechniques. "I wanted as little formal linguistic theory as I could get by with," he told me. "Iwanted the basic linguistic training to do a translation of the New Testament." This changedwhen S.I.L. lost its contract with the Brazilian government to work in the Amazon. S.I.L. urgedthe Everetts to enroll as graduate students at the State University of Campinas (UNICAMP), inthe state of Sao Paulo, since the government would give them permission to continue living ontribal lands only if they could show that they were linguists intent on recording an endangeredlanguage. At UNICAMP, in the fall of 1978, Everett discovered Chomsky's theories. "For me, itwas another conversion experience," he said.In the late fifties, when Chomsky, then a young professor at M.I.T., first began to attract notice,behaviorism dominated the social science. According to B. F. Skinner, children learn words andgrammar by being praised for correct usage, much as lab animals learn to push a lever thatsupplies them with food. In 1959, in a demolishing review of Skinner's book "Verbal Behavior,"Chomsky wrote that the ability of children to create grammatical sentences that they have neverheard before proves that learning to speak does not depend on imitation, instruction, or rewards.As he put it in his book "Reflections on Language" (1975), "To come to know a human languagewould be an extraordinary intellectual achievement for a creature not specifically designed toaccomplish this task."Chomsky hypothesized that a specific faculty for language is encoded in the human brain atbirth. He described it as a language organ, which is equipped with an immutable set of rules--auniversal grammar--that is shared by all languages, regardless of how different they appear to be.The language organ, Chomsky said, cannot be dissected in the way that a liver or a heart can, butit can be described through detailed analyses of the abstract structures underlying language. "Bystudying the properties of natural languages, their structure, organization, and use," Chomskywrote, "we may hope to gain some understanding of the specific characteristics of humanintelligence. We may hope to learn something about human nature."Beginning in the nineteen-fifties, Chomskyans at universities around the world engaged informal analyses of language, breaking sentences down into ever more complex tree diagrams thatshowed branching noun, verb, and prepositional phrases, and also "X-bars," "transformations,""movements," and "deep structures" –Chomsky’s terms for some of the elements that constitutethe organizing principles of all language. ”I’d been doing linguistics at a fairly low level ofrigor," Everett said. "As soon as you started reading Chomsky’s stuff, and the people mostclosely associated with Chomsky, you realized this is a totally different level--this is actuallysomething that looks like science." Everett conceived his Ph.D. dissertation at UNICAMP as astrict Chomskyan analysis of Pirahã. Dividing his time between Sao Paulo and the Pirahã village,where he collected data, Everett completed his thesis in 1983. Written in Portuguese and later© 2008 Dr. Catherine CollierAll Rights Reserved20

published as a book in Brazil, "The Pirahã Language and the story of Syntax" was a highlytechnical discussion replete with Chomskyan tree diagrams. However, Everett says that he wasaware that Pirahã contained many linguistic anomalies that he could not fit into Chomsky’sparadigm. "I knew I was leaving out a lot of stuff," Everett told me. "But these gaps wereunexplainable to me."The dissertation earned Everett a fellowship from the American Council of Learned Societies,and a grant from the National Science Foundation to spend the 1984-85 academic year as avisiting fellow at M.I.T. Everett occupied an office next to Chomsky’s; he found the famedprofessor brilliant but withering. "Whenever you try out a theory on someone, there's alwayssome question that you hope they won't ask," Everett said. "That was always the first thingChomsky would ask."In 1988, Everett was hired by the University of Pittsburgh. By then, Chomsky’s system of ruleshad reached a state of complexity that even Chomsky found too baroque, and he had begun toformulate a simpler model for the principles underlying all languages. Everett faithfully keptabreast of these developments. "Chomsky sent me the papers that he was working on," he said. "Iwas like many of the scholars, in that I made regular pilgrimages to sit in Chomsky's classes tocollect the handout and to figure out exactly where the theory was today." At the same time,Everett says that he was increasingly troubled by the idiosyncrasies of Pirahã. "None of it wasaddressed by Chomskyan linguistics," he told me. "Chomsky’s theory only allows you to talkabout properties that obtain of tree structures.In the early nineties, Everett began to reread the work of linguists who had preceded Chomsky,including that of Edward Sapir, an influential Prussian-born scholar who died in 1939. A studentof the anthropologist Franz Boas, Sapir had taught at Yale and studied the languages of dozensof tribes in the Americas. Sapir was fascinated by the role of culture in shaping languages, andalthough he anticipated Chomsky’s preoccupation with linguistic universals, he was moreinterested in the variations that made each language unique. In his 1921 book, "Language," Sapirstated that language is an acquired skill, which "varies as all creative effort varies--not asconsciously, perhaps, but nonetheless as truly as do the religions, the beliefs, the customs, andthe arts of different peoples." Chomsky, however, believed that culture played little role in thestudy of language, and that going to far-flung places to record the arcane babel of near-extincttongues was a pointless exercise. Chomsky's view had prevailed. Everett began to wonder if thiswas an entirely good thing."When I went back and read the stuff Sapir wrote in the twenties, I just realized, hey, this reallyis a tradition that we lost," Everett said. "People believe they’ve actually studied a languagewhen they have given it a Chomskyan formalism. And you may have given us absolutely no insightwhatsoever into that language as a separate language."Everett began to question the first principle of Chomskyan linguistics: that infants could notlearn language if the principles of grammar had not been pre-installed in the brain. Babies arebathed in language from the moment they acquire the capacity to hear in the womb, Everettreasoned, and parents and caregivers expend great energy teaching children how to say wordsand assemble them into sentences--a process that lasts years. Was it really true that language, asChomsky asserted, simply “grows like any other body organ"? Everett did not deny the existenceof a biological endowment for language-- humans couldn't talk if they did not possess therequisite neurological architecture to do so. But, convinced that culture plays a far greater rolethan Chomsky’s theory accounted for, he decided that he needed to "take a radical reexaminationof my whole approach to the problem."© 2008 Dr. Catherine CollierAll Rights Reserved21

In 1998, after nine years as the chairman of the linguistics department at the University ofPittsburgh, Everett became embroiled in a dispute with the new dean of the arts and sciencesfaculty. Keren was completing a master's in linguistics at the university and was being paid towork as a teaching assistant in Everett's department. Everett was accused of making improperpayments to Keren totaling some two thousand dollars, and he was subjected to an audit. He wasexonerated, but the allegation of misconduct infuriated him. Keren urged him to quit his job sothat they could return to the jungle and resume their work as missionaries among the Pirahã.It had been more than a decade since Everett had done any concerted missionary work – arefection of his waning religious faith. “As I read more and I got into philosophy and met a lotof friends who weren’t Christians, it became difficult for me to sustain the belief structure in thesupernatural,” he said. But he was inclined to return to the Amazon, partly because he hoped torekindle his faith, and partly because he was disillusioned with the theory that had been thefoundation of his intellectual life for two decades. “I couldn’t buy Chomsky’s world view anylonger,” Everett told me, “and I began to feel that academics were a hollow and insignificant wayto spend one’s life.”In the fall of 1999, Everett quit his job, and on the banks of the Maici River he and Keren built atwo-room, eight-by-eight-meter, bug-and snake-proof house from fourteen tons of ironwood thathe had shipped in by boat. Everett equipped the house with a gas stove, a generator-drivenfreezer, a water-filtration system, a TV, and a DVD player. “After twenty years of living like aPirahã, I’d had it with roughing it,” he said. He threw himself into missionary work, translatingthe Book of Luke into Pirahã and reading it to tribe members. His zeal soon dissipated, however.Convinced that the Pirahã assigned no spiritual meaning to the Bible, Everett finally admittedthat he did not, either. He declared himself an atheist, and spent his time tending house andstudying linguistics. In 2000, on a trip to Porto Velho, a town about two hundred miles from thevillage, he found a month-old e-mail from a colleague at the University of Manchester, invitinghim to spend a year as a research professor at the school. In 2002, Everett was hired to a fulltimeposition, and he and Keren moved to England. Three years later, he and Keren separated;she returned to Brazil, where she divides her time between the Pirahã village and an apartment inPorto Velho. He moved back to the United States last fall to begin the new job at Illinois State.Today, Everett says that his three years in the jungle were hardly time wasted.“This new beginning with the Pirahã really was quiet liberating," he told me. "Free fromChomskyan constraints, I was able to imagine new relationships between grammar and culture."It is a matter of some vexation to Everett that the first article on the Pirahã to attract significantattention was written not by him but by his friend (and former colleague at the University ofPittsburgh) Peter Gordon, now at Columbia, who in 2004 published a paper in Science on thePirahã's understanding of numbers Gordon had visited the tribe with Everett in the early nineties,after Everett told him about the Pirahã’s limited "one' "two," and "many" counting system. Othertribes, in Australia, the South Sea Islands, Africa, and the Amazon, have a one-two-manynumerical system, but with an important difference: they are able to learn to count in anotherlanguage. The Pirahã have never been able to do this, despite concerted efforts by the Everetts toteach them to count to ten in Portuguese.During two-month stay with the Pirahã in 1992, Gordon ran several experiments with tribemembers. In one, he sat across from a Pirahã subject and placed in front of himself an array ofobjects-nuts, AA batteries-and had the Pirahã match the array. The Pirahã could perform the taskaccurately when the array consisted of two or three items, but their performance with largergroupings was, Gordon later wrote, "remarkably poor." Gordon also showed subjects nuts,© 2008 Dr. Catherine CollierAll Rights Reserved22

placed them in a can, and withdrew them one at a time. Each time he removed a nut, he asked thesubject whether there were any left in the can. The Pirahã answered correctly only withquantities of three or fewer. Through these and other tests, Gordon concluded that Everett wasright: the people could not perform tasks involving quantities greater than three. Gordon ruledout mass retardation. Though the Pirahã do not allow marriage outside their tribe, they have longkept their gene pool refreshed by permitting women to sleep with outsiders. "Besides," Gordonsaid, “if there was some kind of Appalachian inbreeding or retardation going on, you'd see it inhairlines, facial features, and motor ability. It bleeds over. They don't show any of that.”Gordon surmised that the Pirahã provided support for a controversial hypothesis advanced earlyin the last century by Benjamin Lee Whorf, a student of Sapir's. Whorf argued that the words inour vocabulary determine how we think. Since the Pirahã do not have words for numbers abovetwo, Gordon wrote, they have a limited ability to work with quantities greater than that. "It'slanguage affecting thought," Gordon told me. His paper, "Numerical Cognition without Words:Evidence from Amazonia," was enthusiastically taken up by a coterie of "neo-Whorfian"linguists around the world.Everett did not share this enthusiasm; in the ten years since he had introduced Gordon to thetribe, he had determined that the Pirahã have no fixed numbers. The word that he had long takento mean "one" (hoi, on a falling tone) is used by the Pirahã to refer, more generally, to "a smallsize or amount," and the word for "two" (hoi, on a rising tone) is often used to mean "asomewhat larger size or amount." Everett says that his earlier confusion arose over what's knownas the translation fallacy: the conviction that a word in one language is identical to a word inanother, simply because, in some instances, they overlap in meaning. Gordon had mentioned theelastic boundaries of the words for "one" and "two" in his paper, but in Everett's opinion he hadfailed to explore the significance of the phenomenon. (Gordon disagrees, and for a brief periodthe two did not speak.)Shortly after Gordon's article appeared, Everett began outlining a paper correcting what hebelieved were Gordon's errors. Its scope grew as Everett concluded that the Pirahã’s lack ofnumerals was part of a larger constellation of "gaps." Over the course of three weeks, Everettwrote what would become his Cultural Anthropology article, twenty-five thousand words inwhich he advanced a novel explanation for the many mysteries that had bedeviled him. Inspiredby Sapir's cultural approach to language, he hypothesized that the tribe embodies a living-in-thepresentethos so powerful that it has affected every aspect of the people's lives committed to anexistence in which only observable experience is real, the Pirahã do not think, or speak, inabstractions-- thus do not use color terms, quantifiers, numbers, or myths. Everett point to theword xibipio as a clue to how the Pirahã perceive reality solely according to what exists withinthe boundaries of their direct experience-- which Everett defined as anything that they can seeand hear, or that someone living has seen and heard. "When someone walks around a bend in theriver, the Pirahã say that the person has not simply gone away but xibipio-'gone out ofexperience,' Everett said. "They use the same phrase when a candle flame flickers. The light'goes in and out of experience. To Everett the Pirahã’s unswerving dedication to empiricalreality--he called it the "immediacy-of-experience principle"- explained their resistance toChristianity, since the Pirahã had always reacted to stories about Christ by asking, "Have youmet this man?" Told that Christ died two thousand years ago, the Pirahã would act much as theydid to my using bug repellent. It explained their failure to build up food stocks, since thisrequired planning for a future that did not yet exist; it explained the failure of the boys' modelairplanes to foster a tradition of sculpture-making, since the models expressed only the© 2008 Dr. Catherine CollierAll Rights Reserved23

momentary burst of excitement that accompanied the sight of an actual plane. It explained thePirahã’s lack of original stories about how they came into being, since this was a conundrumburied in a past outside the experience of parents and grandparents.Everett was convinced that the Pirahã’s immediacy-of-experience principle went further still,"extending its tentacles," as he put it, "deep into their core grammar," to that feature thatChomsky claimed was present in all languages: recursion. Chomsky and other experts use theterm to describe how we construct even the simplest utterances. "The girl jumped on the bed" iscomposed of a noun phrase ("the girl"), a verb ("jumped"), and a prepositional phrase ("on thebed"). In theory, as Chomsky has stressed, one could continue to insert chunks of languageinside other chunks ad infinitum, thereby creating a never-ending sentence (“The man who iswearing a top hat that is slightly crushed around the brim although still perfectly elegant iswalking down the street that was recently resurfaced by a crew of construction workers whotended to take coffee breaks that were a little too long while eating a hot dog that was. .."). Orone could create sentences of never-ending variety. The capacity to generate unlimited meaningby placing one thought inside another is the crux of Chomsky's theory-what he calls, quoting theearly-nineteenth- century German linguist Wilhelm von Humboldt, "the infinite use of finitemeans.”According to Everett, however, the Pirahã do not use recursion to insert phrases one insideanother. Instead, they state thoughts in discrete units. When I asked Everett if the Pirahã couldsay, in their language, "I saw the dog that was down by the river get bitten by a snake," he said,"No. They would have to say, 'I saw the dog. The dog was at the beach. A snake bit the dog.'"Everett explained that because the Pirahã accept as real only that which they observe, theirspeech consists only of direct assertions ("The dog was at the beach"), and he maintains thatembedded clauses ("that was down by the river") are not assertions but supporting, quantifying,or qualifying information- in other words, abstractions.In his article, Everett argued that recursion is primarily a cognitive, not a linguistic, trait. Hecited an influential 1962 article, "The Architecture of Complexity," by Herbert Simon, a NobelPrize-winning economist, cognitive psychologist, and computer scientist, who asserted thatembedding entities within like entities (in a recursive tree structure of the type central toChomskyan linguistics) is simply how people naturally organize information. "Microsoft Wordis organized by tree structures," Everett said. "You open up one folder and that splits into twoother things, and that splits into two others. That's a tree structure. Simon argues that this isessential to the way humans organize information and is found in all human intelligence systems.If Simon is correct, there doesn't need to be any specific linguistic principle for this because itsjust general cognition." Or, as Everett sometimes likes to put it: "The ability to put thoughtsinside other thoughts is just the way humans are, because we're smarter than other species.”Everett says that the Pirahã have this cognitive trait but that it is absent from their syntax becauseof cultural constraints.Some scholars believe that Everett's claim that the Pirahã do not use recursion is tantamount tocalling them stupid. Stephen Levinson, the neo-Whorfian director of the Language and CognitionGroup at the Max Planck Institute for Psycholinguistics, in the Netherlands, excoriated Everett inprint for having made the Pirahã sound like the mindless bearers of an almost subhumanlysimple culture." Anna Wierzbicka, a linguist at the Australian National University, was alsotroubled by the paper, and told me, "I think from the point of view of-I don't know-humansolidarity, human rights, and so on, it's really very important to know that it's a question thatmany people don't dare to raise, whether we have the same cognitive abilities or not, we© 2008 Dr. Catherine CollierAll Rights Reserved24

humans." Everett dismissed such criticisms, since he expressly states in the article that theunusual aspects of the Pirahã are not -a result of mental deficiency. A Pirahã removed from thejungle at birth and brought up in any city in the world, he said, would have no trouble learningthe local tongue. Moreover, Everett pointed out, the Pirahã are supremely gifted in all the waysnecessary to insure their continued survival in the jungle: they know the usefulness and locationof important plants in their area; they understand the behavior of local animals and how to catchavoid them; and can walk into the jungle naked, with no tools or weapons, and walk out threeday later with baskets of fruit, nuts, and small game. "They can out-survive anybody, any otherIndian in this region," he said. "They're very intelligent people. It never would occur to me thatsaying they lack things that Levinson or Wierzbicka predict they should have is calling themmindless idiots.”For Everett, the most important reaction to the article was Chomsky's. In an e-mail to Everett lastApril, Chomsky rejected Everett's arguments that the Pirahã’s lack of recursion is a strongcounter example to his theory of universal grammar, writing, “UG is the true theory of thegenetic component that underlies acquisition and use of language." He added that there is "nocoherent alternative to UG.” Chomsky declined to be interviewed for this article, but referred meto "Pirahã Exceptionality: A Reassessment," a paper that was co-authored b David Pesetsky, acolleague of Chomsky's at M.I.T.; Andrew Nevins, a linguistic at Harvard; and CileneRodrigues, a linguistic at UNICAMP.In the paper, which posted last month on the Web site LingBuzz, a repository of articles onChomskyan generative grammar, the authors used data from Everett's 1983 Ph.D. dissertation, aswell as from a paper that he published on Pirahã in 1986, to refute his recent claims about thelanguage's unusual features- including the assertion that the Pirahã do not use recursion. Theauthors conceded that, even in these early works, Everett had noted the absence of certainrecursive structure in Pirahã. (The tribe, Everett wrote in the early eighties, does not embedpossessives inside one another, as English speakers do when they say, "Tom's uncle's carswindshield..."). Nevertheless, they argued, Everett's early data suggested that the Pirahã’s speechdid contain recursive operations.The fact that Everett had collected the data twenty-five years ago, when he was a devotee ofChomsky's theory, was irrelevant, Pesetsky told me in an e-mail. At any rate, Pesetsky wrote, heand his co-authors detected no sign of a particularly Chomskyan perspective" in the descriptiveportions of Everett's early writings, adding, "For the most part, those works are about facts, andthe categorizing of facts."Everett, who two weeks ago posted a response to Pesetsky and his co-authors on LingBuzz, saysthat Chomsky's theory necessarily colored his data-gathering and analysis. "Descriptive workapart from theory does not exist,” he told me. “We ask the questions that our theories tell us toask." In his response on LingBuzz, Everett addressed his critics' arguments point by point anddisputed the contention that his early work was more reliable than his resent research as a guideto Pirahã. "I would find the opposite troubling- i.e., that a researcher never changed their mind orfound errors in their earlier work," he wrote. He added, "There are alternatives to UniversalGrammar, and the fact that NPR"- Nevins, Pesetsky, and Rodrigues- "insist on characterizing theissue as though there were no alternatives, although typical, is either ignorant or purposelymisleading."In a comment on Everett's paper published in Cultural Anthropology, Michael Tomasello, thedirector of the Department of Developmental and Comparative Psychology at the Max PlanckInstitute for Evolutionary Anthropology, in Leipzig, endorsed Everett's conclusions that culture© 2008 Dr. Catherine CollierAll Rights Reserved25

can shape core grammar. Because the Pirahã "talk about different things [than we do], differentthings get grammaticalized," he wrote, adding that "universal grammar was a good try, and itreally was not so implausible at the time it was proposed, but since then we have learned a lotabout many different languages, and they simply do not fit one universal cookie cutter."Steven Pinker, the Harvard cognitive scientist, who wrote admiringly about some of Chomsky'sideas in his 1994 best-seller, "The Language Instinct," told me, "There's a lot of strange stuffgoing on in the Chomskyan program. He's a guru, he makes pronouncements that his disciplesaccept on faith and that he doesn't feel compelled to defend in the conventional scientificmanner. Some of them become accepted within his circle as God's truth without really beingproperly evaluated, and, surprisingly for someone who talks about universal grammar, he hasn'tactually done the spadework of seeing how it works in some weird little language that they speakin New Guinea."Pinker says that his own doubts about the "Chomskyan program" increased in 2002, when MarcHauser, Chomsky, and Tecumseh Fitch published their paper on recursion in Science. Theauthors wrote that the distinctive feature of the human faculty of language, narrowly defined, isrecursion. Dogs, starlings, whales, porpoises, and chimpanzees all use vocally generated soundsto communicate with other members of their species, but none do so recursively, and thus nonecan produce complex utterances of infinitely varied meaning. "Recursion had always been animportant part of Chomsky's theory," Pinker said. "But in Chomsky Mark II, or Mark III, orMark VII, he all of a sudden said that the only thing unique to language is recursion. It's not justthat it's the universal that has to be there; it's the magic ingredient that makes language possible."In early 2005, Pinker and Ray Jackendoff: a linguistics professor at Tufts University, published acritique of Hauser, Chomsky, and Fitch's paper in the journal Cognitio ."In my paper with Ray, we argue that if you just magically inject recursion into a chimpanzeeyou're not going to get human who can put words together into phrases, label concepts withwords, name things that happened decades ago or that may or may not happen decades in thefuture," Pinker said. "There's more to language than recursion." Pinker and Jackendoff, in areference to Everett's research, cited Pirahã as an example of a language that has "phonology,morphology, syntax, and sentences," but no recursion. Pinker, however, was quick to tell me thatthe absence of recursion in one of the more than six thousand known languages is not enough todisprove Chomsky's ideas. "If you had something that was present in five thousand nine hundredand ninety-nine of the languages, and someone found one language that didn't have it-well, Ithink there may be some anthropologists who would say, 'This shows that there's no universals,that anything can happen,'" he said. "But, more likely, you'd say, ‘Well, what's going on with thatweird language?'"Contemporary linguists have generally avoided speculation about how humans acquiredlanguage in the first place. Chomsky himself has long demonstrated a lack of interest in languageorigins and expressed doubt about Darwinian explanations. "It is perfectly safe to attribute thisdevelopment to 'natural selection,'" Chomsky has written, "so long as we realize that there is nosubstance to this assertion, that it amounts to nothing more than a belief that there is somenaturalistic explanation for these phenomena." Moreover, Chomsky's theory of universalgrammar, which was widely understood to portray language as a complex system that arose fullyformed in the brain, discouraged inquiry into how language developed. "This totally slams thedoor on the question," Brent Berlin, a cognitive anthropologist at the University of Georgia, toldme. "It acts as if, in some inexplicable way, almost mysteriously, language is hermetically sealed© 2008 Dr. Catherine CollierAll Rights Reserved26

from the conditions of life of the people who use it to communicate. But this is not some kind ofan abstract, beautiful, mathematical, symbolic system that is not related to real life."Berlin believes that Pirahã may provide a snapshot of language at an earlier stage of syntacticdevelopment. 'That's what Dan's work suggests," Berlin said of Everett's paper. "The plausiblescenarios that we can imagine are ones that would suggest that early language looks somethinglike the kind of thing that Pirahã looks like now."Tecumseh Fitch, a tall, patrician man with long, pointed sideburns and a boyishly enthusiasticmanner, owes his unusual first name to his ancestor, the Civil War general William TecumsehSherman. Fitch attended Brown University and earned a Ph.D. there. As a biologist with aspecial interest in animalcommunication, Fitch discoveredthat red deer possess a descendedlarynx, an anatomical feature thatscientists had previously believedwas unique to human beings andcentral to the development of speech.(The descended larynx has sincebeen found in koalas, lions, tigers,jaguars, and leopards.) Fitch, eager tounderstand how humans acquiredlanguage, turned to linguistics andwas surprised to learn that Chomskyhad written little about the question.But in 1999 Fitch happened to readan interview that Chomsky had givento Spare Change News, a newspaperfor the homeless in Cambridge. "Iread it and all the stuff he said aboutevolution was almost more than he'sever said in any published thing- andhere it is in Spare Change!" Fitchsaid. "And he just made a few pointsthat made me realize what he'd beengetting at in a more enigmaticfashion in some of his previouscomments." Fitch invited Chomskyto speak to a class that he was coteachingat Harvard on the evolutionof language. Afterward, they talkedfor several hours. A few monthslater, Chomsky agreed to collaboratewith Fitch and Hauser on a paper thatwould attempt to pinpoint thefeatures of language which areunique to humans and which allowedHomo sapiens to develop language.Portrait of Xopí: The Pirahã, Everett says, have no numbers,no fixed terms for colors, & no words for left & right.© 2008 Dr. Catherine CollierAll Rights Reserved27

The authors compared animal and human communication, eliminating the aspects of vocalizationthat are shared by both, and concluded that one operation alone distinguished human speech:recursion. In the course of working on the article, Fitch grew sympathetic to Chomsky's ideasand became an articulatedefender of the theory ofuniversal grammar.When Fitch and Everettmet in Porto Velho inJuly, two days beforeheading into the jungle,they seemed, by tacitagreement, to be avoidingtalk of Chomsky. But, onthe eve of our departure,while we were sitting bythe pool at the Hotel VilaRica, Everett mentionedtwo professors who, hesaid, were "among thethree most arrogant peopleI've met."“Who’s the third?" Fitchasked."Noam," Everett said."No!" Fitch cried. "Givenhis status in science,Chomsky is the leastarrogant man, thehumblest great man, I'veever met."Everett was having noneof it. "Noam Chomskythinks of himself asAristotle!" he declared."He has dug a hole forlinguistics that it will takedecades for the disciplinePortrait of Piipaío: The Pirahã’s only ornaments are necklacesto climb out of!"made from seeds, teeth,, feathers, beads, and soda-can pull-tabs.The men argued for thenext two hours, though bythe time they parted for the night civility had been restored, and the détente was still holdingwhen they met in the Pirahã village the next day and agreed to begin experiments the followingmorning.At sunrise, a group of some twenty Pirahã gathered outside Everett's house. They were to be paidfor their work as experimental subjects--with tobacco, cloth, farina, and machetes. "And, believe© 2008 Dr. Catherine CollierAll Rights Reserved28

me," Everett said, "that's the only reason they're here. They have no interest in what we're doing.They're hunter-gatherers, and they see us just like fruit trees to gather from."Fitch went out with Everett into the thick heat, carrying his laptop. The two men, trailed by thePirahã, followed a narrow path through the low underbrush to Everett's office, a small hut, raisedoff the ground on four-foot-high stilts, at the edge of the jungle. Fitch placed his computer on thedesk and launched a program that he had spent several weeks writing in preparation for this tripFitch’s experiments were based on the so-called Chomsky hierarchy, a system for classifyingtypes of grammar, ranked in ascending order of complexity. To test the Pirahã’s ability to learnone of the simplest types of grammar, Fitch had written a program in which grammaticallycorrect constructions were represented by a male voice uttering one nonsense syllable (mi or dohor ga, for instance), followed by a female voice uttering a different nonsense syllable (lee or ta orgee). Correct constructions would cause an animated monkey head at the bottom of the computerscreen to float to a corner at the top of the screen after briefly disappearing; incorrectconstructions (anytime one male syllable was followed by another male syllable or more thanone female syllable) would make the monkey head float to the opposite corner. Fitch set up asmall digital movie camera behind the laptop to film the Pirahã’s eye movements. In the fewseconds’ delay before the monkey head floated to either corner of the screen, Fitch hoped that hewould be able to determine, from the direction of the subjects’ unconscious glances, if they werelearning the grammar. The experiment, using different stimuli, had been conducted withundergraduates and monkeys, all of whom passed the test. Fitch told me that he had little doubtthat the Pirahã would pass. “My expectation coming in here is that they’re going to act just likemy Harvard undergrads,” he said. “They’re going to do exactly what every other human hasdone and they’re going to get this basic pattern. The Pirahã are humans – humans can do this.”Fitch called for the first subject.Everett stepped outside the hut and spoke to a short muscular man with a bowl-shaped haircutand heavily calloused bare feet. The man entered the hut and sat down at the computer, whichpromptly crashed. Fitch rebooted. It crashed again.“It’s the humidity,” Everett said.Fitch finally got the computer working, but then the video camera seized up.“Goddam Chomskya,” Everett said. “Can’t even run an experiment.”Eventually, Fitch got all the equipment running smoothly and started the experiment. It quicklybecame obvious that the Pirahã man was simply watching the floating monkey head and wasn’tresponding to the audio cues.“It didn’t look like he was doing pre-monitory looking,” Fitch said, “Maybe ask him to point towhere he thinks the monkey is going to go.”"They don't point," Everett said. Nor, he added, do they have words for right or left. Instead, theygive directions in absolute terms, telling others to head "up river" or "downriver," or "to theforest" or "away from the forest." Everett told the man to say whether the monkey was goingupriver or downriver. The man said something in reply."What did he say?" Fitch asked."He said, 'Monkeys go to the jungle."Fitch grimaced in frustration. 'Well he's not guessing with his eyes," he said. "Is there anotherway he can indicate?"Everett again told the man to say whether the monkey was going upriver or down. The manmade a noise of assent. Fitch resumed the experiment, but the man simply waited until the© 2008 Dr. Catherine CollierAll Rights Reserved29

monkey moved. He followed it with his eyes, laughed admiringly when it came to a stop, thenannounced whether it had gone upriver or down.After several minutes of this, Fitch said, on a rising note of panic, "If they fail in the recursionone--it's not recursion; I’ve got to stop saying that. I mean embedding. Because, I mean, if hecan't get this-""This is typical Pirahã," Everett said soothingly. "This is new stuff, and they don't do new stuff""But when they're hunting they must have those skills of visual anticipation," Fitch said."Yeah," Everett said dryly. "But this is not a real monkey." He pointed at the grinning animatedhead bobbing on the screen."Fuck!" Fitch said. "If I’d had a joy- stick for him to hunt the monkey!" He paced a little, thensaid, "The crazy thing is that this is already more realistic than the experiments Aslin did withbabies.""Look, "Everett said, "The cognitive issue here is the cultural impediment to doing new things.He doesn't know there's a pattern to recognize."Everett dismissed the man and asked another Pirahã to come into the hut. A young manappeared, wearing a green- and-yellow 2002 Brazilian World Cup shirt, and sat at the computer.Everett told him to say whether the monkey was going to go upriver or downriver.Fitch ran the experiment. The man smiled and pointed with his chin whenever the monkey headcame to rest."The other idea," Fitch said, "is if we got a bunch of the kids, and whoever points first gets alollipop.""That's got an element of competition that they won't go for," Everett said.The computer crashed. Convinced that there was a glitch in the software, Fitch picked up themachine and carried it back to the main house to make repairs."This is typical of fieldwork in the Amazon, which is why most people don't do it," Everett said."But the problem here is not cognitive; it's cultural." He gestured toward the Pirahã man at thetable, "Just because we're sitting in the same room doesn't mean we're sitting in the samecentury."By the next morning, Fitch had debugged his software, but other difficulties persisted. Onesubject, a man in blue nylon running shorts, ignored instructions to listen to the syllables andasked questions about the monkey head: "Is that rubber?" "Does this monkey have a spouse?" "Isit a man?" Another man fell asleep mid-trial (the villagers had been up all night riotously talkingand laughing- a common occurrence for a people who do not live by the clock). Meanwhile,efforts to get subjects to focus were hampered by the other tribe members, who had collectedoutside the hut and held loud conversations that were audible through the screened windows.Steve Sheldon, Everett's predecessor in the Pirahã village, had told me of the challenges he facedin the late sixties when he did research on behalf of Brent Berlin and Paul Kay (an anthropologistand linguist at the University of California at Berkeley), who were collecting data about colorsfrom indigenous peoples. Sheldon had concluded that the Pirahã tribe has fixed color terms-aview duly enshrined in Berlin and Kays book "Basic Color Terms: Their Universality andEvolution" (1969). Only later did Sheldon realize that his data were unreliable. Told to questiontribe members in isolation, Sheldon had been unable to do so because the tribe refused to be splitup; members had eavesdropped on Sheldon's interviews and collaborated on answers. "Theirattitude was who cares what the color is?'" Sheldon told me. "But we'll give him somethingbecause that's what he wants." (Today, Sheldon endorses Everett's claim that the tribe has nofixed color terms.)© 2008 Dr. Catherine CollierAll Rights Reserved30

Sheldon said that the Pirahã’s obstruction approach to researchers is a defensive gesture. "Theyhave been made fun of by outsiders because they do things different,” Sheldon told me. “Withresearchers who don't speak their language, they make fun, giving really bad information, totallywrong information sometimes.”On the third day, Fitch had figured out that he was being hindered by some of the same problemsthat Sheldon had faced. That morning, he tacked up bed sheets over the window screens anddemanded that the tribe remain at a distance from the hut. (Several yards away, Fitch’s cousin,Bill, entertained the group by playing Charlie Parker tunes on his iPod.) Immediately, the testingwent better. 0ne Pirahã man seemed to make anticipated eye movements, although it wasdifficult to tell, because his eyes were hard to make out under the puffy lids, a feature typical ofthe men's faces. Fitch tried the experiment on a young woman with large dark irises, but it wasnot clear that her few correct glances were anything but coincidental. "Lot of random looks,"Everett muttered. "It's not obvious that they're getting it either way," Fitch said.On the fourth day, Fitch seemed to hit pay dirt. The subject was a girl of perhaps sixteen.Focused, alert, and calm, she seemed to grasp the grammar, her eyes moving to the correctcorner of the screen in advance of the monkey’s head. Fitch was delighted, and perhaps relieved;before corning to the Amazon, he had told me that the failure of a Pirahã to perform this taskwould be tantamount to "discovering Sasquatch."Fitch decided to test the girl on a higher level of the Chomsky hierarchy, a “phrase-structuregrammar." He had devised a program in which correct constructions consisted of any number ofmale syllables followed by an equal number of female syllables. Hauser, Chomsky, and Fitch, intheir 2002 paper, had stated that a phrase-structure grammar, which makes greater demands onmemory and pattern recognition, represents the minimum foundation necessary for humanlanguage.Fitch performed several practice trials with the girl to teach her the grammar. Then he andEverett stepped back to watch. "If this is working," Fitch said, "we could try to get N.S.F.money. This could be big-even for psychology."At the mention of psychology-a discipline that often emphasizes the influence of environment onbehavior and thus is at a remove from Chomsky's naturism- Everett laughed. "Now he'sbeginning to see it my way!" he said.The girl gazed at the screen and listened as the HAL-like computer voices flatly intoned themeaningless syllables. Fitch peered at the camera's viewfinder screen, trying to discern whetherthe girl's eye movements indicated that she understood the grammar. It was impossible to say.Fitch would have to take the footage back to Scotland, where it would be vetted by an impartialpost-doc volunteer, who would "score" the images on a time line carefully synchronized to thesoundtrack of the spoken syllables, so that Fitch could say without a doubt whether the subject'seyes had anticipated the monkey head, or merely followed it. (Last week, Fitch said that the data"look promising," but he declined to elaborate, pending publication of his results.)That evening, Everett invited the Pirahã to come to his home to watch a movie: Peter Jackson’sremake of “King Kong.” (Everett had discovered that the tribe loves movies that featureanimals.) After nightfall, to the grinding sound of the generator, a crowd of thirty or so Pirahãassembled on benches and on the wooden floor of Everett’s “Indian room,” a screened-offsection of his house where he confines the Pirahã, owing to their tendency to spit on the floor.Everett had made popcorn, which he distributed in a large bowl. Then he started the movie,clicking ahead to the scene in which Naomi Watts, reprising Fay Wray’s role, is offered as asacrifice by the tribal people of an unspecified South Seas island. The Pirahã shouted with© 2008 Dr. Catherine CollierAll Rights Reserved31

delight, fear, laughter, and surprise – and when Kong himself arrived, smashing though the palmtrees, pandemonium ensued. Small children, who had been sitting close to the screen, jumped upand scurried into their mothers’ laps; the adults laughed and yelled at the screen.If Fitch’s experiments were inconclusive on the subject of whether Chomsky’s universalgrammar applied to the Pirahã, Jackson’s movie left no question about the universality ofHollywood film grammar. Kong battled raptors and Watts dodged giant insects, the Pirahãoffered a running commentary, which Everett translated: "Now he's going to fall!" “He's tired!""She's running!" "Look. A centipede!" Nor were the Pirahã in doubt about what was beingcommunicated in the long, lingering looks that passed between gorilla and girl. "She is hisspouse," one Pirahã said. Yet in the reaction to the movie Everett also saw proof of his theoryabout the tribe. "They’re not generalizing about the character of giant apes," he pointed out.They’re reacting to the immediate action on the screen with direct assertions about what theysee."In Fitch's final two days of experiments, he failed to find another subject as promising as thesixteen-year-old girl. But he was satisfied with what he had been able to accomplish in six daysin the jungle. “I think Dan's is an interesting and valid additional approach to add to the arsenal,"Fitch told me after we had flown back to Porto Velho and were sitting beside the pool at theHotel Vila Rica. "I think you need to look at something as complex as language from lots ofdifferent angles, and I think the angle he's arguing is interesting and deserves more work, moreresearch. But as far as the Pirahã disproving universal grammar? I don’t think anything I couldhave seen out here would have convinced me that that was ever anything other than just thewrong way to frame the problem."On my final night in Brazil, I met Keren Everett, in the gloomy lobby of the hotel. At fifty-five,she is an ageless, elfin woman with large dark eyes and waist-length hair pulled back from herface. She is trained in formal linguistics, but her primary interest in the Pirahã remains mission.In keeping with the tenets of S.I.L., she does not proselytize, or actively attempt to convert them;it is enough, S.I.L. believes, to translate the Bible into the tribal tongue. Keren insists that shedoesn’t know the language well yet. "I still haven't cracked it," she said, adding that she thoughtshe was "beginning to feel it for the first time, after twenty-five years"The key to earning the language is the tribe's singing, Keren said: the way that the group dropconsonants and vowels altogether and communicate purely by variations in pitch, stress, andrhythm--what linguists call "prosody." I was reminded of an evening in the village when I hadheard someone singing a clutch of haunting notes on a rising, then falling scale. The voicerepeated the pattern over and over, without variation, for more than half an hour. I crept up to theedge of one of the Pirahã huts and saw that it was a woman, winding raw cotton onto a spool,and intoning this extraordinary series of notes that sounded like a muted horn. A toddler playedat her feet. I asked Everett about this, and he said something vague about how tribe members"sing their dreams." But when I described the scene to Keren she grew animated and explainedthat this is how the Pirahã teach their children to speak. 'The toddler was absorbing the lesson inprosody through endless repetition--an example, one might argue, of Edward Sapir's culturaltheory of language acquisition at work."This language uses prosody much more than any other language I know of," Keren told me."It’s not the kind of thing that you can write, and capture, and go back to; you have to watch, andyou have to feel it. It’s like someone singing a song. You want to watch and listen and try to singalong with them. So I started doing that, and I began noticing things that never transcribed, andthings I never picked up when I listened to a tape of them, and part of it was the performance. So© 2008 Dr. Catherine CollierAll Rights Reserved32

at that point I said, 'Put the tape recorders and notebooks away, focus on the person, and watchthem.’ They give a lot of things using prosody that you never would have found otherwise. Thishas never been documented in any language I know." Aspects of Pirahã that had longconfounded Keren became clear, she said. “I realized, Oh! That’s what the subject-verb lookslike, that’s what the pieces of the clause and the time phrase and the object and the other phrasesfeel like. That was the beginning of a break through for me. I won't say that I’ve broken it until Ican creatively use the verbal structure--and I can't do it yet."Keren says that Everett's frustration at realizing that they would have to "start all over again"with the language ultimately led to his decision to leave the Amazon in 2002 and return toacademia. "He was diligent and he was trying to use his perspective and his training, and Iwatched the last year that we were together in the village--he just was, like, 'This is it. I'm out ofhere.' That was the year I started singing, and he said, 'Damn it if I'm going to learn to sing thislanguage!' And he was out. It's torment. It is tormenting when you have a good mind and youcan't crack it. I said, 'I don't care, we're missing something. We've got to look at it from adifferent perspective.'" Keren shook her head. "Pirahã has just always been out there defyingevery linguist that's gone out there, because you can't start at the segment level and go on. You'renot going to find out anything, because they really can communicate without the syllables."Later that day, when Everett drove me to the airport in Porto Velho, I told him about myconversation with Keren. He sighed. "Keren has made tremendous progress, and I'm sure sheknows more about musical speech than I do at this point," he said. "There are probably severalareas of Pirahã where her factual knowledge exceeds mine. But it's not all the prosody. That's thething." Keren's perspective on Pirahã derives from her missionary impulses, he said. "It would beimpossible for her to believe that we know the language, because that would mean that the Wordof God doesn't work." Everett pulled into the airport parking lot. It was clear that talking aboutKeren caused him considerable pain. He did not want our conversation to end on a quarrel withher. He reminded me that his disagreement is with Chomsky."A lot of people's view of Chomsky is of the person in the lead on the jungle path," Everett hadtold me in the Pirahã village. "And if anybody s likely to find the way home it's him. So theywant to stay as close behind him as possible. Other people say, 'Fuck that, I’m going to get onthe river and take my canoe.' "© 2008 Dr. Catherine CollierAll Rights Reserved33

Piipaío in a hut: Pirahã huts typically have dirt floors and no walls or furniture except for a raisedplatform of thin branches to sleep on. The Pirahã keep very few possessions in their huts – pots andpans, a machete, a knife.© 2008 Dr. Catherine CollierAll Rights Reserved34

Article I.4 Linguistic Diversity & Reform: CanPractices be Identified?By Adel NadeauDirections in Language and EducationNational Clearinghouse of Bilingual EducationVol. 1, No. 10, Winter 1997A nationwide study of school reform and student diversity, funded through the U.S. Departmentof Education, identified Linda Vista Elementary School as one among eight schools with anexemplary learning environment for limited English proficient (LEP) students (Berman, et.al.,1995). This learning environment came about through the concerted efforts of the Linda Vistastaff and principal, who provided the leadership and vision necessary for sustained school-widereform.The following discussion is written from the perspective of the person responsible forspearheading Linda Vista's reform. In reflecting upon her experiences with reform in a schoolwith a student population 77 percent limited-English proficient, Dr. Adel Nadeau, formerprincipal of Linda Vista, provides both a personalized account of practices implemented at LindaVista and parameters for the reform efforts of other schools serving linguistically and culturallydiverse students. The discussion serves as a valuable commentary in its own right as well ascomplementing the information on Linda Vista in School Reform and Student Diversity: CaseStudies of Exemplary Practices for LEP Students, a report which describes the eight schoolsselected in the nationwide study of exemplary programs serving LEP students. This report andrelated documents may be found at http://www.ncbe.gwu.edu.IntroductionThe current reform movement in American education is more than a passing fad. It represents amajor transition in thinking about the formal education of all students. In describing the processof education reform, Victoria Bernhardt (1994) states: "Schools' staffs must understand from thebeginning that major elements (for reform) are internal rather than external, requiring atransformation of all individuals' thinking about school, students, teaching, and learning." Withinthis context, the language of education reform, expressed by words such as "bottom up,""continuous progress," "collaboration," "inclusion," "vision," and "commitment," begins to takeon a new meaning, one premised on individual empowerment and a shared vision of andcommitment to the academic achievement of all students. This represents a dramatic paradigmshift in the management and delivery of public education services in the United States.We are now, however, in the perplexing stage of devising implementation strategies. Ofparticular concern is the identification of those practices of reform that result in success for poorstudents, those at risk, and those with linguistic differences. Although a strong research base hasbeen established on effective practices for linguistically and culturally diverse (LCD) students(Garcia, 1988, 1991), this knowledge does not appear to have been integrated into the reformagenda. In a discussion paper entitled, For all students: Limited-English proficient students andGoals 2000 (August, Hakuta, and Pompa, 1994), the authors identify issues related to theinclusion of LEP students in the reform effort, issues such as the need for academic standardsthat address primary and secondary language development, assessment of and accountability for© 2008 Dr. Catherine CollierAll Rights Reserved35

LEP students, and equitable organizational structures. Their discussion points to the need for thereform process to incorporate these and other key variables that specifically relate to theinstruction of language minority students.The majority of studies on school reform, however, do not include issues related to languageminority students (Gandara, 1994; Valdez, 1989; Zehler et al., 1994). In a literature review oninstitutional change and services to limited English proficient students, Zehler et al. (1994) foundthat typically there is an assumption about the "universal" teacher and the "universal" student.The implication of this finding is that language acquisition and dual language instructionresearch have not entered the reform picture--that the "universal" connotation assumes anoverriding solution inherent in the innovations, without attention to research and practicalwisdom related to issues of second language acquisition and bilingualism.In contrast to this overriding solution which glosses over the needs of the language minoritystudent, this paper draws on the author's experiences leading a major school reform effort in anurban elementary school with a large limited English proficient student population. Thediscussion provides insight into successful reform practices premised on the language andacademic needs of limited English proficient students. The context for reform as it existed at theauthor's school is presented first. Pedagogical premises that apply to programs for limitedEnglish proficient (LEP) students are outlined next, followed by the reform practices. Theintersecting factors of effective programming for LEP students and reform practices arepresented from the perspective of the author's experience with the reform effort. Concreteexamples of the reform practices are given as well as decision-making strategies that can lead toeffective schoolwide reform and success for all students. Finally, recommendations are made andconclusions drawn that focus on decision-making strategies, schoolwide accountability, and theprocesses that constitute meaningful reform for a culturally and linguistically diverse studentpopulation.Context for ReformBefore becoming Principal in Residence at the U.S. Department of Education from September1994 to May 1995, the author spent seven years as principal leading a successful and sustainedrestructuring effort at Linda Vista Elementary School in San Diego, California. Linda Vista is aninner city school in the San Diego Unified School District serving approximately 1000 studentson two campuses about one-half mile apart. Although the program is nongraded, the maincampus houses what is traditionally known as grades 1-6, while the satellite campus serves theearly childhood center (pre-K, K and some grade 1). The Linda Vista area of San Diego hastraditionally been the most culturally diverse in the city and has undergone numerous changes.Through the 1960s and early 1970s the area was home to roughly equal numbers of African-American, white, and Hispanic families. In the early 1970s, with a large influx of SoutheastAsian refugees, Southeast Asian enrollment at Linda Vista peaked near 70 percent. Recently, thenumber of Hispanics moving into the area has risen. Currently, Linda Vista's enrollment is 37percent Asian, 45 percent Hispanic, 8 percent African American, and 10 percent white. Morethan 94 percent of its students receive free and reduced-priced lunch, making the school eligiblefor schoolwide Chapter 1 funding (now known as Title 1). At least 77 percent of the students arenon- or limited-English proficient. In this diverse population, five major languages are spoken:Hmong, Lao, Vietnamese, Spanish and English. Students from newly arrived immigrant orrefugee families, often with no previous exposure to formal education, form the majority ofLinda Vista's student population.© 2008 Dr. Catherine CollierAll Rights Reserved36

The author came to Linda Vista with an extensive background in second language acquisitionand bilingual education. At that time, the reform movement was not widely known or advocated.Since the San Diego Unified School District was just beginning its process of change, pastpersonal leadership and pioneering experiences were relied upon to inspire the staff to change.An enormous challenge to implement reform was presented to the staff. The challenge meantthat they must raise their expectations for all students, raise their own morale above a marginallevel, and focus on involving parents where previously there had been almost no involvement.Throughout this struggle, a personal evolution in thinking took place that brought experiencewith reform practices and the premises for bilingual instruction together. It became obvious thatunlike the experiences during the period of advocacy for bilingual programs where models andrecipes were the agenda, school reform is a process of inquiry, decision-making and innovation,all born out of the unique setting in which reform is taking place. School reform requiresownership of the changes on the part of the school community; ownership that embraces conflictresolution and accountability as healthy processes.As the outside community began to look at some of the successes in this infant effort, theprincipal began to reflect upon this unique experience with sustained reform. As a result of thisreflection, several broad parameters were discovered that can be identified as reform practices toguide those who would embark upon such an effort. The evolution of this thinking raised animportant question about whether anything goes. As Zehler et al. (1994) point out, is the"universal connotation" the overriding solution? In the case of programs for LEP students, is astaff empowered to design a new curriculum, determine student groupings, or decide on anymode of language use? It was found that in any innovative process, there are certain principleswhich apply. A vision for innovation is created from solid research and practical wisdom. It isderived from a belief system that has the welfare of and a high level of achievement for allstudents as its goal, regardless of economic status, race, or linguistic differences.To include programs for language minority students, based on a framework that incorporatesresearch, practical wisdom and a nonbiased belief system, any schoolwide reform process mustbe based on certain understandings. Principally, educators planning these innovations need tounderstand the underlying assumptions governing second language acquisition in oursociocultural environment. The following premises for dual language programs are based onextensive research and practical wisdom (Miramontes, Nadeau, and Commins, 2000 ).Pedagogical Premises that Apply to Programs for LEP Students1. Knowledge is best acquired when learners actively participate in meaningful activities that aredevelopmentally appropriate.Rationale: Learning is not a piecemeal, behavioristic process, but an internal response toexperiences. The individual derives meaning from outside input when it is presented as anintegrated whole and relates to his or her own experiences (Piaget, 1969).2. The more comprehensive the use of the primary language, the greater the potential tomaximize the linguistically diverse student's academic achievement. Even though circumstancesmay limit the school's capability to fully develop a student's primary language, there are alwaysways to nurture it.Rationale: Language is an important vehicle for thinking (Vygotsky, 1978). The more solid thefoundation in the primary language the greater the chances of academic achievement in thesecond language (Cummins, 1979).3. There is a distinction between developing the primary language as a foundation for thinkingand learning, and using the primary language as a crutch for learning English.© 2008 Dr. Catherine CollierAll Rights Reserved37

4. Second language development creates an added dimension to instructional decision making.Instruction must reflect specific strategies designed to meet the needs of students who aredeveloping second language proficiency (Spener, 1991).5. Second language instruction must be organized to provide students the time, experiences andopportunities they need to fully develop language across a range of social and academic contexts.Rationale: Language is not learned for language's sake. It is a vehicle for thinking, forcommunication and for cultural transmission. Therefore, language is learned in the context ofthese experiences (Halliday, 1975).6. In order to develop bilingual academic proficiency, clear, separate, and meaning-enrichedcontexts for each language must be created during instructional time.7. Decisions regarding transition to formal second language reading and redesignations forprogram exit criteria cannot be made arbitrarily.Rationale: Predictors for academic success in English must be assessed developmentally. Amultifaceted approach that looks at primary language development, and English languageproficiency, as well as other predictors, should be followed (Nadeau and Miramontes, 1988).8. Direct oral language and comprehension support must accompany beginning English literacyand the expansion of English content instruction. (Nadeau and Miramontes, 1988).9. Instructional assessment must be based on the student's first and second language levels ratherthan on grade level or predetermined criteria. An appropriate assessment plan should addresslanguage and literacy development as well as content knowledge.10. Parents and the community need to play a major role in the learning and schooling of theirchildren.Rationale: Academic achievement is known to improve when families are involved in theirchildren's education (Brandt, 1989).11. Instruction must be organized to help students understand and respect themselves and theirown culture as well as the culture of the broader society. Planned cross-cultural interactions arean essential component of programs for all students.12. Sociocultural factors and the political context must be considered in making decisionsregarding every aspect of program planning.Rationale: Sociocultural factors such as language status, socioeconomic level, and teacherexpectations can have a direct impact on educational opportunities, the development ofbilingualism, and student achievement (McDermott, 1976).13. Teachers are decision makers, and as part of a learning community they are responsible forall aspects of the instructional program for linguistically diverse students.© Basic Premises, Miramontes, Nadeau and Commins, 2000.Identifiable Reform PracticesDrawing from the author's experiences with a sustained reform effort, and current research, it isapparent that the issue of reform must be discussed without using the word replication! So muchof the experience and research in reform, thus far, tells us that the process is unique to eachsetting. In fact, if there is one overriding principle, it is that systemic reform proceeds from theroots of the system and continually affects the entire system. In our anxiety to push the reformscale up, the tendency is to revert to recipes, school improvement notions and prescriptions.However, it is possible to avoid prescriptions as we recognize parameters for reform in educationthat might guide us through the process of change. The process is holistic. The parameters are tobe viewed as working together. No one parameter standing alone constitutes a reform process.© 2008 Dr. Catherine CollierAll Rights Reserved38

Following are a description and discussion of some broad parameters that have evolved from theauthor's own experience with a sustained reform effort, and from her experience in working withother educators involved with successful school innovations around the country. These include:decision making, teaching and learning, organization, assessment and accountability. Anexample of each practice from the author's experience at Linda Vista School is presented tocontextualize each parameter.Decision makingIn the decision making process of a change effort, the school organization is inquiry-based.Members of the school community are involved in continued inquiry, questioning and selfassessment.In this system, conflict resolution is healthy, but most discussions are about students,not about the needs of adults. However, the growth of students as well as of the adults is seen asdevelopmental. Both must be taken from where they are in a process of continuous growth.Governance in this organization means that all stakeholders are involved, and not merely byrepresentation. Involvement signifies not just power shifts, but a substantive investment in allaspects of the change process. There are very few hierarchies in this system; thus, parents,classified staff, and teaching staff think and talk about curriculum, instruction and otherimportant issues related to school change.After a thorough process of discussion and inquiry, decisions are made through consensus. It isnot a linear endeavor, but one that requires constant inquiry, assessment and decision making.Members of the reform organization make their decisions in cross-collaborative teams, not onlywithin their own cohort groups. Primary teachers will meet with upper-level teachers, classifiedstaff with teachers, parents and business partners with each of these groups.ExampleThe governance structure at Linda Vista school evolved from an established vision and from thecreation of an innovative program whose primary purpose was to benefit children. Theexperience begets the process. That is, by focusing on students and by jumping in full force,people learned by doing. The infrastructure and programs needed to support the reform effortevolved from the collaborative experience of creating new solutions.The following is an example of the inquiry process established at this school: A schoolwideportfolio assessment process had been established by the fourth year into the restructuring. Astaff development day brought the entire staff together to participate in a protocol that requiredthe teachers to form groups throughout the day to analyze the student portfolios across levels.Discussions were held related to the examination of student work, adjustments in the anchorpapers, and revisions of the rubrics the staff had previously developed. Classified staff acted asreflectors during each session providing feedback to the teachers regarding both content andprocesses. This activity exemplifies a decision making process that achieved not only a trustinginvolvement of all staff, but also a focus on accountability and improvement for the sake of thestudents. Ultimately, parents were also involved in these activities, but as stated earlier, becausethese mostly non-English speaking parents were unfamiliar with this level of decision making, adevelopmental approach was taken until they felt comfortable and competent in suchdiscussions. Learning is a developmental process for adults as well as children.Teaching and learningFundamental change in how teaching and learning are conceptualized is apparent in aprogressive organization. The learner is at the center of the process and is held accountable, butthis accountability is born out of an environment that asks students to continually assess theirown progress. Standards for achievement are apparent to students at all times. The curriculum© 2008 Dr. Catherine CollierAll Rights Reserved39

curriculum or "teaching to the test" is often the result. Redesigning such structures as howstudents are grouped for instruction, how they move up the continuum of learning, and how theflexibility of movement throughout the school day is achieved are but a few aspects to beconsidered when designing a vision for educational change.Again, this portrait of educational reform is to be viewed holistically. A school may have aninnovative instructional program or assessment system, but if student grouping, equity ofservices, or staffing arrangements are highly traditional and rigid, then the innovative practicesmay be constrained. Raising expectations for students by designing challenging standards for allmay not have a chance to flourish if these expectations and standards exist within a gatekeepingorganization.ExampleThere are many instances of organizational change that were instituted at Linda Vista. The mostsignificant ones will be discussed, but they must be viewed holistically. No one change inorganizational practice or unrelated changes will promote and support the foundation for truereform. The Linda Vista staff recognized, early on, that merely deciding to implement anongraded program without also introducing organizational flexibility would not allow a breakfrom the insidious deficit mind-set. Consequently, it was decided that students could move up thelevels of language proficiency at any time they were deemed ready according to the school'sstandards and assessment process. In order to do this, a whole set of organizational changes hadto be implemented. Class size had to be reduced to allow for sufficient room for students to movein and out of levels. Thus, dramatic staffing changes had to be negotiated requiring reallocationof resources to obtain several part-time teachers for the morning program. The focus for classsize reduction was not on achieving equality across classes, but rather on meeting specificstudent needs. For example, an entry level class designed for non-English speaking students, waskept at below 20 students, whereas other classes might be larger.The commitment to offer primary language instruction at whatever level possible led to thecreative assignment of two Southeast Asian teachers. Had the two teachers been assigned toregular classrooms, they would have been able to serve only the group of students in theirclasses. The decision that coincided with the vision for first language development was to assignthese teachers to lead a cadre of primary language instructors that would rotate on a dailyschedule and provide the instruction to students who were regrouped by their primary language.The entire morning schedule had to be arranged to accommodate this program, requiring thecommitment of the entire staff. These are but a few examples of the organizational changes thatwere required in order to realize the vision.Assessment and AccountabilityA state and national vision that strives to help the educational community set high standards forall students, regardless of race, language, disability or socio-economic status (SES), is withoutquestion, a goal that educators and families must embrace. But what the school does with thosestandards is crucial. Unless the school has its own internal mechanism for assessment andaccountability based on the external criteria, the goal of high achievement for all students mayfail.Probably one of the most significant identifiable reform practices is the assessment andaccountability process a school has undertaken. A key factor in this process is the alignment ofcurriculum, instruction, assessment and, ultimately, performance evaluations. Internal to theschool is an assessment system whose purpose is not only to assess student progress, but also tocontinually reevaluate the school's capacity to help all students reach high standards.© 2008 Dr. Catherine CollierAll Rights Reserved41

Through a continuous progress mode, assessments must measure authentic work of students andinvolve them and their parents in the process. Such an internal mechanism may lead tocontinuous improvement in which all staff have an opportunity to examine student work acrosslevels--not just their own students'--with the purpose of continually reevaluating not only studentprogress, but the assessment practices as well.Performance assessment is an individual process. Unlike traditional forms of assessment thatforce placement of students in ranks, along a curve or in a graded system, it evaluates thestudent's progress along his or her own developmental path. Students begin taking responsibilityfor their own learning by being intimately involved in the assessment process.Linked very closely to the schoolwide assessment practices is the accountability challenge.Schools going through the reforming process are facing the dilemma of being held accountablefor improvements in student achievement through standardized test results while beingencouraged to move into authentic and performance-based assessments. Consequently, thechallenge for schools is to learn how to develop program evaluations that incorporate indicatorssuch as those provided by portfolio assessment. An added challenge is how to extrapolateauthentic assessment data to provide accountability information to the public, parents and otheragencies. Can school personnel and their communities learn how to link norm-referenced dataand performance-based assessments, and make sense of the information in a way that will givelegitimacy to the data?ExampleOne of the most significant changes that occurred at Linda Vista was the evolution of aschoolwide assessment system. The evolution took place as a result of the vision that had as itsgoal the high-level achievement of all students through a developmentally appropriate,continuous progress teaching and learning process. As we said at Linda Vista, the experiencebegets the process. It was inevitable that an alternative assessment process would evolve fromthe staff's commitment to continuous progress. The following steps were taken:* Step 1: The staff embarked upon the task of developing language arts standards for all ofthe age grouping and language proficiency levels in the instructional design. This task emergedas the logical starting point since 77 percent of the students were English language learners.Subsequently, math standards were also developed. Although the goals set for all students werebased on California state frameworks and district expectations, benchmarks along the way weretargeted in the school's assessment design. For example, learner outcomes for the non-Englishspeaking student or the student transitioning into formal English reading were delineated tomaintain ongoing monitoring of student progress.* Step 2: Maintaining portfolios for each student was the method chosen for assessingwhether students reached the standards. This would be a school-wide expectation with theappropriate portfolio contents outlined in the assessment document for each level.* Step 3: Rubrics were then developed as a scale of descriptors for all student work. Thesewere designed around the developmental levels: Early Childhood, Primary, Middle and Upper.No grades, numbers, or scale rankings were used.* Step 4: The standard district progress report was changed to reflect the continuousprogress design. On the report, rubrics were checked for each reporting period. The report wasaccompanied by a hard copy of the student's portfolio that contained the appropriate work for thequarter. Each piece of work had an assigned rubric and the teacher decided, often with studentinput, the final rubric for the progress report. A hard copy was given to the parents since all© 2008 Dr. Catherine CollierAll Rights Reserved42

portfolio work was scanned and stored electronically. To exemplify the change in thinking aboutassessment and learning, the new progress report was called, The Growth Record.* Step 5: Teacher and administrator evaluations began to reflect the standards for studentachievement designed by the staff. Schoolwide data were then extrapolated from the portfolioassessment for reporting to the district and to the State School Report Card.At the time the author left Linda Vista School, the staff was prepared to balance existing normreferencedtesting with the data resulting from the schoolwide standards-based assessment. Theyhad matured in their thinking so that they were willing to take a broader view of accountability.They had begun to take a multifaceted approach and put the traditional data in its proper place.The continuing struggle is to make some reasonable connections between the results of the twotypes of assessment, in a way that does not undermine the standards-based approach.RecommendationsThe following are recommendations to the practitioner that suggest a perspective on reformpractices within the context of the basic premises for dual language instruction. These factorsshould be viewed as a whole and together drive the vision for instructional planning and design.The recommendations are:1. In order to examine all aspects of a proposed vision, a necessary step is to analyze the beliefsthat people hold about learning, languages and cultures.2. School staffs must first learn how to make good decisions given their own context.3. Decisions must be based on a schoolwide vision and the premises for dual languageinstruction.4. Models for educational programs provided from outside the school do not instill the necessaryownership required to sustain a change process.5. Assessment practices are what drive the organization and unless the school embraces its owncontinuous progress system, the members of the school community will not hold themselvesaccountable for students' academic achievement.6. In the initial stages of hammering out a vision, the entire staff must be at the table. Parentvoices must be heard in a variety of ways. Traditional representative groups may not reflect allparental views. Many alternative approaches for soliciting the input of all stakeholders may beneeded.ConclusionThe reform agenda in education and the practical wisdom and research related to the instructionof language minority students need to come together to form a comprehensive agenda. Thispaper provides several identifiable reform practices from the author's personal and extensiveexperience in leading a major reform effort. The premises for dual language instruction have alsobeen interwoven into the argument. With these concrete examples from personal experience thereader may conclude that given these two intersecting factors, (practices and premises), wisedecisions can be made -- decisions that are systemic, visionary, involve all stakeholders andensure high standards for all students. The decisions, however, must take into accountappropriate pedagogy, assessment and socio-cultural factors related to the instruction oflinguistically diverse students.Without question, the conversation about educational change has taken on a very differentperspective. The conversation now speaks to high standards for all students, to inclusiveness ofinstruction, to equity and access of resources, and to direct family involvement in the decisionsof the school. The next step is to push the envelope of action to the next level where a trueexamination of practices takes place that will finally ensure success for all students.© 2008 Dr. Catherine CollierAll Rights Reserved43

Article II.1 Teaching and Testing For AcademicAchievement: The Role of Language DevelopmentBy Muriel Saville-TroikeNCBE FOCUS: Occasional Papers in Bilingual Education,Number 4, Spring 1991IntroductionFrom Charles Fries' (1945) definition of mastery of a secondlanguage as control over the pronunciation and grammar within alimited vocabulary, to the recent development of content-basedESL (English as a Second Language)materials emphasizing theimportance of vocabulary in relation to cognitive information,the field of teaching ESL has come a long way in the past fortyyears.However, not all areas of ESL have changed at the same rate,and there has been a "cultural lag" in some areas, perhaps most notably in the area of assessment.Meanwhile, even as a new instructional paradigm is beginning to emerge in ESL, based in parton recent concepts in cognitive psychology such as schema theory, new developments are takingplace and those concepts are already being called into question and replaced by others. Thepurpose of this discussion is to sound a warning about premature fossilization of theoreticalframeworks and methods in ESL--even the latest and most promising ones--and to explore theimplications for assessment of some of these new developments, as well as of what we alreadyknow about first and second language learning. Some basic questions are raised about the needfor special assessment instruments for limited English proficient (LEP) students, while at thesame time calling for fairly radical changes in assessment procedures and interpretation.Focus On LanguageTo understand how language relates to academic achievement, we need to consider thatrelationship in terms of language development as more comprehensive cognitive processes. Ourefforts in ESL and bilingual education in the 1960s and the 1970s were founded largely on thepremise that linguistic differences, and particularly a lack of proficiency in English, are aprimary causative factor in the low academic achievement of students in American schools whoare from minority language backgrounds. This certainly seemed a plausible argument at the time,and still does, especially in those cases where students with limited English-speaking ability arerequired to learn exclusively through the medium of English. Certainly these students are at adisadvantage trying to understand instruction and express themselves in a foreign language,especially when they must compete with other students who have already mastered English. Wehave since learned, however, that such an explanation is overly simplistic, that students'competence in English at the time they come to school does not have as much impact on theirultimate academic success as do some other factors, and that a foreign language of instruction isonly one of the potential barriers to learning for students from linguistically and culturallydifferent backgrounds. Evidence for this comes not only from the negative educational© 2008 Dr. Catherine CollierAll Rights Reserved44

experiences of many groups of students, but also from the positive educational experiences ofothers (see Hakuta, 1990).Let us begin with one example of why our focus on language has generally been too narrow. Atraditional concern in ESL has included contrasting students' native language with English as abasis for assessing “interference" or "negative transfer" so that appropriate remediation couldbe applied. While native language interference is still a viable issue in applied linguistics, there islittle evidence that it has much effect on students' learning to read or on their academicdevelopment.Positive TransferOne thing that has not been adequately recognized is the extent to which positive transfer takesplace across languages, and across contexts of learning for limited English-speaking students.Most important is the extent to which that transfer is social and cultural, as well as cognitive innature.Let us first relate this concept to oral language development. In the earliest stages of a child'sfirst language acquisition, meaning exists in the social and cultural context of interaction.Linguistic forms (such as words and sentences) are first ascribed meaning only because they areembedded in these contexts. With time, through further social interact ional experience andcognitive development, "meaningful context" additionally comes to include the linguistic formsthemselves (Nelson 1981).The knowledge representations that develop and are brought to bear in the communicativeprocess have been labeled "schemata" or "scripts" by cognitive psychologists (Bartlett, 1932;Minsky, 1975; Rinehart and Ortony, 1977; Schank and Abelson, 1977). Scripts are typicallyorganized around are current situation or process such as "going to see Grandma," traveling bybus, or ordering food in a restaurant. They include such matters as knowledge of setting, theidentity and function of props, participant roles and responsibilities, expected activity sequences,rules for interaction, and norms of interpretation.Once they have been acquired, the schemata or scripts that are developed in this process areavailable for the interpretation of meaning in similar events even if the language that is beingspoken by other participants cannot be completely understood. When students begin learning asecond language, they do not start learning all over again, but interpret meaning in terms of whatthey already know--not just about language, but about the context in which it is being used, andabout strategies for social interaction. This means that the process of second language learning isheavily dependent on prior experience and apparently also on the nature and level of firstlanguage development.This transfer phenomenon is easiest to recognize in face-to-face interaction where extra linguisticcontextual cues are abundant. For instance, in my own research I have documented numerousexamples of children who do not share a common language successfully playing with oneanother, negotiating ownership of property, and settling disputes about rights and relationships(Saville-Troike, 1987; Saville-Troike, McClure and Fritz, 1984).We can observe this type of transfer phenomenon in U.S. school settings where there arestudents who have just entered from schools in other countries. Even if the students do notunderstand the language of instruction, those who have had prior school experience enterEnglish-medium classrooms already equipped with a knowledge base for making inferences andpredictions about the meaning of events that will occur there. This preexisting "script for school"accounts in large part for most students’ ability to behave appropriately even when they cannot© 2008 Dr. Catherine CollierAll Rights Reserved45

understand the words others are using, and it provides a meaningful context for the interpretationof new language forms.Because script knowledge is cultural knowledge, however, scripts can be expected to differaccording to social experience. Learning new scripts, or adapting preexisting ones, is thus part ofacculturation or resocialization to a new group and its structure. Efforts to assess limited Englishspeakingstudents need to be sensitive to points where misunderstanding does occur and mayinterfere with academic performance.In the school where we conducted research, Jo Anne Kleifgen and I (Kleifgen and Saville-Troike, in press) analyzed instances of successful communication between the very limitedEnglish-speaking students and their content-area teachers in regular English-medium classroomsto find out what does work in such situations. Based on an examination of videotapes, we foundthat both students and teachers employed a general "top down" cognitive processing strategy,using their understanding of the larger context for the interpretation of particular events andactions. Because of similarities in prior knowledge and experience on both sides, students oftenonly needed to comprehend a single key word to interpret questions, complaints, and directives,and to make an appropriate response. From the standpoint of second language teaching, it isnoteworthy that student errors in pronunciation and grammar had only a minimal effect, if any,on their negotiation of meaning with teachers and English-speaking students when otherdimensions of the situation were understood. Vocabulary knowledge, not surprisingly, was moresignificant than grammar or pronunciation. Even so, students and teachers often bridged lexicalgaps with nonverbal cues (pictures, gestures, and tone of voice), but interpretation again requiredembedding the interaction in known or apprehensible context.The students studied were children of foreign graduate students or visiting faculty at theUniversity of Illinois. Similar social class background, family educational level, andinternationally shared conventions of formal schooling provide the basis for commonalities inscripts between teachers and students. Thus, a relatively high level of positive transfer enablesstudents to function in a new school setting while having limited proficiency in the language ofinstruction--often more successfully, in fact, than native English-speaking students in the sameclassrooms from a less affluent and less well educated social class.Interactional CompetenceOne concept that should be questioned is the common working definition of “comprehensibleinput," which presumes that simplified sentence structure is a significant feature. Our analysis ofclassroom interaction demonstrates that background knowledge is crucial to interpretation ofmeaning when knowledge of language forms is limited, but sentence complexity does not seemto make much difference. Even in the context-reduced processing of written text, Floyd andCarrell (1987) have shown that providing ESL students with supplementary backgroundinformation significantly improves reading comprehension, while simplifying the syntacticstructure has no significant effect. First language readability studies yield similar conclusions: infact, simplifying sentence structure often makes a text more difficult for native speakers toprocess since it reduces redundancy. While students can often negotiate meaning in face-to-faceinteraction even with extremely limited linguistic skills, because of the familiarity or redundancyof the extra linguistic context in which it is situated, their attainment of a high level of academiccompetence requires the ability to decode and encode meaning in context-reduced tasks, such asreading and writing. (Cummins [1980, 1981, 1984] has written extensively on this point,particularly with respect to the different requirements placed on linguistic competence.)© 2008 Dr. Catherine CollierAll Rights Reserved46

While not disagreeing that interpretation of written text requires a higher level of language skills,what should be brought into question is the dichotomization of language competence into CALP(Cognitive Academic Language Proficiency) and BICS (Basic Interpersonal CommunicationSkills) which many have adopted, with common equation of CALP alone with academicachievement. (For a discussion of language competences, see Harley, Allen,Cummins and Swain, 1990.) This is, at least, partially due to the nature of most assessmentinstruments, which abstract tasks from contexts in which they are actually learned and used.Achievement in school is actually heavily dependent on interactional competence--includingdisplay functions which teachers use for continuing informal assessment and calibration ofinstruction (Mehan, 1979). Further, sociolinguistic competence is important in conveying the"good attitude" toward school which receives heavy weighting both in teachers' evaluation of"readiness" and "progress" and in determining students' opportunities to learn. As Cummins(1980) points out, teachers may overestimate students’ linguistic ability to handle contextreducedtasks if they appear linguistically competent in social interaction. On the other hand,however, many students who lack competence in interaction can handle more cognitivelydemanding and context-reduced tasks, but may not be offered the challenge or the opportunity todo so. We will return to this issue, which is essentially one of instructional bias, since it meritsmuch more attention.Development of Strategies in The Native LanguageIn addition to the higher level of language skills required to interpret written text, it is alsoimportant to recognize that academic success requires such strategies as listening or reading forthe main point, generalizing, making logical inferences from known information, andconstructing more complex schemata-- strategies which are not specific to a particular language.Again, once these strategies have been developed in the native language, they apparently transferquite readily to academic tasks in a different language.Among the students I have studied who began school in another country, I have found thatreading achievement in English as a second language is more dependent on reading achievementin their native language than it is on relative oral proficiency in English. This is true even whenthe language the students first learned to read is written in symbols which are quite differentfrom our Roman alphabet, such as Japanese, Korean, and Arabic (Saville-Troike, 1984).Most second language teaching and research has focused on the linguistic factors which makereading comprehension possible, and they are, of course, important. However, as I haveindicated, research in both first and second language on the intersection of backgroundknowledge and reading comprehension shows that prior experience and expectations have asignificant effect on the process. Background knowledge has a direct impact on how readersinteract with what they see on a page. It affects how their thinking is directed as they read alongand what kind of sense they make of a given text. The expectations and interpretive processesreaders bring to the material and the expectations and understandings which they develop ontheir way through the material are directly related to their experience of the world, and theircognitive schemata, attitudes, and values, as well as their previous experience with the printedpage .In addition, academic competence requires knowing how to use language as a tool inacquiring knowledge and in performing analytic processes, but these skills again appear to relatemore closely to language competence in a general sense, rather than to any particular language.© 2008 Dr. Catherine CollierAll Rights Reserved47

Programs often choose to separate languages of instruction, but students who are academicallyengaged probably cannot and should not separate them.In my own research (Saville-Troike, 1984), I have found that most of the students who achievedbest in content areas, as measured by tests in English, were those who had the most opportunityto discuss the concepts they were learning in their native language with peers or with adults,even when they were mainstreamed in English-medium classes. Further, our research on privatespeech (using wireless microphones attached to students; Chen,1987; Saville-Troike, 1988) hasshown that students practice to themselves in their native language what they have been learningin English, providing evidence that their understanding often exceeds their ability to display theirknowledge in English, a point which has more than considerable significance for assessment.First Language Development And Academic AchievementMajor support for the suggestions of Cummins (1980, 1981, 1984) and others concerning theimportance of prior first language development for second language academic achievementcomes from the study of immigrant families(for example, Collier, 1987). There is goodindication that the longer that students are schooled in their native country before immigrating tothe U.S., the higher their school achievement in the U.S. and their learning of English is likely tobe. This finding is one of the most important to emerge in recent years.The now classic study of this phenomenon was reported by Skutnabb- Kangas and Toukomaa(1976) on the basis of a study of Finnish immigrant children in Sweden. They revolutionized theprevailing thought that the younger that children begin school in the new country the better theywould do academically and in learning the second language. Skutnabb-Kangas and Toukomaafound that the optimum time for immigration appeared to be about10-12 years of age. Based onresearch conducted both in Illinois and California, Gonzalez (1986) has found that sixth graderswho had immigrated to the U.S. after two years of education in Mexico consistently did better asa group on the CTBS English reading comprehension test than students who had started schoolin this country. It should be emphasized, however, that what is involved is more than languagealone. Part of the answer also clearly lies in the types of social experiences children have whichcontribute to their knowledge structures. Those which more nearly match the experiences andexpectations of school are going to transfer more readily. Children like those I have studied inIllinois, from well-educated families with extensive literacy-related experiences, are very likelyto succeed in our schools no matter what their entry-level competence in English. Lesseducationally advantaged children have also developed knowledge structures before they cometo school, but the widely held "deficit" position considers their language and culture a barrier tolearning--a source of negative interference-- rather than a resource for potential positive transfer.An alternative which would make their success more likely is to adapt school experiences andallow continuity and transfer of what students already know, and of their interactional andlearning styles. This has been done in a dramatic way in the Kamehameha program for Hawaiianchildren in Honolulu (Au and Jordan,1981) but is not likely to be widely adopted, as institutionsgenerally expect the individual to change to meet their demands, and not vice-versa.Developing The Second LanguageFor limited English speakers who have not yet fully developed their native language skills, thecontext-reduced tasks of reading and writing, or literacy-related processes like inferencing andcomplex schema formation, are obviously more easily fostered in the language students are mostfluent in. Those skills and processes will then transfer to English. Even when we are sure this is© 2008 Dr. Catherine CollierAll Rights Reserved48

sound educational practice, however, we must recognize the powerful influences of culture andpolitics on our schools. There is a false but pervasive belief in our nation that children should"get into English” as soon as possible or they will be retarded in learning. Because this is amatter of faith, based on profound social attitudes and convictions, evidence to the contrary hashad little impact on policy (for discussion, see Krashen, 1991).Unfortunately, initial emphasis ondeveloping English language skills often involves placement and instructional content which isbased on students supposed language proficiency level rather than what would be considered“normal" curriculum content in the larger educational setting, or their level of cognitivedevelopment and prior learning in their native language. This creates a separate (and unequal)curriculum track for LEP students which is often discriminatory in effect, if not in intent. AsMoll (1986)has pointed out: The problem of instructional bias and of watering down thecurriculum is, of course, not limited to non-native English speaking students; it may occur in theeducation of speakers of non-standard English dialects or of students whose language or culturalbehavior does not conform to that of the dominant society. In fact, as Anyon (1980), amongothers, has shown, watering down the curriculum may be viewed as part of a broaderstratification of instruction across social class groups. In the present move toward so-called"sheltered English" programs, well-intentioned as they may be, we are running a great risk thatthe isolation of LEP students from native speakers and regular classes may in fact serve to retardtheir linguistic and academic development. Research by Nagy and others (Nagy, Anderson andHerman, 1987; Nagy, Herman and Anderson, 1985) has shown that a high percentage ofchildren's vocabulary growth during the elementary school years is not based on directvocabulary instruction at all, yet successful reading and academic achievement in content areasdepends heavily on the acquisition of just this other vocabulary.This instructional bias--teaching to children's low level of English--is found even in bilingualprograms, regardless of the children's academic competence in their first language. Moll (1986)also cites evidence that this same phenomenon has more recently become evident in computerinstruction: Poor and limited English proficient students do drill and practice; affluent andEnglish-fluent students do problem solving and programming....Part of the problem is theoverwhelming pressure to make [limited English] students fluent in English at all costs. LearningEnglish, not learning, has become the controlling goal of instruction for these students, even if itplaces the children at risk academically.And again, more than language is involved in this phenomenon. Much of the massive schoolfailure among students from non-English backgrounds must be attributed to attitudes both thoseeducators hold toward minority students and students’ perceptions of themselves and of theschool. Teachers of the advantaged foreign children in Illinois described earlier knew they wouldbe good students before they even met them. On the other hand, in one first grade classroomvisited near the beginning of the school year, the teacher had already determined that a not-soadvantagedgroup of Spanish-speaking children in the class, to quote her, "would not be able tolearn to read this year."Academic AchievementWhy do large numbers of our Spanish-speaking students not succeed in school? Again, I doubtthat the Spanish accent or grammatical interference in their English has much to do with it, atleast directly. As mentioned earlier, Gonzalez (1986) has shown that immigrant students fromMexico who attended school for two years prior to coming here had higher reading scores inEnglish by the sixth grade than did Spanish-speaking peers who began school here. In short,© 2008 Dr. Catherine CollierAll Rights Reserved49

students with two years less instruction in English and here we are not talking about advantagedmiddle class children did better in English than those who had two years more instruction in theU.S. Why should this be, and why should it be that, nationally, blacks consistently average belowHispanics in achievement scores, even though blacks are almost all native speakers of English?The answers are not simple to find, and we should beware of simplistic unidimensionalresponses. Educational programs for non- English-speaking students, whether bilingual or all-English, do not exist in isolation from the schools, school systems, and communities in whichthey are embedded, any of which may exert more effects on program outcomes--for good or ill--than many of the efforts that are expended in instruction, curriculum design, or materialsdevelopment. Ogbu (1978) has argued that the long-term effects of social and economicdiscrimination may negatively affect the cultural attitudes and expectations of minoritycommunities. At the same time, research on school "climate" and the effects of educationalleadership at the school level show that these influences are not wholly deterministic, and thatthe attitudes and behaviors of principals can affect academic results for an entire school. Thefindings of recent "school effectiveness” research (Rutter, 1983) indicate that whole-schooleffects do exist and maybe considerable.It is easy to take refuge in the "home-school discontinuity hypothesis," or the "linguisticmismatch hypothesis," to explain the educational problems of many of the non-English-speakingstudents in our schools. But these simplistic answers--though they are certainly relevant--do notaccount satisfactorily for the academic stratification of blacks and Hispanics in the U.S.--orAsians, for that matter. Here we are in a larger realm of the effect of attitudes on instruction,learning opportunities, motivation, and cognitive demands. Language does not exist in a vacuum,and how it is developed, and for what purposes, lies beyond but is inextricably intertwined withlanguage form and use. The recognition of these issues helps frame the problem for any effort torelate language assessment to academic placement and achievement.Recent developments in the field of cognitive psychology have also begun to emphasizerecognition of the complexities of human knowledge and behavior. One of those working on thecutting edge of this field nationally is Rand Spiro (Spiro, Cousin, Feltovich and Anderson, 1988;Spiro, Vispoel, Schmitz, Samarapungavan and Boerger, 1987), who is challenging theoversimplification and limited scope of previous work in schema theory and arguing that thereare many areas of knowledge which are best characterized as "ill-structured domains," andrequire a more complex approach to understand. He is proposing that we adopt Wittgenstein'smetaphor of landscapes in examining these areas, since they can be looked at from differentperspectives, and may look different depending on the perspective from which they are observed.This view is consonant with other recent developments in science generally, which move awayfrom the older notion that the way to study a phenomenon is to artificially simplify it as much aspossible, and to abstract away from the complexities of natural contexts. While at an early stagein the development of a science this approach may have some heuristic value, there is a seriousdanger that the understandings which result may in fact be an artifact of the simplification, andmay have to be rejected when an analysis is undertaken which more fully acknowledges thecomplexities. The danger is greater in that the illusion that we are dealing with a "well-structureddomain" contributes to development of overly rigid schemata, which have been shown to inhibittransfer and application of knowledge in an "ill-structured domain," such as education. Cziko(1989) has recently come to a similar conclusion, namely, that much of the failure of educationalresearch in particular, and the social sciences more generally, to arrive at valid generalizations© 2008 Dr. Catherine CollierAll Rights Reserved50

arises from the efforts to abstract, simplify, and analyze data indexically rather than in thefullness of their ecological context.Language AssessmentWhat then of testing and assessment of LEP students for academic purposes? Languageassessment in the past, developed largely by linguists working with specialists in measurement,neither of whom have immediate experience in educational contexts, has generally followedpositivistic models and has been focused on language rather than on language in relation toacademic proficiency. If we ask, what is really important to assess in regard to a LEP student'schances for succeeding in a regular English-medium classroom?, we are posing a very differentkind of question than has been asked in the past, and one which, considering the complexities Ihave discussed, requires a very different answer.First of all, it is important to recognize that existing language assessment measures show a verylow productivity with regard to academic achievement, suggesting that they measure the wrongthings from an educationally-significant perspective, and are largely irrelevant for academicpurposes. Such tests reflect the earlier simplistic view that language was the only, or principal,factor affecting academic achievement, so such results are not surprising, and indeed might havebeen expected. Given our present realization of the complexity of factors affecting achievement,what sort of assessment program might we need that would give due recognition to all of thesefactors? Since it is evident that different factors and diverse configurations of factors affectachievement of LEP students differently in various contexts, one approach which might beproposed would be to measure as many of these factors as is feasible, and to examine theirrelation to academic achievement independently in each context. Before this can be done, it willbe necessary to carefully map the areas which are to be assessed. We cannot measure knowledgeor language proficiency directly, since we cannot simply insert electrodes into the brain to do so.Consequently we must construct maps representing the areas of the landscape to be assessed, andthen develop instruments which validly sample the maps.Assessment should be multidimensional. The multidimensionality of the aspects which need tobe considered move us beyond two- dimensional cubes to three-dimensionally interconnectedarrays of these cubes in a model resembling Rubik's famous six-sided cube. Each of these faceswould then be divided into sub-areas: language, for instance, would include subdivisions fordifferent skills in both native and second languages, and academic achievement subdivisions forcognitive processing capacities, content knowledge, and performance skills. The main point ofsuch a model is to recognize not only the complexity of the facets involved, but theirinterconnection as well. This obviously goes beyond what can be portrayed on a twodimensionalpage, but is quite feasible with computer-modeling capabilities. Another approach isto consider what constitutes the ingredients of successful academic achievement among nativeEnglish-speaking children, and how the schools at present routinely measure student progressand use such information in their ongoing operations. Since reading ability in English is thesingle most important skill determining school achievement beyond the third grade, this is amajor criterion in measuring student progress, either in the form of formal tests, or in informalteacher assessment. As has been well-known for a number of years, the most highly correlatedsub score within a reading test with the overall score is that for vocabulary knowledge, so muchso that the vocabulary subtest is often administered as a proxy for the full test. Thus we in ESLare rediscovering what educators have known for some time--that vocabulary knowledge is oneof the most important determinants of academic achievement, and vocabulary tests provide one© 2008 Dr. Catherine CollierAll Rights Reserved51

of the most reliable measures of academic progress. While such tests, either standardized orteacher-made, are clearly a type of language test, note that they are related to normativeexpectations of the types of abilities, skills, and knowledge associated with placement on theacademic scale from success to failure. As such, they do not simply test knowledge of isolatedwords, but rather they are indexical for a larger body of concepts, schemata, and cognitive skillsconsidered central to achievement in the educational enterprise as it is presently defined.ConclusionPerhaps, then, an answer to our search for an adequate and appropriate measure of academiclanguage proficiency has already been found, and all we need to do is adopt some currently usedstandardized reading tests for use with LEP students. Such tests are highly integrative in natureand tap a large proportion of the skills which determine school achievement. While some suchsolution may, in fact, prove to be reasonable, it is not without some caveats and suggestions fornecessary supplementation. As I indicated at the beginning of this discussion, radical changes areneeded in testing procedures and interpretation. For example, Garcia (1987) has shown that LEPchildren may misinterpret English words or reading passages based on erroneous lexical orsemantic associations with their own language, or on different cultural schemata or personalexperiences. This research emphatically showed that scores by LEP students on such tests shouldnot betaken uncritically at face value, but that debriefing interviews afterward are essential tocheck on comprehension and reasons for responses. (This maybe an equally valid point fornative English speakers.) Secondly, we may look to the model of Special Education forassessment and placement procedures, since Federal law and many state plans require thatstudents from non-English language backgrounds must be assessed in their primary language aswell as in English before they are placed into a special program. Despite all of the researchpointing to the importance of the native language in cognitive development, we have failed toinsist that where appropriate (e.g., not where native language loss has occurred or skills aremarginal) all LEP students should have a right to assessment in their native language as well asin English, and that placement judgments should not be based on English performance alone.Further, tests of English language proficiency alone which are not based on or related to standardcurriculum content for native speakers should not be allowed to be used as the basis for academicplacement.© 2008 Dr. Catherine CollierAll Rights Reserved52

Article II.2 Cross-Cultural Communication: AnEssential Dimension of Effective EducationBy Orlando L. Taylor, Ph.D.Copyright © 1987 Revised and reprinted 1990.The Mid-Atlantic Equity Center - 5010Wisconsin Avenue, N.W., Suite 310 -Washington, D.C. 20016 - (202) 885-8517.IntroductionEverything that occurs within a school, andespecially in the classroom, involvescommunication, the act of sharing information.Sometimes communication involves the use oforal or written verbal symbols. On otheroccasions, communication involves varioustypes of nonverbal symbols, including bodylanguage. Communication is the medium for instruction, assessment, interpersonal relationships,group interactions, parent and community relations and counseling. Most behavior problems inschools, and their resolutions, involve some type of communication. In sum, communicationpermeates education. Communication is culture bound. The way an individual communicatesemanates from his or her culture. Of course, a person may know more than one culture or may becompetent in a combination of cultures. Nonetheless, one basic truth prevails: communication isa product of culture. Students with different cultural norms are at risk if teachers have littleknowledge, sensitivity or appreciation of the diversity in communication styles. Such teachersmay perceive differences as problems and respond to students' diversify with negative attitudes,low expectations and culturally inappropriate teaching and assessment procedures. Culturallyand communicatively diverse students, in turn, may respond with low self concepts and lowacademic achievement to a school climate they perceive as hostile. The result is reflected in thesestudents' excessive placements in special education, reduced placements in talented and giftedprograms and high suspension rates.Cultural Diversity in AmericaThe Neglect of Cross Cultural Communication Issues in SchoolsSelf-Assessment on Communication and CultureTable I: What Do I Know About Culture, Communication and Language?Cultural Diversity in AmericaThe United States is currently experiencing radical demographic shifts which are changing thecolors and the cultures of its citizenry. According to recent statistics, one American in fourcurrently defines himself or herself as non white. By the year 2010, because of higher birth ratesand immigration trends, non whites are expected to constitute more than one third of the© 2008 Dr. Catherine CollierAll Rights Reserved53

American people, and upwards of 50 percent of its school aged population. By the year 2050, theaverage U.S. resident will trace his or her descent to Africa, Asia, the Hispanic world, the PacificIslands, the Middle East almost anywhere but white Europe.As diversity increases in our nation's schools, teachers, administrators and other educationpersonnel will be challenged increasingly to become more knowledgeable about the assumptions,attributes, and norms of a range of cultures. These challenges will occur in every dimension ofschool life from the curriculum to the communication that occurs in classrooms. As a result, theissue of cultural diversity will, by necessity, have to be taken into account if effective educationis to be a reality in our nation's schools.The Neglect of Cross Cultural Communication Issues in SchoolsThe failure of many schools to take cross cultural communication issues into account cancontribute to school related problems experienced by specific groups of children. The currentcrisis of African American males in many of the nation's schools demonstrates this point.According to recent statistics, the percentage of African American males who graduate from highschool has decreased since the mid 1970's. A similar trend exists for the percentage that go on tocollege. In 1989, 34 percent of young African American males attended less than four years ofhigh school and only 11 percent attended four years of college or more. In 1990, one out of four(23%) African American males ages 20 29 were in the criminal system, while only 6% of whitemales and 10% of Hispanic males were in the system. Within many schools, it is a welldocumented fact that African American males are disproportionately placed in special educationand speech language pathology programs, and are more likely to be recipients of disciplinaryactions. Recent research has shown that language and communication norms among AfricanAmerican males, particularly those of lower socioeconomic status, are related, at least in part, tothese problems. For example, lower SES African American males are likely to see the speakingof ethnically based English vernaculars, and the use of urban argots, to be markers of masculinityand defiance of white standards. While adherence to ethnic vernaculars allows validation to beachieved within the African American male peerage, their employment within the school settingvirtually guarantees academic problems and, sometimes, social problems. When students'communication norms conflict with the school's communication norms, erosion of academicperformance and acceptable classroom behavior frequently follows. Thus, while the use of Blackvernacular speech by African American males may be perceived as "fresh" by peers, it is viewedby teachers, all too often and incorrectly, as indicators of a slow learning, violent, undisciplined,and obnoxious individual. In order to address these issues, schools must build into their curricula(particularly in the language arts area) the notion that there is a time and place for all language.In this way, respect can be given to students' culturally based vernaculars when used in informal,nonacademic activities, while teaching them the necessity and validity of the school's languagein formal academic settings. Indeed, the creative teacher uses the vernacular linguistic system forcontrastive analysis during the process of teaching standard English. This approach encouragesthe African American male to adapt a bidialectal approach to language use. Techniques forimplementing these approaches to instruction are discussed elsewhere in this booklet. Inaddition, the school must provide examples of strong male images that are able to alternatelyspeak the school's language or the vernacular language, as necessary. Jesse Jackson, Spike Leeand Arsenio Hall represent good examples of highly successful and popular African American© 2008 Dr. Catherine CollierAll Rights Reserved54

males who can serve as positive role models for young students, and who demonstratecompetence in the standard language required in formal situations as well as the communitylanguage when required. Though the condition of the African American male has been describedin detail, many of the concepts presented in this publication can be applied to all culturallydiverse groups.Self Assessment on Communication and CultureThe quiz presented in Table I assesses a person's knowledge of the relationships among culture,communication and language. Before reading further, you may want to take this test.TABLE IWhat Do I Know About Culture, Communication and Language?Directions: Circle A for each statement with which you agree or D for each statement with whichyou disagree.A D 1. One's culture and one's race are usually one and the same.A D 2. Culture consists exclusively of a group's art, music, dance, food, language and dress.A D 3. Cultural groups are generally mutually exclusive of one another.A D 4. Cultural traits tend to have a genetic base.A D 5. In general, people who speak the same language are members of the same culturalgroup.A D 6. People are usually aware of all of the rules of their culture.A D 7. Culture is expressed exclusively by one's verbal behavior.A D 8. The only significant components of one's oral language are vocabulary and rules onpronunciation and grammar.A D 9. Standard English is the correct way to speak at all times.A D 10. There are universal norms for acceptable communicative behavior within the UnitedStates.A D 11. If a student violates the school's cultural or communicative norms, it is almostalways the act of defiance.A D 12. In general, speaking a nonstandard dialect suggests low cognitive development.A D 13. Standard English has more and better structures than other varieties of English.A D 14. Most standardized tests are based on rules of English used by individuals in all© 2008 Dr. Catherine CollierAll Rights Reserved55

linguistic groups.A D 15. Standard English is White English.A D 16. In general, students from poor families do not communicate as well as those frommiddle-class families.A D 17. In general, African American students do not communicate as well as Whitestudents.A D 18. Parents who do not speak standard English should avoid talking to their children toprevent them from developing poor speech habits.A D 19. If students are to learn standard English, they must unlearn any other variety ofEnglish that they speak.A D 20. African American English is by definition a nonstandard variety of English.You can get the answers to these questions from your instructor.Discovering Characteristics Of Other CulturesWhen teachers ask "How can I learn about culture X?", all too often teachers are discouraged tolearn that there is no comprehensive book (or list) available on culture X. The lack of readingmatter on various cultures is, in some ways, a blessing, since knowledge of other cultures isbetter acquired by experience than by study. Thus, teachers should plan to experience culturesdifferent from their own, particularly the cultures represented in their schools and classrooms. Tobe valid, these experiences should incorporate a few basic principles and discovery techniques.As you begin to discover the characteristics of other cultures, remember that cultures varyinternally and are changeable. There are usually many cultural differences within a single race ornationality. It will be useful to keep the following principles in mind:Firsthand experience is necessary to understand many subtleties of any culture.Feelings of apprehension, loneliness or lack of confidence are common when visiting andexperiencing another culture.Differences between cultures are often experienced as threatening.What is logical and important in a particular culture may seem irrational and unimportant to anoutsider.In describing another culture, people tend to stress the differences and overlook the similarities.Stereotyping due to generalizing may be inevitable among those who lack frequent contact withanother culture.Personal observations of others about another culture should be regarded with skepticism.Many cultures often exist within a single race, language group, religion or nationality,differentiated by age, gender, socioeconomic status, education, and exposure to other cultures.All cultures have internal variations.Cultural awareness varies among individuals.One's own sense of cultural identity often is not evident until one encounters another culture.Cultures are continually evolving.© 2008 Dr. Catherine CollierAll Rights Reserved56

Understanding another culture is a continuous process.One should understand the language of a culture to best understand that culture.An understanding of these principles can help you to become a more insightful observer of othercultures. Saville-Troike (1978) suggests a specific set of questions to guide those who seek tounderstand another culture. Some of her questions are presented in Table II . They can providethe basis for interesting, informative and stimulating discussions among teachers, students,administrators and parents.TABLE II Questions to Ask About CultureFamily StructureWho is considered to belong in the family?What are the rights, roles and responsibilities of the members?Life CycleWhat are the important stages, periods and transitions in life?What behaviors are inappropriate or unacceptable for children at various ages?RolesWhat roles are available to whom?How are roles acquired?Interpersonal RelationshipsHow do people greet each other?Who may disagree with whom?How are insults expressed?CommunicationWhat languages and dialects are spoken?What are the characteristics of speaking "well"?What roles, attitudes and personality traits are associated with particular aspects of verbal andnonverbal behavior?Decorum and DisciplineHow do people behave at home and in public?What means of discipline are used?ReligionWhat religious roles and authority are recognized?What should an outsider not know or acknowledge knowing?Health and HygieneHow are illness and death explained?How are specific illnesses treated?© 2008 Dr. Catherine CollierAll Rights Reserved57

FoodWhat is eaten, in what order and how often?What are the rules for table manners, including offering foods, handling foods and discardingfoods?Holidays and CelebrationsWhat holidays are observed? For what purposes?Which holidays are important for children?What cultural values are instilled in children during the holidays?Dress and Personal AppearanceWhat significance does dress have for social identity?What is the concept and value of beauty and attractiveness?ValuesWhat traits and attributes in oneself or others are important? Undesirable?What attributes in the world are important? Undesirable?History and TraditionsHow are history and tradition passed on to the young?How do cultural understandings of history differ from "scientific" facts or literate history?EducationWhat are the purposes of education? What kinds of learning are favored?What teaching and learning methods are used in the home?What are parental expectations for boys versus girls?Work and PlayWhat behaviors are considered "work"? "Play"?What kinds of work are prestigious? Why?Time and SpaceWhat is considered "on time"?What is the importance of punctuality?How important is speed of performance?How are groups organized spatially by age, gender and role?Natural PhenomenaWho or what is responsible for rain, thunder, floods and hurricanes?Are behavioral taboos associated with natural phenomena?© 2008 Dr. Catherine CollierAll Rights Reserved58

Pets and AnimalsWhich animals are valued and for what reasons?What animals are considered appropriate as pets? Inappropriate?Art and MusicWhat forms of art and music are most highly valued?What forms of art and music are considered appropriate for children to perform or appreciate?Expectations and AspirationsDo parents expect and desire assimilation of children to the dominant culture, language ordialect?What cultural values are expected to be maintained despite the degree of formal education?(From Saville Troike, M. A Guide to Culture in the Classroom. Rosslyn, VA: NationalClearinghouse for Bilingual Education, 1978.)Culture, Communication and LanguageNow that your "cultural eye" has been sharpened, let us look at how culture influences students'language skills and their learning of standard English, the language of education. The concept ofcommunicative competence (Hymes, 1962), based on one's knowledge of the rules of languagestructure and language use within a given culture, will be useful. A major responsibility ofteachers at all grade levels is to teach the language and communication skills needed foracademic success, and for career and social mobility. Many students come from cultures whichuse different, though valid, communication and language systems from what is considered"normal" in the classroom. The study of sociolinguistics can help us understand different systemsas a means of improving the quality of our instruction in language and the communication arts.SociolinguisticsTable III: Some Varieties of Nonstandard American EnglishCultural Differences in DiscourseExamples of Cultural and Communicative TendenciesTable IV: Examples of Verbal and Nonverbal CommunicationThe Significance of Culture-Based Communicative Behavior in SchoolTable V: Culture and Communication Conflicts in the ClassroomSociolinguisticsSociolinguists examine social and cultural influences on language behavior. Among the mostimportant concepts to emerge are those relating to dialects and language standards.Sociolinguists have documented the presence of dialects in every language. These dialects, all ofwhich are legitimate, are associated with educational, economic, social and historical conditions.To linguists, the word "dialect" refers to a way of speaking a language, and not to an incorrect© 2008 Dr. Catherine CollierAll Rights Reserved59

way of speaking a language. While all dialects of a given language are linguistically legitimate,some achieve social prestige. In literate, economically developed societies, the dialect spoken bythose with the most formal education, the highest socioeconomic status and the greatest degreeof political power tends to acquire the greatest social prestige. Typically, it becomes the standardfor the culture, for writing and for education. Standard dialects also provide a medium throughwhich persons from different linguistic backgrounds can communicate with one another. Socialand regional variations may exist within standard dialects as long as they conform to specifiedlinguistic rules, largely grammatical in nature. Standard English, therefore, should not beconsidered "Northern English" or "White English," since it is spoken, in one form or another, inall parts of the United States and by some members of all racial and cultural groups. At the otherend of the social spectrum, so called nonstandard dialects are generally spoken by the "havenots:" the powerless, the less educated, the less economically well off and the less sociallyprominent. While legitimate linguistically, these dialects tend to be unacceptable to the "haves"of society. In American English, nonstandard dialects exist within all racial, ethnic and regionalgroups ( see Table III ). Each dialect is a product of distinct social, historical, cultural andeducational factors. All are legitimate in that they represent the concepts, needs and intentions oftheir speakers.TABLE III Some Varieties of Nonstandard American EnglishAppalachian English"He just kept a begging and a crying and a wanting to go out." (He persisted in begging, cryingand wanting to leave.)Athabascan English (Alaska)"Most time we play games." (Most of the time we play games.)African American English Vernacular"He be scared, but I be brave." (He is usually scared, but l am usually brave.)General American Nonstandard English"don't nobody want none." (Nobody wants any.)Keaukaha English (Hawaii)"I no can place that name." (I cannot place that name.)New York City Nonstandard English"She's a good cook, your mother." (Your mother is a good cook.)Southern American Nonstandard English"I mon' rest." (I am going to rest.)Spanish Influenced English"Carol left yesterday. I think is coming back tomorrows." (Carol left yesterday. I think she iscoming back tomorrow.)For a variety of reasons, including negative public attitudes and inadequate teaching models,nonstandard English speakers often do not effectively learn standard English in school. Withoutcompetence in standard English, students will fail academically and face diminished career,social and life options. Many students who do learn standard English do so at a great price:devaluation or rejection of their home or community dialect. When competence in standardEnglish is coupled with rejection of one's own home or community dialect, it may lead to seriouspsychological and identity problems (Chambers, 1983). In the United States, the schools' failure© 2008 Dr. Catherine CollierAll Rights Reserved60

to teach standard English is reflected in the poor performance of nonstandard English speakerson achievement, aptitude and diagnostic tests. Perhaps the most alarming evidence of this failureis the low performance of nonstandard English speakers on tests used to place students inremedial or gifted programs. Virtually all of these tests presume competence in standard English(Oakland, 1977). Many African American children, usually from working class homes orcommunities, speak a nonstandard variety of English. This variety, often referred to as BlackEnglish Vernacular, is thought by many sociolinguists to reflect African influences on AmericanEnglish, and is reinforced by social isolation, segregation and group identity.Cultural Differences in DiscourseIn addition to differences in pronunciation, vocabulary and grammatical structures amongcultural groups, variations also exist in the rules for general discourse in oral communication,covering such specific acts as narratives and conversation. In communicating with one another,teachers and students naturally will follow the assumptions and rules governing discourse withintheir respective cultures. Discourse rules govern such aspects of communication as:• Opening or closing conversations;• Taking turns during conversations;• Interrupting;• Using silence as a communicative device;• Knowing appropriate topics of conversation;• Interjecting humor at appropriate times;• Using nonverbal behavior;• Expressing laughter as a communicative device;• Knowing the appropriate amount of speech to be used by participants; and• Sequencing of elements during discourse.With respect to narratives, Gee (1985) claims that story telling during sharing time in the earlyschool years helps to provide students the foundation for reading and writing instruction in latergrades. Both he and Michaels (1981) report that schools and teachers prefer linear, single topicstory telling, the style that is compatible with strategies encountered in reading and writingactivities. These topic centered stories are characterized by tightly structured sentences thatreflect on a single or small set of highly related topics. The speaker presumes little sharedknowledge with listeners. Topic centered stories, therefore, tend to be very explicit and containgreat detail, emphasizing more telling than showing. Topic centered stories are thought by someto be associated with the field independent cognitive style. I Gee and Michaels note further thatmany working class children, particularly those from oral cultures, tend to prefer a topicassociating narrative style. These story tellers presume a shared knowledge with the audience, domore showing than telling and imply linkages among a wide range of topics which need not bepresented in temporal sequence. Topic associating narrative style is thought by some to beassociated with the field dependent cognitive style. While both narrative styles providemeaningful ways for children to talk about their experiences and realities, research shows thatstudents who tell topic associating stories tend to be called on less and interrupted more oftenthan do children who tell topic centered stories (Michaels, 1981). They also tend to be wrongly© 2008 Dr. Catherine CollierAll Rights Reserved61

considered disorganized or poor thinkers. Finally, they are more likely to be erroneously referredfor psychological assessments or placed in special education.Examples of Cultural and Communicative TendenciesIn this section, we will present examples of cultural and communicative tendencies observableamong African American and white students from several cultural groups in the United States.Tendencies are not universal. In order to avoid generalizations and stereotypes about culturalgroups, variations within cultures must always be considered. Highly educated people of a givencultural group are less likely to reveal indigenous language and communication patterns than lesseducated persons. Moreover, many people communicate in ways influenced by other cultures.Based on a review of literature and anecdotal reports, Taylor (1985) has listed verbal andnonverbal communication styles of working class African Americans as they contrast with thoseof Anglo Americans and middle class persons of other ethnic groups. Some of thesecharacteristics are presented in Table IV. Similar comparisons may be made between othercultural groups in the "typical" American classroom. Unfamiliarity with cultural communicationdifferences can lead to misinterpretation, misunderstanding and even unintentional insult. Forexample, the African American student who shows little reserve in stating his or her feelingsmay be misperceived as hostile, or perhaps as dangerous. The student, meanwhile, may seehimself or herself as an honest person willing to share feelings as a necessary first step inresolving problems. Similarly, the African American student who looks away from speakersduring conversation may be erroneously perceived as showing disrespect or not paying attention.The African American student who freely states his or her position to the teacher may beperceived as challenging the teacher's authority when the student may be demonstrating honestyand pride in the value of his or her opinion. Teachers, like all human beings, have their ownexpectations about communicative behavior. The teacher should 1) recognize any incongruenciesthat may exist between his or her expectations and those of the child; 2) make certain thatbehavioral norms in the classroom are sufficiently broad to embrace all cultural groups; and 3)teach the rules of standard American culture and the reasons for them. Have you witnessed anyof the behaviors identified in Table IV? How did you interpret what you saw? Which behaviorsdo you find represented in your classes? What characteristics could you add to the list from othercultural groups? What in group variations have you observed?Table IV: Examples of Verbal and Nonverbal Communication ContrastsAmong Some African Americans and Some Anglo AmericansSome African AmericansHats and sunglasses may be considered by men asadornments much like jewelry and may be wornindoors.Touching another's hair is generally consideredoffensive.© 2008 Dr. Catherine CollierAll Rights ReservedSome Anglo AmericansHats and sunglasses are consideredutilitarian by men and as outwear to beremoved indoorsTouching another's hair is a sign ofaffection.62

Asking personal questions of a person met for thefirst time may be seen as improper and intrusive.Use of direct questions is sometimes consideredharassment, e.g., asking when something will befinished is like rushing that person to finish."Breaking in" during conversation by participantsis usually tolerated. Competition for the floor isgranted to the person who is most assertive.Conversations are regarded as private between therecognized participants; "butting in" may be seenas eavesdropping and not tolerated.The term "you people" is typically seen aspejorative and racist.Listeners are expected to avert eyes to indicaterespect and attention.Speakers are expected to look at listeners directlyin the eye.Confederate flags and Black lawn ornaments areconsidered offensive and racist.Purposely including a minority person in groupactivities is seen as tokenism.Adoption of dance patterns or music of anothercultural group is suspect or considered offensive.Talking "Black" by outsiders withoutauthorization is an insult.Showing emotions during conflict is perceived ashonesty and as the first step toward the resolutionof a problem.Inquiring about jobs, family and so forth ofsomeone one has met for the first time isseen as friendly.Use of direct questions for personalinformation is permissible.Rules on taking turns in conversationdictate that one person has the floor at atime until all of his or her points are made.Adding points of information or insights toa conversation in which one is not engagedis sometimes seen as helpful.The term "you people" is tolerated.Listeners are expected to look at a speakerdirectly to indicate respect and attention.Speakers are expected to avert eyes,especially in informal speaking situations.Symbols of the Old South, such asconfederate flags and Black lawnornaments, are considered acceptable bymany.Including a minority person in groupactivities is seen as democratic.Adoption of dance patterns or music ofanother cultural group is seen as a free anddesirable exchange.Borrowing of language forms form anothergroup is permissible and encouraged.Showing emotions during conflict isperceived as the beginning of a "fight" andinterference to conflict resolution.© 2008 Dr. Catherine CollierAll Rights Reserved63

The Significance of Culture Based Communicative Behavior in SchoolA variety of cross cultural communication problems can arise in school, and it is important thatteachers not automatically blame the student or the student's family or culture. Problems oftenresult from misunderstandings or value conflicts between teachers and students who are obeyingdifferent culturally based communication rules. Some common problems linked to cultural andcommunicative diversity are presented in Table V.Table V: Problems That May Result From Culture and Communication Conflicts in theClassroomStudent CharacteristicsDiverse CulturalAssumptionsDiverse VerbalBehaviorDiverse NonverbalBehaviorDiverse Story-Tellingand ConversationalRulesSome examples ofInappropriate ResponsesNegative attitudes towardsdialect variationLower student expectationsExcessive interruptionsDiversity ignored in teachingprocessInsufficient attention to studentspeechFrequent misunderstandingsand misinterpretations ofstudentPerception of student asdisorganizedPerception of student as poorthinkerPerception of communicationdifferences as disciplineproblemPerception of communicationdifferences as social insults(unintentional)Some Examples of Possible Impacton StudentLowered self-imageLowered achievement and excessivespecial education placementsLow student participation;disproportionately low placement intalented and gifted programsLowered self-expectationsExcessive speech/language therapyplacementsLowered achievement in oralstandard EnglishLowered achievement and academicself-conceptPerception of frequent social insults(unintentional) from teachers andother studentsFrequent misunderstandings andmisinterpretations from personneland other studentsPerception of negative schoolclimateInterpersonal / intergroup conflictsPoor performance on tests andassessments© 2008 Dr. Catherine CollierAll Rights Reserved64

Using Cross Cultural Communication to Improve RelationshipsIn culturally diverse communities, differences may be expected to exist in the communicationstyles of students, teachers, parents, administrators and noninstructional staff. Perhaps the mostimportant reason for educators to understand cross cultural communication is to improve theirrelations with the diverse groups of students and parents they will encounter. If left ignored,communication differences will inevitably lead to various types of miscommunication whichmay lead, in turn, to conflicts which erode school climate and cause certain groups of studentsusually African American and other nonwhite students to feel unwelcome. The fact that thesecircumstances occur is a tragedy, of course. The greater tragedy, however, is that educators donot always know how to eliminate or minimize this type of discord.Tactics for Removing Cross Cultural Communication BarriersOnce we have established the prerequisites for understanding communication issues, we shouldthen seek to remove cross cultural communication barriers from the school environment. Themajor tactics that might be employed for this purpose fall under two categories:Removing language which appears to stereotype students; andReducing violations of cultural rules during discussions and conversations.With respect to removing stereotypical language, the following strategies might be useful:- Be aware of words, images and situations that suggest that all or most members of a racialgroup are the same.Example: "Why can't Joe ever be on time?" "He's African American, isn't he?"- Avoid using qualifiers that reinforce racial and ethnic stereotypes.Example: "The articulate African American student" implies that African American studentstypically have low verbal skills.- Avoid racial identification except when it is essential to communication.Example: "Judy, an outgoing student" is preferable to "Judy, an outgoing African Americanfemale student."- Be aware of possible negative implications of color symbolism and usage that could offendpeople or reinforce bias.Example: Terms such as "black magic" or "black market" can be offensive.- Avoid language that has questionable racial or ethnic connotations.Example: Phrases such as "culturally deprived," "culturally disadvantaged" and "you people"have racist overtones. With respect to changing communicative behaviors which violate thecultural rules of others, the following strategies may be useful:- Be aware of rules for attentiveness during conversation.Example: The constant maintenance of eye contact while listening during a conversation oftenviolates a conversational rule in working class African American and Hispanic cultures.- Be aware of rules regarding the distance between speakers during conversation.Example: In some cultures, speakers stand close enough to touch often. In other cultures,distance is maintained to denote respect.© 2008 Dr. Catherine CollierAll Rights Reserved65

- Be aware that objects, characters and symbols may reflect different beliefs or values fordifferent groups.Example: The confederate flag and Uncle Remus stories may offend African Americans becausethey reflect the culture of slavery and the Old South.- Be aware that cultures may vary in what they consider humorous or taboo.Example: Ethnic humor is often perceived by many groups as evidence of racial prejudice.Discussion of in group cultural rules and behaviors with outsiders is considered taboo withinmany cultures.- Be aware of different rules for taking turns during conversations.Example: African American children frequently perceive "breaking in" to reinforce or disagreewith another's point to be perfectly permissible, indeed desirable.- Cultures may use different standards for loudness, speed of delivery, silence, attentivenessand time to respond to another's point.Example: Many Native American societies place high value on contemplation and tend,therefore, to feel little responsibility to make immediate responses during conversation.- Be aware of different cultural rules for entering into conversations in progress.Example: African American students tend to consider conversations as private betweenrecognized participants. Therefore, anyone, including the teacher, who "butts in" is viewed as aneavesdropper and rebuked.One way to improve relationships across cultural lines, particularly in the upper grades, is todevelop a unit on "Communicating with One Another." The purpose of such a unit would be toteach students how to communicate more effectively across cultural lines and how to address andnegotiate differences. It is also useful for teachers to brainstorm with one another on how toremove communication barriers. In addition, a well designed staff development program can leadto better relations among staff and generate effective cross cultural communication activities forthe classroom. It can also be useful for teachers to ask parents to identify sources ofmiscommunication and socially offensive behavior or language. Parents may be asked to suggestways that school personnel can improve communication with students, adults and thecommunities. While schools have a responsibility to teach students the behavioral I codes of thesociety at large and to expect students to adhere to them, they have a similar responsibility toreduce culturally induced discipline problems and to avoid misinterpreting cultural differences asbehavioral problems. Patton's (1986) research on cross cultural communication behaviors andtheir relationship to discipline problems at the junior high school level provides excellent readingon this topic. Other readings are Gappa and Pearce's (1983) Removing Bias: Guidelines forStudent Faculty Communication; Kochman's (1981) Black and White Styles in Conflict; andPickens' (1982) Without Bias: A Guidebook for Nondiscriminatory Communication.Teaching Standard English To Speakers Of Nonstandard English DialectsThere is little disagreement that educational success and expanded social and career options arelinked to competence in standard English. An alarming percentage of students who speaknonstandard English are failing to acquire standard English, the language of education.Moreover, many students who do acquire standard English do so while being taught to reject thelanguage of their home, community and peers. In the process, they are denied an effectiveelement of social solidarity, which is an important element of cultural heritage.© 2008 Dr. Catherine CollierAll Rights Reserved66

Why Do Nonstandard English-Speaking Children Fail to Acquire Standard English?Toward More Effective Teaching of Standard EnglishTeaching Standard English from a Cultural PerspectiveWhat Is a Good Approach to Implementing Standard English as a Second Dialect (SESD)Programs?How Can Aides and Parents Help?Why Do Nonstandard English Speaking Children Fail to Acquire StandardEnglish?Many arguments have been advanced to explain the low achievement levels of nonstandardEnglish speaking children, particularly African American children, in acquiring oral competencein standard English. The most tenable argument suggests that the philosophy, assumptions andtraditional classroom methodologies employed in language arts education have failed becausethey have been prescriptive and corrective and have focused too much on language structurerather than on communicative competence (Taylor, 1985). Moreover, traditional teachingmethodologies have not typically been culturally sensitive, nor have they made use ofindigenous, nonstandard dialects. These significant deficiencies are probably due to naiveté, or tonegative attitudes toward language variations by language arts teachers.Toward More Effective Teaching of Standard EnglishIn recent years, a number of educators have begun to devise and implement instructionalstrategies which take into account the various language systems that students bring into theclassroom. In general, these strategies are based on modern sociolinguistic learning theory andon established principles of second language teaching. In 1981, California became the first stateto recognize the importance of indigenous dialects in teaching standard English. Focusingprimarily on the language of African American nonstandard English speaking children,California's State Board of Public Instruction stated in part: Many Black learners come to theschool setting speaking a language that is linguistically different from standard English. Thelanguage they speak is an integral part of the Afro-American culture . . . It is a unique languagewhich serves a uniquely rich culture. However, the school setting and that of the largerAmerican society, including the economic and commercial communities, represent anotherlinguistic sphere in which the student must learn to move and speak successfully. To the extentthat the young student fluently communicates in either language, he increases his opportunitiesin both realms ....Therefore, to provide proficiency in English to California students who arespeakers of Black Language and to provide equal educational opportunities for these students, itis recommended that the State Board of Education and the State Department of Educationhereby recognize: That structured oral language practice in standard English should beprovided on an ongoing basis. That special program strategies are required to address the needsof speakers of Black language. That parents and the general public should be informed ofimplications of educational strategies to address the linguistic needs of Black students.© 2008 Dr. Catherine CollierAll Rights Reserved67

Teaching Standard English from a Cultural PerspectiveTeaching the standard language from a cultural perspective differs from the traditional languageeducation approach in that it does not blame the victim. Standard English instruction from acultural perspective does not presuppose the devaluation or elimination of a learner's indigenouslanguage as a pre-requisite for learning. It recognizes that students need to retain their homedialect where its use is appropriate. Several major requirements for teaching standard Englishfrom a cultural perspective follow. Instruction should:Focus on both the structure of language, and on how to communicate;Maintain an oral basis;Concentrate on the structure of language, situational language requirements and language as avehicle for thinking;Be linked to clearly defined long term goals; andBe integrated across the curriculum.A successful culturally based standard English program recognizes that learning proceeds in anorderly way from the introduction of a particular aspect of language through its mastery. Themodel which has enjoyed the widest use and greatest success was designed in the late 1970s bythe San Diego Oral Language Program. It has been used with modifications in Standard Englishas a Second Dialect (SESD) programs in Dallas, Texas, and in Richmond and Oakland inCalifornia. The model lists several necessary steps for learning a new linguistic system whilepreserving the student's indigenous system. The model includes the following eight steps:1. Developing positive attitudes toward one's own language. The first and continuing job ofthe teacher is to counteract negative evaluations of the students' indigenous language.Lessons on the historical development of various dialects and on language diversity areuseful in accomplishing this goal.2. Developing awareness of language varieties. Students develop a sensitivity to the variousforms of a given language via stories in standard English, poems in different dialects, andrecords, tapes or video recordings of various speakers. At this stage, specific likenessesand differences are emphasized.3. Recognizing, labeling and contrasting dialects. Students learn to recognize differences invarious languages and dialects and to associate specific features with each linguisticsystem.4. Comprehending meanings. Students learn to recognize differences in meanings andintentions when an idea is translated from one language or dialect to another.5. Recognizing situational communication requirements. Students determine the types ofspeech appropriate to various situations.6. Producing in structured situations. Students practice producing successiveapproximations of standard English. Initially, students follow a model at this stage, e.g., ascript, choral reading or poem.7. Producing in controlled situations. Students receive instruction and practice in producingstandard English without a model, e.g., role playing or retelling a story.8. Matching the language to the situation. Students practice speaking appropriately in reallife, spontaneous situations leading to communicative competence.© 2008 Dr. Catherine CollierAll Rights Reserved68

Before beginning to teach standard English from a cultural perspective, the teacher and schoolneed a clear language arts philosophy which embraces modern principles of ethnology,sociolinguistics and second language instruction The philosophy and assumptions statementdeveloped by the Richmond Unified School District in California is a good example of whatschools and teachers may usefully adopt. Finally, before beginning program implementation, theteacher and school community must become thoroughly familiar with the following generalprinciples of second dialect instruction:• Instruction should be preceded by a non biased assessment of each learner's knowledge ofhis or her first dialect and of the second dialect.• Students must feel positive toward their own dialects.• Students must want to learn another dialect. If motivation is not present, the teacher musthelp students discover the advantages of acquiring the second dialect.• Instruction must consider the language goals of students, their families and theircommunities.• Instruction must take into account cultural values associated with learning and teaching.• Instruction must accommodate the preferred cognitive learning styles of the students.Some children prefer a field independent (object oriented) cognitive style. Others prefera field dependent (social oriented) cognitive style. Both are valid, however, schools tendto be more oriented toward the field independent style. See Appendix II for a summaryof these two preferred cognitive styles.• Both the teacher and students must be able to contrast the linguistic and communicativerules of the existing and targeted dialects.• Linguistic and communicative features of the existing dialect should be compared withthose of the targeted dialect.• Instruction should be integrated with students' experiences.• Both the teacher and students must believe that it is possible to acquire a second dialect.What Is a Good Approach to Implementing Standard English as a SecondDialect (SESD) Programs?Minimum standards must be established for evaluating the validity of culturally based SESDcurricula and teaching/learning strategies. An SESD program should:• Permit students to demonstrate their listening skills by summarizing, responding,paraphrasing or following directions.• Allow varied and frequent opportunities for students to communicate with each other.• Provide students opportunities to summarize, analyze or evaluate oral or writtencommunication completed by themselves, their peers or the teacher.• Provide students opportunities to listen and respond appropriately to the teacher, theirclassmates or audio visual materials.• Allow students to use speech for different purposes in a variety of situations, e.g.,persuading, informing, imagining, questioning or asserting.• Teach students how to evaluate the effectiveness of their own communication.• Stress that learning new speech patterns is linked to short term and long term goals.© 2008 Dr. Catherine CollierAll Rights Reserved69

• Underscore the importance of situation, audience or topic during communication.• Indicate that oral communication activities will be included throughout the total curriculum.• Have a clear language and communication focus.• The Richmond and Los Angeles school districts in California and Dade County PublicSchools in Miami, Florida, have developed lessons and support materials for implementingSESD programs.How Can Aides and Parents Help?The teaching of a second dialect cannot be solely the responsibility of the classroom teacher, nomore than it can be limited to the language arts classroom. Teacher aides and parents can supportthe instructional process and assist children's language development in general and SESDinstruction in particular by:• Encouraging children to speak in a variety of situations and before many audiences;• Establishing talking as a frequent, enjoyable and secure activity;• Modeling and expanding students' speech into language appropriate for the topic, situationand audience;• Pointing out what language and communicative behaviors are appropriate as situations occur;• Discouraging teasing about speech;• Not over-correcting students' speech;• Linking corrections of speech to the situation;• Providing the school and teachers with examples of speech used in the home and communityto incorporate in instruction, assessment and teacher training;• Reinforcing writing or reading activities with activities that include talking;• Providing an abundance of verbal stimuli for students irrespective of language orcommunication competencies;• Encouraging students to engage in conversations with a variety of people and on a variety ofsubjects; and• Encouraging students to recount their experiences in narrative form as often as possible andbefore a variety of audiences.Communication Differences, Test Performance and Educational PlacementVirtually all tests and assessment procedures used in schools and classrooms require students tomanipulate information using tools of communication. Often these tests and procedures presumecompetence in standard English and in the communication rules of the educated segment ofsociety. Much discussion and research in the last decade have focused on test bias and onculturally fair methods for assessing students" behavior and knowledge. Indeed, a number ofcourt cases and legislative initiatives have prohibited the use of culturally discriminatory testsand assessment procedures, e.g., Public Law 94 142: the Education for All HandicappedChildren Act. Culturally biased tests can adversely affect students from many cultural groups,particularly those who do not speak standard English. For example, these tests can contribute to:• Lowered expectations of student performance;© 2008 Dr. Catherine CollierAll Rights Reserved70

• Negative attitudes toward low performing students;• Lowered self esteem and expectations by students;• Lowered expectations of student performance by parents;• Inappropriate placement in special education and speech/language therapy programs; and• Lowered probability of placement in talented and gifted programs.While classroom teachers do not typically administer standardized tests, they do use assessmentprocedures in classrooms. Also, because students are placed in classes on the basis ofstandardized test results administered by others, teachers' expectations of students are ofteninfluenced by such test results. Furthermore, teachers typically take part in the referral andevaluation process to determine whether a student needs special education or related services.For these reasons, and others, teachers should understand test and assessment bias and shouldknow what to do about it.Sources of Bias in Tests and Assessment ProceduresWhat to Do About Assessment BiasSources of Bias in Tests and Assessment ProceduresIn general, teachers should be aware of seven major sources of test and assessment bias. They arepresented with examples in Table VI.Table VI: Sources of Communication Biases and Communication Related Biases in Tests andAssessment ProceduresMismatches between examiner and examinee regarding the societal rulesSituational Bias of language, e.g., sarcastic answers to obvious questions. (Examiner: Whattime does the clock say? Examinee: Everybody knows clocks don't talk.)Test directions involve linguistic complexities unfamiliar to the examinee,Directions Bias e.g., "None of the following is true except . . ." is incorrectly interpreted as"all of the following are true except . . ."Examinee is required to exhibit a particular moral or ethical preference,e.g., "One who is dishonest is a) an offender, b) a politician, c) anValue Biasambassador, d) an officer." One might presume each choice to bereasonable.Test presumes that examinee is competent in standard English, e.g.,"Which sentence is ungrammatical? a) They saw Rose. b) You done itLinguistic Biaswrong. c) My brother has never eaten. d) Don't use too much." Anonstandard English speaker might see each as grammatical.Test procedures or requirements are inconsistent with examinee's cognitiveand/or learning style, e.g., "Select the best answer to the following . . ."Format BiasOne might consider answers to be either right or wrong, thereby no suchthing as a best answer if all are correct.CulturalExaminer erroneously interprets cultural practices of examinee, e.g., aMisinterpretation child who exhibits silence as a natural reaction to an unfamiliar adult© 2008 Dr. Catherine CollierAll Rights Reserved71

Stimulus Biasexaminer is diagnosed as nonverbal, or a child who does not respondquickly to test items is labeled unknowledgeable.Examiner presents stimuli rich in objects and analytical materials to anexaminee that prefers a field dependent cognitive style, and is more likelyto be responsive to materials rich in social content.Teachers need to be aware of how communicative factors can influence test performance and ofthe impact their evaluations of students' performance can have. For example, if students fail tounderstand test directions, they may respond incorrectly to a whole series of questions. If theyare not competent in standard English, students may answer questions incorrectly because ofinaccurate analyses of their content. If their cognitive learning style is field dependent orsynthesizing rather than field independent or analytical, they may have difficulty with formatswhich seek best answers to questions when all answers contain an element of truth. Furthermore,students may hold different values than those assumed by the teacher or they may obey differentcommunicative rules for verbal behavior in the social situation which test questions address.Finally, the teacher may misinterpret a student's culturally based communicative style and drawerroneous conclusions about the student's knowledge.What to Do About Assessment BiasIt is not reasonable to expect that teachers will develop their own culturally fair tests andassessment procedures. However, teachers can contribute to a more positive assessmentenvironment for culturally diverse students. According to Taylor and Payne (1983), teachers canaddress the assessment bias issue in the following ways:• Ensure that all students understand the test directions.• Ensure that the test does not presume linguistic knowledge that students may not have unless,of course, the test is designed to assess knowledge of a specific language or dialect.• The test should not require students to articulate values which they do not share.• The test should be compatible with the preferred learning styles of the student.• Situational and interact ional constraints must be considered from a cultural perspective inevaluating oral communication behavior on tests.• In the scoring or interpretation of test performance, the student's response style must beconsidered.• When referring students for special testing or placement in special education, be certain thatresults of evaluations are not flawed by cultural differences in behavior and language.• Do not include culturally discriminatory items on tests.• Accept as correct those test answers which are considered valid within the student's culture,even if the particular answer is not listed in the manual as being correct.• Consider two scores for tests which have a cultural component, a culturally unadjusted scoreand a culturally adjusted score.• Support efforts to decrease reliance on standardized tests for assessments, using alternativeassessment procedures where possible and appropriate.• Support efforts to involve parents and the community in assessing test findings.© 2008 Dr. Catherine CollierAll Rights Reserved72

New standardized tests and assessment procedures are needed which are culturally valid. Untilthose tests are available, teachers should be skeptical of the results of most standardized testsadministered to culturally and linguistically diverse children and should exercise caution in usingthese results for making placement decisions.Communication Differences and Discipline ProblemsTable VII: Explanations of BlameSources of Cultural and Communicative IncongruitiesWhat Do We Do About Cultural and Communicative Sources of Student Misbehavior?Most educators would agree that schools have a responsibility to establish an educationalenvironment that encourages positive learning experiences for all students. Many would alsoclaim that schools should be expected to teach students the social behaviors which are consideredacceptable by the society at large, as well as the sanctions which are likely to be applied whenacceptable standards are violated. In light of these dual and sometimes conflicting imperatives,the disproportionate suspension and expulsion rates for minority students throughout the nationare cause for considerable concern. In 1979, for example, the National Advisory Committee onBlack Higher Education and Black Universities and Colleges reported that while students fromminority ethnic groups comprised approximately 25 percent of the nation's public schoolpopulation, they comprised 40 percent of those suspended. A 1982 Fact Sheet on InstitutionalRacism revealed that while 16 percent of the nation's students were African American, theycomprised 29 percent of all suspensions and 27 percent of all expulsions. In a 1984 report by theMinority Relations Monitoring Committee of Montgomery County, Maryland, it was stated that:. . .Nowhere is the school system more vulnerable to the charge of being "unfair" in dealing withits students than in the manner in which it manages student behavior . . . Clearly the difference insuspension rates of White and African American students in many of the senior high schools isso broad as to constitute a serious threat to the opportunity for an equal education.Many reasons are given to explain the high rate of discipline problems among minority students,particularly African Americans. Some blame the institution; others blame the student or victim.(see Table VII)Table VII Explanations of BlameInstitutional ExplanationsVictim-Focused ExplanationsNegative school climateLow student motivationRacial and ethnic biasCultural disrespect for rulesInadequate classroom management Unfamiliarity with rules of school and the mainstreamInappropriate educational placements societyInappropriate teacher expectations Lackadaisical discipline standards in student's homeDifferential applications of rules for and communitystudent conduct© 2008 Dr. Catherine CollierAll Rights Reserved73

Before drawing any conclusions relative to the cause of the high rate of discipline problems andsuspension and expulsion rates for non white students, it would be useful to explore the majortypes of student behaviors that teachers report as being unacceptable. Mitchell (1983) reports thatfive of the nine most frequently reported behaviors involve cultural and communicative issues,e.g., verbal threats, classroom disruptions and disrespect. You may recall that the claim wasmade that every human encounter positive or negative reflects cultural assumptions upon whichcommunication rules are based. This claim leads us to advance two important hypotheses aboutcross cultural communication and school discipline problems:Discord is inevitable among students when they have different perceptions of the rules forcommunication and what is appropriate and acceptable behavior.Teachers and other school personnel are more likely to perceive students' behaviors, includingtheir communicative behaviors, as being disruptive and discordant when those behaviors divergefrom the norms of the school.Schools reflect culture, mainly the culture of the dominant society. Patton (1986) claims thatschools are a micro culture with a distinct system of preferred and required communicativebehaviors. This culture, she asserts, is more congruent with some cultures than with others.Typically, it reflects the values and norms of the immediate community. Students with culturaland communicative norms which are incongruous with the schools" norms are more likely, sheargues, to engage in unacceptable behavior.Sources of Cultural and Communicative Incongruities Which Lead toDiscipline ProblemsResearch suggests that several categories of communicative behavior are consideredunacceptable in many, if not most, school environments. While some of these unacceptablebehaviors do not result in sanctions, many do. Examples of behaviors which almost always arepunished include:• Challenging the teacher's authority;• Using obscene language in class;• Using obscene language with other students;• Not listening quietly when the teacher is presenting a lesson;• Moving around the room when the teacher is presenting a lesson;• Interrupting another student;• Seeking assistance from another student on a test;• Not waiting until one person has finished speaking before taking a turn;• Ignoring the teacher's directions;• Responding in a loud voice;• Socializing in class;• Being late to class;• Not walking away during discord;• Using physical means to settle a conflict; and• Showing emotion during discord.© 2008 Dr. Catherine CollierAll Rights Reserved74

Each of the behaviors listed above may be in conflict with communication rules for a givencultural group. The teacher's task is to determine through reading, discussion and observationwhich perceived behaviors could have a different cultural significance for the groups representedin the classroom.What Do We Do About Cultural and Communicative Sources of StudentMisbehavior?While the majority of students, including those from minority groups, learn, accept and conformto the rules of schools, cultural differences may play an important role in apparent disciplineproblems in the school. These differences cannot account for all student misbehavior, yet manydiscipline problems may result from students' failure to know, accept or conform to school normswhich are either divergent or incongruous with their own cultural or communicative norms.School personnel should be open to examining the causes of perceived misbehavior in theclassroom. In those cases where school or classroom norms are intolerant of cultural differences,revisions of school norms would seem to be most appropriate. Where school norms are generallycongruent with norms across the cultural spectrum of the school community, the school has aresponsibility to teach those norms to the student. One cannot expect that all students come toschool with prior knowledge or acceptance of school norms. In any case, schools, like allinstitutions, must establish reasonable rules of conduct, but these rules should be sensitive to thecultural assumptions, values and communication expectations of the total school community.The following activities are useful in considering cross cultural communication issues foraddressing school discipline problems:1. Devise a plan for teaching all students the acceptable communicative behaviors of theschool and the justification for them.2. Devise a plan for teaching students the nature, origin and need for rules to govern schooland classroom communicative behavior, while simultaneously recognizing the validity ofother behaviors in different settings, such as students' homes or communities.3. Determine which school or classroom norms may be in conflict with cultural orcommunicative norms of specific cultural groups.4. Engage in discussions with colleagues on how school or classroom norms might berevised to make them more sensitive to the various cultural groups in the schoolcommunity.SummaryThis booklet has reviewed several basic concepts relating to culture, communication andlanguage and has shown how communication issues affect human behavior in general and schoollife in particular. Also, suggestions have been made relative to instruction, assessment,interpersonal relationships and discipline. You might now re-take the introductory test fromTable I. Any change reflected in your score will give you a measure of what you have learnedabout culture and communication. To put into use what you have learned, consider working withyour colleagues to:1. Integrate cross cultural communication topics and materials into the curriculum;© 2008 Dr. Catherine CollierAll Rights Reserved75

2. Educate those who administer standardized tests that different styles of communicationare used by various cultural groups;3. Review and revise classroom tests to eliminate cultural bias; and4. Address cross cultural communication issues in all phases of school life.Suggestions for classroom utilization have been made throughout the booklet. The rest is up toyou. Cross cultural communication is an ongoing process. Mistakes are inevitable, but sensitivityto cultural and communication issues can enhance the quality of education for all students.Some Attributes of Field Independent and Field Dependent Cognitive Styles (Adaptedfrom Ramirez and Castaneda, 1974).Student CharacteristicsOverall characteristicsRelationship to peers:Personal relationship toteacher:Instructional relationshipto teacher:Characteristics ofcurriculum that facilitatelearning:Field Independent CognitiveStyleFocuses on parts, rather thanon the wholeIs reality oriented to objectsand analyses of discreteelementsDemonstrates topic centerednarrative stylePrefers to work independentlyLikes to compete and gainindividual recognitionIs task oriented and inattentiveto social environment whenworkingRarely seeks physical contactwith teacherInteracts with teacher to tasksat handLikes to try new tasks withoutteacher's helpIs impatient to begin tasks;likes to finish firstSeeks nonsocial rewardsDetails of concepts areemphasized; parts havemeanings of their ownMathematics and scienceconcepts are emphasizedEmphasis placed on discoveryField Dependent Cognitive StyleFocuses on the whole, rather than thepartsIs reality oriented to relationships andsocial attributesDemonstrates topic associatingnarrative styleLikes to work with others to achievea common goalLikes to assist othersIs sensitive to feelings and opinionsof othersOpenly expresses positive feelingsfor teacherAsks questions about teacher's tastesand personal experiences; seeks tobecome like teacherSeeks guidance and demonstrationfrom teacherSeeks rewards which strengthenrelationship with teacherIs highly motivated when workingindividually with teacherPerformance objectives and globalaspects of curriculum are carefullyexplainedConcepts are presented in humanizedor story formatConcepts are related to personalinterests and experiences© 2008 Dr. Catherine CollierAll Rights Reserved76

Article II.3 Hard Work Hypothesis: Is Doing YourHomework Enough to Overcome the Effects ofPoverty?Stephen KrashenMulticultural Education 12 (4): 16-19, 2005It is well-established that the effects of poverty are devastating forchildren in school. More generally, scholars have documented thatlow "socio-economic status" (SES), whether measured by familyincome, parent education, or parent occupation, is usually the mostpowerful predictor of achievement and test score performance,sometimes swamping all other factors (White, 1982). This advantagetranslates to life success; children of the wealthy are far more likelyto become wealthy, become professionals, and attain positions ofpower than children of the poor. (Simonton, 1994). Simonton, in fact,concludes that "the log cabin myth is just that, pure myth" (p. 157).There have been some recent challenges to this generalization, however, claims that somechildren, especially Asian immigrant children and the children of Asian immigrants, do very welleven though they come from high-poverty backgrounds. Two such cases are the "Boat People" ofVietnam, who arrived in the US in 1978, and the Hmong.The Boat People and the Hard Work HypothesisThe Boat People, according to Caplan, Choy and Whitmore (1992) came only with "the clotheson their backs." Nevertheless, their children did well in American schools. Caplan et. al., (1989,1992) examined a subset of these children (n = 355) in grades K through 12 and reported thattheir overall grade point average was 3.02, nearly exactly a B, and they did especially well inmath: A group of high school students from their sample (n = 97) scored at the 72 nd percentile onthe math CAT.How did they do it? Hard work and family values, according to Caplan et. al., with lots ofhomework (a specific time set aside for homework every evening, with older children helpingthe younger ones. Caplan et. al. document this: Those in high school averaged three hours andten minutes homework per evening, those in junior high two and a half hours. The US averagefor junior high and high school is 1.5 hours.The Hmong and the Hard Work HypothesisThe Hmong, immigrants from Laos, also appear to support the Hard Work hypothesis, thehypothesis that hard work (homework) can overcome the disadvantages associated with poverty.As a group, the Hmong are among the poorest of immigrant groups and are among the leasteducated. For the 14,000 Hmong in California in 1990, median household income was $16,000per year, compared to the then national average of $36,000 (University of Wisconsin, 2000).Only 3% of the Hmong in California had graduated college (University of Wisconsin, 2000);© 2008 Dr. Catherine CollierAll Rights Reserved77

compare this to the finding that 42% of the Chinese/other Asian group studied by Portes andRumbaut (2001) had graduated college.Yet the Hmong do well in school, and it is undeniable that they are very hard workers. Table 1,from Portes and Rumbaut, reveals, in fact, that the Hmong are the champion homework studentsof their entire sample, with nearly half reporting two or more hours of homework per day.Table 1: Background and Homeworkschool engagement homeworkMexico 52% 14%Nicaragua 57% 21%Vietnam 59% 45%Hmong 62% 48%Chinese/other 61% 38%All data from Portes and Rumbaut, tables 4,1, 8.4School engagement: percent who feel grades are very importantHomework: percent who report doing two or more hours per day of homeworkTheir homework time appears to pay off in terms of better grades: Rumbaut (1997) reported thatfrom the sample of children of immigrants in US high schools studied by Portes and Rumbaut(2001), more time spent in homework resulted in better grades (table 2).Table 2: Homework and GPAHours GPA: 1992 GPA: 1996less than 1 2.3 2.31 hr to 2 hrs 2.6 2.62 to 3 hrs 2.9 2.84 or more 3 3.1From: Rumbaut (1997)More Evidence for the Hard Work Hypothesis: Background Counts WhenSES is ControlledTable 3 presents a multiple regression analysis from Portes and Rumbaut (2001), examiningpredictors of grades and standardized test results for children of immigrants in high school.Multiple regression is a very useful statistical tool that allows us to examine the impact ofdifferent predictors, holding the others constant. For example, table 3 tells us that high schoolstudents who have higher socio-economic status (SES) score higher on a test of reading. The"regression coefficient" for SES is 6.88. SES was measured on a five-point scale (-2 to +2); for© 2008 Dr. Catherine CollierAll Rights Reserved78

each point higher in SES, students scored 6.88 higher on the reading test. Thus, students from thehighest SES group scored about 34 points higher in reading than students from the lowest group..The regression coefficient represents the impact of SES without influence of other factors, thatis, when interpreting the impact of SES, we can pretend students were identical in all other ways.In table 3, SES is a strong predictor of all three measures, reading, math and grades. Note,however, even after controlling for SES, as well as for other predictors, such as the SES of one'sfriends, background is still a significant predictor. Being Mexican, for example, predicts lowertest scores (15 percentiles lower in reading) as well as a quarter of a grade lower grade pointaverage. Being Chinese/Korean predicts 3/4 of a grade higher GPA, 13 percentiles higher inreading, and a spectacular 23 percentiles higher in math.Table 3: Multiple Regression: from Portes andRumbaut (2001).Predictor READING MATH GRADESAge (range 12-18) -3.24*** -4.10*** -.10***Sex (1 = female) 2.63* 1.62* .33***Region (1 = CA, 0 = FL) 3.85* -8.46** 0.27Parental SES (range -2 to +2) 6.88** 5.8*** .18***Intact Family (1 = both parents) 0.9 3.45** .18***US Born 6.06*** 0.56 -.16***Long Term Resident* 6.23*** 0.02 -.12*Fluent Bilingual (1 = fluent bi.) 2.3* 1.93* .06*Limited Bilingual -12.14*** -7.14***Parent-Child conflict (1-4) -2.15** -3.12*** -.16***Second-generation friends (1-3) 1.88* 3.84* -.06**Inner City (1 = inner city, 2 = not inner city) -2.07* -6.21** 0Average Student SES (free/reduced lunch) .2*** .10** -0.06Chinese/Korean 12.97** 23.22*** .76***Columbian 0.07 -0.45 -0.05Cuban, private school 7.14* 5.6 .25**Cuban, public school -1.64 -3.48* -.14*Filipino -1.69 3.39 .17*Haitian -8.29** -7.18* 0Laotian/Cambodian -12.41** 2.1 .47***Mexican -14.7*** -11.52*** -.25**Nicaraguan -3.71 -2.62 -0.1Vietnamese -3.21 15.79*** .50***West Indian 1.07 -4.94 -0.01© 2008 Dr. Catherine CollierAll Rights Reserved79

2 0.29 0.25 0.24* = moderate effect, ** = strong effect, *** = very strong effectLength of residence in US: 1 = less than 5 years; 2 = 5-9, 3 = 10 or more, 4 = native born.Range of grades = 0 to 5, mean = 2.52A look back at table 2 gives a plausible explanation: Chinese/other Asian background studentsdo a lot more homework than Mexican background students, and are more engaged in school.This seems to say that homework counts, that social class is not everything, and suggests thathomework can at least make up for some of the disadvantages poor children have.The Counterevidence: Another look at the Boat People. How well did they really do?A closer look at research on the Boat People reveals that they did well, but were not spectacular.As noted above, their overall GPA was 3.02, but much of this was due to high performance inmath: Without math, their average drops to 2.64 (Caplan et. al., 1989, p. 67).For the 96 high school students studied, CAT math scores were very high, but their CAT scoresfor language and reading were a more ordinary 46th percentile. This is very good for studentswho have only been in the US f and a half years, but not spectacular. (It should be noted thatCaplan et. al. reported the results for CAT reading and language tests combined. They areseparate tests, and the reading test is considered more demanding and a better test of academiclanguage.)No comparison group was included; there was no comparison with students of similar SES, withsimilar length of residence, from the same or other countries. In addition, we have no idea howmuch education the children had before arriving in the US, and what the quality of theireducation was. Quality of education in the first language is a strong predictor of success(Krashen, 1996).The Boat People: Were they really low SES?Caplan and colleagues provide somewhat conflicting information. It is very clear that the BoatPeople suffered from considerable poverty when they arrived in the US, but it is not clear whatSES they had in Vietnam. In one place, we are told that "the vast majority of the parents... camefrom low SES backgrounds, the poorer and less educated segments of their society (Caplan,Whitmore and Choy, 1989, p. 212), but elsewhere we are told that they "... are much more urban,more highly educated, and skilled in jobs more related to urban environments that the majority inthe society they left behind" (p. 27). Walker-Moffet (1995) in fact, concludes that the BoatPeople study "really tells us is that students from well-educated backgrounds succeedacademically regardless of their status as refugees" (p. 12).There is some concrete data: Whitmore, Trautman and Caplan (1989) inform us that 26% of theparents in their sample had completed high school. This is considerably higher that the 9%reported for the Hmong living in California in 1990 (University of Wisconsin, 2000) butconsiderably less than the US average of 78% for 1990 (Digest of Educational Statistics, 1966,table 8).Caplan et al, 1992, however, tell us that about half of the parents read to their children, either inVietnamese or English, suggestive of at least moderate levels of literacy. Caplan et. al. (1989)mention that a multiple regression analysis was done, and it showed no impact of previousparental SES on school performance of the children, but no details are provided.A closer look at the Boat People, thus, shows that their school performance, while respectable,was not miraculous. A closer look at their backgrounds reveals the possibility that they did notbelong to the lowest SES group, and that at least some of the parents were literate and educated.© 2008 Dr. Catherine CollierAll Rights Reserved80

This case does not provide strong support for the Hard Work hypothesis. It is, at best, asuggestive case that lacks a great deal of crucial data.The Hmong: Grades yes, tests no.In contrast to the Boat People, the Hmong clearly are low SES with respect to both income andparental education. Recall that the hard-working Hmong children, according to table 2, exceedall groups in the amount of homework done and are among the high-poverty groups. Does all thehard work pay off?Another examination of table 3 reveals that being a member of the group Laotian/Cambodiandoes indeed predict a higher GPA and the effect is substantial: Laotian/Cambodian-originstudents, after controlling for SES and other factors, earn a half-grade higher GPA. BeingLaotian/Cambodian, however, does not increase scores on the standardized test of math, andresults in 12 percentiles less on reading. Their success, in other words, is limited to good grades.Walker-Moffett (1995) interprets this result as showing that the Hmong try very hard, and pleasethe teacher, but do not accomplish as much as their high grades indicate.Additional evidence for this interpretation comes from Lee (1995), in her ethnographic study ofAsian-American high school students:"At Academic High I found that Asian American students were rewarded for being quiet, polite,and respectful. On several occasions I learned from teachers that they had given passing gradesto Asian American students who had not earned them. In each case, the student who was passedwas described as a quiet and polite student ... similarly, in her study on Hmong students,Goldstein (1985) [unpublished dissertation] found that Hmong students were promoted to thenext grade based on their behavior rather than on their academic performance" (pp. 62-63).I interpret this result as showing that for less privileged children, hard work will take you only sofar: It may result in higher grades, but the effect may be limited to higher grades. The effects ofpoverty and having less-educated parents is very strong. As discussed below there is a lot thatcan be done for these children, but simply telling them to try hard in school and get good gradesis not the complete solution.DiscussionMy conclusions are consistent with other results showing the powerful effect of low SES.Krashen (2002) concluded that despite claims made by Ed Trust, a re-analysis of their data forCalifornia showed that high poverty schools that were also high performing, as defined byperformance on standardized tests, were nearly non-existent.My conclusions are also consistent with research showing that high performance on standardizedtests of reading is consistently associated with access to print (Krashen, 1993, McQuillan, 1998),and that children of poverty have far less access to print (Neuman and Celano, 2000). Thisrelationship holds because access to print results in more recreational reading, and morerecreational reading results in more literacy development (Krashen, 1993). Massive amounts oftraditional homework cannot build literacy nearly as effectively as hours of pleasure reading. Theobvious part of the cure for children of poverty is more access to print, via better school andpublic libraries.Higher SES is also associated with better education in the country of origin. For students whoarrived in the US during school age, children of poverty rarely have had quality education in theprimary language, a factor that has a strong effect on performance on school in the US. In fact,some middle class immigrant children not only have a strong background when they arrive, but© 2008 Dr. Catherine CollierAll Rights Reserved81

are also provided with tutoring (Krashen, 1996). Bilingual programs are designed to provide thisbackground rapidly for those with limited English proficiency.This analysis also forces us to reconsider the effects of homework. More time devoted tohomework appears to be related to better grades but not necessarily to higher performance onstandardized tests. This should not be interpreted as saying that homework is useless; obviously,the value of homework depends on the assignment and degree of engagement. In the case ofreading, for example, the research cited above on the impact of pleasure reading suggests thatoutside of school activities that engage the student in interesting texts and that encourageadditional free voluntary reading will be highly effective. This is only possible, however, ifreading material is easily available. But homework that focuses on skill-building will havelimited value.It would also be of interest to determine which aspects of SES are crucial: The case of the BoatPeople suggests that parental education is more relevant than current income. Better educatedparents, it is predicted, will be more likely to read to their children, provide access to readingmaterials, and be better prepared to help with homework. If this help is done in the firstlanguage, it is a form of de facto bilingual education. And of course we should be looking atmore than standardized test results to measure true academic success.© 2008 Dr. Catherine CollierAll Rights Reserved82

Article II.4 Language Acquisition and theBilingual Exceptional ChildBy Julia Scherba de ValenzuelaAn excerpt reprinted by permission from Educating Culturally AndLinguistically Diverse Students: A Professional DevelopmentResource Series, BUENO Center for Multicultural Education,University of Colorado, 2002.OBJECTIVES• To emphasize the importance of understanding languageacquisition in bilingual children• To discuss popular fallacies about bilingual children withexceptionalities• To define basic concepts of language• To overview language acquisition as a general process• To define second language acquisition• To compare and contrast first and second language acquisition• To examine some models of second language acquisition• To discuss current controversies in second language acquisitionIntroductionUnderstanding the process of language acquisition is crucial to the appropriate education andassessment of culturally and linguistically diverse (CLD) students. The purpose of this chapter isto provide a framework for considering how language development is different for bilingualchildren. This will hopefully lead educators to consider how bilingualism can effect the academicperformance and assessment of bilingual students and how it plays a part in the education ofCLD students with exceptionalities.In this chapter, “language acquisition” is used as a central concept, rather than “second languageacquisition.” This is done so with the intent of shifting the focus from examining how English isacquired to a more general examination of how bilingual students acquire language. There areseveral reasons for this shift in focus: 1) many bilingual students in this country begin theprocess of developing English before their native language is fully established and their languagedevelopment can most accurately be called “bilingual”; 2) understanding the effects of secondlanguage development on first language development will help us understand where students are,in terms of their overall language competence; and 3) incorrect assumptions about students’native language competence and development can lead to inappropriate referrals and erroneousassessment procedures. We cannot simply talk about acquisition of English as a second languagewithout considering the general process of language acquisition for bilingual students.There are many popular misconceptions about bilingualism, language acquisition, and bilingualstudents. Some people erroneously believe that students with exceptionalities cannot learn two(or more) languages. Others believe that encouraging the parents of CLD students, with andwithout exceptionalities, to speak with their children at home in English, is in the best interests of© 2008 Dr. Catherine CollierAll Rights Reserved83

the students. Another common fallacy is that acquiring more than one language is “difficult” andcan lead to academic problems. Some teachers have been heard to suggest that their bilingualstudents don’t speak any language to a real extent and are “semilingual.” We will return to thesemyths at the end of this chapter. By then, the intervening information will have provided youwith enough understanding about the normal process of language acquisition that the faulty logicof these beliefs should be obvious.Regardless of whether the above beliefs lack grounding, they are very common and caninfluence how educators assess and educate CLD students. Without understanding how languageacquisition works, the continuation of these fallacies is understandable. Therefore, educatorsworking with bilingual students must understand the language acquisition process for effectiveimplementation of assessment and selection of appropriate intervention techniques.Language AcquisitionTheories of Language AcquisitionWhy do we study language acquisition? Depending on the orientation of the researcher, the studyof language acquisition can be either important in and of itself or as a mechanism forunderstanding other areas of research such as theoretical linguistics, cognitive psychology,neuroscience, or child development. Some want to know what the process of first and secondlanguage acquisition tell us about human development, learning, and organization of the brain.Others want to know about language acquisition so they can make recommendations abouteducational programs. Regardless of their research focus, all of these researchers must take aposition on some fundamental arguments about the development of language and its relationshipto other mental functions.Linguistic theory can be divided into several main camps depending on 1) whether language isviewed as either separate from or as an outgrowth of cognition and 2) whether languageacquisition is believed to be directed by a preprogrammed, language-specific area of the brain orwhether it is guided primarily by experiential factors. Noam Chomsky addressed both of theseissues when he proposed that human brains are “hardwired” for language acquisition via whathas come to be called the “Language Acquisition Device” or LAD (Chomsky, 1965). This isconsidered a “nativist” theoretical position. The development of this theory was in directopposition to B. F. Skinner’s ideas about the development of language as shaped by behavioralreinforcement received from a child’s environment (Skinner, 1957). Chomsky demonstrated thatthe language available in a child’s environment is far too complex and the reinforcement (i.e.praise) far too inconsistent to account for a behavioral model of language development. Althoughthe extent to which grammatical structures are specifically preprogrammed is still hotly debated(Bowerman, 1994), most child language specialists today agree that, at least to some extent, thehuman brain is predisposed to 1) attend differentially to language input, 2) process that inputaccording to some preset principles, and 3) formulate unconscious rules for languagecomprehension and production.This perspective of language as biologically driven has been traditionally considereddiametrically opposed to sociocultural perspectives that recognize the role of the environment inshaping language development. However, many child language researchers are coming to theconclusion that this does not necessarily need to be the case. Some have come to take theposition that “it is perfectly consistent to believe that, while much of language development isgoverned by the operation of powerful innate principles, some important aspects of earlylanguage development are significantly influenced by the child’s language experience” (Harris,© 2008 Dr. Catherine CollierAll Rights Reserved84

1992, p. XI). These researchers, while acknowledging the existence of the biological basis oflanguage acquisition, also acknowledge that “interpretation and meaning are necessarilyembedded in cultural systems of understanding. If language is a meaning-making system andspeaking and listening are meaning-making activities, then accounts of these phenomena must atsome point draw on accounts of society and culture” (Ochs, 1988, p. 4). In recognition of theinfluence of culture, Schieffelin and Ochs (1986) refer to the process of language acquisition inthe sociocultural context as “language socialization.” Ochs defines socialization as “the processby which one becomes a competent member or society” (p. 5) and language socialization as“socialization through language and socialization to use language” (p.14). This perspectiverecognizes that linguistic competence involves much more than the ability to comprehend andproduce grammatically correct utterances. Language competence involves the ability to selectbetween a variety of possible options of linguistic form and content according the social contextand cultural norms, as well as interpreting subtle meanings that require extensive social andcultural knowledge. An example is knowing when “Would you like to sit down?” becomes anindirect command, a real question, or a comment on inappropriate behavior.The study of second language acquisition clearly fits within the above theoretical debate.Understanding how second language learners best acquire a new language may shed moreinformation about the relative influence of cognition, language-specific brain functions, andenvironmental influences. Unlike the process of first language acquisition in young children, theprocess of second language acquisition can be observed under a variety of circumstances. Therelative influences of age, personality, social context, type of language input and other factorscan be better isolated and explored. This information can, in turn, inform our understanding ofhow language acquisition as a generic process unfolds.The Process of First Language AcquisitionIt is a remarkable fact that, without special training or carefully sequenced linguistic input,every normal child acquires a natural language. The universality of language in our speciesstands in glaring contrast to the much more selective attainment of comparable cognitive skills,such as the ability to perform arithmetic calculations. A related fact is that every child in alinguistic community succeeds in converging on a grammatical system that is equivalent toeveryone else’s, despite considerable variability in linguistic experience. Moreover, childrenacquire language quite rapidly and with few wrong turns, considering the number of potentialpitfalls that exist (Crain, 1994, p. 364).Precursors to language production.Language acquisition is indeed a remarkable process. Children acquire the language(s) heardaround them and seem to do so in strikingly similar ways the world over. This process is evenmore remarkable when we consider what a child must know and be able to do to produce herfirst word, for example, “mama.” To answer this question, we need to distinguish between thereal use of a word for communicative purposes and the imitation of a string of sounds. It is easyfor first time parents, eagerly anticipating Joey’s first word, to interpret /mama/ as mother and/dada/ for daddy when the child is really just engaging in babbling or sound play. In order to call/mama/ or /dada/ a word, it should occur in consistent contexts, spontaneously (withoutimitation), and appear to indicate a communicative intent. Clearly, this can be difficult todetermine. However, if Joey says /dada/ when repeating sounds made during play with hismother and produces this consonant-vowel combination along with others, such as /baba/, /gaga/,and /dadada/, then we can probably suspect that it is not a real word. On the other hand, if Joey© 2008 Dr. Catherine CollierAll Rights Reserved85

sees his father coming into the room and spontaneously looks at him and says /dada/, and inaddition, has done this before in a similar context, then there is a good likelihood that for him,/dada/ is now a word that means “Daddy.”For Joey to do this, what must he know and be able to do? He must have a variety of linguistic,cognitive, social, and perceptual competencies. It is amazing to consider all that a child must becapable of to produce even one real word. This is as true for the child acquiring a secondlanguage as it is for her first.Linguistically, Joey has already developed quite a bit of knowledge about the phonology of hisfirst language. Studies have indicated that within the first year of life, children have alreadylearned what sounds and sound combinations are possible in their native language. Using theexample presented in the beginning of this chapter, native-English speaking infants learn that aword cannot start with “mb.” Joey has also learned how to produce a variety of sounds correctlyin English. Although his pronunciation is not adult-like, it is well-enough developed that manyearly words are grossly intelligible.Joey has also started understanding how language functions. Although first words can assume avariety of grammatical functions, young children have at least a basic understanding of what aword is and what different functions they can perform. Joey also understands the meaning ofvarious words, even though his understanding may not be exactly like that of an adult. To at leastsome extent then, Joey understands the words “mama” and “daddy.”In addition to understanding the early words that he produces, Joey understands their underlyingconcepts. To identify a mother, Joey has to understand what a mother is. His understanding ofthis concept, again, may not be exactly like that of an adult (he may conceive of a mother as anyfemale adult caregiver rather than one unique individual) but he clearly does have some idea.This means that Joey can also differentiate between individuals and can typically differentiatebetween genders and age groups. Joey must be able to understand a lot of what goes on aroundhim and differentiate and interpret relationships between objects, actions, sounds and people inhis environment.To produce one word, a child must also have acquire quite a bit of social and cultural knowledge.Words have social functions - they can be used to label, request, greet, query, and deny, to namejust a few. Therefore, by saying “mama”, Joey demonstrates his knowledge about the socialfunctions of language and his understanding of social relationships. He knows that words are tobe used interactively with others and he knows when and how to initiate and continuecommunicative interactions. He knows how to take conversational turns and he expects thatothers will respond to him in specific ways.Joey must have some useful auditory capabilities if he is producing his first words orally.Although deaf and severely hearing impaired children do begin to babble, they stop doing soaround the age of six months. Therefore, production of oral speech is dependent of the ability tohear. Joey is also able to process auditory input. He can distinguish between human speech andenvironmental sounds. Visual acuity is not necessary to the development of oral language,however, the language development of blind children has been studied and found to be affectedby their blindness. Clearly, children can do and know a lot, even before they ever produce theirfirst words.Language comprehension.What does it mean to understand a word? Language comprehension is not an all or nonephenomena. This idea is important for educators to understand. Children may be able to“comprehend” a word or phrase in some contexts and not in others. Children may first© 2008 Dr. Catherine CollierAll Rights Reserved86

comprehend a word within a specific routine or context (such as “jacket” during a gettingdressed for going outside routine) and with an accompanying gesture or linguistic cue (such aspointing to or naming an item using known words, such as “doggie book”). We can say a childreally understands a word when she is able to follow a direction or identify an item, when theitem(s) involved are not present and the context is not routine. For example, a child demonstratessome amount of comprehension when she puts her shoes on after being asked to during adressing routine. She demonstrates greater comprehension when she is asked to “show me howyou put your shoes on” during a non-dressing situation, such as play testing. With secondlanguage learners, we must be careful that we do not either over- or under-estimate theirreceptive language abilities by failing to analyze the linguistic, social, situational, and/or gesturalcontext of comprehension.The process of language acquisition.Language acquisition does not progress randomly. Although there is significant individualvariation, children appear to acquire language in a fairly systematic fashion. The sequence ofacquisition of syntax and morphology, the manner in which these develop, and the developmentof phonology all appear to proceed in a somewhat consistent and systematic fashion.Evidence for the biological foundation of grammatical acquisition came with the discovery thatgrammatical morpheme acquisition appears to progress in an orderly fashion (Brown, 1973).Brown observed the grammatical development of three children, Adam, Eve and Sarah, over aperiod of several years and discovered that the order of appearance of the first 14 grammaticalmorphemes was the same for all three children. Table 6-1 lists these morphemes in their order ofappearance. Another classic study also found striking similarities in the sequence ofmorphological acquisition among native-English speaking children (deVilliers & deVilliers,1973).Figure 6-1Brown’s first 14 Morphemespresent progressive -ing (without auxiliary)“in”“on”regular plural –sirregular pastpossessive – suncontractible copula (to be as main verb)articlesregular past –edregular 3 rd personirregular 3 rd personuncontractible auxiliarycontractible copulacontractible auxiliaryIn addition to determining the sequence of acquisition of grammatical morphemes in English, theresearch by Brown (1973) and deVilliers and deVilliers (1973) was also important inestablishing parameters for when these morphemes should appear, in terms of stage of language© 2008 Dr. Catherine CollierAll Rights Reserved87

development. This research described children’s level of language development according to ameasure of utterance length - Mean Length of Utterance (MLU). A determination of MLU ismade by analyzing a 100 utterance language sample for the average length of child-producedutterances, in terms of morphemes. For example, the utterance 'Bobby hitting' has a morphemecount of 3, as 'hitting' includes both the verb 'to hit' and the present progressive morpheme -ing.This type of measure of language development, rather than age, is considered a better predictorof grammatical development, as children of similar chronological ages can vary greatly in theirlanguage acquisition. Therefore, researchers have attempted to determine in what order and atwhat stage of linguistic development different language forms should appear.Until recently, most research has focused on similarities in grammatical development, with theaccompanying perception that all normal children follow a relatively fixed and stable pattern ofgrammatical development. However, more recently, investigators have begun to question thisassumption, with research that indicates a greater variability than previously assumed (Lahey,Liebergott, Chesnick, Menyuk & Adams, 1982). Given the original research indicating veryconsistent patterns of grammatical acquisition among middle-class native-English speakers, thistrend was assumed to hold for non-dominant culture English speakers, as well as childrenacquiring languages other than English.A considerable amount of research on the grammatical development of native-Spanish speakershas been performed, with the goal of finding a consistent order of acquisition of morphemes suchas been found among middle class, native English speakers (e.g. González, 1978; Kvaal,Shipstead-Cox, Nevitt, Hodson & Launer, 1988; Olarte, 1985; Vivas, 1979). However, in a metaanalysisof these studies, Merino (1992) found that no more than very gross generalities could befound, such as the acquisition of the present tense before the subjective. These findings haveimportant implications for the language assessment of bilingual and non-English speakingchildren.These studies that indicate that the order and developmental age of appearance of morphemesmay not be as consistent as previously thought should warn educators and assessment personnelagainst making judgments about grammatical development for the purposes of determiningspecial education placement. We need to be very cautious when making comparisons ofindividual children against assumed benchmarks of “normal” development. These guidelines ofmorphological development should be seen as generalities and deviance from this observedprogression should not be taken as indications of a language problem.In addition to research focused on the stage and sequence of grammatical development,researchers have also looked at the process by which children acquire grammatical forms. Thisprocess was assumed to be a matter of modeling, imitation and reinforcement when Skinner’sbehavioral model of language development was in vogue. However, researchers have sinceobserved that children seem to go through a process of hypothesis testing. Children may acquirea form of a word, such as “went” without acquiring the rule for irregular past tenses. Later, whenthey acquire the rule for formation of regular past tenses, they may over generalize the rule,forming productions such as “goed”. As they fine tune their system of grammatical rules andacquire the irregular past tense, they will once again correctly produce “went”. During thisperiod of over-extensions, children may be very resistant to corrections of forms that they havenot yet acquired. Table 6-2 provides an example of a hypothetical, but typical mother-childinteraction that demonstrates children’s attention to the truthfulness, rather than the grammaticalcorrectness of the input. In fact, observations of mother-child dyads indicate that parents respond© 2008 Dr. Catherine CollierAll Rights Reserved88

more to the truthfulness of the child’s productions, rather than to the grammatical form. This ismore evidence of why the behavioral model of language learning has been rejected as untenable.Figure 6-2Example of Grammatical Over-extensionChild:Mom:Child:he falled downno Timmy, he fell downyeah, he falled downSimilar to grammatical acquisition, phonological acquisition appears to follow a general patternof development, with some sounds typically appearing earlier than other. In general, sounds suchas ‘p’, ‘b’, and ‘m’ appear very early, even across different cultures. Other sounds, such as ‘s’,‘th’, ‘r’, and ‘l’, typically appear much later. However, as with all generalities, these patternsmay not hold for individual children. Although a normally developing English-speaking childmay not correctly produce ‘r’ in all phonological contexts until age 5 or 6, some 3 year olds canproduce this sound without difficulty. This is one more reason that we must take a child’s totallanguage abilities into consideration when evaluating their language competence: language form,content, and use all must be assessed.Input influences.Even if we accept a strong nativist position, the influence of external input on the process oflanguage acquisition is undeniable. A child growing up in China learns to speak Chinese, notbecause of a physical predisposition, but because that is the language of his environment. If thatsame child had been born in the United States, of monolingual English speaking, Chinese-American parents, his native language would be English. Obviously the language of input is thelanguage that will be acquired.A child raised in a home where American Sign Language (ASL) is the dominant language, willacquire that code. In addition to the language of input, the mode of input, signed or spoken, willdetermine the language acquired. Sign language, just like spoken language, is a complex,systematic, rule-governed code. The linguistic universals discussed in the beginning of thischapter apply to sign languages just as to oral languages.The language and mode of input are unarguable examples of the influence of external factors.The influence of other factors though, are more questionable. Some researchers have found thatthe way that parents speak to their children may affect early vocabulary development (Beals &Tabors, 1995; Harris, 1992). However, what long-term affect different maternal styles will haveis still unknown. At one time, the typical way that middle-class English-speaking motherscommunicate with their children was thought to a universal pattern. The elevated pitch,expressive intonation and the tendency to treat even very young infants as capablecommunicative partners has been termed “motherese.” However, by the 1980s evidence fromcross-cultural studies has begun to reveal that what was once considered a universalcommunication style with young children, and necessary for facilitation of language acquisitionis not found in all cultures (Ochs, 1982, 1988; Schieffelin & Ochs, 1986). Although there areconsiderable differences in the ways that adults communicate with young children and theexpectations they have for children’s participation in conversation, children the world overacquire language fluently and competently.© 2008 Dr. Catherine CollierAll Rights Reserved89

What Does This Mean for Students with Exceptionalities?To understand how all of this information applies to students with or suspected as havingdisabilities, we now return to the fallacies introduced at the beginning of this chapter.1) Fallacy: Students with exceptionalities cannot learn two (or more) languages.The reality is that many CLD with disabilities must learn a second language. If a child withdisabilities speaks a home language other than English, she must acquire a second language toparticipate in the school environment. Although research does suggest that some children mayacquire a second language more slowly, especially if they exhibited language difficulties in theirnative language (Kessler, 1984), this should not dissuade educators from assisting their students’second language acquisition as much as possible. Therefore, the real question becomes, shouldthe language of instruction for CLD students with disabilities be the student’s first or secondlanguage. Studies suggest that, just as for students without disabilities, a second language is bestacquired from a firm and well-developed first language foundation (Perozzi, 1985; Perozzi &Sanchez, 1992). This research suggests that grammatical forms are most quickly and accuratelyacquired in English when they have first been taught in the student’s native language. Thissupports a bilingual approach to special education with CLD students.2) Fallacy: Parents of CLD students, with and without exceptionalities, should speak with theirchildren at home in English.This advice, although popular, is incorrect for several reasons. As discussed above, students willbest acquire a second language if their first language is well-established. Second, asking parentswho may not be able to provide an adequate language model in English to restrict the use of theirmore proficient language is absurd. Parents will neither be able to stimulate their child’slanguage development nor will they be able to communicate easily for social purposes with theirchild. Wong Fillmore (1991b, p. 343) makes the following poignant observation:When parents are unable to talk to their children, they cannot easily convey to them their values,beliefs, understandings, or wisdom about how to cope with their experiences. They cannot teachthem about the meaning of work, or about personal responsibility, or what it means to be a moralor ethical person in a world with too many choices and too few guideposts to follow. What is lostare the bits of advice, the consejos parents should be able to offer children in their everydayinteractions with them. Talk is a crucial link between parents and children: It is how parentsimpart their cultures to their children and enable them to become the kind of men and womenthey want them to be. When parents lose the means for socializing and influencing their children,rifts develop and families lose the intimacy that comes from shared beliefs and understandings.3) Fallacy: Acquiring more than one language is “difficult” and can lead to academic problems.Cummins’ additive bilingualism enrichment principle and the research on the cognitive benefitsof bilingualism clearly suggest that bilingualism is not a burden for students. In fact, in manyparts of the world, it is a common part of daily life. When fluently bilingual parents areencouraged to raise their children monolingually, as in the case of a 1995 child custody case inwhich Texas State District Judge, Samuel C. Kaiser equated raising the child of a bilingualmother in a Spanish-speaking home as tantamount to child abuse, beliefs about bilingualism as acognitive deficit are reinforced. Regardless of the cognitive benefits, bilingualism is of socialbenefit in our global village and can only have positive outcomes when students leave school andseek employment.4) Fallacy: Some bilingual students don’t speak any language to a real extent and are“semilingual.”© 2008 Dr. Catherine CollierAll Rights Reserved90

This idea of “semilingualism” can be compelling when we do not understand languageacquisition. Educators may confuse students undergoing a “silent period” as demonstrating alack of ability to communicate. Remember all that a child has to know to be able to say even oneword in his first language. Even those children who demonstrate little expressive language in theschool environment (in their first or second language) bring with them a wealth of informationabout the phonology, morphology, syntax, lexicon, and language use patterns of their nativelanguage. Additionally, standardized tests developed based on standard dialects in English andthe student’s native language may fail to identify their communicative competence.There are three important concepts that relate to this issue of language competence in bilingualchildren: language attrition, semilingualism, and code switching. Language attrition is arecognized phenomena in which an individual loses all or part of her native languagecompetence. It can happen naturally as a result of immigration and the lack of opportunities tocommunicate in a particular language. For adults who have immigrated without their families toa new country and rarely return for visits, this may occur over time. It can also happen to youngchildren who are exposed to a new language at school before their first language has been wellestablished or where there is a significant discrepancy between the social prestige of the twolanguages. Language attrition is a common phenomena in the United States among children fromdiscriminated and dominated ethnic groups, such as Hispanics. The preliminary results of theNational Association for Bilingual Education (NABE) No-Cost Study on Families indicate thatthe early exposure to English, by enrollment of non-English background children in preschoolprograms that are not conducted exclusively in the children’s home language, results in a shift inuse of the native language at home and leads to language attrition (Wong Fillmore, 1991b).Merino (1982) found that among the native-Spanish speaking children in her research project,language attrition occurred by the fourth grade, and sometimes even earlier. She also found thatlanguage attrition initially affects later developing skills and abilities. The dynamic relationshipbetween increasing second language proficiency and first language loss can result in a temporarystage in which the child appears limited in both languages.The problem with applying the label of “semilingualism” to these children is twofold: 1) thisterm suggests a difficulty in acquiring language and does not recognize that children may havelost language skills they once possessed, and 2) this term implies a resultant cognitive deficit.Although bilingual children from non-dominant culture backgrounds do have a higherpercentage of below average academic performance, there is no evidence that this stems from acognitive problem brought about by their bilingualism. Inappropriate academic programs andhome-school incongruities have been suggested as reasons for these academic problems. In fact,one of the major problems facing our school system today is the lowered academic achievementof all culturally different youth, regardless of whether they speak a home language other thanEnglish or not.Sometimes the term semilingual is applied because educators observe students speaking whatthey consider to be a mixture of two languages. However, research has indicated that what mayappear to a monolingual speaker to be a random hodgepodge of two different languages is in facta systematic and socially governed interplay between two separate and well-developed linguisticsystems (Genishi, 1981). In fact, that students are able to switch back and forth between twodistinct codes is a sign of linguistic maturity. Code-switching refers to “the use of two or morelinguistic varieties in the same conversation or interaction” (Scotton & Ury, 1977, p. 5). This caninvolve switching between social styles or registers or between different languages. Most peoplecode switch as a regular part of social interactions, but we don’t even recognize what we are© 2008 Dr. Catherine CollierAll Rights Reserved91

doing because it is such a normal part of communication. When we are talking with a friend andusing informal speech and then start speaking more formally when the boss walks by, we areengaging in “code-switching.” Research suggests that bilingual children are able to differentiatebetween their two languages at a very early age, even as young as age two (Lindholm & Padilla,1978; Meisel, 1987). Therefore, we can conclude that the use of two different languages, as wellas a range of social styles, merely makes a bilingual individual able to communicate in a widerrange of social contexts. Depending on the social situation and the community norms, bilingualcode-switching may be more or less prevalent. However, regardless of the frequency of codeswitching,it should not be taken as evidence that a child is “semilingual.”SummaryIn this chapter, the foundations for first and second language acquisition were reviewed. Centralconcepts in linguistic and child language were presented and discussed in terms of theirimplications for second language acquisition. The similarities and differences between first andsecond language acquisition were delineated. Three current theories of second languageacquisition, those developed by Krashen (1981), Wong Fillmore (1991a), and Cummins (1981)were summarized and their central tenets were compared with the research literature forfeasibility. In light of the information presented on linguistic fundamentals and first and secondlanguage acquisition, the four common fallacies which were presented at the beginning of thechapter were revisited. These were considered with regard to CLD students both with andwithout disabilities. The information presented in this chapter should allow the reader to makemore informed observations and assessment and intervention decisions regarding CLD studentsin the schools.© 2008 Dr. Catherine CollierAll Rights Reserved92

Article III.1 A Brief Description of Second LanguageAcquisitionBy Dr. Kathy Escamilla & Elizabeth GrassiFrom the Professional Development Resource Series, “SecondLanguage Acquisition”, BUENO Center, University of Colorado,Boulder (2000)This essay serves as a summary of several of the seminal theoriesof second language acquisition. It is meant to be used as asupplemental resource for those users who are looking for a moredetailed description of the theories mentioned throughout themodule. While the perspectives discussed here are theoretical innature, they are grounded in the needs of the classroom teacherwho is increasingly likely to be faced with growing numbers ofEnglish language learners in her classroom. Teachers of secondlanguage students are faced with many challenges and often havequestions about how to best teach these students. Common questions include:• How long does it take to learn a second language?• What do we know about second language learners?• What influences the learning of a second language?• What can I do in my classroom to facilitate the process of second language acquisition?• What should I expect the second language learners in my classroom to be able to do?This paper will address the above questions through an analysis of two second languageacquisition theories which have greatly influenced second language teaching today: NativistTheory, and Environmentalist Theory. Nativist theory explores the linguistic aspects of languageacquisition and provides an answer to the question of how people acquire a second language.However, this theory does not adequately address the environmental factors which can affectlanguage acquisition. Environmentalists suggest that social and psychological factors are equallyas important as linguistic factors in second language acquisition. This article provides a briefdescription of both theories and discusses the principal researchers within each camp and thecontributions they have made to the teaching and understanding of second language acquisition.Nativist TheoryThe term nativist “is derived from the fundamental assertion that language acquisition isinnately determined, that we are born with a built-in device of some kind that predisposes us tolanguage acquisition.” (Brown p.25). In 1965, Noam Chomsky, a linguist, proposed the theorythat all people have an innate, biological ability to acquire a language. He theorized that peoplepossess a Language Acquisition Device (LAD), a sort of neurological wiring that, regardless ofthe language to be acquired, allows a child to listen to a language, decipher the rules of thatlanguage, and begin creating with the language at a very young age. With the LAD they are ableto make or understand utterances that they have not previously heard. Their first language isacquired with no direct instruction, no practice, no drills and with no apparent difficulty.Chomsky suggests that, if provided with the correct input, the LAD predisposes all people to theacquisition of a second language in basically the same manner.© 2008 Dr. Catherine CollierAll Rights Reserved93

Most of us cannot remember learning our first language – it just seemed to happen automatically.However, for many children learning a second language, the process does not seem natural orautomatic, and it can be associated with many negative experiences and memories.Stephen Krashen (1977,1981,1982,1985) developed his Monitor Theory based on Chomsky’sconcept of a LAD. The Monitor Theory is composed of four hypotheses that provide aframework for teaching a second language:• The Input Hypothesis;• The Natural Order Hypothesis;• The Affective Filter Hypothesis; and• The Acquisition vs. Learning Hypothesis.These hypotheses lay the foundation for the communication-based teaching strategies that havebecome popular with many instructors today. We will briefly describe each hypothesis, and thenprovide a checklist summarizing the methods and tools necessary to put Krashen’s theory intoaction.Input HypothesisKrashen proposes that children only comprehensible input to activate the LAD and beginacquisition of a second language. Comprehensible input is language (either written or heard) thatis understood by the second language learner. Input becomes comprehensible when the teacheruses strategies such as:Showing pictures or visuals to accompany new vocabulary words and communicativeconcepts. For example, a math lesson on multiplying and dividing negative and positivenumbers should incorporate pictures of the positive (+) and negative (-) signs, as well as thesymbols for multiply and divide. A history lesson on Columbus sailing to the New World shouldincorporate pictures of Columbus, the ships, maps and the Taino Indians. In wood shop, ratherthan just telling the students what the tools are and what they can do, the teacher can hold up thetool or point to the tool as he/she says the name and demonstrate the tools capabilities.Additionally, the teacher can provide second language students with pictures of the tools next towhich the students can write the name and use of the tool to use as a study guide.Incorporating gestures, drama and music into the lessons,Listening to rhythms and music and physically acting out situations make a lesson morecomprehensible and memorable. To better illustrate the experiences of the first English colonistsin Virginia, the teacher might have students act out the situation using their new vocabulary andusing facial expressions to demonstrate the feelings of the colonists. The American states can betaught through songs such as “The United States.” which rhythmically introduces the students tothe name of each state. Carolyn Graham has produced a number of books for English languagedevelopment which use these rhythmic “jazz chants” to emphasize different grammar rules.Designing lessons with hands-on activities and manipulatives. Science and math are idealsubjects for incorporating hands-on activities and manipulatives. In math class, the teacher candemonstrate the concept of perimeter and area by having students place a string around differentsections of grid paper. The string represents the perimeter and the squares within the grid paperrepresent the area. To study the planets of the solar system in science class, the teacher can helpstudents measure out a scaled representation of the distances between planets outside on theplayground. Students can then represent each of the planets and find out why different planets© 2008 Dr. Catherine CollierAll Rights Reserved94

take different lengths of time to complete their orbits. An English or history lesson can utilizepictures and props to support the new vocabulary and the plot of a story. To check forcomprehension, the teacher might have students place the pictures in order and describe thepictures using the newly learned vocabulary. In geography, the teacher or the students can designa floor puzzle of the United States consisting of 50 pieces, each one a different state. Studentscan work jointly on putting the puzzle together. As well, the teacher or students can design afloor map of the world and students can stand on the different continents as the teacher says theirnames.Repeating new vocabulary. The teacher repeats vocabulary crucial to the understanding of thelesson in a variety of ways: 1) games where students place the written vocabulary word on thecorresponding picture or write the word as the teacher holds up the picture; 2) board games suchas trivial pursuit, vocabulary, scrabble, boggle or hangman. 3) computer assisted lessons thatincorporate the new vocabulary.Translation. When the second language group is homogeneous, the teacher can quickly translatekey concepts to make the lesson comprehensible. However, the teacher should not come to relyupon translation as a common teaching tool in her second language classroom, as the studentswill learn to direct their attention to the translation rather than the target language.Krashen posits that without comprehensible input, the second language learner is left with agroup of words that are perceived as incomprehensible noise and can not be processed in theLAD. As Met (1994) states, “By enabling students to match what they hear to what they see andexperience, teachers can ensure that students have access to meaning. Experiential, hands-onactivities make input comprehensible.”Comprehensible input should be adjusted as the child acquires more and more language. “Theacquirer understands (via hearing or reading) input language that contains structure ‘a bitbeyond’ his or her current level of competence.” (Krashen 1981a:100) Krashen refers to thisconcept as i+1, where “i” symbolizes the child’s present stage of acquisition and the “1”symbolizes the more advanced input the teacher will provide the child so that she may progressbeyond the present stage. For example, if a student in a Chinese language class has alreadylearned the weather expressions (its hot, its cold, its snowy etc..) the teacher can introduce theseasons using the weather terms as descriptors. In winter, it is cold and snowy. In summer, it ishot and sunny. The seasons (the more advanced concepts referred to as “1”) becomecomprehensible by describing them using known vocabulary words (“i”) .Ideally, comprehensible input should be attended to in more than just the English languagedevelopment classroom. Northcutt-Gonzalez and Schifini developed what they refer to as the“sheltered approach” to instruction, where materials in content area classes are madecomprehensible by incorporating the strategies of comprehensible input outlined above. Tharp(1989,1991, 1992, 1994) discusses the principal of Contextualization in which teachers tap intostudents’ prior knowledge to create a comprehensible content area lesson. For example, a targetlanguage science lesson on electricity will yield differing results if it is taught to a secondlanguage student who has previously studied batteries and circuits in school in her country oforigin (prior knowledge) versus a second language student who may not have even grown up© 2008 Dr. Catherine CollierAll Rights Reserved95

with electricity in her house. The student who has previous experience with electricity will havethe capability to transfer that prior knowledge to the present lesson and therefore, will have abetter understanding of the lesson and the new vocabulary. The student who has had only alimited exposure to electricity in her personal experience will most likely find the new materialincomprehensible and, therefore, difficult to acquire. It is important that teachers assess the priorknowledge of their students and then, build upon the complexity of the lesson from the point ofprior knowledge to the introduction of more advanced concepts. The principal ofContextualization is very similar to Krashen’s notion of “i+1.” The “i” in this situation is thestudent’s previously acquired academic knowledge. The “1” would symbolize the newknowledge the teacher will build onto the prior knowledge.It is important to note that the sheltered approach to content area instruction is best suited for ahomogeneous group of English language learners. If the Sheltered class consists of a mixedgroup of English-only students and English language learners, the lesson is more difficult toimplement. These two groups have different needs and the English-only students will tend todominate the class due to their superior English skills.Making a lesson comprehensible also involves assessing a student’s level of academicvocabulary. Cummins (1979) discusses two forms of language developed in the acquisitionprocess: Basic Interpersonal Communication Skills (BICS) and Cognitive Academic LanguageProficiency (CALP). BICS is the first type of language a student acquires and is often referred toas “playground vocabulary” or survival vocabulary. It is the language that a person needs tofunction in society or to socialize with family and friends. It usually takes a student 2-5 years toacquire a high proficiency in BICS. Students with advanced BICS skills can converse about aseemingly endless number of topics and are familiar with the target language slang and idiomaticexpressions. However, these same students may have great difficulty in academic areas such asscience, math and history that require a different type of vocabulary (i.e. CALP). Rarely dostudents discuss academic subjects using academic vocabulary while socializing with family orfriends. Therefore, unless specifically instructed in the academic vocabulary necessary tounderstand the lesson, second language students are prone to a low proficiency in this cognitiveacademic language, which can lead to academic failure. To ensure the success of secondlanguage students, it is important for content area teachers to directly instruct second languagestudents (using comprehensible input strategies) in the academic vocabulary and languagepatterns necessary to comprehend the content area lesson. The following would be consideredexamples of academic vocabulary:Math - add, subtract, divide, multiple, integer, graph, etc.;History - democratic, vote, president, constitution, etc.;English - plot, protagonist, character, outline, scene, etc.; andScience - beaker, Bunsen burner, electricity, atoms, etc.Thus, while teachers need to be aware of providing comprehensible input, they must also attendto the degree of Cognitive Challenge. As Tharp (1994) warns, “At-risk students, particularlythose of limited Standard English proficiency, are often “forgiven” any academic challenges, onthe assumption that they are of limited ability; or they are “forgiven” any genuine assessment ofprogress because the assessment tools don’t fit.” Teachers should not “water down” thecurriculum to make it comprehensible. At-risk students require instruction that is cognitivelychallenging and demanding of analysis and critical thinking, not just drills and repetition.Teachers should set challenging standards for their students and then prepare lessons thatfacilitate students’ ascension to these standards. A cognitively challenging lesson in history, for© 2008 Dr. Catherine CollierAll Rights Reserved96

example, should involve the same concepts as those introduced in the corresponding mainstreamhistory class (i.e. Manifest Destiny, the Constitution), but, the second language history teacherwould incorporate strategies to make the lesson more comprehensible (i.e. visuals, drama) anddirectly instruct the students on the academic vocabulary necessary to understand the lesson. Ifthe teacher uses comprehensible input in the lessons while simultaneously developing CALPvocabulary, the majority of students will meet the cognitive challenge presented by the teacher.Checklist for Teachers Implementing the “Input Hypothesis” in the Classroom• Prepare visuals, realia, and manipulatives (drawings, photographs, real life objects,counting devices, etc.).• Determine the prior knowledge students bring to your lesson and plan the lesson fromthat point.• Plan content area lessons that incorporate the same concepts as the mainstream classesbut modify the curriculum by adding devices to make it comprehensible to yourstudents.• Decipher the academic vocabulary the students require to understand the lesson andplan a strategy to teach the students this vocabulary.Natural Order HypothesisKrashen hypothesizes that every person learning a language will acquire that language in apredictable order. For example, students learning English, regardless of their cultural andlinguistic background, will acquire the plural “s” (girls) before the third person singular “s”(likes). Despite the time a teacher spends practicing the grammatical aspect of third personsingular “s” with the students, the students will not use the grammatical aspect in target languageconversations until they have naturally acquired it. The natural order of acquisition is notaffected by instructional sequences. Krashen suggests that providing students with meaningfulcomprehensible input that contains grammar, but focuses on communication, will enablestudents to naturally acquire the necessary grammar. Krashen expanded the “Natural Order”hypothesis when he teamed with Tracy Terrell to develop the four stages of BICS languageacquisition.Stage 1- Pre-ProductionStudent Characteristics and Examples of Student WorkDuring the pre-production stage, the second language student actively listens to the languageinput. Much like a one year old baby, the student is developing comprehension in the secondlanguage. At this level, student ‘s reading and oral production will be marked with a transferenceof their first language pronunciation. For example, the letters of the alphabet represent differentsounds in different languages. A Spanish speaker will pronounce the letter “i” as “e,” and “v” as“b.” Because the Asian languages do not differentiate between the phonemes [l] and [r], thestudents will mix these letters when speaking or reading. Given time, however, the students willacquire the sounds of the English alphabet and slowly change their pronunciation. As well, theformat of stories differs among languages. For example, a Spanish speaker is accustomed tolonger sentences and will, therefore, often ignore the English punctuation when reading aloud toproduce a story formatted more like his/her native language. Again, given time, the secondlanguage student will acquire the format of English language story telling and writing. A stage 1student will be unable to correctly produce verbs in the past or future tenses and will rely on the© 2008 Dr. Catherine CollierAll Rights Reserved97

present tense for communication. A stage 1 student will have great difficulty writing in the targetlanguage.Length of stageThe length of this stage varies greatly, although most students spend 3-6 months in stage 1.Some students are ready to produce language immediately and will mimic the sounds of theteacher and the target language group until they can produce words in the target language. Otherstudents are more hesitant to produce language and will spend a longer period listening beforethey attempt to speak. However, a student’s reluctance to speak in the target language is not anindication that he/she is not acquiring the second language. Students at this stage can followdirections, point, draw, and act out situations in the target language - all without speaking.Teaching StrategiesEffective teaching strategies at Stage 1 include Total Physical Response (TPR), NaturalApproach, and cooperative grouping.Assessment TechniquesAssessment techniques involve a number of modifications, especially that of substituting purelywritten examinations for visual and oral assessments. Students may be assessed by:• placing pictures in order to show the history or sequence of a story,• matching vocabulary words to pictures,• touching or pointing to the vocabulary word (Show me a “saw.” Where is the“chalkboard?”).• drawing concepts and adding words to the drawings (Draw the “Stamp Act” and label itusing these words - “fire,” “stamp”,” British,” and “Colonists”).Each of these assessment techniques require oral instruction and support from the teacher. TheStage 1 learner will not understand solely written or abstract oral instructions.Stage 2 Early ProductionStudent Characteristics and Examples of Student WorkAt the early production level, students began producing utterances of one word. Students willrepeat words they have often heard and that are comprehensible. It is not uncommon for studentsat this stage to produce memorized phrases such as “My name is..” My address is,” “Sit down!, “and “Leave me alone.” Many teachers are fooled by the student's ability to produce a perfectlyformed phrase. However, use of a memorized phrase does not indicate that the student is capableof manipulating and forming original sentences in the target language. The student has notnecessarily learned the function of the separate words in the phrase, but has learned the functionof the phrase as a whole. In other words, the whole phrase is utilized in the same manner that thestudent would utilize a single word. At stage 2, students will still have transference of nativelanguage pronunciation similar to stage 1. Toward the end of stage 2, students will begin toincorporate different verb forms into speech but will over-generalize the grammar rules. Forexample, once the student attributes “ed” with forming the past tense, he/she will use “ed” toform all past tense – regular and irregular. Students will produce utterances such as “She goed tothe store,” or “She doed her homework.” Although the teacher might view theseovergeneralizations as a negative, they actually indicate that the student is progressing quite wellin the target language. Given time and adequate comprehensible input, the student willeventually differentiate between the regular past tense and the irregular and correctly apply thegrammar rules.Length of Stage© 2008 Dr. Catherine CollierAll Rights Reserved98

As with all stages of BICS acquisition, length of time spent at this stage varies; some studentspass through this stage rather quickly and move on to complete sentences while other studentswill communicate in one word utterances for a few months before progressing to stage 3. Thetypical amount of time spent at this stage is 6 months to one year. As long as the teacher providescomprehensible input that continues to progress beyond the students’ present capabilities (i+1),the student will continue to acquire the L2.Teaching StrategiesEffective teaching techniques at stage 2 include T.P.R., Natural Approach, cooperative grouping,drama or acting out of concepts, reading books with illustrations, Sheltered Approach instructionin the academic content areas of math, science and electives, and Instructional Conversation.Tharp advocates the use of Instruction Conversation to encourage skills that will enhanceinteraction with the target culture. “Basic thinking skills - the ability to form, express, andexchange ideas in speech and writing - are most effectively developed through dialogue, throughthe process of questioning and sharing ideas and knowledge that happens in the instructionalconversation” (Tharp 1989, 1994). Instructional Conversation involves eliciting responses fromstudents beyond the “known answers” established in many classrooms. In the majority ofclassrooms, teachers tend to do most of the talking which focuses on a “predetermined anddecontextualized inventory of skills and topics.” (Pease-Alvarez and Vasquez 1994) Studentopportunities to talk are limited and do not mimic the type of talk utilized in student’s home orsocial interactions. Instructional Conversation, on the other hand, focuses on eliciting authenticdialogue between students and the teacher. This type of dialogue not only encourages thepractice of new and different vocabulary, but enlightens the teacher as to the beliefs, ideas andculture of the students. If Instructional Conversation is utilized in a classroom of mixed cultures,other students will also gain an understanding and, hopefully, acceptance of the culturesrepresented in the classroom.Assessment TechniquesAssessment modifications are similar to those of stage 1 except the teacher can now introducemore written words into the exams. For example, the student can:• match written vocabulary words to short written descriptions (1. Columbus - c. a man fromItaly who wants to discover a New World.)• find the written definition of a vocabulary word in a multiple choice test (as long as thevocabulary utilized is not more advanced than the student’s present knowledge).• match a written word to a picture.• answer “who,” “what,” “where,” and “when” questions requiring one word answers (Whogave Columbus the ships to travel to the New World? When did Columbus land in the NewWorld? What type of people did Columbus find in the New World?).• carry out performance assessments that focus on higher order thinking skills whileminimizing linguistic demands.The student will still require oral assistance while taking the test and can rarely performeffectively on a totally written exam. As well, the teacher should expect significant spelling andgrammar mistakes at this stage as listening, speaking, reading, and writing all precede writtenproficiency.Stage 3 Speech EmergenceStudent Characteristics© 2008 Dr. Catherine CollierAll Rights Reserved99

During the speech emergence stage, students begin to construct simple sentences until theyacquire enough language to produce increasingly complex phrases. Transference of firstlanguage pronunciation will begin to subside although accent will continue to be an issue,especially for older learners. Students at this stage begin to use differing verb forms (past,present, future) and can engage in more lengthy conversations. However, teachers should expecta multitude of grammar rule overgeneralizations. For example, students will produce sentencessuch as:• “doesn’t can” instead of “can not”• “doesn’t goed” or “didn’t went” instead of “didn’t go”• “I have many money” instead of “I have a lot of money”• “He have little apples” instead of “He has a few apples”• “She are going to the store” instead of “She is going to the store.”All of these overgeneralizations indicate great progress on the part of the second languagestudent. The above examples demonstrate that the student has acquired rules for negatives, pasttense, future tense, and count and non-count nouns. The student requires more time and morecomprehensible input to sort out the grammar rules and begin to utilize them correctly.Length of StageAs with the prior two stages, students progress through this stage at varied times. The typicallyamount of time spent at this stage is 1-2 years.Teaching StrategiesTeaching strategies at stage 3 include Natural Approach, Sheltered Instruction in content areaclasses, cooperative grouping, drama, reading, writing, and Instructional Conversation. Todevelop better communicative competence as well as cultural awareness, it is beneficial to thesecond language learner to have ample contact and interaction with the target language group atthis stage.To encourage interaction between the second language students and the target culture, Tharpadvocates the use of Joint Production Activity. Tharp highlights the frequency with which jointactivities are used within families, communities and jobs to help children or novices learn byworking jointly on a project with more experienced others. However, in K-12 education, jointactivity is rarely practiced. A Joint Production Activity is an activity performed among mixedgroups: novice student--expert student groups; student--teacher groups; second language student--target language student groups; and expert community member--student groups. Such anactivity is designed to yield an end product that is meaningful to all who are involved inproducing it and includes all those involved in the planning. Joint Production Activities havemany advantages in the second language classroom:• they enhance opportunities for SLLs to acquire new vocabulary and communicativestrategies;• they assist second language students in learning new linguistic skills by interacting withexperts;• they increase the opportunity for SLLs to hear language used in meaningful conversationsand increase the opportunity for SLLs to test their own hypothesis of language usage and;• they help SLLs develop relationships within the target language culture which enhancefeelings of acceptance and understanding and help to eliminate social and psychologicaldistances.Assessment Techniques© 2008 Dr. Catherine CollierAll Rights Reserved100

Assessment techniques at stage 3 can include more written work, although the vocabulary shouldbe modified. Assessment techniques include:• Multiple choice and matching answers,• Short answer essay type questions incorporating “who,” “what,” “where,” “ when,” and“how” questions.• Continued reliance on high cognitive ability – low linguistic ability assessment such asperformance tasks.At this stage, teachers can expect exam answers to be grammatically incorrect with spelling andpunctuation mistakes. For example, a stage 3 answer to the question, “Was Columbus a goodgovernor?” would look something like this: “Columbus, he no good governor because he treatbad the Taino Indians. They no like he and he take theys money and make they slave.” Thestudent has the correct answer although the answer is riddled with grammar and spellingmistakes. At this stage, the teacher needs to be clear about the criteria he/she wishes to grade; theknowledge of the content area lesson or the English language. If knowledge is being graded, thenthe teacher should not mark down for spelling and grammar errors. If language is being graded,then it is appropriate to mark for spelling and grammar.Stage 4 Intermediate FluencyStudent CharacteristicsAt intermediate fluency the student can now manipulate the language to create original sentencesand engage in more complex conversations. The student should have a good grasp on differentverb forms (past, present, future, etc..) although grammatical mistakes are still quite common.The student is ready to acquire advanced writing skills and perfect his/her oral and reading skills.Length of StageThe student tends to stay in stage 4 longer than the other stages. This stage involves morecomplex language skills and the acquisition of advanced academic vocabulary should beemphasized. It is not unusual for a student to be at this stage for 2-3 years.Teaching StrategiesThe teacher should continue to provide the child with comprehensible input, both oral andwritten. At this stage it is beneficial for the child to spend a significant amount of time withtarget language peer models. Instruction should emphasize the teaching of academic languageusing sheltered techniques, the Natural Approach, Instructional Conversation and JointProduction Activities.Assessment TechniquesAssessment techniques are similar to those of stage 3 but can also include modified mainstreamexams that incorporate the more advanced “why” questions. These exams should use wordingthat is straight forward with no slang or idiomatic expressions. The teacher should be aware thatgrammar and spelling mistakes are still common at this stage.Affective Filter HypothesisKrashen claims that all people possess a “filter” which moves into one of two positions, low orhigh. A low position allows language to enter the person’s LAD and be acquired. A highposition prohibits language from entering the LAD thereby restricting acquisition. A lowaffective filter exists when the student feels comfortable and non-threatened in the learningenvironment. A high affective filter exists when a person is too pressured by outside factors torelax and allow the acquisition process to occur. Outside factors include a stressful learning© 2008 Dr. Catherine CollierAll Rights Reserved101

environment such as too much instructional focus on error correction, pronunciation and form, ora humiliating learning situation where the child is forced to produce language before he/she isready. To ensure acquisition of the second language, it is important that the teacher maintain arelaxed and enjoyable learning environment.Checklist for Teachers implementing the “Affective Filter Hypothesis” in the ClassroomCreate a learning environment that is comfortable for all studentsAttempt to model the correct grammar form rather than explicitly correcting the student everytime a mistake occurs.Focus the majority of the lesson on meaningful communication and content area subjects.Acquisition vs. Learning HypothesisKrashen asserts that two separate systems underlay second language performance. The firstsystem is the acquisition system, which naturally occurs when a person receives plenty ofcomprehensible input, has a low affective filter, and the focus of the language lesson is oncommunication and meaningful use of the language. If these criteria are met, the language entersthe learner’s LAD and is acquired into the mind - a totally unconscious process. The advantageto acquiring a language is that the language becomes part of the linguistic system of the learnerand can be automatically used in conversations and communication with the target culture group.The learning system is activated when the learner is conscious of the language and is focused onthe form and rules of the language. Learning a language encourages the student to focus onediting and planning the language rather than communicating with the language. Learning occursmost often in a grammar-based, drill and practice type instructional setting. Although learning isan important aspect of second language acquisition, Krashen believes that in order to fully uselanguage in a communicative setting, the second language student must first acquire the languagebefore learning is introduced. Learning too early in the process will interrupt the acquisitionprocess and produce learners that focus on form and editing and are stilted in their ability tocommunicate fluently.While Krashen’s Monitor Theory offers numerous linguistic explanations for the acquisition of asecond language, it fails to completely address the social or psychological aspects of learning asecond language. These additional factors are important in second language learning. Too oftenteachers are faced with linguistically capable students whose feelings of alienation, fear orfrustration toward the target culture prevent them from acquiring high proficiency in the secondlanguage. The Environmentalist theory of second language acquisition specifically addresses theaffective aspects of second language learning.Environmentalist TheoryEnvironmentalists posit that environmental/outside influences over the learner play a substantialrole in acquisition of a second language. The principal environmentalist theory, Schumann’s“Acculturation Model,” suggests that a learner’s social and psychological distance from thetarget language group influences that individual’s ability to develop proficiency in the targetlanguage.Social distance refers to the social proximity of two cultures that come into contact with oneother. For example, the culture of Americans and Canadians are quite similar. Therefore, thesocial distance between them is minor. However, American and Mexican cultures are very© 2008 Dr. Catherine CollierAll Rights Reserved102

different. Therefore, the social distance between these two cultures is quite great. Schumann(1976) describes social distance as consisting of eight factors, described below, having positiveor negative effects on the acquisition of a second language.Positive Social FactorsSocial DominanceThe second language learner’s cultural groupfeels dominant or of equal status to the targetlanguage group.Integration PatternThe second language learner desires assimilationor acculturation into the target language group.CohesivenessThe second language learner’s cultural groupencourages ample contact with the targetlanguage group.EnclosureThe second language learner’s group requirescontact with the target language group to goabout daily life.SizeThe second language learner’s group is small andencourages inter-group relations.Cultural CongruenceThe target language group and the secondlanguage group are culturally congruent(similar).AttitudeThe attitude of the two groups toward oneanother is positive.Length of ResidenceThe second language learner intends to residewithin the target culture for an extended periodof time.Negative Social FactorsSocial DominanceThe second language learner’s group feelssubordinate to the target language group.Integration PatternThe second language learner desirespreservation of his/her own culturalidentity.CohesivenessThe second language learner’s group iscohesive and tends to discourage contactwith the target language group.EnclosureThe second language learner’s group has itsown churches, newspapers and leaders andis not dependent on the target culture fordaily living.SizeThe second language learner’s group islarge and tends to facilitate only intragrouprelations.Cultural CongruenceThe target language group and the secondlanguage group are incongruent(dissimilar).AttitudeThe attitude of the two groups toward oneanother is negative.Length of ResidenceThe second language learner only intendsto reside within the target culture for alimited period of time.Schumann hypothesizes that the greater number of negative social factors affecting a secondlanguage learner’s (SLL) relationship with the target language (TL) group, the more difficult itwill be for the second language learner to acquire the target language. For example, in a newculture, a SLL can face a number of social pressures:• alienation from the target culture because he/she feels subordinate to the TL group (- socialdominance),• encouragement only to associate with people from his/her culture ( - enclosure) ,© 2008 Dr. Catherine CollierAll Rights Reserved103

• origin from a culture that is very different from the target culture ( - cultural congruence),• a decision to stay in the target culture for only a limited period of time (- length of residence).These social pressures all work against the SLL and make acquiring a high proficiency in thetarget language a difficult task. On the other hand, if the learner feels:• accepted into the target culture and shares equal status with the target culture (+ socialdominance),• associates often with the target culture (+ enclosure),• is highly motivated to learn the language (+ attitude),• and plans to reside in the target culture for a significant amount of time (+ length ofresidence),• the learner will have greater ease in acquiring the language.In addition to social distance, Schumann (1978) describes psychological distance which consistsof three factors: 1) culture shock, 2) language shock, and 3) motivation. These psychologicalfactors, as well as the social factors mentioned above, affect an individual’s ability to acquire asecond language.1) Culture shock is the second stage of acculturation into a new society. The excitement andeuphoria over the newness of the situation eventually wears off and the feeling that more andmore cultural differences are intruding into the SLL’s own image of self and security arise. Atthis stage, the SLL often complains about the target language (TL) culture and seeks solace incountrymen or situations similar to his/her own culture. It is not uncommon for SLLs totemporarily reject the target language and culture at this stage. However, if the SLL receivesassistance and support during the culture shock phase and is encouraged to continue studying theTL, he/she will eventually acquire proficiency in the TL.2) Language shock occurs when the target language is so different from the second languagelearner’s own language that the learner passes through a “shock” period very similar to that ofculture shock. The language system is so overwhelming that the second language learner canreject the target language for a period of time. Seville-Troike refers to the “saturation point” aspart of language shock. Although required to attend school all day in the target language, newlyarrived second language learners can reach a saturation point of learning in the new languageafter 30-45 minutes of instruction. Students who have reached this saturation point will stoppaying attention, act out in class and become frustrated and discouraged. The saturation point canalso occur in second language learners who have been in the target culture for 6 months to ayear. These students hit a plateau in their language learning; they seem to stabilize at a set pointand can not progress. These students also act out in class and become very discouraged. It isimportant for teachers to recognize when a student is at the saturation point and remove thepressure to produce. Eventually, the student will move beyond the saturation point on his/herown and begin to absorb more language. The second language learner needs encouragement andsupport to successfully overcome language shock and continue acquiring the target language.3) Motivation is subdivided into two categories: instrumental and integrative. Instrumentalmotivation refers to the desire to learn the second language for reasons such as furthering acareer, translation, gaining employment, or college admissions. Integrative motivation refers to aSLL’s desire to integrate into the TL culture. In this case, the SLL must learn the target languageto become part of that society. Research by Lambert (1972) and Spolsky (1969) found that SLLswith integrative motivation scored higher on second language proficiency tests than those withinstrumental motivation. However, studies by Yasmeen Lukmani (1972) have found that inIndia, instrumental motivation played an even greater role in language acquisition than did© 2008 Dr. Catherine CollierAll Rights Reserved104

integrative motivation. Basically, motivation in general is psychologically essential to acquiringa second language. If the learner is involuntarily in the target culture and has no motivation tolearn the target language, proficiency is unlikely.Regardless of the competence and qualifications of the SL teacher, if the learner feels greatsocial or psychological distance from the target culture, acquisition of the TL will be negativelyaffected. One of the manners in which a teacher can prevent social or psychological distance isby helping students become familiar with and enjoy the target culture. Swain, Canale, Long andGee hypothesize that language acquisition is enhanced through interaction with people of thetarget language and culture. Swain argues that opportunities for output (student speech) are asimportant to the acquisition of a second language as opportunities for input. While inputprovides students with vocabulary and grammar, output provides students with opportunities totest their hypothesis of the linguistic and discourse rules of the language. Gee hypothesizes asocial component to language acquisition. “Acquisition is a process of acquiring somethingsubconsciously by exposure to models, a process of trial and error, and practice within socialgroups without formal teaching. It happens in natural settings that are meaningful and functionalin the sense that acquirers know that they need to acquire the thing they are exposed to in orderto function and that they in fact want to so function” (1992 pg. 113).ConclusionOur understanding of the process of second language acquisition continues to evolve. Thesedeveloping conceptions in turn influence our beliefs as to what is best practice for the teaching ofEnglish language learners in our schools. Recommendations for teachers have changed over theyears along with our understanding of the research on classroom practice. In this module ourgoal is to present the latest knowledge on both the theory and practice of teaching Englishlanguage learners. As we discussed in the introduction, teachers of second language learnersoften have questions about how to best teach these students, questions like: what can I do in myclassroom to facilitate the process of second language acquisition, and what should I expect thesecond language learners in my classroom to be able to do?In this article we have tried to address such questions by outlining two popular second languageacquisition theories: Nativist Theory, and Environmentalist Theory. We have argued that nativisttheory provides an answer to the question of how people acquire a second language, but that itfalls short in addressing the importance of the environmental factors encountered by the secondlanguage learner. Environmentalist theory highlights the social and psychological factors as wellas the linguistic factors in second language acquisition. This article is meant to introduce thereader to both of these theories as well as to some of the principal researchers within each camp.Much of the remainder of this module is dedicated to the further exploration and elaboration ofthese ideas.© 2008 Dr. Catherine CollierAll Rights Reserved105

academic content. In earlier decades in the United States, we emphasized teaching the secondlanguage as the first step, and postponed the teaching of academics. Research has shown us thatpostponing or interrupting academic development is likely to promote academic failure. In aninformation driven society that demands more knowledge processing with each succeeding year,students cannot afford the lost time.Cognitive development. The fourth component of this model, the cognitive dimension, has beenmostly neglected by second language educators in the U.S. until the past decade. In languageteaching, we simplified, structured, and sequenced language curricula during the 1970s, andwhen we added academic content into our language lessons in the 1980s, we watered downacademics into cognitively simple tasks. We also too often neglected the crucial role of cognitivedevelopment in the first language. Now we know from our growing research base that we mustaddress all of these components equally if we are to succeed in developing deep academicproficiency in a second language.Interdependence of the four components. All of these four components-sociocultural,academic, cognitive, and linguistic-are interdependent. If one is developed to the neglect ofanother, this may be detrimental to a student's overall growth and future success. The academic,cognitive, and linguistic components must be viewed as developmental, and for the child,adolescent, and young adult still going through the process of formal schooling, development ofany one of these three components depends critically on simultaneous development of the othertwo, through both first and second languages. Sociocultural processes strongly influence, in bothpositive and negative ways, students' access to cognitive, academic, and language development.It is crucial that educators provide a socioculturally supportive school environment that allowsnatural language, academic, and cognitive development to flourish.Research Evidence to Support the ModelFirst and second language acquisition: A lifelong process. To understand the processesoccurring in language acquisition during the school years, it is important to recognize thecomplex, lifelong process that we go through in acquiring our first language and the parallelprocesses that occur in second language acquisition. Development of a complex oral languagesystem from birth to age five is universal, given no physical disabilities and no isolation fromhumans. But the most gifted five-year-old entering kindergarten is not yet half-way through theprocess of first language development. Children from ages 6 to 12 continue to acquire subtlephonological distinctions, vocabulary, semantics, syntax, formal discourse patterns, and complexaspects of pragmatics in the oral system of their first language (Berko Gleason, 1993). Inaddition, children being formally schooled during these years add reading and writing to thelanguage skills of listening and speaking, across all the domains of language, with each age andgrade level increasing the cognitive level of language use within each academic subject. Anadolescent entering college must acquire enormous amounts of vocabulary in every discipline ofstudy and continue the acquisition of complex writing skills, processes that continue through ouradult life as we add new contexts of language use to our life experience. As adults we acquirenew subtleties in pragmatics, as well as the constantly changing patterns in language use thataffect our everyday oral and written communication with others. Thus first language acquisitionis an unending process throughout our lifetime (Berko Gleason, 1993; Collier, 1992a). Secondlanguage acquisition is an equally complex phenomenon. We use some of the same innateprocesses that are used to acquire our first language, going through developmental stages andrelying on native speakers to provide modified speech that we can at least partially comprehend© 2008 Dr. Catherine CollierAll Rights Reserved108

(Ellis, 1985; Hakuta, 1986). However, second language acquisition is more subject to influencefrom other factors than was oral development in our first language. When the context of secondlanguage use is school, a very deep level of proficiency is required.Academic second language proficiency: How long? Cummins (1989) popularized foreducators the concept of different levels of language proficiency needed depending on thecontext of language use, basing his theories on the work of many other researchers before him.Given the level of language development needed to succeed in an academic context, my coresearcher,Wayne Thomas, and I have been exploring the "how long" question for the past tenyears, following Cummins' initial examination (1981) of long-term academic achievement ofimmigrants in Canada. In the Thomas and Collier series of studies (Collier, 1987, 1989, 1992b;Collier & Thomas, 1989; Thomas & Collier, 1995), we have carefully controlled for a widevariety of student background variables and instructional treatments, to examine studentperformance on many different types of outcome measures across time. The measures we areusing are the academic achievement measures used by school systems to monitor students'progress in school, including standardized tests and performance assessment measures inlanguage arts, reading, mathematics, science, and social studies. In contrast to a typical languageproficiency test, these are not static measures. Instead, they change with each succeeding gradelevel, because the academic and cognitive work expected with each additional year of schoolingbecomes increasingly more complex. Therefore, results on these tests are very different from theresults on a language proficiency instrument that uses the same form each time it is administered.We choose to use these tests because they are the ultimate measures of academic proficiency in asecond language. When students being schooled in a second language reach deep enoughproficiency levels in a second language to compete at the typical level of native speakerperformance (expressed on a standardized test as 50th percentile or normal curve equivalent[NCE]), this is a major achievement, because native speakers are not sitting around waiting fornon-native speakers to catch up with them. During the school years, native speakers' firstlanguage development is continuing at a rapid rate. For non-native speakers, the goal ofproficiency equal to a native speaker is a moving target (Thomas, 1992).In our studies we have found that in U.S. schools where all instruction is given through thesecond language (English), non-native speakers of English with no schooling in their firstlanguage take 7-10 years or more to reach age and grade-level norms of their native Englishspeakingpeers. Immigrant students who have had 2-3 years of first language schooling in theirhome country before they come to the U.S. take at least 5-7 years to reach typical native-speakerperformance (similar to what Cummins [1981] found). This pattern exists across many studentgroups, regardless of the particular home language that students speak, country of origin,socioeconomic status, and other student background variables. In our examination of large datasets across many different research sites, we have found that the most significant studentbackground variable is the amount of formal schooling students have received in their firstlanguage. Across all program treatments, we have found that non-native speakers being schooledin a second language for part or all of the school day typically do reasonably well in the earlyyears of schooling (kindergarten through second or third grade). But from fourth grade onthrough middle school and high school, when the academic and cognitive demands of thecurriculum increase rapidly with each succeeding year, students with little or no academic andcognitive development in their first language do less and less well as they move into the uppergrades.© 2008 Dr. Catherine CollierAll Rights Reserved109

What about students schooled bilingually in the U.S.? It still takes a long time to demonstrateacademic proficiency in a second language comparable to a native speaker. But the difference instudent performance in a bilingual program, in contrast to an all-English program, is that studentstypically score at or above grade level in their first language in all subject areas, while they arebuilding academic development in the second language. When students are tested in their secondlanguage, they typically reach and surpass native speakers' performance across all subject areasafter 4-7 years in a quality bilingual program. Because they have not fallen behind in cognitiveand academic growth during the 4-7 years that it takes to build academic proficiency in a secondlanguage, bilingually schooled students typically sustain this level of academic achievement andoutperform monolingually schooled students in the upper grades (Collier, 1992b; Thomas &Collier, 1995). Remarkably, these findings apply to students of many different backgrounds,including language majority students in a bilingual program. For example, in Canada, Englishspeakingstudents who receive all their schooling bilingually, typically begin to reach nativespeakernorms on academic tests given in their second language (French) around fifth or sixthgrade, and when tested in their first language, they outperform monolingually schooled students(Collier, 1992a; Genesee, 1987).Role of first language. Many studies have found that cognitive and academic development inthe first language has an extremely important and positive effect on second language schooling(e.g. Bialystok, 1991; Collier, 1989, 1992b; Garcia, 1994; Genesee, 1987, 1994; Thomas &Collier, 1995). Academic skills, literacy development, concept formation, subject knowledge,and learning strategies developed in the first language will all transfer to the second language. Asstudents expand their vocabulary and their oral and written communication skills in the secondlanguage, they can increasingly demonstrate their knowledge base developed in the firstlanguage.Furthermore, some studies indicate that if students do not reach a certain threshold in their firstlanguage, including literacy, they may experience cognitive difficulties in the second language(Collier, 1987; Collier & Thomas, 1989; Cummins, 1981, 1991; Thomas & Collier, 1995). Thekey to understanding the role of the first language in the academic development of the secondlanguage is to understand the function of uninterrupted cognitive development. When studentsswitch to second language use at school and teachers encourage parents to speak in the secondlanguage at home, both students and parents are functioning at a level cognitively far below theirage. Whereas, when parents and children speak the language that they know best, they areworking at their actual level of cognitive maturity. Cognitive development can occur at homeeven with non-formally-schooled parents through, for example, asking questions, solvingproblems together, building or fixing something, cooking together, and talking about lifeexperiences.Role of input and interaction in language development. In our current research (Thomas &Collier, 1995), we have also found that classes in school that are highly interactive, emphasizingstudent problem-solving and discovery learning through thematic experiences across thecurriculum are likely to provide the kind of social setting for natural language acquisition to takeplace, simultaneously with academic and cognitive development. Collaborative interaction inwhich meaning is negotiated with peers is central to the language acquisition process, both fororal and written language development (Ellis, 1985; Enright & McCloskey, 1988; Freeman &Freeman, 1992; Goodman & Wilde, 1992; Swain, 1985; Wong Fillmore, 1991).Sociocultural context of schooling. Research from anthropology, sociology, sociolinguistics,psycholinguistics, and education has provided insights into the powerful and complex influence© 2008 Dr. Catherine CollierAll Rights Reserved110

that sociocultural processes have on language acquisition. Just a few examples are provided here.Among our new arrivals to the U.S. are undocumented as well as legal refugees seeking refugefrom war, political oppression, or severe economic conditions. These students bring to ourclasses special social, emotional, and academic needs, often having experienced interruptedschooling in their home countries. Students escaping war may exhibit symptoms of posttraumaticstress disorder, such as depression, withdrawal, hyperactivity, aggression, and intenseanxiety in response to situations that recall traumatic events in their lives (Coelho, 1994). Studiesof these refugees' adaptation to life in the U.S. and success in school have emphasized theimportance of a bicultural schooling context, integrating first language, culture, and communityknowledge into the curriculum, as well as the importance of parents' maintenance of homelanguage and cultural traditions (Caplan, Choy & Whitmore, 1992; Tharp & Gallimore, 1988;Trueba, Jacobs & Kirton, 1990).External societal factors in the U.S. may have major influence on language acquisition forschool. Examples are the social and psychological distance often created between first andsecond language speakers, perceptions of each group in inter-ethnic comparisons, culturalstereotyping, intergroup hostility, subordinate status of a minority group, or societal patterns ofacculturation vs. assimilation forces at work. Majority-minority and inter-ethnic relations, as wellas social class differences are at the heart of these factors influencing second languageacquisition and success in school. Researchers such as Ogbu (1993), Oakes (1985), andMinicucci and Olsen (1992) have found extensive evidence of institutionalized structures in U.S.schools that deny access to the core curriculum through tracking, ability grouping, and specialprograms that segregate language minority students. Segregated transitional bilingual classes andEnglish as a second language (ESL) classes can sometimes heighten the social inequities andsubconsciously maintain the status quo in majority-minority relations (Hernandez-Chavez, 1984;Spencer, 1988).The negative social perception of these classes that both English-speaking and language minoritystudents have often developed in U.S. schools has led to second-language students' socialisolation, denying them the critical conditions that Wong Fillmore (1991) says must be presentfor second language acquisition to take place. To break the cycle of special classes beingperceived as remedial in nature, they must be a permanent, desired, integral part of thecurriculum, taught through quality instruction that encourages interactive, problem-solving,experiential learning, through a multicultural, global perspective (Frederickson, 1995). Schoolscan serve as agents of change or places where teachers, students, and staff of many variedbackgrounds join together and transform tensions between groups that currently exist in thebroader society.Research-based Recommendations for EducatorsIn our current research (Thomas & Collier, 1995), when examining interactions among studentbackground variables and instructional treatments and their influence on student outcomes, wehave found that two-way bilingual education at the elementary school level is the most promisingprogram model for the long-term academic success of language minority students. As a group,students in this program maintain grade-level skills in their first language at least through sixthgrade and reach the 50th percentile or NCE in their second language generally after 4-5 years ofschooling in both languages. They also generally sustain the gains they made when they reachsecondary education, unlike the students in programs that provide little or no academic supportin the first language. Program characteristics include: (1) integrated schooling, with English© 2008 Dr. Catherine CollierAll Rights Reserved111

speakers and language minority students learning academically through each others' languages;(2) perceptions among staff, students, and parents that it is a "gifted and talented" program,leading to high expectations for student performance; (3) equal status of the two languagesachieved, to a large extent, creating self-confidence among language minority students; (4)healthy parent involvement among both language minority and language majority parents forcloser home-school cooperation; and (5) continuous support for staff development, emphasizingwhole language approaches, natural language acquisition through all content areas, cooperativelearning, interactive and discovery learning, and cognitive complexity of the curriculum for allproficiency levels.In our research, we have also found significant differences between "traditional" vs. "current"approaches to language teaching for students schooled in the U.S. for kindergarten throughtwelfth grade. In the long term, students do less well in programs that focus on discrete units oflanguage taught in a structured, sequenced curriculum with the learner treated as a passiverecipient of knowledge. Students achieve significantly better in programs that teach languagethrough cognitively complex content, taught through problem-solving, discovery learning inhighly interactive classroom activities. ESL pullout in the early grades, when taught traditionally,is the least successful program model for students' long-term academic success. During GradesK-3, there is little difference between programs, but significant differences appear as studentscontinue in the mainstream at the secondary level.When first language instructional support cannot be provided, the following programcharacteristics can make a significant difference in academic achievement for English languagelearners entering U.S. schools at the secondary level: (1) second language taught throughacademic content; (2) conscious focus on teaching learning strategies needed to develop thinkingskills and problem-solving abilities; and (3) continuous support for staff developmentemphasizing activation of students' prior knowledge, respect for students' home language andculture, cooperative learning, interactive and discovery learning, intense and meaningfulcognitive/academic development, and ongoing assessment using multiple measures.We have found that for young children and adolescents in Grades K-12, uninterrupted cognitive,academic, and linguistic development is essential to school success, and neglect or overemphasisof one of these three components may affect students' long-term growth. Our data show thatextensive cognitive and academic development in students' first language is crucial to secondlanguage academic success. Furthermore, the sociocultural context in which students areschooled is equally important to students' long-term success in second language schooling.Contrary to the popular idea that it takes a motivated student a short time to acquire a secondlanguage, our studies examining immigrants and language minority students in many differentregions of the U.S. and with many different background characteristics have found that 4-12years of second language development are needed for the most advantaged students to reachdeep academic proficiency and compete successfully with native speakers. Given the extensivelength of time, educators must understand the complex variables influencing the second languageprocess and provide a sociocultural context that is supportive while academically and cognitivelychallenging.© 2008 Dr. Catherine CollierAll Rights Reserved112

Article III.3 How Children Acquire Language: A NewAnswerBy Dr. Laura Ann PetittoReprinted by permission.How do babies acquire language? What do babiesknow when they start to speak? Prevailing viewsabout the biological foundations of language assumethat very early language acquisition is tied tospeech. Universal regularities in the timing andstructure of infants’ vocal babbling and first wordshave been taken as evidence that the brain must beattuned to perceiving and producing spokenlanguage, per se, in early life. To be sure, a frequentanswer to the question "how does early human language acquisition begin?" is that it is the resultof the development of the neuroanatomical and neurophysiological mechanisms involved in theperception and the production of speech. Put another way, the view of human biology at workhere is that evolution has rendered the human brain neurologically "hardwired" for speech. Overthe past 20 years I have been investigating these issues through intensive studies of hearingbabies acquiring spoken languages (English or French) and deaf babies acquiring signedlanguages (American Sign Language, ASL, or Langue des Signes Québécoise, LSQ), ages birththrough 48 months. The most striking finding to emerge from these studies is that speech, per se,is not critical to the human language acquisition process. Irrespective of whether an infant isexposed to spoken or signed languages, both are acquired on an identical maturational timecourse. Further, hearing infants acquiring spoken languages and deaf infants acquiring signedlanguages exhibit the same linguistic, semantic, and conceptual complexity, stage for stage. Ifsound and speech are critical to normal language acquisition how then can we account for thesepersistent findings? In order for signed and spoken languages to be acquired in the same manner,human infants at birth may not be sensitive to sound or speech, per se. Instead, infants may besensitive to what is encoded within this modality. I propose that humans are born with asensitivity to particular distributional, rhythmical, and temporal patterns unique to aspects ofnatural language structure, along specific physical dimensions (temporal "sing-song" prosodicpatterning and bite-sized, maximally-contrasting syllable segments--both levels of languageorganization that are found in spoken and signed languages). If the input language contains thesespecific patterns, infants will then attempt to produce them--regardless of whether they encounterthese patterns on the hands or on the tongue. One novel implication here is that languagemodality, be it spoken or signed, is highly plastic and may be neurologically set after birth. Putanother way, babies are born with a propensity to acquire language. Whether the language comesas speech, sign language, or some other way of having language, it does not appear to matter tothe brain. As long as the language input has the above crucial properties, human babies willattempt to acquire it.© 2008 Dr. Catherine CollierAll Rights Reserved113

Timing MilestonesDeaf children exposed to signed languages from birth, acquire these languages on an identicalmaturational time course as hearingchildren acquire spoken languages.Deaf children acquiring signedlanguages do so without anymodification, loss, or delay to thetiming, content, and maturationalcourse associated with reaching alllinguistic milestones observed inspoken language. Beginning at birth,and continuing through age 3 andbeyond, speaking and signing childrenexhibit the identical stages of languageacquisition. These include the (a)"syllabic babbling stage" (7-10months) as well as other developmentsin babbling, including "variegated babbling," ages 10-12 months, and "jargon babbling," ages 12months and beyond, (b) "first word stage" (11-14 months), (c)"first two-word stage" (16-22months), and the grammatical and semantic developments beyond. Surprising similarities arealso observed in deaf and hearing children's timing onset and use of gestures as well. Signing andspeaking children produce strikingly similar pre-linguistic (9-12 months) and post-linguisticcommunicative gestures (12-48 months). Deaf babies do not produce more gestures, even thoughlinguistic "signs" (identical to the "word") and communicative gestures reside in the samemodality, and even though some signs and gestures are formationally and referentially similar.Instead, deaf children consistently differentiate linguistic signs from communicative gesturesthroughout development, using each in the same ways observed in hearing children. Throughoutdevelopment, signing and speaking children also exhibit remarkably similar complexity in theirutterances.The Discovery of Manual BabblingIn trying to understand the biological roots of human language, researchers have naturally triedto find its "beginning." The regular onset of vocal babbling--the bababa and other repetitive,syllabic sounds that infants produce--has led researchers to conclude that babbling represents the"beginning" of human language acquisition, albeit, language production. Babbling--and thusearly language acquisition in our species--is said to be determined by the development of theanatomy of the vocal tract and the neuroanatomical and neurophysiological mechanismssubserving the motor control of speech production. In the course of conducting research on deafinfants' transition from pre-linguistic gesturing to first signs (9-12 months), I first discovered aclass of hand activity that contained linguistically-relevant units that was different from all otherhand activity at this time. To my surprise, these deaf infants appeared to be babbling with theirhands. Additional studies were undertaken to understand the basis of this extraordinary behavior.The findings that we reported in Science revealed unambiguously a discrete class of handactivity in deaf infants that was structurally identical to vocal babbling observed in hearinginfants. Like vocal babbling, manual babbling was found to possess (i) a restricted set of© 2008 Dr. Catherine CollierAll Rights Reserved114

phonetic units (unique to signed languages), (ii) syllabic organization, and it was (iii) usedwithout meaning or reference. This hand activity was also wholly distinct from all infants'rhythmic hand activity, be they deaf or hearing. Even its structure was wholly distinct from allinfants' communicative gestures. The discovery of babbling in another modality was exciting. Itconfirmed the hypothesis that babbling represents a distinct and critical stage in the ontogeny ofhuman language. However, it disconfirmed existing hypotheses about why babbling occurs: Itdisconfirmed the view that babbling is neurologically determined by the maturation of thespeech-production mechanisms, per se. Specifically, it was thought that the "baba," CV(consonant-vowel) alternation that infants produce is determined by the rhythmic opening andclosing of the mandible (jaw). But manual babbling is also produced with rhythmic, syllabic(open-close, hold-movement hand) alternations. How can we explain this? Where does thiscommon structure come from? A new series of studies is currently under way to examine thephysical basis of this extraordinary phenomenon (see Optotrak studies below, "The Physics ofManual Babbling").The Physics of Manual BabblingWhere does the common structures in vocal and manual babbling come from? Is manualbabbling really different from all babies' other rhythmic hand movements? I have hypothesizedthat the common structure observed across manual and vocal babbling is due to the existence of"supra-modal constraints," with the rhythmic oscillations of babbling being key. Both manualand vocal babbling, alone, are produced in rhythmic, temporally-oscillating bundles, which Ihave hypothesized may, in turn, be yoked to constraints on the infant's perceptual systems. Thenext challenge then was to figure out how to study it. I recently conducted a new study of manualbabbling with my colleague at McGill, David Ostry, and students Siobhan Holowka de Belle,Lauren Sergio, and Bronna Levy. We used the powerful "OPTOTRAK Computer Visual-Graphic Analysis System. The precise physical properties of all infants' manual activity weremeasured by placing tiny Light-Emitting Diodes (LEDs) on infants' hands and feet. The LEDstransmitted light impulses to cameras that, in turn, sent signals into the OPTOTRAK system.This information was then fed into the computer software that we designed to provide us withinformation analogous to the spectrographic representation of speech, but adapted here for thespectrographic representation of sign. Thus, for the first time, we were able to obtain recordingsof the timing, rate, path movement, velocity, and "info" for all infant hand activity, and to obtainsophisticated, 3-D graphic displays of each. This work is presently in press in Nature (2001).Bilingualism and Early Brain DevelopmentI. Bi-lingual hearing babies acquiring a signed and a spoken language from birth, and bi-lingualhearing babies acquiring two different signed languages from birth (and no speech): Presently, anadditional test of the hypothesis that speech is critical to the acquisition process is underinvestigation in my laboratory, testing two critical populations: (1) "bi-lingual" hearing infantswho are being exposed to signed and spoken languages (i.e., one parent signs, one parentspeaks), and (2) "bi-lingual" hearing infants who are being exposed to two distinct signedlanguages (ASL and LSQ), but who are receiving no spoken language input whatsoever. Withregard to group (1), bi-lingual, signing/speaking children achieve all linguistic milestones in bothmodalities at the same time (e.g., vocal and manual babbling, first words and first signs, firstgrammatical combinations of words and signs, respectively, and beyond; see Petitto et al., in© 2008 Dr. Catherine CollierAll Rights Reserved115

press, Journal of Child Language). Further, infants in both groups (1) and (2) exhibit theirlinguistic and semantic-conceptual milestones on the identical overall maturational time courseas seen in monolingual children (Petitto, 2000), with their specific developmental patterns beingidentical to that which has been observed in the typical case of bi-lingual hearing babies exposedto two spoken languages (e.g., spoken French and spoken English; more below).II. Discovery of common timing in bi-lingual and mono-lingual children: In the course ofconducting the above research on the maturational timing mechanisms in hearing babiesacquiring signed & spoken languages from their bilingual parents, we discovered that our younghearing controls--bilingual children learning spoken French and English--were achieving allmajor linguistic milestones in each of their respective languages on the identical time course, andon the identical time course as monolinguals (Petitto, 1994, 1997). Significance: Prevailingresearch on very young bilinguals, however, had reported that bilingual babies under 20 monthsexhibited language delay and confusion relative to monolingual babies because they ostensiblyhad a single, fused representation of their two native languages, which they were only able tosort out over the first three years of life. By contrast, my findings suggested that very youngbilingual babies have highly distinct representations of their two native languages quite probablyfrom birth. I have further advanced an hypothesis stating what mechanisms in the human brainmay enable the very young baby to differentiate between its two native languages from birth, andI have offered the field an explanation as to why the perception of "delay" and "confusion" inyoung bilinguals has prevailed, both among scientists and the public (see Petitto et al., 2001,Journal of Child Language).© 2008 Dr. Catherine CollierAll Rights Reserved116

Article III.4 Toward a Sociocultural Model of SecondLanguage AcquisitionBy Cory A. Buxton & Dr. Kathy EscamillaFrom the Professional Development Resource Series, “SecondLanguage Acquisition”, BUENO Center, University of Colorado,Boulder (2000)IntroductionCulturally and linguistically diverse (CLD) students in ourschools today face considerable challenges to their academicsuccess. These students need to develop interpersonal skills, adaptto a new culture, master subject area content, and acquireacademic language proficiency for each subject -- all in theirsecond language.We highlight key aspects of the evolution toward the development of a sociocultural model ofsecond language acquisition. We begin by briefly discussing the lengthy history of bilingualeducation in the United States. Next, we consider several of the theories that continue to holdsway over much of the work in bilingual education and second language acquisition. Third, wediscuss some of the teaching practices that have proven to be successful in working with CLDstudents. Finally, we present a sociocultural model of teaching and learning, grounded in thework of Roland Tharp and his colleagues (D'Amato & Tharp, 1997; Tharp, 1989; Tharp, 1997;Tharp & Dalton, 1994; Tharp & Gallimore, 1988), and discuss how this model can be modifiedand applied to second language acquisition in order to enhance the success of culturally andlinguistically diverse students in our schools.At this point, we should pause to make clear who we are referring to as culturally andlinguistically diverse students. CLD students are those students who come from a cultural, social,and linguistic background that is different from the received norms of mainstream EuropeanAmerican culture in this country. CLD is often used to refer to those students who speak a natallanguage other than English. Some of these students may be Limited English Proficient (LEP),while others may be fully fluent in English. The term CLD is generally not used for thoseindividuals of culturally diverse heritage who learn English as their first language and may ormay not be fluent in their heritage language. The term English Language Learner (ELL) is oftenused synonymously with CLD.Historical Trends in Second Language AcquisitionThere is a forgotten legacy of bilingualism and linguistic diversity in this nation. From the timeof the earliest European settlements in the United States until the late 19 th century, an acceptanceand an appreciation of bilingualism could be found in schools and in society at large (Crawford,1991; Kloss, 1977). Since there was no official language policy in the US during our earlyhistory, schools catered to the language needs and desires of the local populous. For example, inthe mid-1800’s, many Midwestern schools offered instruction partly or entirely in German onaccount of the preponderance of German-speaking settlers. The same was true of French inLouisiana and Spanish in New Mexico (Crawford, 1991; Lessow-Hurley, 1991).© 2008 Dr. Catherine CollierAll Rights Reserved117

In all of the above mentioned educational settings, schools were practicing what would today berecognized as maintenance bilingual education. Instruction began in the first language, and thenEnglish was gradually added in the upper elementary grades, while the student’s first languagecontinued to be supported and maintained. Access to these educational opportunities wassupported by state laws. However, “Americanization” efforts in the 1880’s, and later, anti -German sentiment during World War I, lead to a rise in language restrictionism (Castellanos,1983; Kloss, 1977)After World War II, cultural deprivation theory in educational psychology and culturalanthropology, along with political ideologies such as McCarthyism, created a belief thatlanguages other than English and cultures other than the mainstream (white) American culturewere inferior (Castellanos, 1983; Crawford, 1991; Darcy, 1963). The role of the school in thispolitical and educational environment was to remediate cultural deprivation and Americanize theimmigrant. Submersion models, commonly called “sink or swim,” became the norm in schools,with the emphasis being placed on learning English as quickly as possible, with no attentiongiven to the development of skills and knowledge in the first language, and no culturalaffirmation. It is not surprising that disproportionate numbers of language minority studentsbegan ending up in special education classes, and that they continue to do so in schools thatcontinue practicing this model (Baca & de Valenzuela, 1998; de Valenzuela, 1998; Figueroa,1993).It the early 1960’s, bilingual education was reborn in this country due to the immigration ofmiddle-class, educated Cubans following the Cuban Revolution (Castellanos, 1983; Gonzalez,1979). A highly successful, two-way model, bilingual education program was implemented inCoral Way, Florida in 1969. It is important to note that there was a great deal of political andpublic support for Cuban immigrants and their desire for bilingual education because they wereresisting the Communist revolution in Cuba. Programs modeled on the successful experiences ofthe Cubans, using first language instruction and maintenance bilingual education, have continuedto spread over the last several decades. However, an ever-present and growing resistance tobilingual education has flourished as well.During the Reagan Era, the English Only movement, through the political lobbying of the groupU.S. English, gained power and advocated a legislative agenda that directly attacked bilingualeducation. This movement was complimented by the cultural conservatism movement,exemplified by authors such as E.D. Hirsch (1996) and William Bennett (1987). These writerscontended that pluralistic education movements, such as bilingual education, would fragmentand divide Western society. Battles between the proponents and opponents of bilingual educationshow every signs of intensifying during the next few years. Political issues, such as the EnglishOnly movement of the 1980’s and 1990’s, coupled with a backlash against affirmative action andanti-immigrant sentiments, have fueled the controversy about bilingual education. Such politicaldebates have manifested themselves in the form of Propositions 63, 209, 227, and 287 inCalifornia. These propositions made English the state’s “official language,” severely limitedbilingual education, attempted to deprive many immigrants of the right to a public education, anddismantled affirmative action (Padilla, 1990). No doubt, bilingual education will continue to beinfluenced by the larger political and social context.Current Theories of Second Language AcquisitionOur understanding of what language is, how it develops and how it is used by differentindividuals at different times and in different settings, continues to evolve. What is important to© 2008 Dr. Catherine CollierAll Rights Reserved118

emember is that the fundamental goal of language is to communicate one’s intentionssuccessfully to another person. When communication is taking place, proper language usage isoccurring, regardless of our value judgments based on factors such as speech style or adherenceto reified linguistic norms. Our view is that no one language is any better than another. Alllanguages have developed over time to allow their speakers to effectively communicate theneeds and ideas that are important to them.Having an understanding of the basic components of language is an important first step in theconsideration of the more complex ideas surrounding second language acquisition. A lack ofbasic knowledge about language may be at the root of many of the misunderstandings aboutsecond language acquisition. Such misunderstandings have negatively influenced school policyand pedagogical practices.The process of second language acquisition is both similar to and different from the process ofnatal language development (Wong Fillmore, 1991a). Children whose development is influencedby two language systems must be understood vis-à-vis two languages and cultures. It isinappropriate to focus only on their development in English. The extent to which these studentshave had a chance to develop their native language can have important implications for secondlanguage acquisition. Studies suggest that students acquire a second language better if they havea firm conceptual foundation in their native language (Cummins, 1979; Cummins, 1994; Perozzi,1985; Perozzi & Sanchez, 1992).Debate continues among modern linguists as to the specific neurological and cognitive processesthat take place during language acquisition. However, it is generally agreed upon that the humanbrain is predisposed: to attend to language input differently than it does to other sounds, toprocess that input according to some preset principles, and to formulate unconscious rules forlanguage comprehension and production (Chomsky, 1993). What this implies is that humans’approach to learning language is fundamentally different than our approach to learning any otherkind of skill, task, or concept.Thus, an understanding of the process of second language acquisition can only be developedthrough the study of language learners and not through the study of other cognitive tasks. Thequestion of social setting is also critical to understanding how first and second languagesdevelop. Studying language development in one or more languages requires that researchersconsider psycho and sociolinguistic variables as well as individual and situational factors(Hakuta, 1986). These factors make the study of language acquisition, whether in the first orsecond language, messy and complex, but also valuable and fascinating.Based on the above ideas, there are a number of current theories that have been developed in anattempt to explain various aspects of the process of second language acquisition. Five suchtheories are discussed briefly here.Stephen Krashen is one of the foremost theorists in second language acquisition. He hasdeveloped an number of interrelated hypotheses based on the idea that individuals learn languagebest in authentic settings, when they are made to feel relaxed and comfortable, and when thesecond language input that they are receiving is only slightly more advanced than their currentlevel of proficiency (Krashen, 1994).Lilly Wong Filmore (Wong Fillmore, 1991a) focuses on the importance of social, linguistic, andcognitive settings that encourage native speakers and language learners to interact in meaningfulways. She claims that the social environment and the states of language influence languageacquisition and development, including language loss.© 2008 Dr. Catherine CollierAll Rights Reserved119

Jim Cummins has postulated three basic principles related to second language acquisition(Cummins, 1979; Cummins, 1994). These have come to be known as: the conversational/academic language proficiency principle, the linguistic interdependence principle, and theadditive bilingual enrichment principle. It is the first of these three principles that has given riseto the now famous distinction between basic interpersonal communication skills (BICS) andcognitive academic language proficiency (CALP).Saville-Troike has focused on the social settings and academic skills necessary for maximumacademic achievement of CLD students (Saville-Troike, 1984; Saville-Troike, 1988). Amongother things, she found that a well-developed vocabulary is the most important aspect of oralEnglish proficiency for academic achievement. However, an academic focus on structuralpatterns, such as basic grammar rules, appears to contribute little toward meeting students’immediate academic needs. At the root of her theory is the belief that students need opportunitiesto discuss academic concepts in their native language if they are to master content areaknowledge. Thus, social interaction between students is not sufficient for development ofacademic English language skills.Finally, Roland Tharp’s work on sociocultural theory (D'Amato & Tharp, 1997; Tharp, 1997;Tharp & Dalton, 1994), and his development of five principles for cross-culturally compatibleeducation, while not explicitly a theory of second language acquisition, can be readily applied tocreating environments that encourage language development and bilingualism. Tharp’s theoryand its application are discussed in more detail in the final section of this article.Best Practices for Second Language LearnersFrom the perspective of this model, every student brings with him/her to the classroom a widerange of prior skills and knowledge. In order to take advantage of this rich background, as wellas to provide students with the skills and knowledge that they need to be successful in ourschools, teachers must focus on authentic and meaningful language use. Authentic language usetakes place when students are active participants in language use focusing on topics that aremeaningful to them.One important factor in fostering authentic language usage is to focus on the social organizationof the classroom. Research with CLD students (Garcia, 1991; Miramontes, Nadeau, & Commins,1998) demonstrates the importance of teachers making the shift from the traditional classroommodel where the desks are in straight rows and teacher is in front, to a model of classroomorganization emphasizing student centers, and a mixture of guided, group, and independentwork. Not surprisingly, the same organizational models have been shown to be successful withdominant culture students as well (Resnick, 1987a; Resnick, 1987b). When such a shift in socialorganization occurs, it is almost impossible for a teacher not to change her pedagogy to onewhich is more closely aligned with the sociocultural model of teaching and learning.Assessment of the following conditions can be used to provide a rough estimate of whether aschool is providing an adequate learning environment for CLD students that is aligned with whatwe know about best practices: 1) classrooms should be student-centered and offer a nonthreateningenvironment that is informal and literature-rich; 2) classroom climate should becomfortable in terms of space, temperature, and freedom from distractions, and offer access totechnology, curriculum, and supplies that allow for plentiful and appropriate learning choices;and 3) to be equitable, CLD students should be participating in all activities the school offersincluding honors classes, programs for the gifted and talented, and all manner of extracurricularactivities.© 2008 Dr. Catherine CollierAll Rights Reserved120

The best practices models can be thought of as constituting a rough continuum, from those thatpay the least attention to the role of the student’s first language, such as immersion and Englishas a Second Language (ESL) instruction, to those models that give the first language a prominentrole, such as developmental bilingual education and dual language. While each of these modelshas some merit in certain settings, the sociocultural approach to second language acquisition thatwe advocate calls for instruction that supports and develops the first language as the way to besttake advantage of a student’s prior knowledge base (Collier, 1995; Wong Fillmore, 1991b).A Sociocultural Model for Second Language AcquisitionThe sociocultural model that is presented is meant to address the complexities of secondlanguage acquisition, including the challenges faced by CLD students and the importance ofcreating a classroom environment that does as much as possible to help these students cope,survive, and thrive within the challenges of school in an alien culture and a foreign language. Asmentioned earlier, the model we present is based largely on the work of Roland Tharp (D'Amato& Tharp, 1997; Tharp, 1997; Tharp & Dalton, 1994) and his colleagues. Tharp’s model iscomposed of five standards for culturally congruent education. These standards are meant toprovide a framework for creating a classroom environment (both social and academic) thatserves as a starting point for addressing the challenges we have discussed.While Tharp does not consider himself a cultural anthropologist, his model can be categorizedwithin the anthropological framework known as cultural difference theory. The five principlescan be expanded and strengthened by looking to a newly emerging framework from culturalanthropology known as practice theory. Therefore, our sociocultural model for second languageacquisition starts with Tharp’s five standards, and adds a sixth, awareness of culturalproductions. Together, these six principles are meant to push students to succeed academically,while at the same time helping to create a safe and comfortable social environment where theywill not be intimidated and will not have their prior knowledge devalued.We begin here by briefly outlining the framework of cultural difference. Cultural differencetheory is grounded in the idea that cultural differences arise when groups face differenthistorical, social, and economic conditions (Holland & Eisenhart, 1990; Levinson & Holland,1996). Children learn the culture of the group through child-rearing practices in the home and thelocal community. Different child-rearing practices across groups lead to consistent patterns ofbehavior, language use, thinking, and feeling for individuals within groups. However, thesepatterns differ from the behavior, language use, etc. of members of other groups coming fromdifferent historical, social, and economic conditions. Further, the relationship between groups israrely symmetrical. Some groups enjoy greater power, prestige, and status that other groups.One of the strengths of cultural difference theory is that this view lends itself to developinginterventions, based on cultural differences that can be successful in certain settings. Forexample, strategies that address uses of time, space, motivation, and language that are morecompatible with, and which build upon community-specific features, can help engage students inschool who are otherwise prone to disengagement and failure (Heath, 1983; Moll, 1992).Tharp’s five standards have been derived from working with such diverse populations as nativeHawaiian elementary school students (Tharp & Gallimore, 1988), adult Hawaiian vocationaleducation students (D'Amato & Tharp, 1997), and Native Americans on the Navajo and Zunireservations (Tharp & Dalton, 1994). The effectiveness of these principles comes from studyingand making use of broad cultural patterns of behavior. The principles that have emerged asseemingly cross-culturally compatible are as follows:© 2008 Dr. Catherine CollierAll Rights Reserved121

Joint productive activity is the process of both teacher and students working together to producesomething that is of value to everyone involved.Developing the language and literacy of instruction in all activities means that there is an overtfocus on learning the discourse of the content area concurrent with learning the concepts.Contexualizing schooling in students’ lived experiences refers to making connections (in anovert manner) between the content being learned and the students’ personal prior experiences,both within and beyond school.Teaching for cognitive complexity means making sure that every student is cognitivelychallenged, which implies individualization of tasks so that everyone is working on a task that isappropriate for his or her knowledge and ability.Instructional conversation refers to the teacher engaging in conversation with an individual orgroup of students, where the topic of conversation is of importance to all involved, and theteacher does not already know the answer to the question he or she is asking, and additionally,has an authentic reason for wanting to know the answer.Despite the various successes of educational approaches grounded in cultural difference theory,in recent years cultural anthropologists have begun to critique the underlying framework onseveral counts. These critiques can provide insight into how Tharp’s principles can be enhanced.The first criticism is that cultural difference theory pays insufficient attention to power relations(Mehan, 1993). For example, how can the gap between poor, urban Latino culture and theculture of a discipline of power, such as math or science, be successfully negotiated when thereis such a great power differential between the two? The second critique is that cultural differencetheory focuses almost exclusively on the structural (macro) level when defining culture, byfocusing on class, race, and/or gender issues. This overlooks the more micro level influencessuch as peer groups, or the culture of the individual institution (Geertz, 1973).The third critique deals with cultural difference theory's inattention to issues of identity andindividual agency (Minick, 1993). From this perspective, there is no room for an individual todevelop a cultural identity that is at odds with his or her structurally defined culture of origin,while in reality, such individuals clearly do exist. The final argument is that cultural differencetheory paints a monolithic perspective of culture. For example, all students from the samecultural group are thought to have developed the same patterns of thinking, talking, and actingduring childhood, and can likewise be aided by the same classroom interventions (Eisenhart &Graue, 1993).Simply by examining one's own cultural background and the variety of factors that effect it, itbecomes clear that cultural difference theory is an oversimplification of how culture helps shapean individual’s identity in practice. For this reason, we turn to a second anthropological tradition,practice theory, which addresses some of the more intricate dynamics in the interactions betweenstructural features and individual agency.Practice theory is concerned with a number of different aspects of culture, including: 1) thesocio-historical development of cultural groups and their members over time (Levinson &Holland, 1996); 2) the interplay of cultural productions (resistance) and cultural reproductions(conforming voluntarily or involuntarily to the cultural and structural norms) (Levinson, 1996);3) power relations within and across cultural groups (Luykx, 1996); and 4) the dialectic betweenstructural (macro-level) features and individual agency (Holland & Eisenhart, 1990). Practicetheory explores how historical persons are formed in practice, within and against larger societalforces and structures. These structures provide the (tacitly understood) framework that governsthe functioning of schools, as well as other social institutions.© 2008 Dr. Catherine CollierAll Rights Reserved122

By focusing on the dialectic between structure and agency, practice theory gives equal weight tomacro-level structures (i.e. gender, class, race) and micro-level factors (i.e. an individual’sagency and capacity to resist). From this perspective, culture is seen not as a set of characteristicsas it is in cultural difference theory, but rather as a process, continually being constructed inpractice. Individuals are not seen as passively accepting the conditions in which they live andfunction. Instead, they are given agency to counter the determining structures, though they muststill do so while functioning within those structures. Through agency, individuals may have theopportunity to create a different vision for the culture in which they function, and ultimately, acton that vision in ways that challenge the status quo. However, at the same time, there arepressures on the individual to conform to the culture as currently practiced.A consideration of practice theory gives rise to the need for a sixth principle in our socioculturalmodel - awareness of cultural productions. As stated above, cultural productions are the actionsthat an individual takes both within and against the cultural confines in which he/she findshim/herself. This perspective implies that each CLD student will react in different ways to theacademic and social settings in which he/she finds him/herself. Students from nearly identicalcultural backgrounds may react to their school environment in very different ways based on arange of factors, including their interpretation of the power structure in place in the classroom orschool, their interactions and place in the social hierarchy of their peer group, or the opinion ofthe value of education held by individual family members.Given these individual idiosyncrasies, each student will determine the arenas in which she willattempt to conform to the culture of the classroom and school, and those arenas in which she willresist the system. The importance of these avenues of resistance cannot be overstated, because itis through this resistance that transformation of the cultural norms that govern the educationsetting can occur.Thus, awareness of cultural productions is an important principle in considering culturallycompatible academic settings for CLD students, because it calls into question the entire notion ofcultural compatibility. While there are, without question, certain cultural norms that will befound in the majority of the members of a given cultural group, the very nature of being a CLDstudent means that one will be thrust into a dynamic cultural milieu that will challenge and alterthe beliefs inherited from one’s heritage culture. An awareness of cultural productions meansthat a teacher will not make assumptions about a student’s academic needs, desires orpreferences based on cultural overgeneralizations. Rather, the teacher will attempt to get to knoweach student as an individual, to understand why the student accepts and rejects the aspects of theschool’s culture that he or she does, and perhaps even work with the student to transform thoseaspects of the social and academic setting that he or she opposes.ConclusionAs a framework, we have outlined what we mean by a sociocultural model of second languageacquisition. This model takes pieces from several areas of study: existing research done inunderstanding the specifics of how humans process and learn language, principles grounded inbeliefs about cultural difference, and emerging theories related to how individuals work bothwithin and against structural norms to both conform to and resist their cultural and linguisticidentities.© 2008 Dr. Catherine CollierAll Rights Reserved123

Article IV.1 Preserving Home Languages and Culturesin the Classroom: Challenges and Opportunitiesby Lourdes Díaz Soto, Jocelynn L. Smrekar, and Deanna L. NekcoveiDirections in Language & Education, National Clearinghouse forBilingual EducationNo. 13, Spring 1999IntroductionDecades of research document the powerful academic and socioaffectivebenefits of a strong home language base and affirmation ofhome language and culture as a valuable resource (Soto, 1998). Yetpolicymakers across America continue to consider ways ofeliminating diverse languages and cultures by proposing learningenvironments where English is the only language of instruction andculture is a minor issue. Examples of the political pressures affectingour educational system can be seen in the passage of California'sProposition 227, "English for the Children," earmarked at eliminating bilingual educationprograms; in the Wall Street Journal (April 10, 1996) where anti-bilingual advocates callthemselves "bilingual education abolitionists"; in a Texas courtroom where the home languageissue became a child custody issue (NY Times, August 30, 1995); and in the U.S. Congress whereEnglish-only legislation continues to be debated. For educators hoping to instill respect forcultural and linguistic diversity through bilingual and multicultural content, this climate ofcontroversy offers both challenges and opportunities.What are the implicit challenges (daily realities) and opportunities (practical implications) ofincorporating language and culture in today's classrooms? This article explores these issues asthey relate to culturally and linguistically diverse language learners, first describing the dailyrealities faced by teachers, students, families and communities, and then offering practicalsuggestions for all classroom educators.The Challenges: Daily RealitiesThe increased linguistic and cultural diversity in the American classroom is a reality that manyeducators feel compelled to address. Demographic data from the U.S. Census show that, as of1990, 14 percent of all children aged 5 to 17 (about 6.3 million) were reported as not speakingEnglish at home. Moreover, the number of school-age children who did not speak English athome had increased significantly over the previous decade (National Association for BilingualEducation, 1992). Similarly, recent state-reported enrollment figures indicate that the number ofEnglish language learners in grades K-12 increased by almost 2 million during the ten yearperiod between 1986-87 and 1996-97, from 1,553,918 to 3,405,915 (Macias, 1998). It is evidentfrom these trends that educators will continue to face daily realities involving the interrelatedissues of language and culture.Yet, the fact that so many children in our classrooms represent multiple languages and multiplecultures is not a grim reality offering only challenges, but rather a unique opportunity toexchange valuable wisdom among learners, families, educators, and communities. Nor does suchdiversity threaten the cohesiveness of our nation and its democratic ideals; the diverse© 2008 Dr. Catherine CollierAll Rights Reserved124

intergenerational wisdom shared by culturally and linguistically diverse families can support andstrengthen America's goals. To this end, educators can choose to affirm, acknowledge, andrespond to the importance of children's home languages and cultures (NAEYC, 1996).Challenges and opportunities arise from multiple levels of the surrounding environment, levelswhich can be likened to different perspectives on the part of each player (Bronfenbrenner, 1989).For our purposes, the players are the teachers, children, families, and communities that make upa school or district. The following vignettes and discussion will help to illustrate the multipleexperiences of these different groups.Player 1: The TeachersAn early education center in an urban East Coast community serves over 150 linguistically andculturally diverse preschool children. The center's lead teacher voiced these concerns:We service very young children who are either still learning their native language, who aremostly English speakers, or are children who seem to lack dominance in any language. At ourlast staff meeting, we struggled with the language issue because it was not clear to us how weshould proceed.As the staff struggled with finding an effective approach for their diverse students, theyultimately asked parents and other family members what they thought was the best way to teachsecond language learners. While most were ambivalent as to the appropriate method, everyfamily wanted to do what was best for their children. One Asian grandmother stated, forinstance, "I would like all children to learn English quickly." Yet, she was cautious andconcerned when she recalled how the family lost the ability to communicate with their teenagechildren.From experiences such as the one described above, teachers are learning that an important step inteaching language learners is to allay concerns that they will not learn English quickly enough orsucceed academically if they continue to speak their native language at home. The reverse isactually true (Collier & Thomas, 1989; Cummins, 1981) children who have learned concepts athome in their first language can readily transfer this knowledge to the second language. Themajor challenge for teachers and families is to work together to address the combined issues oflanguage shift and lack of respect for home languages. In the vignette above, teachers andparents responded to the challenge by continuing the dialogue and implementing a curriculumthat reflected their unique community of languages and cultures.Player 2: The ChildrenYoung children are caught in the often confusing political and policy making decisions of adultswho find it difficult to respect home languages and home cultures. A Spanish-speaking childabout to enter kindergarten recently asked:Do you know if I can speak Spanish in school? Marcy told me that if you use Spanish words inschool...the teacher punishes you. She said three `Mejicanos' could not go out to play... becausethey kept forgetting. Can you find out if I will be punished too?Unfortunately, accounts of children being singled out and punished for speaking their nativelanguage have been reported by educators across the country. The practice of punishing childrenin response to speaking a language other than English does not consider their backgrounds andexperiences (Bredekamp & Coppolo, 1997). In addition, punishing children for home languageuse is a violation of their linguistic human rights (Skutnabb-Kangas & Phillipson, 1995).Educators are in the position of being able to give native language speaking children support and© 2008 Dr. Catherine CollierAll Rights Reserved125

encouragement as they are attempting to make sense of their world while learning tocommunicate in the second language.Player 3: The FamiliesLinguistically and culturally diverse families also struggle with issues of language shift and lackof respect for home languages and cultures (Soto, 1997). One poignant example involves aLatino family who, like many other families, received the following advice from a wellintentionedKindergarten teacher:La missy nos dijo (the teacher told us) that we should speak English-only at home. She warnedus that if Graciella could not speak English (fluently) by the end of the year, she would be heldback. We told Abuelita (Grandmother) not to talk to Graciella because she would only confuseher. But Graciella spends so much time with Abuelita that we wonder if that is such a good idea.What do you suggest we do?Giving advice to speak English-only at home can have tragic outcomes for culturally andlinguistically diverse families (Wong Fillmore, 1991). Home languages and home cultures are atthe heart of the communicative process for families; thus, teachers who provide advice that iscongruent with research evidence, mainly that a strong home language base actually facilitateschildren's second language learning, are in many ways providing a gift to their students.Player 4: The CommunitiesThe Wall Street Journal article (April 10, 1996), referred to earlier, documents the intentions of agroup of English-only proponents ("bilingual education abolitionists") in a Pennsylvania schooldistrict. Hoping to dismantle the district's bilingual education program, the group initiated aseries of public meetings to discuss the issue. Debate over the educational goals of thecommunity became increasingly divisive, with the school board calling for police protection tomaintain order. This was unwarranted, as the bilingual families who were participating in themeeting were not acting aggressively, but had simply knelt and prayed to demonstrate theirbelief in the value of the bilingual program for their children.Preserving Home Languages and Cultures in the ClassroomWhat is interesting for educators to note about such bitter community battles over multiculturaleducation is that they contrast sharply with what happens in most nations of the world. Schoolsin Denmark, Sweden, France, Greece, Germany, Hungary, Scotland, Canada, Australia, Africa,and England, have long included second language learning opportunities for even their youngestchildren. Teachers are also expected to have second language skills; in Luxembourg, forexample, the typical teacher is competent in French, German, English, and Letzeburgesch (Peck,1993).The opportunity for communities to be knowledgeable about languages and cultures was outlineda decade ago by the President's Commission on Foreign Language and International Studies(1980):The United States is blessed with a largely untapped resource of talent in the form of racial andethnic minorities who, by being brought into the mainstream of educational and employmentopportunities ... can be expected to make rapid, new, and valuable contributions to America'scapacity to deal persuasively and effectively with the world outside its borders (p. 24).© 2008 Dr. Catherine CollierAll Rights Reserved126

Daily Realities RecappedThe above vignettes illustrate the challenge of acknowledging and responding to children's homelanguages and cultures in the midst of so much controversy and at the multiple levels ofchildren's learning environments. Out of this challenge, however, arises the opportunity foreducators to provide their students with the gift of their home language and culture, and to helpthem reap the academic and socio-affective benefits of such an approach.The Opportunities: Practical ImplicationsWhat are the practical implications for teachers working with diverse learners? Althoughselecting effective methods for addressing diversity is complex, and existing research sometimescontradictory, there are strategies that will enable children to value and appreciate their family'sidentity, to feel secure about using their home language, and to respect the cultural diversity oftheir classmates. Below is a list of practical applications for teachers that recognize and honorchildren's multifaceted linguistic and cultural abilities:Preserving Home Languages and Cultures in the ClassroomAccept the premise that children are members of diverse family and community systemsbringing multiple gifts of language, culture, and wisdom (Soto, et al., 1995; Soto, 1996)* Work toward the preservation of home languages and cultures in order to enhance familycommunication and children's identities (Chang, 1993)* Learn and share knowledge about the wisdom gained from culturally diverse families,including consensus building* Maximize cultural courtesies that demonstrate respect for the community, the family, andthe child* Create bonds among caregivers and families that build relationships and model extendedfamily valuesBecome an ethnographer; keenly observe children's language capabilities, both in the nativelanguage and the second language (Genishi, 1989)* Observe and record authentic language use (Soto, 1991)* Create a portfolio documenting language growth and progression in both formal andinformal contexts (Salinger, 1996)Encourage and accept children's attempts to communicate since errors are a normal part oflanguage learning (Bredekamp & Coppolo, 1997)* Adopt a learner-centered teaching style: responsive and nurturing (Smrekar, 1994)* Disseminate information to families that helps them to understand the role homelanguage plays in child development and family cohesion* Afford opportunities for teachers, children and families to share linguistic and culturalknowledgeProvide many opportunities for children to communicate (Smrekar, 1994)* Arrange the classroom with spaces that encourage social interaction* Schedule ample pretend, free play, and dramatic experiences in order to facilitate peersocial interaction* Foster optimal language learning with good language role models in both the home andthe second language (Wong Fillmore, 1992)Plan authentic integrated activities that allow children to value and appreciate their identities(Gutwirth, 1997; York, 1991)© 2008 Dr. Catherine CollierAll Rights Reserved127

* Routinely integrate accurate (not exotic) art, music, children's literature, games, andfoods representing a variety of cultures; afford children the opportunity to participate in planningand preparation of activities* Schedule trips, classroom visits, art, music, and storytelling activities guided by parentsand elders that reflect children's home cultures and languagesOrganize the physical environment to reflect the diversity of cultures in the classroom (Morrison& Rodgers, 1996)* Furnish the classroom with multicultural dolls, authentic art, children's literature, music,authentic dramatic play props, games, and posters that ensure an aesthetically pleasing (notcommercially oriented) learning environment* Establish learning centers that can be utilized in multiple ways (e.g. dramatic play) andreflect various roles and occupations; develop prop boxes of multicultural items (Boutte, VanScoy, & Hendley, 1996)* Choose accurate materials that avoid stereotypical associations or inaccuratecontemporary portrayals* Display family portraits, collages, life-size drawings of the children and familiesBuild lines of communication among linguistically and culturally diverse families and educators(Cummins, 1996; Smrekar, 1994; Soto, 1997; Wong Fillmore, 1991)* Invite parents and elders to share stories, wisdom, and cultural traditions* Plan occasions that will warrant parents' and elders' participation such as children'smulticultural dramatic play events, community feasts and celebrations* Encourage parents to maintain and share their home language and customs* Share suggestions with colleagues during professional development activities, facultymeetings and informal interaction so that they, too, can become well-informed and advocates fortheir studentsInitiate dialogues and advocacy work about language and culture among and acrossorganizations and leaders capable of addressing the specific policy needs of young children(Soto, et al., 1995)* Strengthen communication among nationally recognized organizations involved inmeeting the needs of bilingual/bicultural children* Promote community education that articulates the very real needs of children and familiesin America* Write letters to newspapers, politicians, organizations, and leaders offering themopportunities to work on behalf of culturally and linguistically diverse learnersConclusionThe political and educational challenges discussed above offer substantial opportunities fordeveloping socially competent communities responsive to issues of social justice. Ourwillingness to view children's home language and culture as an important resource is anopportunity to enhance our nation's mission for a democratic society. Our ability to implementthe kinds of educational strategies described in this article affords educators, families andcommunities a myriad of opportunities to share their collective wisdom.Let us pretend for a moment that as a nation, we accept the challenge to preserve children's rightsto be culturally and linguistically competent by affirming children's home language and homeculture. Imagine the wisdom that America can reap from such diverse linguistic and cultural© 2008 Dr. Catherine CollierAll Rights Reserved128

traditions. At a time in our history when we are concerned with our youth, with violence andracism, with the environment and popular culture, it is in our best interest to share knowledgeacross communities, knowledge that helps us to implement collaborative and consensus buildingapproaches, to preserve and respect our environment, to gain peace and harmony in our dailylives, to be considerate of the needs of future generations, to appreciate group success andcommunity spirit, to develop child-rearing practices that are child-centered and reveredendeavors. Collaboratively, we can reach a better place. Together, we can help to achieve anincreasingly democratic and just nation.© 2008 Dr. Catherine CollierAll Rights Reserved129

Article IV.2 Toward A Better Understanding of CodeSwitching and Interlanguage in BilingualityLuisa DuranThe Journal of Educational Issues of Language Minority Students, v14 p69-88, Winter 1994.IntroductionThis paper attempts to help teachers involved in bilingual instruction come to a betterunderstanding of two important aspects of the whole process of bilingual development.Developmental knowledge should aid instructional knowledge, and instructional knowledgeshould aid development.The two major aspects of the bilingual development process which will be outlined here are thenotion of interlanguage and the notion of code-switching. It is important that these nonnormativeaspects of bilingual development be understood better from linguistic perspectives.There seems to be much misunderstanding of these two phenomena and how to treat them in theclassroom.Before a focus on interlanguage and code-switching is done, it is important to summarize whatwe know about language in its broadest and most varied forms, that is, in its monolingual,bilingual, multilingual forms, and althea stages of each.Commonality of All Language FormsThe language development process is a creative process if it is anything at all. The latestaccounts on language development across cultures underscore the notion of an orderly andsystematic linguistic process which is half imitation and half invention (Pfaff, 1987; Berko-Gleason, 1993; Edmundson, 1985). At the same time that language development is orderly andsystematic it is also complete yet not totally complete. While it does arrive at significantconsolidations, accomplishments, and achievements (what we may call stages), Durkin et al.(1986) say that these "will be subjected to addition, elaboration, refinement, re application andre-organization in the years ahead" (p. ix). Acquisition or development as both a synchronic(short-term) and as a diachronic (long-term) process is now better understood by those who studylanguage. Because we have come to understand both the continuous and discontinuouscharacteristics of language development we have also begun to focus our attention on both thenormative and the non-normative patterns of language behavior along this more holisticcontinuum. Nonnormative forms of language are now a major focus of linguistic study inmonolingual, bilingual, and multilingual development. What was and still is many times thoughtof by lay people as "strange," "deviant," "confusing," or “broken" language is being appreciatedby linguists, psycholinguists, and sociolinguists as part of the total natural language developmentand acquisition process. Continued research efforts in this area, along with the traditionalnormative focus, are yielding a portrait of language development which may be more accurateand more helpful to many but especially to those concerned with how to support the naturalcapacities and learning strategies of language learners. This may lead us to a more holistic androbust explanation and description of the language development or the languagelearning/acquisition continuum, spiral, or cycle across the life span, across cultures, and acrosssituations, contexts, and domains.© 2008 Dr. Catherine CollierAll Rights Reserved130

We will now look at two examples of non-normative linguistic behavior found in bothdeveloping bilinguals and in more developed ones. The term non-normative is used here to meanthose aspects of language behavior which are not seen as normal as other language forms bymost lay people. In bilinguality there is the more general notion which linguists call crosslinguisticinteraction or cross linguistic influence. While several particular forms of crosslinguisticinteractions are found in the literature, I will review only two in this paper,interlanguage and code-switching, and only in an introductory way.The Interlanguage NotionThe notion of interlanguage is central in the explanation of bilingual-learner language or SecondLanguage Acquisition (SLA) (Hamers &Blanc, 1990). Interlanguage is the result of theinteraction among the many language acquisition device factors in any two ( in multilingualsituations) languages developing more or less simultaneously. This linguistic behavior isdescribed similarly by many researchers in second-language acquisition and bilinguality studies.The term was first coined by Larry Selinker in 1972, and many others have since expanded thenotion. According the Hamers and Blanc (1990), between the choice of one language or the otherthere exists for the bilingual speaker a whole range of intermediary strategies which include themodification of either code and the relative use of both (see figures 1-8 in appendix ii).Interlanguage maybe viewed as an adaptive strategy in which the speaker tries to speak theinterlocutor’s L1 although he has little proficiency in it. This strategy uses simplification,reduction, overgeneralization, transfer, formulaic language, omissions, substitutions, andrestructurings (Selinker, 1972).Ellis (1985) explains interlanguage as the theoretical constructwhich underlies the attempts of SLA researchers to identify the stages ofdevelopment through which L2 learners pass on their way to L2 or near L2proficiency. He sayslearners do not progress from zero knowledge of a target rule to perfect knowledge of the rule.They progress through a series of interim or developmental stages on their way to targetlanguage competence. Interlanguage is described by many as permeable, dynamic, changing, andyet systematic (Corder, 1975; Selinker, 1972). It may undergo relative fossilization and relativechange, but it reveals an underlying cognitive process even though its surface structure seems theopposite because it does not match conventional forms of what is linguistically correct.According to Andersen (1984), interlanguage goes from a nativization process to adenativization one. For Andersen there has been too much focus on the acquisition of targetlanguage features rather than on the process of interlanguage construction itself. For him secondlanguage acquisition by any other name can and should be characterized in positive terms andprinciples that govern the construction of meaning-to-form and function-to-form relationships.For Klein (1986) interlanguages or learner-language varieties should be viewed not only assystematic as well as variable but also as creative with rules unique to itself and not just aborrowed form of the other languages. He says that however imperfect from a normative point ofview, these means represent the learner's current repertoire and, as such, a learner variety of thetarget language. He concludes that the process of language acquisition can be construed as aseries of transitions from one variety to the next. Pit Corder (1981) defines the structuralproperties of interlanguage as: (a) a simple morphological system, (b) a more or less fixed wordorder, (c) a simple personal pronoun system, (d) a small number of grammatical function words,(e) little or no use of the copula, and (f) the absence of an article system. He describes theinterlanguage notion as "transitional language" and presents a model which is inaccurate invarious respects but is always more or less complete. For Corder it is a working model, a© 2008 Dr. Catherine CollierAll Rights Reserved131

grammar, a system which can be used quite effectively for communicative purposes. Itpresupposes that the language learner at all points of his learning career “has a language" (seeappendixes i and ii). He acknowledges that the notionof interlanguage is difficult for teachers whose attitude toward it is as a deviant, distorted, ordebased form of some standard or target language. Grosjean (1985) also points to theincomprehension and misconceptions by many of the interlanguage behavior of developingbilinguals or of the language behavior specific to bilinguality. He says that when these speakershave used intermediary speech modes that were seen as errors, the speakers themselves wereperceived as confused, unbalanced, semilingual, alingual, anomalous, and so on. While thecomplexity of the interlanguage is not the purpose of this paper to detail, it is introduced as anatural and normal functional process. For the language learner it may only be non-normative,when others who do not understand its normative place, call it such.The Code-switching NotionAnother non-normative or "quaint" linguistic behavior of bilinguals or developing bilinguals iscode-switching and its related concepts such as language transferring and language borrowing.As with the interlanguage construct it is not the purpose of this paper to outline in any refinedway the complexity of code switching and its related notions. Nor is it the purpose of this paperto outline the relationship which exits between the interlanguage notion and the code-switchingnotions, although my hope is that someone is doing or will do this for us soon. It is the intent ofthis paper to try to see them both as interlinguistic phenomena specific to bilinguals. Thesebilingual phenomena are in much need of explanation and understanding for those who must dealwith them in the instructional context or other contexts where this is an important matter. Whileboth interlanguage and code-switching are probably strongly related and may appear more orless concurrently in the language life of the developing bilingual I will use the term codeswitchingfor that point in the developmental time of bilingual learners when they are consciousof such behavior and then choose more or less purposefully to use or not to use it. For ourpurposes only, if there is a beginning, middle, and culminating phase in becoming bilingual Iwould associate the interlanguage notion with the earlier stages of developing bilinguals, andcode switching (including mixing, transferring, and borrowing) with the middle and later phasesof bilingual acquisition. However, while interlanguage is the language constructed beforearriving at more ideal forms of the target language, code switching may occur during and afterthe interlanguage phase. It may be that with children who are simultaneously developing two ()languages from infancy that interlanguage and code-switching may be less distinguishable or lessdiscernible one from the other. In any case, both seem to be a natural cross linguistic outgrowthof becoming or having become bilingual.Code-switching is the use of two languages simultaneously or interchangeably (Valdes-Fallis,1977). It implies some degree of competence in the two languages even if bilingual fluency isnot yet stable. Code-switching may be used to achieve two things: (a) fill a linguistic/conceptualgap, or (b) for other multiple communicative purposes (Gysels, 1992). While in some places andcases code switching is the exception, in many multilingual and bilingual communities it is andshould be seen as the norm (Swigart, 1992; Goyvaerts & Zembele, 1992). It appears that wherecode-switching is the norm it is perceived as fluid, unmarked, and uneventful, and where it is theexception it will be perceived as marked, purposeful, emphasis-oriented, and strange. How iscode-switching explained by those who study it? Gumperz (1982) describes code-switching asdiscourse exchanges which form a single unitary interactional whole:© 2008 Dr. Catherine CollierAll Rights Reserved132

Speakers communicate fluently, maintaining an even flow of talk. No hesitation pauses, changesin sentence rhythm, pitch level or intonation contour mark the shift in code. There is nothing inthe exchange as a whole to indicate that speakers don't understand each other. Apart from thealternation itself, the passages have all the earmarks of ordinary conversation in a singlelanguage. (p. 60)This bilingual state of affairs is described in terms of the heavily interactive nature of the twolanguages. Traditionally code-switching was seen and still is seen by many as a random processhat could be explained by interference. Today it is considered as rule-governed behavior and as acommunication strategy (Corder, 1981).Grosjean and Soares (1986) studied language processing in the mixed language mode inFrench/English and Portuguese/English. They state that a bilingual has the choice of activatingboth, thus code-mixing, or of deactivating one and activating the other in a monolingual context;however, there is never total deactivation of one language when the other is more prominent inthe situation. They propose a base or matrix language and then the bringing in of the otherlanguage by either code-switching through the word, phrase, clause, or sentence level or throughborrowings. The interaction procedure is still unclear in terms of linguistic processing theory.They propose with Obler and Albert (1978) a general language monitoring device that is flexible,rapid, and automatic, and they agree with Paradis (1980) that the bilingual has two languagelexicons, each of which is connected to one conceptual store (Paradis does not posit a bilingualmonitoring device). If there is such a device it uses all the information it can to indicate asquickly as possible which language is being spoken: prosodic information (fundamentalfrequency, duration, rate, amplitude, stress pattern, etc.);segmental information (phoneme andsyllable characteristic); syntactic and semantic rules; knowledge of the speaker and of the topic;pragmatic factors, as well as the constraints imposed on code-switching and borrowing by thetwo languages in question. This means, of course, that the device is constantly receivingfeedback from the higher level processors. The device is always active, but especially so whenthe speaker is in a bilingual speech mode and the probability of language mixing is high.Grosjean and Soares (1984) explain that the challenge for psycholinguists interested in studyingmixed language processing will be to explain how communication in mixed language takes placeso rapidly and so efficiently despite what they have seen as some rather intricate operations andstrategies.Sankoff and Poplack (1980) refer to an "equivalence constraint" or rule which states thatbilinguals in uttering sentences may use constituents of one language at one point and those ofanother at another point as long as the order of these constituents is shared by the two languages(at least in the study of Spanish/English mix modes).Woolford (1983) views code-switched sentences as resulting from a mixture of phrase structurerules extracted from the two languages. She argues that phrase structure rules of the twolanguages can be freely mixed in the construction of the tree structures of code-switchedsentences.Sridhar and Sridhar (1980) assume that there is a basic language bilingual discourse and proposethe terminology of guest and host languages to describe code-switched utterances. They arguethat intrasentential code-switching is a case where guest elements, which have their own internalstructure, occur in the sentences of the host language, obeying the placement rules of the hostlanguage or the matrix language (at least as they saw it in the study of Kannada/English mix).Poplack (1980) discusses two grammatical constraints on code-switching: (a) a free-morphemeconstraint which states that a switch cannot occur between a lexical form and a bound morpheme© 2008 Dr. Catherine CollierAll Rights Reserved133

unless the former has been phonologically integrated into the language of the latter and (b) theequivalence constraint rule which states that the word order immediately before and immediatelyafter a switching point should exist in the two languages to make it possible for a switch to takeplace. The two languages involved can then be interchanged freely.Chana (1984) describes code-switching as the juxtaposition within the same speech exchange ofpassages of speech belonging to two different grammatical systems or subsystems. The items aretied together prosodically as well as by semantic and syntactic relations equivalent to those thatjoin passages in a single language.Lipski (1982) speaks of a bilingual grammar especially during intrasentential code-switching inwritten text: Type I is the monolingual text in which there are some Spanish words withinEnglish literature or vice-versa; Type II text is the bilingual text in which lines from the twolanguages alternate with switches at the phrase or sentence boundaries. Type III text is thebilingual text which would include intrasentential code-switches-- the most highly developed. Heproposes a bilingual grammar constructed of a finely integrated blend of two languages. He saysthat code-switching provides evidence on two related planes: linguistic and psychological. Thepsychological includes the situational variables that permit a switch to occur, and the linguisticincludes factors that facilitate the switch and the precise form that a switched utterance takes.Code-switching seems to serve important communicative and cognitive functions. However inthe majority of communities studied some social stigma has been attached to this mode ofspeaking by both in/out groups.Gibbons (1983) studied language attitudes and code-switching in Hong Kong between Cantoneseand English. They indicate that when Chinese speakers use English with one another they givean impression of status and westernization. When they use Cantonese they give an impression ofChinese humility and solidarity. However, a mix was considered ill-mannered, show-off,ignorant, not good-looking, aggressive, and proud from the Cantonese point of view.Stevens (1983) found similar attitudes between French and Arabic in Tunisia with Arabic thefavored language. Chana (1984) studied listeners' evaluative reactions to code-switched speechby a speaker who was also heard as a perfect Punjabi and as a perfect English speaker. When heused the code-switched form he was considered less fluent, less intelligent, and less expressivethan when he used only Punjabi or only English.Ten years later the code-switching research shows that both users and nonusers of this linguisticform do not admit or allow its acceptability. The findings by language researchers about thelinguistic and communicative integrity of this language form seem not to have been widelyunderstood or accepted yet. A special issue on code-switching in The Journal of Multilingual andMulticultural Development summarizes (a) the wide use of code-switching in different contextsand with different languages-in-contact, (b) the major purpose of code-switching as being both asocial/discourse phenomenon as well as a structured linguistic one, (c) a comparison to therelated notion of borrowing, (d) a description of code switching patterns not as an aberration butas systematic and logical, and(e) the negative view many users and nonusers of code-switchingstill hold about it. In this special issue an important statement drawn from Monica Heller (1992)is that "the absence of code-switching can be as significant as the presence of it" (p. 124). If it issomething which happens naturally in the scheme of bilinguality it must serve importantfunctions for the language learner/user. This natural language function is in direct conflict withnormative or conventional forms and attitudes about what is "good language" and thus, it is notappreciated or supported. People may go to great lengths in some cases from denying it tofinding all sorts of reasons and ways to discourage its use and to eradicate it. Perhaps is it time to© 2008 Dr. Catherine CollierAll Rights Reserved134

question the effects of such attitudes and behavior on the whole bilingual language developmentcontinuum.ConclusionBecause code-switching (and interlanguage) do not sound conventional, and because we do notunderstand the role they play in natural language development and usage, and because we havelittle control over them, we tend to see them as aberrations. Because we do not understand wellthe linguistic and communicative rules and purposes which explain them as natural and creativeoutgrowths of being bilingual we cannot accept them as a pattern unique to bilinguals. Becausehey are neither fish nor fowl we may see them solely as alingualism, semilingualism,interference, confusion, or fossilization instead of as new and alternative forms created bycognitive/conceptual synthesis of two languages. Whether code switching is used to fill a gap orif it is a conscious desire to mix the two languages to create new forms, the language created inmost code-switches has internal linguistic consistency and validity for the learner's deepstructure. While the surface structure also has bilingual consistency and validity to thosecommunicating with it, for language separatists and language purists it is otherwise. Firstlanguage purists fear that the use of the second language with the first will either keep the firstone from growing or debase it or cause confusion in the speaker's mind. Second language puriststhink the same. It may just be that both interlanguage and code-switching are needed for manyreasons, one of which may be in order not o debase, erase, or cause cognitive confusion to eachlanguage. Wouldn't that surprise us all?In stable bilingual/multilingual contexts confusion does not seem to be the case. In fact, forbilinguals, it seems to be a case of a three-pronged language growth. It is important that peoplecharged with the language growth of bilinguals lend support to all three forms in order that theylend support to the development of the two separate languages. I have not come across anystudies which outline the development of bilinguality without the use of interlanguage, codeswitching,and other cross-linguistic phenomena. We do not know empirically for sure whetherand where and how the use of these cross-linguistic creations helps or hinders the separatedevelopment of all the languages in question. We do know that bilinguals eventually do learn toseparate their languages but that sometimes they choose not to. As with the larger bilingualeducation vs. monolingual education story, the story of cross-linguistic influence may show ahistory first of negative effects, next of neutral effects, and later still of some positive effects.It could be that even if code-switching and interlanguage are not intentional goals we might aimfor directly, at the very least it may be that we come to understand the concepts in more positivelinguistic, cognitive, communicative, and developmental terms. Both seem to be highlyincriminated in the growth process when more than one language is developing. We have not yetstudied the effects that stifling or allowing code-switching and interlanguage may have on thetotal linguistic, conceptual, and social development of the bilingual individual. We have notstudied the effects of such on bicultural development and ultimately on cultural restoration andmaintenance or on cultural shift and assimilation. My assumption is that perhaps where there is amore or less balanced code-switching (that is where as much of one language is intertwined inthe utterance of the other) in appropriate, natural, and needed code-switching domains that it maybe one of the strongest signs on the part of code-switchers to simultaneously give equal time,equal value, and equal space to both languages in both the cognitive and cultural sense. Thisshould not detract from total time, total value, and total space given to each language in other© 2008 Dr. Catherine CollierAll Rights Reserved135

linguistic events where code-switching is not needed, intended, or appropriate. For bilinguals andothers with multiple languages there will always be many combinations of language use.Viewing and constructing the world from one cultural point of view may appear to be morenormative and refined and therefore more conventionally accepted. The same constructs can beviewed, however, from two or more worldviews in a rich bilingual/multicultural environment. Inthis case one language might help the other, and sometimes both together may create a new idea,image, thought, behavior, outlook, organization, and adaptation, and thus move culture to newadaptive places in the dynamics of cross-cultural life.Finally code-switching and interlanguage seem to have a function of facilitating and supportingthinking and communication, no matter how the outward information may appear. It remains forus to see how we might approach such facilitation in the bilingual classroom.Humans are capable of constructing many changes, and once having constructed them to usethem or abandon them, and so it is with the non-normative forms of language calledinterlanguage and code-switching. Someday they may come to be seen as normative and assignificant in the life of bilinguals.© 2008 Dr. Catherine CollierAll Rights Reserved136

Article IV.3 Culture Change: Effects On ChildrenDr. Catherine CollierDIVERSITY Vol.2 No.1, Portland, OR: Portland State UniversityThere is a growing concern in the United States about the increasingnumbers of aliens in American society. People worry about how weare going to absorb them, about whether they can be assimilated intoour society, and about what our society will be like if all of thesepeople from different cultures become members of ours. Concernsabout culturally and linguistically different populations are as old ashumanity itself, as old as the first awareness of the other.This concern is especially evident at this time in American history.There are not many areas of U.S. society that have not been affectedby this new influx of culturally and linguistically different people. Those who have retaineddistinct cultural and linguistic identities have special social and educational needs. Many ofthese needs have to do with culture change, sometimes referred to as acculturation, and its effectson the education and development of culturally and linguistically different (CLD) children.There is almost always some degree of resistance to change as most cultural groups do notlightly give up valued practices, whether economic, religious, or communicative. In recentyears, many researchers have studied the effect of language and culture change upon thedevelopment of thinking and social skills among CLD children (Ben-Zeev 1977, Cummins 1978,Wells 1981, Collier 1985).The concern is especially acute for CLD children with learning and behavior problems. Theneeds of the child experiencing problems in the classroom may be both negatively and positivelyaffected by the process of changing from the home culture and language to the culture andlanguage of the mainstream classroom. In this paper, we will discuss culture change amongvarious cultural groups as well as some of the effects of this process upon children's developmentand education. This will include comments regarding children experiencing learning andbehavior problems in the classroom.The common concept of acculturation or culture change is that of the melting pot, the completeassimilation of one group into another. However, assimilation is just one of the possible resultsof the complex process of culture change. Although most social scientists would agree withEdward Spicer (1961) that every contact (between cultures) involves some degree of social andcultural integration, there are several ways to look at what happens during this contact andintegration process. Before one can understand the dynamic process of culture change, one mustconsider first what it is that is changing, i.e., culture. Culture is a very broad and complex termusually viewed as the shared concept of reality or patterns of interaction, communication, andsocialization held in common by a particular group of people.Another aspect of culture is that a group of people, in addition to sharing behavioral patterns andvalues, also share a common sense of identity. There is an identifiable boundary betweenmembers and non-members in the particular culture. This self and external identity becomesespecially meaningful in the establishment and longevity of ethnic groups. For example, bothmembers and non-members of the Navajo culture would recognize fluency in the Navajolanguage as one possible indicator of Navajo-ness, but probably only members would recognize© 2008 Dr. Catherine CollierAll Rights Reserved137

a lack of speculation as a Navajo trait. Traditional Navajos rarely speculate about motives orpast or future happenings. They generally report exactly what they see and hear withoutinterpretation.In a study by Henry (1947) Hopi and Navajo children were asked to make up stories about a setof ambiguous pictures. The Navajos described what they saw and did not try to explain thepictured activity. The Hopi children explained what they thought the people were doing and whythey were doing it. They also volunteered what they thought had led up to the activity and whatmight happen in the future. This type of different view of reality has obvious implications in theuse of particular curriculum materials or instructional techniques.Many cultural elements or indicators are shared by different cultures, especially those in temporalor geographic proximity to one another. This sharing is frequently a result of the process ofculture change, a dynamic source of some elements within a particular culture at a particularpoint in time. Culture is dynamic and no two individual members of the cultural group shareexactly the same system of cultural knowledge. Social scientists all have a slightly differentview of culture, depending on their peculiar perspective. As used in this paper, culture is acognitive construct as defined by Goodenough (1957).(Culture is) whatever it is one has to know or believe in order to operate in amanner acceptable to its members. Culture is not a material phenomenon; it doesnot consist of things, behavior, or emotions. It is rather an organization of thesethings. It is the form of things that people have in mind, their models ofperceiving, relating, and otherwise interpreting them.As used in this paper, culture change is a type of adjustment to the presence of two (or more)cultures. Just as goods and services may be exchanged by the two cultures, so may values,languages, and behavior patterns. Padilla (1 980) proposes that theme are three stages of culturechange: contact, conflict, and adaptation, He states that any measurement of culture change mustconsider each of these three stages at both the group and individual level. Linton (I 940) statedthat directed and non-directed situations of contact, i.e., the purpose of the contact, must also beconsidered. The history, persistence, purpose of the contact, the nature of conflict andadaptations to this contact, as well as the individual's exposure to the second culture, interpersonaland intrapersonal conflicts, and personal adaptations, must all be considered.It is important to consider the nature, purpose, duration, and permanence of the contact. Forexample, less culture change may be expected when the purpose of the contact is mutuallydesired trade, as in the 19th century between the Tlingits and the British in SE Alaska, asopposed to the forced exchange between the Russians and Eskimos in SW Alaska during thesame time period.Where there is deliberate extermination of beliefs over a long period of time we find even greaterculture change. For example, the Moravian missionaries engaged in a systematic destruction ofYup'ik Eskimo culture. There is almost always some degree of resistance to change as mostcultural groups do not lightly give up valued practices, whether economic, religious, or communicative.This conflict may be manifested in many ways, whether as psychological stress orphysical aggression, but will always lead to some form of adaptation. Adaptation is in this sensea reduction of conflict and may take several forms.Another effect of culture change is acculturative stress. This stress is common though not inevitableduring culture change, Berry (1 970) stated that acculturative stress is characterized by deviantbehavior, psychosomatic symptoms and feelings of marginality, Berry also found thatvariations in stress and culture change patterns were dependent to some extent upon the cultural© 2008 Dr. Catherine CollierAll Rights Reserved138

and psychological characteristics of the culture group, and the degree and nature of previouscontact with culturally diverse groups. This type of stress has particular implications foreducators as the side effects look a lot like the indications of learning disabilities commonly usedto place children in special education classes.The influence of child rearing and training patterns is of special importance in the growth anddevelopment of children. Socialization and family interaction patterns are affected by culturechange in the same way that language, economic subsistence, and other cultural elements are.There is also an interactive effect among these elements in regard to the growth and developmentof children. There are effects upon their language, cognitive style, personality, and self concept.These are of concern in the child's development of his or her full potential and especially whenthe child may have special learning and behavior problems in the mainstream school system.In one study of Hispanic culture change, Szapocznik and Kurtines (1980) found that stress fromthe culture change process could result in emotional and substance abuse problems. Thisoccurred when family members adjusting to life in mainstream American culture did not integratetheir home culture and language with that of the mainstream community. Although livingin a community with two languages and cultures, both the parents and teenagers in the studyrejected one culture and tried to identify exclusively with only one culture. These families werecompared with families where parents and teenagers were bilingual and developed cross-culturalmethods of adapting to their new communities. The bilingual and cross-cultural families hadfewer incidents of substance abuse and dysfunctional interactions than the "monocultural'families.In summary, the differing culture change patterns of culturally and linguistically different childrenmust be taken into consideration in their assessment and instruction in mainstream schoolsand classrooms. The interactive effects of a particular child's culture change and the child'sspecial needs must be considered in developing appropriate educational programs. In the nextissue of Diversity, these special needs and the implications of culture and language change uponinstructional intervention will be addressed.© 2008 Dr. Catherine CollierAll Rights Reserved139

Article IV.4 Culture Change: Effects on Children(Part II)In the last issue, Dr. Collier shared information about the process of culture change, focusing onhow it affects the children in various cultural communities. In this issue, she will continue thatdiscussion with the results of a research project. The goal of the project was to find out howculture changes influence the treatment of culturally and linguistically different children inpublic schools. The summary findings were: 1) in mainstream schools, these children arereferred to special education programs much more often than other children, and 2) thesechildren's referral and placement in special education programs was based on teachers'attitudes about what they perceived as learning and behavior problems.This study of five public school districts in Colorado in 1984 and 1985 examined the assessmentand instruction of culturally and linguistically diverse children. When identifying andresponding to children with special needs, what parts do culture change and teacher's attitudesplay? The study highlighted some important differences between those children referred tospecial education programs and those not referred. We also examined how we might predict achild's placement in a special education program by looking at these characteristics. Thechildren we studied were Latino/Chicano students enrolled in bilingual or English as a SecondLanguage programs. In the study, 49 percent of the children had been referred to specialeducation programs. Of those children, 55 percent were actually placed in special educationprograms.There were several major findings. Features of each student's education were evaluated, andonly one reliable way of predicting a child's referral and placement in special education wasfound. This was the teacher's level of concern about what they saw as learning and behaviorproblems. Groups of children with different levels of cultural adaptation and culture changecharacteristics were referred to special education programs at different rates. Placed childrenmore often came from schools with low minority enrollment. Those children who were referredto special education programs, but not placed, were the least culturally adapted and had thelowest achievement levels in all subjects.The schools participating in this study provided information on Latino/Chicano students randomlyselected from their K-6 bilingual and ESL programs. The students were compared on 15variables related to culture change and educational achievement. These features were selectedfrom cultural and linguistic factors, and also from education variables that are usually consideredin the placement of any child in special education.We wanted to look at these variables because prior research showed that children that had beenreferred or placed in a special education program should be much different from those that hadn'tbeen referred, especially in relation to achievement and ability. In previous studies, the culturaland linguistic differences between mainstream and minority students became an additional factorin whether or not a child was referred or placed in special education. In our study, however, allof the children were from the same cultural and linguistic background. As they were also fromthe same socioeconomic background and age range, it was expected that the children's linguisticand cultural characteristics would be more or less the same. In theory, we were expecting to findthat the differences between referred and nonreferred children would be in their educationalprofile, and not their cultural and linguistic profile.© 2008 Dr. Catherine CollierAll Rights Reserved140

Contraryto what wewere expecting, the educational profiles of thechildren referred to specialeducationand those not referredwere relatively the same. The difference wasin their culturalprofiles. There was also a strong link between certainculture change variables and educationvariables. Although "academic" concerns were cited as the primary reason for referral, theree wasno statistically significant difference in academic testt scores in any subject. There was ameaningful effect size between nonreferredand placedchildren inachievement scores. Thismeans that there was a great difference in their achievement, but it was not "statistically"significant.It became clear that the number of minority students enrolled in a school had a direct impact onstudents' educationalachievement. In our study, the nonreferred students that attended schoolswith highh minority enrollment achieved higher test scores than those 'students who attendedd lowminority enrollment schools.Some ofour results concerned all of the children in our study. All of the children with moreyears of bilingual instruction hadbetter educational achievement than children with less bilingualinstruction. Those who had spent more thanfour years in the U.S. consistently scored higher onacademicc testing, and those who had adapted culturally performed betteron educationalachievement.Prior research showed that whenminority enrollment ina school increases andwhere bilingualservices exist, the proportion of culturally and linguistically different children referred to specialeducationdecreases (Finn, 1982). Our research supported this finding, and affirmedthatbilinguall education improves educational achievement for all of these children.Why do these differential placements occur? Maybe when there are more minority studentsenrolled in a school, the quality of the alternative programs, like bilingual instruction, increases.These trends may also be related to the presence of role models and improved self-concepts,among other things. There is also the possibility that teachers with few minoritystudents intheirclasses refer the onesthey have to special education programs too often, and thatt teachers used toworking with many minority students resistreferring those with special learning and behavior© 2008 Dr. Catherine CollierAll Rights Reserved141

problems. Teachers may be less willing to risk censure for referring minority children withlearning and behavior problems.Other prior research tells us that in referring a child to special education, the child's educationalachievement and overall ability may not be as important as we usually think. The teacher'sdegree of concern about the child's special needs may be a more important factor (Ysseldyke &Algozzine, 1981). The results of our study indicate that referred children and those not referreddo not differ in achievement or ability, but teachers have very subjective degrees of concern.Although they didn't differ much on their education profile, all children were referred for "academic"reasons rather than for "behavior" or other reasons. The implications are that regularclassroom teachers need improved training in the identification of learning problems among theculturally and linguistically different. Having more and better alternative programs may be indicated.Also, assisting teachers in utilizing more effective instructional strategies prior to referralmay also help. As a result of this study, such a teacher assistance and child intervention processwas developed and is available.In conclusion, culturally and linguistically different children are being referred to specialeducation programs, for reasons other than educational underachievement. The results andconclusions of this study provide guidance in developing appropriate training for schoolpersonnel in the identification, referral and instruction of culturally and linguistically diversepopulations in public schools.© 2008 Dr. Catherine CollierAll Rights Reserved142

Article V.1 Assessment in ESL & Bilingual EducationGary R. Hargett, Ph.D. NWREL’s Comprehensive Center, Region X. August 1998. Northwest RegionalEducational LaboratoryIntroductionAssessment plays a key role in every aspect of programs for limited-English-proficient (LEP)levels of service, monitoring their progress, improving the programs that serve them, anddeciding when a special program is no longer needed. This paper offers a nontechnicaldiscussion of why, when, and how to assess LEP students. It does not pretend to be exhaustive,but it provides a foundation in the issues program planners should consider in selecting theirassessment approaches.The paper is structured around the purposes of testing in bilingual education and English as asecond language. Different instruments, including specific published tests, are listed and brieflydescribed for language proficiency assessment, achievement testing, and assessment for specialeducation. However, readers should take time to read the sections of the paper on “Asking theRight Testing Questions,” “When to Assess,” and “Defining Second-Language Proficiency” inorder to put those instruments in perspective. In fact, readers are encouraged to take the time toread the entire paper because it embeds various theoretical, instructional, and data-managementissues within the context of different instruments and assessment strategies.Two terms used throughout the paper have sparked some controversy: “limited Englishproficient (LEP)”and “alternative program of instruction.” Many people object to the termlimited English proficient because it defines students in terms of their limitations, what theycannot do as opposed to what they can do. The term “English-language learner” has begun toreplace it, and occasionally, it is used in this paper. However, LEP is used here because it iscommonly recognized and is used in the Improving America’s Schools Act and other federal andstate official documents. An “alternative program of instruction” for LEP students does notimply alternative education in the same sense that it is often used for at-risk students. The term isused by the U.S. Office for Civil Rights and in this paper to refer to strategies that overcomebarriers to equitable education based on language differences. The “alternative program” refersto instructional strategies, not to curriculum. LEP students are entitled to access to the samecurriculum and other programs and services as all other students.Asking the Right Testing Questions“Asking the right questions” means being clear about what purposes a test will serve before it isselected and used. Two “what” questions should guide the selection process:What information about a student is needed?What will that information be used for?These two “what” questions must be answered before the more common “how” questionsteachers usually ask regarding how to assess English proficiency, how to assess native-languageliteracy, how to assess academic achievement, etc.© 2008 Dr. Catherine CollierAll Rights Reserved143

What Information Is Needed About a Student?Tests provide information about a student that is not otherwise available from records or simpleobservation.They also document that information for an audience—such as the public, parents, orpolicymakers—in a way that makes sense to the audience. In either case, a test provides certainkinds of information, and we cannot wisely select our assessment strategies for either purposeunless we know what that information is. When we want to assess students, who are limited intheir English proficiency, examples of the kinds of needed information may include:• Can the student participate in the oral language of a mainstream classroom?• Can the student read and write English at levels similar to his or her mainstream grademates?• Does the student need an ESL or bilingual program?• If the student needs an ESL or bilingual program, what should his or her placement be?• Does the student read and write the native language at grade level?• Are the student’s academic skills near grade level in the native language?• What specific aspects of English grammar or vocabulary does the student lack?• Is the student progressing in oral or written English?The trick becomes matching the right assessment strategy to the school’s purposes, that is,getting the answers to the school’s questions about the student. The right strategy might be apublished test, application of a scoring rubric, or simply the teacher’s informed observation of astudent’s performance. In any case, the strategy must meet this one criterion: The assessmenttasks must represent the kind of information the school wants to know about the student. Each ofthese kinds of information places certain requirements on the test. Figure 1 illustrates this pointby listing requirements a test should meet to answer the questions listed above.Figure 1 Requirements a Test Should MeetQuestionCan the student participate in the orallanguage of a mainstream classroom?Can the student read and write English atlevels similar to his or her mainstream grademates?Does the student need an ESL or bilingualprogram? If the student needs an ESL orbilingual program, what should his or herplacement be?Does the student read and write the nativelanguage at grade level?Are the student’s academic skills near gradelevel in the native language?RequirementsThe tasks must simulate the oral language of amainstream classroom.Ask the student to read or write something atthat level.Give tasks that represent a range of difficulty,from grade-level performance to little or noEnglish proficiency.Staff need to know what that language’sgrade-level expectations are; tasks shouldrequire the student to read and writesomething at that level.Give tasks that represent that language’sgrade-level standards in the academic areas ofinterest.What specific aspects of English grammar or Tasks should pinpoint specific grammatical© 2008 Dr. Catherine CollierAll Rights Reserved144

vocabulary does the student lack?Is the student progressing in oral or writtenEnglish?structures or vocabulary.Tasks should cover a range of difficulty,spanning at least the student’s initial abilitylevel and the level he or she is expected toreach after instruction.t RequirementsIf the school wants to know whether the student can participate in the oral language of amainstream classroom, the tasks must simulate the oral language of a mainstream classroom.1. To find out whether the student can read and write English at levels similar to his or hermainstream grade mates, ask the student to read or write something at that level.2. To find out whether the student needs an ESL or bilingual program, and what his or herplacement should be, give tasks that represent a range of difficulty, from grade-levelperformance to little or no English proficiency.3. If the school wants to find out whether the student reads and writes the native language atgrade level, staff need to know what that language’s grade-level expectations are; tasksshould require the student to read and write something at that level.4. To find out whether the student’s academic skills are near grade level in the nativelanguage, give tasks that represent that language’s grade-level standards in the academicareas of interest5. To find out what specific aspects of English grammar or vocabulary the student lacks,tasks should pinpoint specific grammatical structures or vocabularyIf the school wants to know whether the student is progressing in oral or written English, tasksshould cover a range of difficulty, spanning at least the student’s initial ability level and the levelhe or she is expected to reach after instruction.Assessments that tell whether a student is limited English proficient, or performing similarly toage or grade mates in English or another language, are explicitly or implicitly standardreferenced.In other words, they set criteria for success according to the assessment’s purpose(for example, the child is proficient in English, the child reads and writes his or her nativelanguage at grade level, etc.). Assessments that pinpoint specific weaknesses are not standardreferencedbut offer diagnostic capabilities. Some of the previous questions imply standards, andsome imply the need for diagnostic information. Test selectors must know which kind ofinformation they want and select the right instrument accordingly. Examples of both kinds ofinstruments are given later in this paper.Appropriate Courses of ActionAssessment information is not valuable if it does not lead to a course of action for the student’s(and teacher’s) benefit. For ESL and bilingual education, courses of action include:1. Placing the student in a program that develops his or her English while providingacademic instruction2. Placing the student at the correct level of an ESL curriculum3. Providing the student with instructional services that match his or her academicpreparation in the native language4. Reassigning the student into a different component or level of the alternative programaccording to new assessment results5. Improving the ESL or other instructional services if students do not show progresstoward English proficiency or academic achievement© 2008 Dr. Catherine CollierAll Rights Reserved145

6. Designing instruction to address specifically identified weaknessesAll of these applications of assessment information presuppose an alternative program ofinstruction for a school’s LEP students. All assessment decisions should be made in the contextof such a program that systematically identifies potential LEP students, assesses their Englishproficiency, places them in appropriate levels of service, monitors their progress in English andacademic skills, and exits them when their English proficiency warrants. Readers who do nothave such a program in place should refer to the Northwest Comprehensive Assistance Center’sBilingual/ESL Education Project, which provides guidance on legal requirements for servingLEP students and describes various program options. It also details legal requirements regardingassessment of LEP students.When to AssessThe decisions on when to assess are guided partly by legal requirements and partly by goodeducational practice. The various assessment strategies described in this paper should be selectedaccording their appropriateness for different purposes within an alternative program ofinstruction for LEP students. Initial assessment should occur when potential LEP students areidentified as they first register for school. They should be assessed for English-languageproficiency in listening, speaking, reading, and writing to determine whether lack of Englishproficiency is likely to affect their opportunity to learn, indicating a need for an alternativeprogram of instruction. For initial assessment, select an English proficiency test or assessmentstrategy that represents a range of difficulty, from age- or grade-level performance to little or noEnglish proficiency. If using a published test that states guidelines for determining limitedEnglish proficiency or English fluency, look for evidence that those guidelines were establishedthrough field-testing of the instrument.Good teaching involves ongoing assessment to make sure students are progressing, whether inEnglish proficiency, English or native-language literacy, or academic subjects. Ongoingassessment provides formative information for teachers to use in designing instruction to addressspecifically identified weaknesses. When standards and expectations are clearly communicatedto students, it also allows students to become partners in assessing their own progress. Ongoingassessment can use less formal instruments than those used initially in identifying LEP students.LEP students served in alternative instructional programs should be assessed annually forprogress in English proficiency. Instruments and procedures used should, to the extent possible,be standard across the program; this allows the aggregation of data across grades and buildingsto communicate to the public, parents, or policymakers. It also provides program leaders withpatterns of success or failure among program components so they can improve instructionalservices. Instruments and procedures used in the annual assessment do not have to be the sameones used for initial LEP identification.They should, however, have similar qualities—i.e., they should represent a range of difficulty,from age- or grade-level performance to little or no English proficiency—so the program canshow growth from initial assessment and track the student toward full proficiency and eligibilityfor program exit.Assessing Language ProficiencyHow to assess English proficiency is the main question facing ESL and bilingual educators, oncethey have addressed the “what” questions. They are legally obligated to assess the Englishproficiency of students with a primary or home language other than English to determine© 2008 Dr. Catherine CollierAll Rights Reserved146

whether they are LEP and to provide appropriate services if they are so identified. Schools alsoare obligated to systematically monitor the students’ progress in English. At the classroom level,teachers want to know students’ levels” of proficiency in order to gear instruction appropriately;they want to know what specific aspects of English different students are finding problematic inorder to tailor instruction; and they want evidence that, in fact, their students are makingprogress.Several instruments and strategies are available to meet these various purposes. Some arecommercially published, some are available in the public domain, and some can be constructedby teachers. They rely on different kinds of logic for scoring, and often they represent differentfacets of English proficiency. These instruments and strategies are most effective whenadministered and interpreted by teachers or other staff people who have insight into what we aremeasuring when we measure proficiency in English as a second language. Therefore, it isimportant to take a few minutes to develop an understanding of what language proficiencymeans.Defining Second-Language ProficiencyEveryone agrees that language is used for communicating thoughts, ideas, or information.Communication requires at least two people, and it requires that those two people have acommon system that regularly represents the thoughts, ideas, or information they wish to sharethrough language; one speaks or writes, the other listens or reads. Anyone who has studied oreven encountered a second language immediately recognizes that the second language differsfrom his or her own in that it uses a different vocabulary to represent individual thoughts, ideas,or information.A person who has tried to learn a second language solely by learning word lists soon comes faceto face with the reality that simply stringing words together does not necessarily communicate.Sometimes words in a second language appear in a different order from the native language, andmany other rules govern the relationships among words. For example, different languages attachvery different inflections to the ends of verbs to indicate past time, while some do nothing at allto the verbs. Each language has a systematic grammar, shared among its users, that clarifies therelationships among the words in an utterance. Every natural human language gives physicalform to its words through speech sounds (with a few notable exceptions, such as American SignLanguage). Different languages use different sounds, and different languages may use the samesounds differently. The sounds of a language are also systematically governed by rules ofphonology that are shared by its speakers. The most commonly encountered languages alsorepresent their words in written forms, and written languages also follow elaborate rule systems.To be proficient in a second language means to effectively communicate or understand thoughtsor ideas through the language’s grammatical system and its vocabulary, using its sounds orwritten symbols.Any or all components of this definition may comprise the kind of information you wish to learnabout students when assessing their language proficiency. In one way or another, languageproficiency tests assess the student’s communication ability or knowledge of the components oflanguage. When deciding how to assess, educators must first decide which aspect of languageproficiency they are interested in. Those aspects might include the ability to communicate orallyand in writing similarly to non-LEP grade mates; to use correct or clear pronunciation; or toexhibit knowledge of specific grammatical structures.© 2008 Dr. Catherine CollierAll Rights Reserved147

They must then select the appropriate assessment tool specifically designed to obtain thatinformation.In recent years, most ESL or bilingual educators have taken a stronger or weaker form of theposition that language cannot be “taught.” Rather, its acquisition can only be facilitated. What isimportant, then, is not discrete grammar or vocabulary but communicative competence.Individual educators may make up their own minds on this point. However, our definition oflanguage proficiency shows that effective communication happens with command of alanguage’s grammar and vocabulary, and spoken or written representations of them, whetherthey can be explicitly taught or not. It is also important to remember that current educationalreform movements set standards for writing, and correct use of English conventions figuresprominently in those standards. English-language learners may produce well-organized andinteresting writing, but they will likely be penalized for the kinds of errors that stem fromincomplete command of English’s linguistic rules.Five SuggestionsBefore considering the various instruments and strategies available for assessing languageproficiency, five suggestions are offered that may affect how they are used. First, many of theseinstruments—even ones that have quite specific instructions for the examiner—rely to an extenton the examiner’s personal judgment in scoring an item right or wrong or assigning a rating.Remember that a fair test is not necessarily one on which the student can perform well and get ahigh score; a fair test is one that accurately assesses a student, providing a score that reasonablyrepresents the student’s true ability. In individually administered tests, there’s often a temptationto prompt the student, to give credit for being “in the ballpark,” or to say, based on otherinteractions with the student, “I know he really knows that.” This attempt to be “fair” can in factpenalize the student by failing to identify him or her for needed services. Follow establishedtesting procedures as rigorously as possible. Otherwise, the test may not give a true reading, andwill mean different things for different students, undermining the purpose for giving the test tobegin with. Second, several of the instruments or procedures have other-language forms or caneasily be adapted for other languages besides English. It is useful to assess students’ abilities intheir native language, especially when the alternative program has the capability to providenative-language instruction and development.The fact that a student with a home language other than English is limited in English does notmean that student is proficient in the home language. The student may profit from additionalnative-language development. In some cases, it may be that the student’s English, thoughlimited, is stronger than the home language. Two students, testing at the same level in Englishand with the same native language, may be candidates for different kinds of program servicesdepending on the strength of their native language.Third, do not rely on one instrument to meet all needs. Select a variety of instruments orstrategies for the most complete picture of a given student or a group of students. Instrumentsthat work well for initial identification and program exit may not be useful in helping a teacherplan for a given class of students or for individualizing instruction. Similarly, instruments thatprovide rich individual information may not lend themselves to aggregation for programevaluation or to communicating progress to a public audience. And sometimes an instrumentgives information that doesn’t match other sources of information. If a student tests as LEP on anEnglish proficiency test but reads and writes English as effectively as monolingual Englishspeakinggrade mates, the latter information might override the test score.© 2008 Dr. Catherine CollierAll Rights Reserved148

Fourth, ESL and bilingual program staff should enlist the support of the district’s testingspecialists.Initial identification and annual progress assessment especially should be integrated into thedistrict’s broader assessment program. It is a mistake to leave ESL and bilingual testing solely tothe personnel assigned to that program. Appropriate services for LEP students, includingidentification and monitoring, are the district’s legal responsibility. Integration helpsinstitutionalize that responsibility. Also, ESL and bilingual program staff may not have thetechnical expertise needed to interpret and report testing data. For example, some tests providedifferent kinds of scores—such as level scores, scale scores, and percentiles—which areanalyzed and reported in unique ways. The district’s testing specialists should be involved in theanalysis, interpretation, and reporting of program wide data.Fifth, make sure all educational personnel who serve LEP students understand the implicationsof the assessment results and what limited English proficiency means. It does not mean thestudent lacks innate ability or cannot learn. In many cases, it does not mean the student is not atgrade level; many LEP8 students have records of academic success, but they cannot demonstratetheir knowledge in English. The student’s identification as LEP simply means the student needshelp in acquiring English proficiency and should be taught in ways that account for his or herlinguistic differences.Instruments for Assessing Language ProficiencyThis discussion of language proficiency assessment instruments concentrates on Englishproficiency assessment because that is the one constant in all ESL and bilingual educationprograms. Some of the instruments have non-English forms, and some can easily be adapted toother languages. Implications and applications for other languages are discussed in the context ofeach instrument. Many more instruments exist than are listed here. The tests selected forinclusion in this paper illustrate the matching of tests to various purposes, represent differentapproaches to language assessment, and are commonly used among ESL or bilingual programsin the Northwest. Unfortunately, instruments for assessing language proficiency do not lendthemselves to any logical taxonomy. It would be very helpful to say, “These are the instrumentsto assess oral proficiency; here are pragmatic, integrative tests of communicative competence;this list contains discrete-point tests that isolate problematic grammar; this test is appropriate forelementary school program placement, that one for secondary schools.” In real life, theinstruments spill over across categories and imperfectly fit others, so this discussion willconsider each instrument in turn and discuss what it assesses, how it is scored, and how it can beused to answer questions about students’ language abilities.This discussion is not intended to prepare the reader to administer the instruments, nor toevaluate their technical quality, nor to corroborate or dispute their claims. For publishedinstruments, technical or examiner manuals are available to provide technical data and detailedinstructions on administration, scoring, and interpretation. Readers who are interested in aspecific instrument should request additional information from the publisher and determine forthemselves whether it suits their programs. In some cases, examples are given of how a test’sdata can be reported. Those examples are not necessarily unique to that test. For example, if anexample is given of the reporting of NCE data, that example could apply to any test that yieldsthat kind of score.© 2008 Dr. Catherine CollierAll Rights Reserved149

1. Language Assessment Scales, Oral (LAS­O)Published by CTB\Macmillan\McGraw-HillThe LAS-O is published in three different forms: the Pre-LAS, intended for young childrenapproximately four to six years of age; the LAS-O I, for elementary grade levels; and the LAS-OII, for secondary levels. (A LAS-A, for adult second-language learners, is also available.) TheLAS-O is individually administered. It presents a variety of oral language tasks, includingsupplying correct vocabulary for pictures of objects; responding to comprehension questions ortasks; and producing oral language in response to picture stimuli. Four categories of tasks arediscretely scored on the basis of right or wrong answers. A fifth category, accounting for half thescore, utilizes a rating scale.The LAS-O is scored to classify students into five different proficiency levels subsumed withinthe broader designations of non-English speaking, limited English speaking, and fluent. Ittherefore can meet a number of purposes. It can be used as part of the initial identification ofLEP students needing an alternative program of services. Depending on how the alternativeprogram is structured, it can place students into different instructional groupings. It can be usedto track annual progress in oral English proficiency. And it can form part of the criteria forredesignating students as no longer limited in oral English proficiency and possibly eligible forprogram exit.The LAS-O provides two principal kinds of scores: proficiency levels (1-5) and converted scores(1-100). The proficiency levels are useful for categorizing students as eligible for the ESL orbilingual program, and for producing data that depict the characteristics of a school’s LEPpopulation. For example, they can be used to construct a table such as that in Figure 2, drawnfrom a small program serving 50 students in kindergarten through grade three. The figure showsthe distribution of the students by grade and LAS proficiency levels. The reader can tell at aglance that initially, low-proficient students were more concentrated at lower grade levels. Avisual comparison of the pretest and posttest data shows there was general progress becausemore students at all grade levels scored at higher proficiency levels on the posttest.However, proficiency-level scores have limitations. They can understate or overstate studentprogress, particularly for individual students. A level score spans a range of converted scores.For example, converted scores of 55 and 63 both fall within proficiency level 2. But based onlevel scores, a student who advanced from 55 to 63 would show no growth. Similarly, a studentmight gain only one or two points, essentially making no progress, yet could show advancementfrom one proficiency level to another. It is a good idea to collect, record, analyze, and report bothkinds of scores.The LAS-O is also available in Spanish. It is useful to assess students’ abilities in their nativelanguage, especially when the alternative program has the capability to provide native languageinstruction and development. The fact that a student with a home language other than English islimited in English does not mean that student is proficient in the home language. The studentmay profit from additional native- language development, and in some cases, it may be that thestudent’s English, though limited, is stronger than the home language.2. Language Assessment Scales, Reading and Writing (LAS R/W)Published by CTB\Macmillan\McGraw-HillThe LAS R/W is published in three different levels to accommodate grades two through highschool. Each of the three levels contains a variety of reading and writing tasks, some scored on aright-or-wrong basis, some judged according to a rating scale. Scores on the different sectionsare combined into standard scores, 0-100, which in turn can be interpreted according to threereading/writing competency levels, representing non-reader/writer; limited reader/writer; and© 2008 Dr. Catherine CollierAll Rights Reserved150

competent reader/writer. All versions of the LAS R/W can be group administered. It is availablein both English and Spanish.Because the LAS R/W provides scores that indicate whether a student is performing at acompetent level in reading and writing, and it spans a range from non-reader/writer to competent,it can be used as part of the initial identification process, to track annual growth, and as one ofthe criteria for exit from an alternative program of instruction.3. Woodcock­Muñoz Language SurveyPublished by Riverside Publishing Co. (Houghton Mifflin)The Woodcock-Muñoz Language Survey is intended to provide information on a student’scognitive and academic language proficiency, that is, the extent to which the student commandsthe kind of language typically required in school. It is individually administered. It has bothEnglish and Spanish forms, both consisting of four subtests, two of which comprise a score fororal language ability, two a reading and writing ability score. Together, all four comprise a broadlanguage-ability score. All items are scored as right or wrong; there are no productive tasks—such as original writing or spoken discourse—scored on a rating scale. It is constructed torepresent the language abilities expected at all ages or grade levels from preschool throughcollege.The Woodcock-Muñoz yields several different kinds of scores. It provides levels, called CALP(cognitive academic language proficiency) levels 1-5. The levels designate negligible English orSpanish; very limited English or Spanish; limited English or Spanish; fluent English or Spanish;and advanced English or Spanish. (Levels can be expressed at intermediate values; for example,a 3-4 is a student between levels 3 and 4.) It also yields grade equivalents, age equivalents,percentiles, normal-curve equivalents, scale scores (called W scores), and standard scores.When both the English and Spanish forms are used, they can provide a comparative languageindex, indicating which of the two languages a student has a better command of. TheWoodcock-Muñoz can be used as part of the initial identification of LEP students needing analternative program of services. Depending on how the alternative program is structured, it canplace students in different instructional groupings; it can be used to track annual progress inlanguage proficiency; and it can form part of the criteria for redesignating students as no longerlimited in English proficiency and eligible for program exit.The variety of scores provided accommodates various interpretation and reporting needs. Forexample, if CALP levels are reported in both English and Spanish, and a school’s program hasbilingual capabilities, those scores can be used for placement in different components. Considerthe hypothetical example of data from 71 students at the same, or perhaps adjacent, grade levelswho took both the English and Spanish versions, shown in Figure 3 on the following page.Several students are clearly limited in English and much stronger in Spanish, making themlogical candidates for native-language instruction that builds on their native-language abilities.Others are limited in both Spanish and English and could profit from further, intensivedevelopment in both. Some are limited in English, but their English is clearly stronger than theirSpanish. Those students need an alternative program of instruction, but they may not becandidates for instruction in Spanish, particularly if resources for that are limited.As stated previously, level scores have limitations. For example, they are difficult to use intracking progress for groups of students. For that purpose, NCE scores or scale scores may bepreferable. Figure 4 presents a hypothetical example of an elementary ESL program usingWoodcock- Muñoz Broad English NCE averages by grade level to represent gains made overone year’s time, from 1996-1997. This kind of display can be used to show that by and large,© 2008 Dr. Catherine CollierAll Rights Reserved151

students advanced in their broad English abilities but remained quite low and still in need of analternative program of instruction.Article V.2 Measurement of AcculturationBy Catherine Collier, Ph.D. Excerpted from The Multicultural Handbook forSchool Psychologists. Lawrence Erlbaum and Associates. (2007)Acculturation Screening ToolsEarly acculturation studies conceptualized acculturation as a process taking placealong a single, or unilinear, continuum (Szapocznik et al. 1978). “According tothe unilinear model, acculturation occurs when a person moves from one end ofa continuum, reflecting involvement in the culture of origin, to the other end of the same continuum,reflecting involvement in the host culture.”(Kim & Abreu, 2001). Many of the current instrumentsused to measure acculturation are dual-cultural unilinear measurements based on a single continuum,with one end reflecting high adherence to the indigenous culture and the other indicating highadherence to the dominant culture. Szapocznik and his colleagues (Szapocznik & Kurtines, &Fernandez, 1980) developed the first bilinear measurement model of acculturation in which onecontinuum represented either cultural involvement or marginality while the other continuum reflectedeither monoculturalism or biculturalism. There have been several formal and informal screeningtools developed to measure and monitor the level and rate of acculturation. In the United States thesetools have been primarily developed to study the adaptation of Spanish-speaking students fromMexico, Cuba and Puerto Rico. There are also specific tools to assess the acculturation of individualsfrom Asian, African American and Native American backgrounds. Table 1 compares the most recentof these tools by language focus, age range, and other factors. Most of these focus on thepsychological impact of acculturation upon youth and young adults adapting to American society andhave been used to identify students at risk for dysfunctional sexual relationships and substance abuseproblems.© 2008 Dr. Catherine CollierAll Rights Reserved152

Table 1Comparison of Recent Acculturation Screening ToolsTitle/ AuthorAcculturation QuickScreen (AQS)/C. CollierAcculturation RatingScale for MexicanAmericans(ARSMA) Rev./I. Cuellar (originalauthor S. Orozco)Scale To AssessAfrican AmericanAcculturation/L. Snowden,A. HinesBehavioralAcculturation Scale/J. SzapocznikLatino YouthAcculturation Scale(LYAS)/ M. Pillen,R. Hoewing-RobersonOrthogonal CulturalIdentification Scale(OCIS)/Oetting & BeauvaisShort AcculturationScale for HispanicYouth (SASH-Y)/A. Barona, J. MillerSocietal, Attitudinal,Familial, &EnvironmentalAcculturative StressScale (SAFE) – Rev./D. Chavez (originalauthor A. Padilla)Suinn-Lew AsianSelf-IdentityAcculturation Scale(SL-ASIA)/R. SuinnCurrentEdition/Original Pub.Date;Publisher2002/ 1985CCDES1995/ 1991HispanicJournal ofBehavioralSciences1999Journal ofBlackPsychology1978/1975WestviewPressAge/GradeRange6-20 yrs/1st-12thCollegeageAdultsAdults/HighschoolPurpose Ethnic Focus LanguageFocusMeasures level andrate of acculturationto public schools inUS & Canada.Measurespsychologicaladjustment oncultural orientationrange - MA toAngloMeasuredpsychologicalacculturationcharacteristicsMeasures adherenceto home culture vs.transition to N.American culture.1992 5th-8th Measures level ofacculturation forLatino youth1991 All Measuresidentification with atraditional culturevs. identificationwith other cultures.1994 HispanicJournal ofBehavioralSciences1997/1985HispanicJournal ofBehavioralSciences1992/1987Educational &PsychologicalMeasurement5th-8th8-10 yrs.oldStudentsModification of anexistingacculturation scalefor Hispanic adultsMeasures level ofacculturative stressModeled after asuccessful Hispanicscale to apply toAsian subjectsAll indigenous,immigrant,refugee, &migrant groupsin US &Canada.MexicanAmericanAfricanAmericanCubanAmericanPuerto Rican/MexicanLatinoAll ethnicgroupsHispanic youthLatinoEnglish andall languagesand dialectsBilingualSpanish/EnglishEnglishEnglishSpanish/EnglishEnglishSpanish/English(assumed)Spanish/English(assumed)Results5 levels ofacculturation withspecificrecommendationsfor strategies tofacilitate integrationRevised scalegeneratesmultidimensionalacculturative typesEthnic orientationwithin sampleindicated, withnotable variationson 10 scale itemsIntergenerationaldifferences shownto develop asyounger familymembersacculturate fasterthan older onesFour key factorsappear to driveacculturation:family identity,self/peer identity,customs, and food.Has been validatedfor: NativeAmerican youth andadult MissionIndians, as well asother populationsAnalysis indicatedrelevance ofExtrafamilial &Familial LanguageUse, & EthnicSocial RelationsU.S.-born Latinosexperiencedsignificantly moreacculturative stressthan mainstreampeersAsian English Reliability &validity datareported for 2samples of AsiansubjectsResearch ExampleCompareddifferences amongHispanic studentswho were or mightbe referred andplaced in specialeducation.Comparedadjustment tocollege.Assessed reliabilityand validity socialintegrationmeasures.Used to studyChinese-Canadianyouth in Manitoba,Canada.Used to comparewith high-riskbehavior rate.Compared ethnicidentification ofAsian Americanuniversity studentsvs. acculturationusing the Suinn-Lew ScaleN/ACompared US- bornLatino &Euroamericanchildren age 8-10- Has also beenspecifically adaptedfor KoreanAmericanadolescents)Re-validated withChinese- andFilipino-Americans18 years and older© 2008 Dr. Catherine CollierAll Rights Reserved153

Acculturation of Asian AmericansThe reliability and validity of the Orthogonal Cultural Identification Scale (OCIS) was evaluated in asample of Asian American university students (Johnson, Wall, Guanipa, Terry-Guyer, Velasquez,2002) using the OCIS to measure the strength of the students’ ethnic identification and the degree ofthis strength as a factor in the students’ level of acculturation. In this study, the OCIS was comparedwith the Suinn-Lew Asian Self-Identity Acculturation Scale. The OCIS and the Suinn-Lew scale werefound to measure related but separate constructs. The Suinn-Lew scale is a dual-cultural unilinearmodel of measurement. A monocultural unilinear system of measurement is the Asian Values Scale(Kim, Atkinson, & Yang, 1999) used to measure an individual’s adherence to Asian cultural valuesendorsed more highly by Asian Americans than by European Americans.Acculturation of African AmericansThe African American Acculturation Scale (AAAS) is used to measure changes in and retention ofbeliefs, practices and attitudes within the African American population. A shorter version has beendeveloped and validated (Landrine & Klonoff, 1995). Another tool for measuring the acculturationof African Americans is the SAAAA, Scale to Assess African American Acculturation (Snowden &Hines, 1999). Both the AAAS and the SAAAA use a dual-cultural unilinear model of measurementwith a continuum of scores ranging between immersion in the African American culture toimmersion in the European American culture.Acculturation of Hispanic/Latino AmericansAs part of a substance abuse prevention project, a tool was developed in both English and Spanish tomeasure levels of acculturation among members of a Latino youth population for later comparison ofacculturation with high-risk behaviors. The Latino Youth Acculturation Scale (LYAS) appears tomeasure levels of acculturation in terms of four factors: family identity, self/peer identity, customs,and food (Pillen & Hoewing-Roberson, 1992). The Short Acculturation Scale for Hispanic Youth(SASH-Y) is a modification of an existing acculturation scale for Hispanic adults and provides ameans to measure the impact of language use within and outside the family settings as well as ethnicsocial relations (Barona & Miller, 1994). The Acculturation Rating Scale for Mexican Americans(ARSMA) has been used several times to examine the acculturation of Mexican students at thecollege level, especially in conjunction with the Minnesota Multiphasic Personality Inventory(MMPI). The ARSMA consists of 20 questions measuring preferences and behavioral tendencies ona five-point continuum from Mexican to Anglo orientation (Orozco, 1991; Orozco & Freidrich, 1992;Dana, 1996). There is also a revised version available (Cuellar, 1995), which provides multifactorialassessment. The Bicultural Scale for Puerto Ricans (BSPR) is a bilinear tool using factor analysiswhich examines reference culture, language preferences and usage, values, child-rearing practicesand more (Cortes, Rogler, and Malgady, 1994).Acculturation of Native AmericansThe Navajo Family Acculturation Scale (Boyce & Boyce, 1983) is a dual-cultural unilinear system ofmeasurement designed to measure the modernity of the physical home environment, involvement ofthe family in the elements of a traditional lifestyle, and the extent of family contact with the Angloworld. The Rosebud Personal Opinion Survey (Hoffman, Dana, & Bolton, 1985) is a dual-cultural© 2008 Dr. Catherine CollierAll Rights Reserved154

unilinear model, which measures the behavioral dimension of acculturation. The AcculturationQuick Screen for Native Americans (Collier, 2000) uses a modified bilinear model to measure theadaptation of indigenous people in the Americas to public schools.The Acculturation Quick ScreenThe AQS is discussed in depth here because it is not specific to any one language or ethnic groupand is used to measure the relative level and rate of acculturation of students to public schoolculture throughout the United States and Canada. It is a specific tool for school psychologists andgeneral education personnel, which measures adaptation to academic instructional settings. TheAQS is based on research (cited above) on the factors that predict the degree of successfulintegration for those who are experiencing “culture shock”. Since students acculturate to newenvironments at different rates it is often difficult to tell who is still experiencing difficulty andwho is not. The AQS measures this acculturation and leads directly to strategies to addressculture shock. The AQS can also be used to monitor the rate of acculturation of diverse students.Most diverse students will acculturate gradually over several years and at a steady rate relative tothe elements measured on the AQS. Those who do not show change year-to-year may have someunidentified difficulty or may be having some other destabilizing stressful experience. Thus theAQS can be used to separate difference from disability concerns when diverse learners exhibitlearning and behavior problems. It can also be used to monitor the adaptation progress beingmade by migrant, immigrant and refugee students and provide an early warning system foreducation personnel when something is not working correctly or most effectively for thesestudents within the building or district. In one study in a school district in Washington State, anevaluation team used the AQS to monitor individual students participating in the English as asecond language (ESL) service program (Collier & Pennington, 1997). When data wasdisaggregated by language population, it was found that the Spanish speaking ESL students werenot adapting at the same rate as the Russian speaking students in the program. Russian speakingstudents showed gains on average over 11% each year whereas Spanish speaking students onaverage were making only 8% per year. These AQS scores added to the body of evidence thatHispanic students in the study population were not doing as well in school as the Russian,Ukrainian, and Georgian refugee students. The performance and achievement differences werealso found in an increased referral and placement rate of Hispanic students into SpecialEducation in relation to Russian speaking students. Using the AQS data as an indication of somesystem weakness rather than assuming there was some inherent disabling element among theHispanic students, the researchers found that the intervention program available to non-Spanishspeakers within the ESL and general education program was more intensive and comprehensive.Spanish speakers did not have access to the same level of support both during and after schoolthat Russian speakers were receiving. The district used this information to modify their serviceplans; research is continuing.The AQS measures the student’s approximate level and rate of acculturation to mainstreamAmerican school culture. It is not intended for use in isolation nor as a predictive tool, butprovides a useful piece of supplemental assessment information when students from diversecultural and linguistic backgrounds enroll in a school district. It should be part of the screeningroutinely done to determine eligibility for newcomers or other special language/cultureassistance services. In addition, it is recommended for use as part of the information gathered tomake instructional decisions during the “pre-referral intervention” period or for“prevention/intervention” instructional activities. It may be used to plan the selection of specific© 2008 Dr. Catherine CollierAll Rights Reserved155

intensive learning and behavior interventions for culturally/linguistically diverse students ratherthan referring them to an evaluation and staffing. The AQS is also useful for substantiatingdecisions to modify testing evaluation and assessment procedures. In general, students scoring atthe lower acculturation levels should not be assessed with standardized assessment instrumentswithout case-specific modification of administration and interpretation. The AQS providesdocumentation and guidance to school psychologists and other evaluation specialists workingwith culturally/linguistically diverse students.The AQS should be administered at least four weeks after students have entered the school. Thiswill allow the user to assess their language abilities and to obtain previous school records. Thisfirst AQS will be the baseline from which to measure rate and level of acculturation. Studentsshould be assessed every year at the same time to obtain an ongoing record and documentation oftheir rate of adaptation to the school system. The results map against a range from lessacculturated to more acculturated on a 48-point scale. The AQS measures five levels ofacculturation: Significantly Less Acculturated, Less Acculturated, In Transition, MoreAcculturated, and Significantly More Acculturated. The average rate of acculturation is between10% to 12% each school year, depending upon type of program offered to students. Studentsfrom diverse cultural and linguistic backgrounds will vary in their rate and level of acculturationto public school culture, but all are affected by various factors. These include individualcharacteristics of the students and their school.Although any student may be measured with AQS, it is most useful for students who come froma cultural or linguistic background that differs significantly from the mainstream of his or herparticular public school. For example, the AQS will be useful with an American student from anethnically, linguistically or racially diverse background who may be demonstrating learning orbehavior difficulties. It also provides a significant profile for placement of refugee or immigrantstudents.The AQS is based on research conducted in rural and urban school districts concerning thereferral and placement of limited English proficient children of migrant, refugee and immigrantfamilies. The original study population was a random sample of students in grades K-8 who wereidentified as Hispanic and limited English proficient by their districts. The results of the studyshowed that perceptions of acculturation contributed to decisions to refer and place thesestudents. Additionally, it was found that student characteristics could be used to accuratelyidentify and monitor level and rate of acculturation among at-risk students. Later the AQS wasmodified for use with other linguistically and culturally diverse populations. The AQS has beenused by school systems to monitor and plan assistance to at-risk students from Native American,Asian American, African American, and other culturally and linguistically diverse populations.The information needed to complete the AQS is as follows:1. Number of years the student has been in the United States.2. Number of years the student has been in the current school: This should be actualcumulative time in this school to the extent possible.3. Number of school years the student has received direct instruction in bilingual or Englishas a Second Language classes: This should be actual cumulative time in this instruction to theextent possible. Using school records, interviews with parents or other teachers, identify howmuch time, if any, the student has spent in direct instruction in bilingual education or English asa second language (ESL) classes. For example, a student in a dual language program receiving atleast an hour each day of direct instruction in ESL in the content area would have received a fullyear in ESL/Bilingual instruction by the end of the school year. On the other hand, a student in© 2008 Dr. Catherine CollierAll Rights Reserved156

an English-only instructional program, receiving a half hour or less pull out assistance inlanguage transition once or twice a week is clearly not receiving a year’s instruction in a year’stime. Determining the extent of cumulative time may be difficult, but the rule of thumb is thatpull out, limited time assistance, accumulates at about half the rate of longer time, dailyassistance. This may be “Newcomers” classes or classes for students of limited Englishproficiency.4. Degree of language proficiency in the native language or dialect: Using formal or locallanguage tests, observations, informal assessment instruments, and interviews with parents andschool personnel, identify the relative level of proficiency the student has in the native languageor dialect. This may be a recognized linguistic variation of English, distinct for this student’sethnic background and/or traditional community.5. Degree of proficiency in English: Using standardized or local language tests, observations,informal assessment instruments, and interviews with parents and school personnel, identify therelative level of proficiency the student has in “standard” English or the English used in a schooland by the majority of school personnel. This represents the “second language” or secondarycommunication modality to which this student is acculturating.6. Degree of bilingual proficiency: Using formal or local language tests, observations, informalassessment instruments, and/or interviews with parents and school personnel, identify therelative level of balanced proficiency the student has in both English and the student’s nativelanguage or dialects. This may be a recognized linguistic variation of English, distinct for thisstudent’s ethnic background and/or traditional community.7. Ethnicity or national origin: Using information from intake personnel, or intakeinformation, interviews with the student, parents and teachers, identify how the student views hisethnicity or national origin. This gives an indication of his or her cultural awareness and selfidentity.The multiple terms reflect differences in self-identification to some extent, though dueto economy of space, only a few of the many variations of self-reference are included. EastAsian refers to populations and countries around or to the north and east of the Bay of Bengaland West Asian refers to populations and countries around or to the north and west of theArabian Sea. Moreover, this item gives an indication of how prepared the school system is toadapt to the needs of this particular student and how likely it is that the curriculum/system is tobe effective for this student. Research has shown that, in general, public schools and schoolpersonnel are less prepared to work effectively with particular groups of diverse students,particularly when their presence in the schools is infrequent or of “low incidence”. This does notmean an individual teacher, specialist or school is not doing as well as possible, but curricula andteacher training programs show measurable differences in their responsiveness to particularethnic/linguistic populations. This item is a broad indicator of issues that make acculturationmore stressful for particular students in mainstream public school systems.8. Percent in school speaking student’s language/dialect: Using information from district andbuilding level personnel, identify how many of this student’s particular cultural community areenrolled in the school. This is specific to the student, i.e., if the student is Navajo, how manyother Navajo students are in the school. If this is the only Navajo student in the school, theevaluator would note a nearly 0% enrollment even if there are several students from otherAmerican Indian tribes in the school. In other words, counting “Native Alaskans” or “NativeAmericans” is incorrect here; how many students come from this student’s particularcultural/linguistic community is important. This will also give an indication of the degree towhich this student is at risk for stress factors such as alienation and isolation, and possible© 2008 Dr. Catherine CollierAll Rights Reserved157

discrimination by other students. However, the primary reason for this and item number 7 is asan indication of how prepared the school system is to adapt and acculturate to this particularstudent.SummaryThe key learning of this chapter is that students who are in the process of adapting to a newculture/social environment may behave in a manner that is similar to learning disability or otherinhibiting factor. Examples and preliminary strategies are provided. The evaluation of diversestudents for eligibility in special education programs must include the assessment andconsideration of these observable consequences of culture shock that may be confused with thecommonly used criteria for learning or emotional disabilities. Where students are significantlyless acculturated as measured on any of the various tools described in this chapter, assessmentand evaluation personnel must modify their evaluation procedures, and choice of assessmenttools to reflect the level of acculturation. The interpretation of evaluation findings must includediscussion of the impact of the students’ level and rate of acculturation as a factor contributing toall evaluation findings. This will help optimize the allocation of limited resources within theschool and the district, while providing each student with the most appropriate assistance.© 2008 Dr. Catherine CollierAll Rights Reserved158

Article V.3 Assessment of English Language LearnersBy John J. Hoover, Ph.D.An excerpt reprinted by permission from Educating Culturally And Linguistically DiverseStudents: A Professional Development Resource Series, BUENO Center for MulticulturalEducation, University of Colorado, 2002.IntroductionImproving assessment capabilities and identification of students' instructional needs is critical forany learner. The need for identifying appropriate ways to address persistent educationalunderachievement of students from culturally diverse backgrounds has been a recurring themeover the past 15 years (Cummins & Sayers, 1995). The widespread inappropriate uses of thestandardized testing with students from diverse backgrounds is well documented as is theconcern that these types of tests often underestimate students’ academic progress and potential.Additionally, Cummins (1995) wrote that the use of standardized tests with students from diversebackgrounds is inappropriate due to their heavy emphasis on lower-level thinking skills at theexpense of assessing higher-order thinking and creative writing. This emphasizes the need formore authentic assessment, in addition to or instead of standardized testing for students fromculturally diverse backgrounds.It is of paramount importance that exceptional students with language and cultural differences beidentified early and their unique learning needs be addressed as effectively and comprehensivelyas possible. The learning and behavior problems of some students may be due to a differentsociolinguistic and cultural background, their adjustment to a new sociocultural milieu, thepresence of a disability, or a combination of these factors.The learning and behavior problems exhibited by many English Language Learners (ELLs) areoften similar to the problems which elicit referrals for all students: a lack of or seriously deficientacademic achievement; social and classroom behavior which is disruptive to instruction; and/orother problems which are difficult for the teacher to handle in the general classroom setting. Asa result, ELLs are often over-identified, under-identified, or mis-identified for special educationservices (Gonzalez, Brusca-Vega, & Yawkey, 1997).Changes in the traditional assessment process are necessary to determine the most appropriateinstructional needs of ELLs. Problems which appear to indicate a disability may actually be themanifestation of cultural, experiential, or sociolinguistic differences rather than a disability.When assessing learning and behavior problems in ELLs, one must consider these possibilities,as well as the legal requirements in student assessment.Assessment Process and ComponentsAlthough various models for assessing ELLs exist, each contains the four major components ofReferral, Formal Assessment, Staffing, and Placement. While the primary emphasis should be onthe prereferral process, should these efforts not produce the desired results then it may becomenecessary to move into the formal process of assessment. The assessment process discussedincludes suggested adaptations for effectively assessing ELL students (California Department ofEducation, 1988; Ortiz & Garcia, 1988; Ortiz, 1990; De Leon, 1990). Four major elements ofthe assessment process are:© 2008 Dr. Catherine CollierAll Rights Reserved159

1. Learning/Behavior Need is Recognized (Awareness Stage)2. Pre-Referral Intervention in Inclusive Settings(Teacher Assistance Team - TAT)3. Formal Referral and Assessment4. Staffing, Placement (IEP), and MonitoringOne major purpose of formally assessing ELL students is to determine (a) whethercultural/linguistic factors are the major contributors to students apparent learning and behaviorproblems, or (b) whether cultural/linguistic factors and the presence of a disability contribute tostudents' learning and behavior problems. When these variables affect the student's learning, it isessential to assess the student's academic and behavioral functioning. The information obtainedfrom all assessment procedures should be used to develop the individualized educational plan(IEP) documenting appropriate content, strategies, and classroom settings necessary to provideappropriate education. To accomplish this, the assessment process (outlined above) mustgenerate and emphasize specific adaptations needed to develop meaningful instruction for ELLstudents.Cultural and Language DiversityAreas of special concern in the assessment of ELL students include six cultural factors:Language Function (Communicative/Academic), Acculturation, Conceptual Knowledge,Thinking Abilities, Cultural Values/Norms, and Teaching/Learning Styles (O’Malley & Pierce,1996). Knowledge about students in these six areas is critical as they form the foundation foreffective assessment and instruction for ELLs. Addressing these six cultural, language, andcognitive factors will help the practitioner ascertain whether potential learning/behavior needsexhibited by the ELL student are due to cultural factors, some other problem (e.g., limitedprevious schooling) or disability, or a combination of these.Test Taking SkillsAchievement test scores for ELLs frequently reflect their lack of effective test taking skills ratherthan content knowledge or higher level thinking abilities. An understanding of specific testtaking skills of ELLs must be determined in order to best identify test taking strategies thatshould be taught to students in school and to educators of ELLs in their ProfessionalDevelopment. The following test-taking skills are essential to effectively completing variousforms of assessment (Hoover & Patton, 1995).TEST PREPARATION SKILLS1. Know major topics to be covered on test2. Know the type of test to be taken (e.g., multiple choice, essay)3. Anticipate potential test questions4. Make a list of main topics to be covered on the test and indicate how well you know eachtopic (e.g., Finding the main idea, Sequencing, Writing a complete sentence)5. Practice completing sample items to those on the testTEST COMPLETION SKILLS -OBJECTIVE TESTS1. Read each question carefully2. Review all test questions prior to answering questions3. Respond to more difficult items last4. Know whether it is better to leave an answer blank or guess5. Make certain all responses are accurately recorded6. Narrow possible correct answers7. Logically eliminate obviously wrong answers© 2008 Dr. Catherine CollierAll Rights Reserved160

8. Change answers only for a very good reason9. Review your test prior to turning it in10. Correct previous test taking errorsTEST COMPLETION SKILLS -ESSAY TESTS1. Content in answer directly relates to the question2. The facts in each response are clear and organized3. Proper grammar, punctuation, and sentence structure are used4. Responses are proofread prior to turning in the test5. More difficult questions are answered last6. First sentence in response directly responds to the question7. Previous test taking errors are corrected8. Questions worth the most points are identified and given appropriate amount of timeTEST REVIEW SKILLS1. Each question was clearly understood prior to answering2. Reasons for incorrect answers are explored3. Carelessness in answering the items is considered4. Amount of time preparing for the test was sufficient5. Test preparation items were followed6. Review correct responses7. Identify test-taking skills used during the testThe extent to which an ELL possesses these test-taking sills directly influences and affectsassessment results.Prereferral InterventionThe term "prereferral" refers to the time period following an indication by a teacher or otherconcerned person that the student has some kind of learning, communication, or behaviorproblem, but before formal referral for assessment occurs. Prereferral intervention serves as acritical educational step in minimizing biases and inequalities in assessing ELLs (Gonzalez,Bursca-Vega, Yawkey, 1997). It is estimated that 75-90 percent of ELL students referred forspecial education eventually are placed in special education (Stefanakis, 1998). Characteristicssuch as physical appearance, levels of acculturation, and/or differences in language or culturalbackground often disproportionately influence placement and programming decisions, ratherthan clarifying specific needs or disabilities.A key element of prereferral intervention is the implementation of curricular interventions, priorto formal referral, which attempt to modify the instructional environment so that the students areable to function more effectively. Prereferral intervention should accomplish at least two goals:1) assist to clarify and better understand the educational need or concern; and, 2) assist to solvesome or all of the suspected problems through needed adaptations or modifications to teachingand learning. Within prereferral, the teacher requests assistance and begins gatheringinformation about the student’s cultural, language, and behavioral needs. This includesaddressing the six cultural, language, and cognitive factors previously discussed.The requested assistance may be provided by a fellow teacher, or by other specialists, such as thebilingual education specialist, the speech pathologist, or the resource or special educationteacher. Ortiz (1992) wrote that these problem-solving teams or Teacher Assistance Teams(TAT) work well for prereferral intervention of ELLs. If current language proficiencyinformation is available and indicates that the student has limited English proficiency, the teacher© 2008 Dr. Catherine CollierAll Rights Reserved161

should then modify the use of various assessment and interventions techniques to accommodatethe student’s language and cultural background.The educator should implement a series of interventions to facilitate the collection of informationand to try out possible resolutions to the student’s learning and behavior needs. Duringprereferral of an ELL, the educator should consider several relevant variables including: (a) timefor adjustment, (b) familiarity with the school system and language, and (c) cultural differences.After attempting various prereferral interventions and documenting their results, the TeacherAssistance Team may decide that a formal comprehensive evaluation is necessary. Once a formalreferral for assessment is made, the legal constraints of IDEA (1997) regarding staffing andformal assessment must be applied.Authentic AssessmentOver the past three decades two themes have emerged that highlight many issues in the area ofassessment for English Language Learners (ELL): 1) accountability; and, 2) underachievementin education (Cummins & Sayers, 1995). The problems associated with use of formal,standardized testing with ELLs are well documented (Baca & Cervantes, 1998; Cummins, 1995;Gonzalez, Brusca-Vega, & Yawkey, 1997). Four specific types of authentic assessment areappropriate for use with ELLs (O’Malley & Pierce, 1996):1. Curriculum-Based2. Performance-Based3. Portfolio4. Self-AssessmentWhen used in conjunction with prereferral interventions as well as selected standardizedassessment techniques, authentic assessment provides more relevant and meaningful informationconcerning the educational needs of ELLs.Assessment TechniquesA variety of assessment techniques exist and may effectively be used in the assessment of ELLs.HOW these techniques are applied and how results are interpreted will determine theireffectiveness with ELLs. When used in the overall assessment process, including prereferralinterventions, these alternative and authentic assessment techniques provide valuable andmeaningful information about the student. They include:1. Observations2. Interviews3. Rating Scales/Surveys4. Review of Records5. Work Sample Analysis6. Journaling7. Task Analysis8. Language Samples9. Rubrics10. Analytic TeachingEach of these assessment techniques should be considered in the comprehensive and effectiveassessment process for ELLs, with results being compared and corroborated through the varioustechniques.SummaryThe assessment of ELL students has been a subject of legislation for many years (IDEA, 1977;BEA, 1994; Dyrcia S. et al. V. Board of Education of the City of New York, 1979; Larry P. v© 2008 Dr. Catherine CollierAll Rights Reserved162

Riles, 1979; Jose P. v Ambac, 1983). This legislation has led to several legal guidelines that mustbe followed when assessing ELLs. If not completed appropriately, assessment, and its results,may be irrelevant to programming and may merely confirm the initial impressions of theclassroom teacher. Information about cultural, language, and cognitive factors and theircontributions to students’ learning and behavior needs must be obtained prior to referral forformal assessment. Appropriate placement and programming decisions should be made withknowledge of the extent to which cultural and language factors contribute to the learning andbehavior needs of an ELL.© 2008 Dr. Catherine CollierAll Rights Reserved163

Article V.4 Alternative Assessment for TransitionalReadersBy Carol Beaumont, Julia Scherba de Valenzuela, Elise TrumbullBilingual Research Journal, 26: 2 Summer 2002 214Equitable assessment for English language learners remains one ofthe greatest challenges of educational reform. The complexinteraction of language, culture, educational background, andopportunities to learn make equitable, large-scale assessmentsparticularly difficult to achieve (Figueroa, 1990; García & Pearson,1994; Solano-Flores & Nelson-Barber, 2001; Ulibarri, 1985). Often,English language learners (ELLs) are assessed through the samemechanisms as their English-only counterparts. This practicecompromises the accuracy of results and may lead to inappropriateinterpretations about English language learners’ proficiency inreading and academic content areas (García, 1991; Miramontes,Nadeau & Commins 1997; Valdés & Figueroa, 1994).The purpose of this paper is to describe student performance on a specially designed assessmentof reading comprehension created for native Spanish-speaking students in their first year ofEnglish instruction. This study was guided by the following questions: (a) How do scoring andstudent evaluation practices in this assessment account for linguistic characteristics of first-yearEnglish language learners? (b) What evidence exists that reflects first-year English languagelearners’ ability to process English texts at an inferential level? (c) How can this transitionassessment inform curriculum and instruction for these students?Assessment Reform and English Language LearnersIn the decade-long debate on school reform, some educators favor new methods of assessment toremedy persistent problems in the American schooling. They have suggested that complex,cognitively demanding assessments can drive improvements in curriculum and instruction (e.g.,Darling-Hammond, 1994; Mitchell, 1992; O’Day & Smith, 1993). They have also pointed outthe overrepresentation of ELLs and other students from non-dominant cultures among poorperformers on standardized tests(Figueroa, 1990; Valdés & Figueroa, 1994; Williams, 1996) andthe consequent reinforcement and extension of social and educational inequalities (Darling-Hammond, 1994, Durán, 1989; Oakes, 1985).ELLs’ lack of full English proficiency is an obvious reason for their poor performance onstandardized tests (LaCelle-Peterson & Rivera, 1994; Cummins,1989; Genesee & Hamayan,1994), but it reflects a limited understanding of the factors contributing to a student’sperformance on an assessment or standardized test. To begin to grasp the relationship betweenEnglish proficiency and test performance requires a deep understanding of the developmentalnature of second language acquisition, the difference between surface fluency and the ability toparticipate in content area instruction in English, the time required to achieve academic fluency,the kinds of language conventions used on typical assessments and tests, and the language skillsnecessary for success on standardized tests. Because the length of time required for full English© 2008 Dr. Catherine CollierAll Rights Reserved164

proficiency is often underestimated (Collier; 1988;Cummins, 1989; Valdés, 1998), and studentsmay appear to be more proficient with English than they are, they may be expected to take testsin English long before they are fully proficient with the kind of academic language needed toperform well. All of these factors should be considered in designing appropriate assessment andinstruction.Standardization itself rules out any contextualization of assessment, meaning that linguisticdifferences among students cannot be accounted for adequately. The practice of assessingEnglish language learners with the same mechanisms as their English-only counterparts mayseriously compromise the validity of results and lead to misleading interpretations and unfairdecisions affecting their futures (August & Hakuta, 1997; García, 1991;LaCelle-Peterson &Rivera, 1994; Miramontes, Nadeau & Commins, 1997;Valdés & Figueroa, 1994).English-language assessment prompts that make extensive use of complex or idiomatic languagepenalize English language learners who may access important concepts in their first language,but not yet in English, or they may access them more slowly in English (Abedi, 2001; Abedi,Leon, & Mirocha,2001, Figueroa & García, 1994, García, 1991, Heubert & Hauser,1999).Misinterpretations of the directions or text of an assessment task can lead to flawedconceptualization of the problem to be solved and consequent failure to devise a correct solution(Durán, 1989). In such cases, teachers or others who grade or score an English language learner’sperformance may falsely underestimate that student’s level of understanding or skill.A move to standards-based performance assessments and other forms of “authentic” assessment(Valencia, Hiebert, & Afflerbach, 1994; Wiggins, 1993)does not ensure assessment validity forELLs. These kinds of assessments are subject to the same sources of error, particularly giventheir increased language demand in comparison to multiple choice or short answer tests(August& Hakuta, 1997; Shepard, 1993). Farr and Trumbull (1997) point out that, “good instruction andassessment should look different in different environments, depending on the students served”(p. 2).Assessments designed for native English speakers will simply not meet standards of validity forEnglish language learners. According to the Standards for Educational and PsychologicalTesting (American Educational Research Association, American Psychological Association, &National Council on Measurement in Education, 1999):For all test takers, any test that employslanguage is, in part, a measure of their language skills. . . . Language dominance is notnecessarilyan indicator of language competence in taking a test, and some accommodation may benecessary even when administering the test in the more familiar language. Therefore, it isimportant to consider language background in developing, selecting, and administering tests andin interpreting test performance. (p. 91)Specific research studies on the validity of performance-based assessments for ELLs havefocused primarily on accommodations (e.g., increasing the time allotted for completion, allowingthe use of dictionaries, or modifying the language of test prompts) rather than on major changesin the instruments themselves (Abedi, 1997; Abedi, Lord, Hofstetter, & Baker,2000; Kiplinger,Haug, & Abedi, 2000; Kopriva, 1997; Olson & Goldstein,1997; Sweet, 1997). Some have madeefforts to make fairer the scoring of ELLs’ responses (Wong Fillmore & Lara, 1996). However,little research literature exists on performance assessments for English language learners.© 2008 Dr. Catherine CollierAll Rights Reserved165

Second Language Learning and Reading ComprehensionThe theoretical foundation for much of the literature on second language reading is the constructof linguistic interdependence based largely on Cummins’ (1981) construct of a CommonUnderlying Proficiency (CUP),whereby skills transfer from one language to another. The theoryof linguistic interdependence indicates that well-established skills in the first language transfer tothe second language. The theory of linguistic interdependence incorporates both content andprocesses in reading (e.g., subject matter knowledge, conceptual knowledge, higher-orderthinking skills and reading strategies). Support for the transfer phenomenon in studiesinvestigating reading behavior has been widespread (Fitzgerald, 1995; Krashen, 1996).Evidence cited for the transfer hypothesis centers around (a) use of similar strategies to read inboth languages (e.g., Pritchard, 1990); (b) predictability of level of reading proficiency in thesecond language on the basis of reading proficiency in the first (e.g., Calero-Breckheimer &Goetz, 1993; Carrell, 1991;Droop & Verhoeven, 1998; García, 1998; Geva, Shany, & Himel,1992;Saville-Troike, 1984); and (c) an apparently closer correlation between readingproficiencies in the two languages than between reading and oral proficiency in the secondlanguage (Geva, Wade-Woolley, & Shany, 1993;Saville-Troike, 1984).Some research has focused on specific language skills and the degree to which they may transferfrom reading in one language to reading in another. Durgunoglu, Nagy, and Hancin-Bhatt (1993)showed that young children’s phonological awareness and word recognition skills in Spanishwere predictive of phonological awareness and word recognition in English. However,Heubertand Hauser (1999) conclude that transfer is not automatic: “It occurs only whenconditions for the emergence of the analogous second-language skills exist, and it can be aidedby explicit support for the process of transfer”(p. 224). Others caution that when cognitive andlinguistic factors are taken into account to the exclusion of macro social factors (i.e., status oflanguages and relations among groups of people), evidence of variance in literacy success isoverlooked (Gee, 1999).Information on text processing among bilinguals suggests that several factors may createdifficulties for students reading in their second language:(a) the degree to which readingcomprehension strategies are well developed in the first language (Jimenez, García, & Pearson,1996; Langer, Bartolome, Vasquez, & Lucas, 1990; Thonis, 1983; Westby, 1989; Laberge &Samuels,1974; Jimenez, 1997; Bialystok, 1991); (b) characteristic features of the Englishlanguage and the specifics of English text structures (Kucer & Silva, 1995);(c) inaccurate orincomplete background knowledge (Jimenez, García, &Pearson, 1996; Kucer & Silva, 1995;Flood & Menyuk, 1983; Vellutino, Scanlon,& Tanzman, 1990; García, 1991; Westby, 1989;Bartolome, 1994; Edelsky, 1983);(d) vocabulary limitations (Wong Fillmore, 1989; Gibbons,1991; Nagy, García, Durgunogulu, & Hacin-Bhatt, 1993; García, 1991; Baker, 2000); and(e)speed of processing (Fitzgerald, 1995; Jimenez, García, & Pearson, 1996).When reading comprehension is assessed through writing, these difficulties are compounded(Gutiérrez, Baquedano-López, Alvarez, & Chiu,1999; Reyes, 1992). As writing developmentmay tend to lag somewhat behind reading development in English language learners (DeAvila,1997), assessment of reading through writing could underestimate reading proficiency. Also,time limits fail to take into account the slower rate at which English learners process texts(Fichtner, Peitzman, & Sasser, 1994; Peitzman, personal communication, April 7, 2000; Carrell,1991; August & Hakuta, 1997; García& Pearson, 1991; Mercado & Romero, 1993; Pérez &Torres-Guzmán, 1996).© 2008 Dr. Catherine CollierAll Rights Reserved166

Finally, when second language readers have the opportunity to use their dominant language todiscuss texts written in their second language, they reveal deeper understandings than they wereable to demonstrate when required to talk about them in English (Díaz, Moll, & Mehan, 1986).Itseems reasonable to assume that these factors may contribute to English language learners’ poorperformance on alternative assessments, unless they have been developed with English languagelearners in mind. These issues in the assessment of bilingual readers and particularly transitionalstudents, point to the need for continued exploration of effective assessment instruments that takeinto account the unique profile of bilingual readers.Background to the StudyThis study took place in an ethnically diverse urban school district in the West San FranciscoBay. Of the 8,500 students in the district, 49% were minority and 95% of those were Mexican-American. The district’s bilingual program had been in place for many years and followed themulti-year program design based on California’s “State Program for Students of Limited EnglishProficiency,” where students receive primary language instruction in content areas for four years,in addition to daily English as a second language instruction. Students begin English content areainstruction in science and/or math in fourth grade and language arts instruction in English in fifthgrade. Students who enter after kindergarten are placed in primary language or English contentarea instruction based on their school background and language proficiency test scores.During the study the district was in the fourth year of a major standards-based reform effort inlanguage arts, mathematics, and social studies,K-8. The district had changed from astandardized, norm-referenced test to a Performance-Based Assessment (PBA) given at second,fifth, and seventh grades. In general, English learners had scored poorly in the first few years ofthe PBA. Many of the district’s bilingual teachers had been involved in scoring the districtassessment and believed the English learners’ poor performance was indicative of aninappropriate instrument rather than a reflection of the students’ abilities. They were particularlyconcerned that the reading level, vocabulary, content, and thematic focus of the readingselections were inappropriate for these students. After trying several alternative approaches andaccommodations, teachers in the bilingual department proposed the development of analternative assessment specifically designed for the transitional bilingual students. This paperdiscusses the results of the first administration of this teacher-developed alternative assessment.MethodParticipantsThe primary participants in the study were 89 fifth-grade students in transition, completing theirfirst year of English language arts instruction. All were from lower socio-economic status (SES),Mexican-American homes where Spanish was the primary language. While the criterion fortransitioning to English instruction was grade level reading in Spanish, in practice many studentswere transitioned before that level of proficiency was reached. The test development workgroupconsisted of 10 teachers, the director of bilingual education, and two consultants from theWestEd Regional Educational Laboratory. The teachers were either state certified bilingualteachers or Specially Designed Academic Instruction in English (SDAIE) teachers who hadparticipated in specialized training in effective techniques for teaching transitional students. Thedirector of bilingual education and the regional laboratory consultants all had extensiveinstructional experience with English language learners and alternative assessment and advanceddegrees in linguistics and education.© 2008 Dr. Catherine CollierAll Rights Reserved167

Data SourcesThe data consisted of 89 performance-based assessments completed by the transitional studentsand field notes and audiotapes from test development meetings, scoring sessions, and interviewswith the workgroup members. The summary and interpretation of events were triangulated withinterviews with the project participants upon completion of the analysis. Project participants andone of the consultants read a draft of this manuscript and provided clarifications to thebackground and test development sections of this paper and feedback on the analysis of the tests.The Transitional Performance-Based Assessment (TPBA)The assessment was developed over an 18-month period during which workgroup members metto discuss relevant language issues; look for appropriate reading materials; create, field test, andevaluate test items; complete a final assessment measure; and test, score, and interpret studentperformance. The transitional assessment instrument, which was framed in second languageacquisition theory and Vygotsky’s construct of assisted performance in assessments (1978),paralleled the district-wide Performance-Based Assessment (PBA), (as required by the districtassessment department):(a) students were required to read and respond to two English texts onrelated topics, one fiction and one nonfiction; (b) test items required constructed writtenresponses targeting the fifth-grade language arts standards; and(c) the texts were given in Englishto evaluate a transitional program for students participating in English-only classroominstruction.The workgroup considered the following most important: (a) adjust the reading level to make thetexts accessible to transitional readers; (b) test more than one standard per item, enablingstudents to build on their knowledge across items; (c) design holistic scoring procedures tocomplement the integrated structure of the assessment; and (d) use of bilingual teachers forscoring.In adjusting the reading level of the texts, the workgroup members considered three factors. Theworkgroup chose shorter texts than those used for the district PBA to allow bilinguals enoughprocessing time. The workgroup also believed that the students could not be assessed fairly usingfifth grade reading material since they had had only one year of English reading instruction.They selected texts they believed were age-appropriate in content but were less dense andvocabulary-dependent and culturally more familiar than the texts used in the district PBA. Aclassroom teacher administered the test in two periods: one for pre-reading activities to activatebackground knowledge and set the purpose for reading, and the other for administering the testitself. All classroom teachers used the same pre-reading activities, as did the English onlyteachers in administering the PBA.Since many items of the TPBA measured multiple standards, the group chose holistic scoring toevaluate responses. In addition to changes in the content and structure of the test, which provideda kind of built-in scaffolding, other features of the test such as graphics, formatting, and preassessmentactivities provided for assisted performance consistent with fair assessment practicefor English language learners. Strategies included development of necessary vocabulary,activation of background knowledge, and providing the opportunity to formulate ideas orallybefore writing.District bilingual teachers scored the assessment during a two-day period approximately twoweeks after it was given. Consultants who had read all of the tests developed a rubric and chosenanchor papers from student work, facilitated scorer calibration. During the calibration processteachers read the anchor papers, assigned a score based on the rubric, and recorded evidence© 2008 Dr. Catherine CollierAll Rights Reserved168

for their decisions, noting issues such as language or test formatting for later debriefing. Duringthe subsequent discussion, teachers explained their rationale for assigning certain scores, anddisagreements were resolved by reference to the rubric and advisement with the consultants.After calibration, pairs of teachers scored each test independently, checked their scores with eachother, discussed discrepancies, and reconciled their scores. Differences that could not bereconciled (25% of the cases) were refereed by one of the consultants.Data AnalysisAll 89 tests were read several times, and 5 of the 11 test items were chosen for in-depth analysisbecause they required inferential thinking beyond basic identification and summarization tasks,and allowed for more detailed analysis of reading comprehension. Each student’s responses tothe five items were then re-read multiple times without reference to pre-determined analyticcategories. As patterns began to emerge from the readings, tentative analytic categories wereidentified. The tests were then read again to see how successfully each category captured aspectsof student performance. They were then coded according to the evaluation scheme thatdeveloped. As examples of various response patterns were logged and preliminary interpretationswere made, the categories were refined and examples re-examined for consistency.Seventy-four percent of the students received holistic scores in the 2–3range. According to therubric, this score range reflects a processing of English texts primarily at the literal level. Levels2 and 3 are described as “limited” and “adequate,” respectively. However, because the tests werescored holistically, many individual examples of more abstract or inferential processing thatmight provide more specific information about the nature of transitional reading comprehensionwere not reflected in the score distribution. To further investigate the validity of this measure inassessing transitional students’ reading comprehension skills, the holistic scoring wassupplemented by more in-depth item analysis by this paper’s authors. This analysis wasundertaken to gain more specific information about the conditions under which higher level,inferential responses were elicited.The analytic categories included phonological, syntactic, and semantic aspects of language aswell as deeper levels of text processing regarding character and author motivation, text themes,and ability to assume other points of view (Westby, 1989). Although, originally, surface featuresor “conventions” of the written responses were not scored, for the purposes of this study they arepart of the analysis here to amplify scorer response.Results and DiscussionAnalysis of the students’ performance revealed four significant findings:(a) Characteristics ofsecond language learning in the areas of phonology, syntax, and semantics were present tovarying degrees in l00% of the students’ responses; (b) specific knowledge of the linguisticcharacteristics of transitional students was necessary for effective interpretation and accuratescoring of student responses; (c) the transitional students demonstrated the ability to comprehendand interpret English texts at inferential levels when assessed on the transitional assessmentmeasure; and (d) the transitional performance-based assessment results suggested to teachersmany possible changes they could make in their instruction to meet the literacy needs oftransitional students.Finding 1: Second Language Learning CharacteristicsIn the first level of analysis, phonological, semantic, and syntactic features of the students’responses were analyzed to determine the role they played in the students’ ability to express© 2008 Dr. Catherine CollierAll Rights Reserved169

meaning and the scorer’s ability to interpret meaning. Characteristic transitional languagefeatures such as phonological and syntactical evidence of Spanish language influence, semanticconfusion, limited vocabulary, and limited variety in sentence structure appeared as expected.Two of these features, phonology and semantics, are discussed below.Phonological featuresPhonological indicators of the influence of Spanish appeared frequently. Spelling patterns suchas writing “starbing” for starving, “kip” for keep,” slipping” for sleeping, “mit” for meet, “veri”for very, “way” for why, “many” for money, “nating” for nothing, and “so posto” for supposedto, reflect the orthographic influence of Spanish, particularly in the vowel substitutions.Other spelling patterns such as promus/promise, humpbell/humble, and concidaret/consideratereflect the students’ attempts to sound out words in English using a one-to-one match betweensound and symbol, a strategy that works well in Spanish but causes problems when applied toEnglish, given the inconsistency of English orthography. Not only the inconsistency of Englishorthography but also the variations in how similar sounds are represented between the twolanguages pose problems for Spanish readers. For example, the /h/ sound in English can berepresented in Spanish by the letters “g,” when followed by “e” and “i,” and “j.”The following examples, taken from responses to several items, illustrate these phonologicalpatterns as they occurred in connected text:‣ “He geib ibriting hi had hi dident kip nating” [He gave everything he had. He didn’t keepnothing];‣ “I help some homeless people the was slipping in thee strits” [I help some homeless peoplethey was sleeping in the streets.];‣ “Yo yahto knoe way I dont gib many bicas wen I diden hab many they !deden Help my!”[You have to know why I don’t give money because when I didn’t have money they didn’t helpme!];‣ “!Stil you so posto gib many the piople in this vilig ar yor famali!” [Still, you supposed togive money the people in this village are your family!].When numerous errors such as these occurred within connected text, interpretation was ofteninitially difficult for the scorers. However, when scorers read the papers aloud and could hearthe Spanish language influence as reflected in the students’ approximations of English words, theeffect of phonological variations on meaning was reduced. This supports the earlier finding thathaving knowledgeable, bilingual scorers was essential to equitable assessment of students’reading/writing abilities.SemanticsAt the semantic level, vocabulary limitations and confusions appeared in three commonly seenpatterns: (a) the use of phrases in lieu of a single adjective;(b) the frequent use of nonspecificvocabulary such as “nice,” “good,” and “bad”; and (c) incorrect use of word forms.Phrase substitutions appeared primarily in responses to item 1 of one reading passage and item 2of another, which asked for descriptions of character traits. In describing the main characters, forexample, phrases such as “do not do good to beggars” and “not nice to poor people” appeared inplace of single adjectives such as “mean” or “stingy” and “gives things to homeless” or “helpsthem live” instead of adjectives such as “generous” or “helpful.”Interestingly, these phrases often suggested a higher level of text comprehension than did singleadjectives because they indicated the students had not simply searched the text for words to fillin the blank but rather understood what the characters’ actions conveyed.Finding 2: The Role of Bilingual Scorers© 2008 Dr. Catherine CollierAll Rights Reserved170

As mentioned above, when the transition students took the district PBA, the English-only scorershad difficulty scoring the tests because of their unfamiliarity with transition students’ writing. Inthis case the bilingual scorers were able to interpret unconventional responses based on theirknowledge of the second language acquisition process. In their evaluation of test responses, theywere able to consider how Spanish influenced a student’s understanding and use of Englishorthography, syntax, and vocabulary. Because the test assessed reading comprehension and notwriting, it was critical that scorers were able to see beyond problems in form to the underlyingcontent students were communicating. Although we did not have a control group of English-onlyscorers, we believe the examples below show the importance of bilingual scorers for accuratetest interpretation.Certainly, teachers who do not know the phonological and orthographic rules of Spanish cannotbe expected to infer children’s intentions in choosing certain spelling patterns, and hence, thetarget words they may be aiming for(Fashola, Drum, Mayer, & Kang, 1996). Nor can they beexpected to disentangle syntax that makes sense from the perspective of another language withwhich they are unfamiliar.As demonstrated above, the influence of Spanish was evident at several levels of language formand content. It was at the semantic level, however, that translations from Spanish to Englishcould have caused potential misinterpretation of student responses. For example, one studentwrote, “I like wasting money on poor people.” As several bilingual scorers recognized, thisvocabulary usage suggests that the student translated his/her response from the Spanish“gastar,” which can mean either to spend or to waste. The student almost surely meant to say, “Ilike spending money on poor people,” which would be more in keeping with the intendedresponse to the question.Another example of the semantic influence of Spanish in this item occurred when a studentbegan a dialogue by having one character try to persuade another to give some money to thepoor. After getting nowhere trying to convince her to be more generous, the student writes thatthe first character says, “I think no one can change you. Of your form your act.”Again, an English-only teacher scoring the test might wonder what “of your form your act”really means. A bilingual scorer, however, might realize the student is most likely translating thenotion of “manera de ser” or “form of being,” which refers to character traits, not transitorybehavior. From this perspective, it is clear that the student was able to take two points of viewand sustain them through a dialogue, demonstrating a well-developed sense of character.On another item that asked students to describe character traits, one student argued that the poorbeggar woman “should be shy for that.” This word usage suggests that the student wastranslating from the Spanish word vergüenza, which means to be embarrassed or ashamed aswell as to be shy. The student most likely wanted to say “She should be ashamed of herself forbegging.” These answers and many others like them could have been misinterpreted if thescorers had not been aware of the Spanish equivalents that were implied by the students’answers.There was, of course, a relationship between form and meaning and many of the discussionsamong scorers revolved around how to score consistently and fairly for meaning rather thanform. The scorers recognized that students who had more control over form were able to expresstheir meaning more clearly, especially in items that required extended responses. In looking attext processing patterns suggested by the nature of the students’ answers, the researchers’ itemanalysis built on the effort to interpret meaning that the scorers demonstrated.Finding 3: Comprehension of English Texts© 2008 Dr. Catherine CollierAll Rights Reserved171

Looking beyond the surface features of the students’ written responses, we next examined textprocessing patterns suggested by the nature of their answers. The transitional assessmentaddressed a number of reading comprehension skills including summarizing, identifying mainidea, and locating information in text. The items chosen for in-depth analysis, however, werethose that required more inferential responses, such as identifying character traits, notingcharacter development, determining the author’s purpose, and contrasting points of view. Asmentioned earlier, the distribution of holistic scores after the first test administration reflected thetransitional students’ processing of English text primarily at the literal level, according to thedistrict rubric. One might expect this for first year transition students, given their still-developingskills in decoding English. However, item analysis gave more specific information about whatstudents could do, what they had difficulty doing, and the conditions under which students wereable to go beyond basic, literal answers to expand on text and make connections, generalizations,and extensions.Character traitsThe early items in the assessment required students to identify traits of the main characters and toprovide text-based justifications for their choice of adjectives. By requiring textual support, theitem gave an indication of whether the adjectives used were meaningful to the students or weresimply copied from the text in a “match to sample” strategy (this strategy of lifting words fromthe text that match those in the comprehension questions has been identified as a commonapproach to comprehension test questions used by second language readers (Anderson & Roit,1996; Cole & Griffin, 1983). Three frequently occurring response patterns to this item wereidentified. In the first type of response, difficulty with text processing was evident from themismatch of adjectives and textual support. In these examples, students may have been able toidentify adjectives by lifting key words from the text but, when asked to justify their answers,were unable to do so. For example, the students might identify “stingy” as an appropriatedescription of a rich character in the story, and then fail to use textual excerpts that demonstratethat they understood the meaning of stingy and what actions proved the character deserved thatlabel.The second type of response involved the use of nonspecific adjectives coupled with appropriatetext support that would earn a 3 (adequate understanding) on the rubric. This could be the resultof difficulty with text in-depth processing, suggesting that students got only a general idea of thecharacter’s traits. However, as suggested above, the use of vague descriptors such as “nice,”“bad,” or “good” may also be a reflection of limited vocabulary rather than limited textprocessing. In describing one character, students used the vague descriptor “nice” in 20% of theresponses. However, in 90%of those cases the students provided appropriate text support thatsuggests that they did understand the basic description of the character and what evidence wasavailable to support that description. These three responses are illustrative examples of this typeof response: “Nice”: (a) he was nice because he made a home fur them; (b) because he helped thehomeless by giving them food; (c) when the señora gave her [referring to the beggar] the onionskins she humpbull [humbly]said muchas gracias.For the first two definitions, more specific adjectives that would have been appropriate include“helpful,” “generous,” or “kind,” but “nice” is not inappropriate, and the evidence given isaccurate. In the third definition, “nice” is used in the sense of “polite” and, again, the evidence isappropriate. The third definition gave the strongest suggestion of accurate text processing byindicating that students had understood the meaning of the adjectives they used and chose thesedescriptors based on the text. The examples below demonstrate a clear match between specific,© 2008 Dr. Catherine CollierAll Rights Reserved172

ather than generic adjectives and textual support, and would have received a 4 or 5 on therubric—“clear” or “thorough” understanding of the text. Sixty-four percent of the studentsresponded with this type of response:‣ Greedy: Quickly the rich señora threw a cloth over her plate to hide the food piled.‣ Selfish: There were plenty of meat lots of corn and tortiallas. But the rich señora did not wantto waste such fine food on a beggar.‣ Stingy: Every day poor people came to her house to beg for food and every day she sent themaway with nothing.Some differences were noted in the responses to the character trait items for two characters. Inone question students were asked to copy passages from the text that supported their adjectives,but for the other they were asked to use their own words. The match-to-sample strategy forfinding textual support would not be as effective in the latter case. Perhaps as a result of beingfreed from copying exact information from the text, students used 18different adjectives todescribe the character in the second question, including sophisticated adjectives such asconvincing, successful, powerful, intelligent, active, and considerate, but only six differentadjectives to describe the character in the first question (stingy, bad, mean, selfish, greedy, rude).This stronger use of vocabulary in the second passage may also be a reflection of the text, whichprovided much more detail about the character than the fictional piece did about its two maincharacters. This level of analysis is important in the assessment of language learners where it isoften difficult to ascertain which factors plays a role in the students’ responses.Character developmentSubsequent items required the students to expand on this basic sense of character to make morecomplex inferences. Asking students to write what the character is thinking about the things shedid in her life assessed understanding of character development. Asking students to imagine acharacter’s final thoughts as she drops into a sea of darkness required them to attribute feelings,plans, motivations, and thoughts to the character. Two general types of responses wereidentified. In the first type, students inferred that the character regretted her actions because oftheir impact on her. In the second type, students extended their understanding a bit by suggestingthat the character felt remorse for hurting other people. Both types of responses demonstrateinference and would have received scores of 3–5 on the rubric.Type 1 examples (regret):1. I tink tat she was tinking about tat she did not give food to the poor people and that she slamdthe door to the old laydy, and god puneshd her becues she did not give food to the poor peopleand she was stingy2. If I have gave that old woman more that old onion, now I should be in heaven.In these examples the students use details from the story to frame their answers, a response thatwould be considered a type of restatement of the text and thus earn a score of 3. Although thistype of response does reflect inference, it is a very basic level given the obvious result of heractions.Students who went beyond this level ascribed more complex motivations to the Rich Señora andby so doing, indicated their ability to go beyond literal thinking to, “understand connectionsamong story elements and overall meaning,” as level 4 of the rubric described, or “makeplausible interpretations or generalizations,” as characterized by the following examples.Type 2 examples (remorse):© 2008 Dr. Catherine CollierAll Rights Reserved173

1. I had been nicer with the beggar womans. I had think of it before. But why was I mean. Poorwomans I hope somebody saves me. So I can help beggar womas. Specialy the old one. Now amgoing to help peeple. But first some one help me.2. She was thinking that she was greedy to the poor people and mean and she was going to helpthe poor pleople and she was sorry.3. I was very selfish I had never given those dry skins I should given meet, corn and tortillas tothe old women, I should given food the poor peple ho came for food. I’m sorry, I’m sorry, sorry,sorry, sorry, I’m sorry!4. Oh I should have never rejected beggars I-I-I wish I could live again and fixed my errors. Andnever give poor people only a slice of old onion. I wish I could just stay poor put have givenpoor people food before getting poor.Author’s point of viewIn another item, students are asked to step out of the story and imagine the author’s point of viewand purpose in response to the question, “Write why you think the author made this happen tothe Rich Señora.” The analysis of responses to this item revealed that 93% of the students had asense of the author’s purpose and were able to imagine the author’s point of view,71% atprimarily a literal level, and 22% at a more inferential level. Two general types of responseswere identified. In the first type of response students demonstrated a “basic understanding ofstory elements” (score of 2–3 on the rubric) and assumed that the author was telling a tale ofretribution for misdeeds as shown in the following examples:1. Because when the beggar woman came to the rich woman’s house the rich woman threw acloth over her plate and she didn’t wanted to give food to the people because she didn’t want towasted food for a beggar.2. I think the author made this happend to her because she was berry stingy and mean and shewas berry rich she had enouhgf food to give to poor and for her.These examples illustrate some of the differences in students’ responses and suggest differentlevels of text processing. Although both are essentially restatements of obvious information inthe text—literal interpretations—in the first example, the student used details from the story todescribe the character’s behavior. In the second example the student generalized from thebehavior to character traits—“she was very stingy and mean.” Both are examples of building onearlier test items where students are asked to describe the character and give examples of actionsthat prove the character was as the student described.The next level of response illustrates the students’ ability to go beyond a simple text theme ofretribution and restatement of the text to imagine the author’s interaction with the reader,suggesting that the author wanted to convey a moral or lesson.Type 2 examples:1. I think the author made this happen to the rich señora because it was a lesson to everybody toknow that there are people less fortunate then you so don’t be so gretty because you never knowwhat will happen next.2. I think that the author made this happen is for the people to see that if you want to recibesomething you also have to share. I also think that happen for people to see that been selffish isnot good.Contrasting characters’ points of viewIn the comparison portion of the assessment, the students are asked to construct a dialoguebetween the main characters of the fiction and nonfiction pieces. The task requires students totake the point of view of both characters alternately and sustain a dialogue that reflects each of© 2008 Dr. Catherine CollierAll Rights Reserved174

their imagined points of view. There were three general types of responses to this item. The firsttype of response demonstrated minimal text comprehension or inappropriate departure from thetext. The second and third types of responses were of an inferential nature, giving “evidence thatthe reader is beginning to interpret text by making connections between details and largermeanings” (score 4)or “make plausible interpretations or generalizations” (score 5). Eighty-fourpercent of the responses to this item were at an inferential level. The examples below illustratethese two higher levels of text processing.Type 1 and 2 examples:Example 1Trevor: Do you know how much te people need that food? You are rich, yo’u should at least givea little meat the people are staving. Please give them something.Rich Señora: Yeah Yeah I’m not going to waste my food let them work so they could buy whatthey need and I’m not giving nothing to anybody.Trevor: You have plenty of food they dont have nothing dont you have any sipithy for them.Rich Señora: Ok, ok, I’ll just give some little of meat thats all.Example 2Trevor: Wy did you youst gave that poor woman a skines onion. Whyare you so neine to people.be more nise to them.Rich Señora: but I just hate when thet women comes beggin for things. the should feall sheshould be shy for that.Trevor: but dirint you now if that lady had a famaly thats why you should help.Rich Señora: No I can’t I keape my richnes and my good to myselfok.In both examples the students used textual information to create a dialogue that was plausible forthe characters. They went beyond restating the text to extend the characters in appropriate ways.For example, Trevor’s question, “Why?” asks the Rich Señora to reflect on her actions. Sheresponds that she thinks the beggar should be ashamed of herself for begging and Trevorsuggests she may have a family and needs help.There are a number of possible explanations why this item drew high level responses from somany students. According to the TPBA work group, the alternative assessment was constructedso that the items built on each other throughout the text, providing a kind of internal scaffolding.From the workgroup’s point of view, the students actually learned from, as well as performed, onthe test and as they went through the items. This scaffolding allowed them to see connectionsbetween plot and character that were then reflected in this item. The task was also an engagingone in which students had more interpretive freedom. According to the teachers, the task alsoresembled those used in instruction as students worked toward the standards in their languageArts classes. They were familiar with projecting character traits through dialogue and using“thought bubbles” to indicate the character’s exact words. Although the item came at the end ofthe assessment when performance could have been affected by fatigue, the opposite appeared tooccur. The text processing patterns suggested by the students’ answers indicate that students canmake inferences and apply them to elaborate context even though their ability to expressthemselves clearly is still developing.Finding 4: Implications for InstructionUnlike the district PBA where transitional students’ performances revealed global difficulties,the transitional performance-based assessment provided more specific information about thestudents’ strengths and difficulties. The phonological, syntactic and semantic patternsdemonstrated in the students’ work reflected the influence of their first language, Spanish.© 2008 Dr. Catherine CollierAll Rights Reserved175

When the papers were read out loud, it became clear that students were using Spanishsound/symbol relationships to spell English words. While sounding out words is often useful inSpanish, it is a much less reliable strategy in English. Students would benefit from explicitinstruction in the similarities and differences between Spanish and English sound systems andtheir orthographic representations. In addition, more emphasis should be placed on theconceptual basis for English spelling patterns, as well as memorization of high frequency sightwords. Attention to how morphemes are combined to make different grammatical categories inEnglish is clearly needed (e.g., root word plus ing). Understanding the conceptual basis of words(e.g., root words, word families) would be more appropriate than memorizing lists of words. Thesemantic content of the students’ answers indicate that students need instruction to increaselexical variety, vocabulary specificity, and understanding of word forms. A study ofSpanish/English cognates would be one useful strategy, as would study of synonyms and theuse of a thesaurus.Analysis of students’ text comprehension patterns indicated that students were able to read at aninferential level with an appropriately leveled text, graphic support, and scaffold test structure.Students were able to use text information to take a variety of viewpoints and make plausibleextensions of text information and connect it to personal experience. The use of open-endedquestions gave specific information on what the students were getting from the text. In additionto revealing the students’ strengths, the test answers also indicated that students need more workon text analysis skills, such as identifying text passages that support ideas, building from recallof detail to inferences, and developing metacognitive skills for processing texts (e.g., usinggraphic organizers to track multiple characters, and comprehension monitoring).While the influence of Spanish was an obvious and expected feature of transition students’language, it is important to look beyond this developmental process in language acquisition tothe instruction students receive in both languages. Anderson and Roit (1996) point out thatinstructional practices can exacerbate the difficulties second language readers experience inprocessing text. They suggest that typical instruction for language minority students emphasizeshigh frequency nouns, verbs, and adjectives but ignores vocabulary that “carries much of thelogic of the language, such as negatives, conjunctions, prepositions and other abstract words” (p.298) that would enable students to process texts more accurately. Another difficulty bilingualstudents may experience once they are able to decode English text is the task of distinguishingimportant from unimportant text segments. Students struggling with text comprehension mayfocus on noticeable or dramatic portions of a text while failing to understand the relevance ofparticular text portions to the whole. This difficulty may also be instructionally induced when, inthe desire to encourage discussion, teachers allow students to engagein lengthy discussions about trivial aspects of a text.DiscussionThe development of the Transitional Performance Based Assessment was an effort on the part ofone school district to adjust its assessment practices to the local context. While the projectaffected only 89 students out of the nearly 600 who take the PBA at fifth grade, the issues raisedare pertinent to the assessment of all students, particularly those from linguistic and culturalbackgrounds.The creation of a specific test for transition students provided a great deal of useful information.Findings from the research project concluded that the TPBA showed that students whotraditionally performed poorly on district-wide assessments could successfully engage with© 2008 Dr. Catherine CollierAll Rights Reserved176

English texts and demonstrate progress toward the district’s language arts standards when theassessment took into account their unique language and literacy profiles and planned for theirlanguage development needs. The transition assessment, with its emphasis on assistedperformance, enabled the students to demonstrate their ability to comprehend English text. Eventhose students whose overall performance was low in the holistic scoring demonstrated ability togo beyond concrete responses on certain items.This detailed analysis of student performance showed that while difficulties with languageprocessing clearly affected performance, the specific difficulties with vocabulary, syntax,spelling, and text comprehension did not preclude engagement with text or larger ideas. In herreflection on the results of the assessment, one of the consultants noted:It’s so seductive to say “you better get the basics, the parts, before you get the whole” rather thanrecognizing that it’s kind of going backend forth between the two. That kids can deal with bigideas long before they have all the parts. A kid may have a lot of insight into character and misssome kind of a sequential thing. . . .The teachers got so much from the kids and I think it’sbecause of the format [of the alternative assessment], its nature, the kinds of things we ask kidsto do. We asked them to do fairly interesting and demanding things but [with] a lot of latitude forpersonal interpretation.The presence of knowledgeable scorers capable of interpreting students’ responses was a keyfactor in revealing the students’ strengths and weaknesses. Whereas similar performances oftransitional students on the district PBA in previous years might have been marked“unscoreable” due to the difficulty with conventions and language expression, these tests wereinterpretable by readers familiar with the developmental nature of second language acquisition ingeneral and the influence of Spanish in particular.As districts and states grapple with the challenges of equitable assessments for language minoritystudents, many questions remain unanswered. Specific questions about test construction andadministration exist alongside larger questions of the role of language in learning and theresponsibility of schools to prepare all students for equal participation in society. In spite of itsunanswered questions and unmet challenges, this assessment project can inform other districtsfacing similar challenges of equitable assessment. Its results give us a regard for the complexitiesof making sense of the performance of English language learners on assessments in English.Questioning the Rationale of the TPBARegardless of the strengths shown in the students’ performance, many critics argue that while wemay be sympathetic to these features of second language acquisition, the students musteventually perform proficiently in English and efforts such as these to document their strengthsin the transition phase diverts attention from the pressing need to move these students forward inEnglish proficiency. Goldenberg (1996) for example, questions the appropriateness of programstructures that are specifically designed for students making the transition from primary languageto English reading. He cautions “perhaps reifying transition—making it into an explicit phase ofstudents’ school program—is precisely the wrong thing to do” (p. 357). He contends that thewide variability in the rate and manner of second language acquisition argue against structuringformal transitional programs.Secondly, he suggests that schools that fail to provide a gradual shift in language instruction andinstead abruptly move students to English language instruction before they are prepared may inlarge part socially construct the “transition” period of second language acquisition. Herecommends instead a model of gradual transition that would provide primary language support© 2008 Dr. Catherine CollierAll Rights Reserved177

for longer periods of time and more gradual exposure to English, possibly circumventing theclustering of intense language needs during a particular phase of second language readingdevelopment. While the district under study here did have such a program, many districts withfewer language minority students may not be able to provide carefully sequenced programs.Those who oppose the creation of special assessments for transitional students often advocatereporting students’ performance on one assessment used for all students but removing the highstakes associated with those scores for students in the transition process. That adjustment doesnot satisfactorily answer the need for valid information about such students, nor does it addressthe ethical issues associated with putting students through potentially frustrating and painfulexercises from which they gain little, if any, benefit. In this case, however, for a number ofreasons, including the nature of the assessment and the scoring procedures, the district PBA leftteachers disheartened with the dismal performances of their students without receiving anyuseful information for instruction. In contrast, the TPBA gave direction for instruction, asdiscussed in the previous section.This study was designed to shed light on the performance of transition students on a test ofreading comprehension specifically designed with their linguistic profiles in mind. Wespecifically wanted to know whether an alternative test would yield more information relevantfor instruction than the district-wide test had done. In considering the generalizability of thefindings, a number of issues emerged.While the above approach yielded much information on the form and quality of responses toselected texts, it did not provide comparative information. A useful next step would be to givesubsets of students, matched for particular characteristics, similar types of performanceassessments using different texts to help determine the influence of the texts themselves onstudent performance.It would also be of interest to have a subset of students take both the district PBA and thetransitional PBA to more specifically compare performances. Information on the transitionstudents’ educational backgrounds, level of Spanish oral fluency and literacy, and oral Englishfluency levels would enable us to make more inferences about the relation between these factorsand English reading comprehension. Ethnographic methodology could be used to gaininformation on instructional practices that may have contributed to the prevalence ofperformance at the literal level.In addition to these specific research design challenges, the issue of assessing reading throughwriting poses particular dilemmas for transition students and is a critical limitation of theapproach of both the PBA and the TPBA. It is likely that additional means of assessing reading(think-alouds, portfolios, exhibitions, oral discussions) would yield more definitive informationon students’ ability to meet the standards and on the reading process itself(see Fitzgerald, 1995for a review of various methodologies used to study second language reading).Another issue raised by these findings is the feasibility of creating assessment instrumentsspecific to particular groups of students. The cost of test development and the difficulty infinding knowledgeable scorers would present significant problems to many districts. However,as discussed above, the staff development model adopted in this district demonstrated aneconomical way of accomplishing the test development at the same time that teacher expertisewas well used and further developed. The use of knowledgeable district employees kept the costmanageable and provided professional development opportunities for the teachers whoseexpertise had previously been underutilized. For less equipped districts, pairing bilingual andEnglish-only teachers for scoring purposes may be one solution to this dilemma (possibly even© 2008 Dr. Catherine CollierAll Rights Reserved178

using inter-district teams). In these cases significant attention would need to be paid to theestablishment of inter-rater reliability.The work group realized that one assessment measure could not definitively answer complexquestions of why particular response patterns occurred, especially when the assessment wasscored holistically. The clustering of scores at the 2 or 3 level may have been due to processingfactors, but could also have been influenced by instructional factors, writing ability, theparticular texts chosen, or a combination of factors. Without more information about thestudents’ previous Spanish instruction, more information about results on multiple measures ofreading comprehension, and/or a comparison group using other texts, it is still difficult to answerthese questions.Perhaps the most important result of having a specially designed assessment for students intransition was the opportunity to break the cycle of failure and the culture of discouragementover the school performance of language minority students. The Transitional Performance BasedAssessment allowed teachers to see what their students were able to do and gave theminformation on how to build on their strengths and, through appropriate instruction, guidestudents to more proficient use of English and fuller access to educational opportunities.© 2008 Dr. Catherine CollierAll Rights Reserved179