Prefixation in English and Catalan - Departament de Filologia ...

More documents

Recommendations

Info

Because Reinhart’s marking procedures apply only for the arguments on a V’s Θ-grid, it will only be at the verbal node, and not before, that arguments will get index 1 ([+] cluster), 2 ([-] cluster) or no marking at all (mixed cluster). That means that the Θ-roles of As and Ns cannot follow Reinhart’s marking system, since this is not applicable to them. With respect to the relationship between the Θ-roles of As and Ns and their syntactic position (whether they are external or internal arguments), I follow the regularities (see (10)) already established in other work (for example, Williams 1981b). That is, a [-c-m] role on an A will be external. The same role on a N will be internal and the R-role, 9 which is associated with Ns, will be external. The resulting pattern is as follows: (10) a. For As: i. [-c-m] → external b. For Ns: i. [R] → external ii. [-c-m] → internal As will be seen, such approach seems problematic at first sight, because the external argument of an A (This apple is edible) is internalized when it is on the V’s node (I ate the apple). However, this apparent internalization is explained if we adopt the view according to which only Θ-roles percolate, and the notions external or internal are determined by the category the Θ-roles are associated with (see Neeleman & Schipper (1992) for a similar view). In other words, the A will force an argument specified as [-c-m] to be external, whereas the same role on a V will be given index 2, which will determine internal merging. 2.2 Two analyses of reflexives: unergative vs. unaccusative 9 The source of the R-role is to be found in Williams (1981b), who notes that Ns also have external Θroles. In sentence (i), (i) I consider that [destruction of a city by evil forces] the predicative NP destruction has two internal arguments: the Theme a city and the Agent evil forces, but it also has an external argument which has no counterpart in the verbal system, i.e. that, which he gives the label R. That is, destruction of a city by evil forces is predicated of that. “The label R is meant to suggest ‘referential’, since it is this argument position (R) that is involved in referential uses of NPs as well” (p. 86) 9
As for the treatment of reflexives (quite abundant in my Catalan survey, cf. Appendix B), we have just seen that, according to Reinhart, they are unergative entries, which have been derived by reducing the internal argument of a transitive V (see Reinhart & Siloni (1999) and Reinhart (2000, 2001) for details). However, Reinhart is somehow forced to stipulate that reflexivization is the result of reducing the internal argument, because she already has an external reduction operation for expletivization (recall that this is how she derives unaccusative and unergative entries; cf. section 2.1, pp. 7-8). Similarly, one could also stipulate that reflexivization is the outcome of reducing the external Θ-role, and that se is the obligatory marker (in Romance languages) that results from the reduction operation. In fact, this is roughly the unaccusative approach to reflexives, which has also been defended (see Grimshaw (1990), Sportiche (1998), for instance). According to this approach, the subject of reflexives, like the subject of unaccusative verbs, is the underlying object. Within the unaccusative analysis of reflexives, there are two different variants: the lexical and the syntactic. While the former assumes that the external argument is lexically absorbed, the latter assumes that the clitic se is the external argument present in syntax. On theoretical grounds, there is no reason for choosing one approach (reflexives as unergatives vs. reflexives as unaccusatives) over the other. The data of my Catalan study will, however, suggest that the unaccusative approach to reflexives is the one which seems to be on the right track. 2.3 Θ-percolation and inheritance As far as the Θ-percolation approach is concerned, the basic idea is that the thematic information of a complex word is derived from the different elements that form the word, irrespective of whether they are prefixes or suffixes. This view of Θ-percolation is in conflict with the RHR, which states that only the head is able to transfer its features. The data analysed in my study will show that the strict RHR (Williams 1981a) has to be abandoned, in favour of the Rel. RHR (Di Sciullo & Williams (1987): 25-28), according to which the head for a specific feature is the rightmost element that contains the feature in question. To illustrate this, consider the Latin word in (11), which according to the Rel. RHR will have two heads, given that both bi and tur are the rightmost elements with respect to the features they are marked; i.e. the former is specified as [+future] and the latter is marked with the feature [+passive]. 10
Page 1 and 2: Argument Structure and Morphology:
Page 3 and 4: TABLE OF CONTENTS TABLE OF CONTENTS
Page 5 and 6: 2a) are typically category-changing
Page 7 and 8: (5) a. dobbelenV ‘to gamble’ Ag
Page 9 and 10: 2 THEORETICAL BACKGROUND This secti
Page 11: how the marking procedures work and
Page 15 and 16: The fact that the suffix -ize forms
Page 17 and 18: may be considered negative by some,
Page 19 and 20: First, the classification of suffix
Page 21 and 22: unknown to derivation. 16 This is w
Page 23 and 24: 3 RESEARCH METHODS Dictionaries hav
Page 25 and 26: To confirm the results from the dic
Page 27 and 28: ‘A boat crossed the Atlantic’ (
Page 29 and 30: form (endur-se
Page 31 and 32: (32) V {[+c]1, [-c-m]2} / \ en V {[
Page 33 and 34: . Ell encaputxà la Maria. He EN-ho
Page 35 and 36: (39) coixíN ‘cushion’ R encoix
Page 37 and 38: Although the dictionary lists the u
Page 39 and 40: (45) a. The people outside chanted
Page 41 and 42: a. The reporter enlarged the pictur
Page 43 and 44: syllables), and strengthenV/lengthe
Page 45 and 46: (58) a. caseN R b. encaseV [+c] [-c
Page 47 and 48: meaning (e.g. list, trench), which
Page 49 and 50: 4.3 Discussion This section is main
Page 51 and 52: the speaker intended on that occasi
Page 53 and 54: had no function to perform, i.e. th
Page 55 and 56: HK’s treatment of deadjectival Vs
Page 57 and 58: 5 CONCLUSIONS In this thesis, I con
Page 59 and 60: As for the remark made by several a
Page 61 and 62: a) V-to-enV Enact (act) 1 When a go
Page 63 and 64:
Entreat 1 If you entreat someone to
Page 65 and 66:
Enrich (rich) 1 To enrich something
Page 67 and 68:
ORIGIN: not given. Cipher (n.) 1 A
Page 69 and 70:
Enfold (fold) 1 If something enfold
Page 71 and 72:
Lighten 2 (v.) 1 When something lig
Page 73 and 74:
ORIGIN C19: from Spanish, via L. fr
Page 75 and 76:
Derivative: entrapment. ORIGIN C16:
Page 77 and 78:
d) This section presents those Vs w
Page 79 and 80:
ORIGIN ME: from OFr. enduire, partl
Page 81 and 82:
Ensconce (sconce) (often ensconce o
Page 83 and 84:
Velop: non-existing. Environ formal
Page 85 and 86:
Intermediate Vs which have meanings
Page 87 and 88:
A → [en+A]V Transitive [+c] [-c-m
Page 89 and 90:
B) Trans [+c] (usually [+c+m]) [-c-
Page 91 and 92:
Tela → telar → entelar Verí
Page 93 and 94:
Appendix C This appendix includes t
Page 95 and 96:
Results: English questionnaire When
Page 97 and 98:
Results: Catalan questionnaire Of a
Page 99 and 100:
Di Sciullo, A. M. 1997. “Prefixed
Page 101 and 102:
Marantz, A. 1997. “No Escape from
Page 103:
Varela, S. (ed.). 1993. La formaci
show all

Prefixation in English and Catalan - Departament de Filologia ...

You also want an ePaper? Increase the reach of your titles

Delete template?

Save as template?