Javanese and Madurese DPs display a great deal of fluidity in the ordering
of their constituents. The number of possible permutations makes description
difficult and a coherent analysis seem nearly impossible in a deterministic
theory such as that underpinning the Minimalist Program. The present paper attempts
to account for some of the variations, showing that making use of some proposals
already available in the literature allows some inroads into an account of the
Javanese and Madurese data. Perhaps more importantly, the data from both languages
require treating some adjectival modifiers not as phrases but as X° adjuncts
to the N head. Just such an analysis has been proposed on various occasions
for prenominal adjectives in English, including Stowell 1981, Sadler & Arnold
1994, and Baker 2003. The Javanese and Madurese data thus provide new evidence
in support of that analysis.
2003. "Extreme Locality in Madurese
wh-Questions," Syntax 6:237-259.
This article investigates the apparently three types of wh questions in Madurese—wh
in situ, overt wh movement, and partial wh movement—and argues that the
three are actually instantiations of a single wh in situ strategy and that the
relation between the wh element and its operator obeys extreme locality. An
analysis is proposed in which overt long distance and partial wh movement are
actually reflexes of A-movement to subject position in a proleptic object construction.
Madurese wh phenomena are situated in term of Cole & Hermon’s (1998)
typology of wh movement. It is further suggested here that movement across clause
boundaries is illicit in Madurese.
2003. (William D. Davies & Stanley Dubinsky)
"On Extraction from NPs," NLLT 21:1-37.
This paper sheds new light on the conditions governing extraction from NPs.
A close examination of wh-extraction out of object NPs reveals that previously
unnoticed semantic factors play a greater role than has been recognized. In
particular, we find that NPs lacking `participant' structure do not permit wh-extraction
at all, and that certain NPs permit extraction even when they are definite.
At the same time, the prohibition on wh-extraction from subject NPs is shown
to be a purely syntactic phenomenon which arises from the particular way
in which the Extended Projection Principle is satisfied in English.
2000. "Events in Madurese Reciprocals," Oceanic Linguistics 39:123-143.
This paper examines the two manners of forming reciprocal meaning in Madurese, each of which is based on elements that are used for nonreciprocal meaning. Although the two constructions appear to have little in common formally, I argue that they do share an important semantic property that makes them particularly suited to forming reciprocals. This property is that they are used to describe multiple events, precisely the type of situation that obtains in reciprocal meaning. Therefore, these constructions are selected rather than other candidate constructions that show plural activity but crucially are not used for discrete multiple events. The properties of the various candidate construction are explicitly contrasted in some detail. Finally, I demonstrate the special syntactic characteristics of the reciprocals that distinguish them from the nonreciprocal uses, properties that indicate that the reciprocals have been grammaticized.
1999. "Madurese and Javanese as Strict Word Order Languages," Oceanic Linguistics 38:152- 167.
Although much of the published literature on Madurese and Javanese characterize them as having fairly strict SVO order, the colloquial languages allow major constitutents of a clause to occur in almost any order, a point Uhlenbeck (1975) makes quite clear in his study of Javanese word orders. Based on this relative freedom of word order, Uhlenbeck concludes that the basic unit of syntactic analysis for Javanese clauses should be the "sentence segment" rather than the phrase structure constitutents of transformational theories. Here is it argued on the basis of dislocation structures, a restriction on the specificity of subjects, and a restriction on word order in questions that, in fact, the freedom of word order in Madurese and Javanese is only apparent and that many word orders diverging from the basic SVO are a result of dislocation structures. This analysis thus challenges the conclusion that phrase structure representations are inappropriate or unsightful for these languages.
1998 (William D. Davies & Tamar I. Kaplan). "Native Speaker vs. L2 Learner Grammaticality Judgments," Applied Linguistics 16:183-203.
Although many SLA syntax studies have used non-native speaker (NNS) grammaticality judgments, what underlies NNS grammaticality judgments has recently come under increasing scrutiny. Some researchers (Ellis 1991; Goss, Ying-Hua, and Lantolf 1994), using speak-aloud protocols and oral group work, have presented evidence that indicates NNSs use very different strategies in making decisions about the grammaticality of particular sentences than do native speakers (NSs) of the same language. NNSs' use of strategies such as guessing, operationalizing learned rules, translation, and analogy calls into question whether or not their grammaticality judgments are a reliable indicator of their operational grammar. However, heretofore oral protocol data have not been available for NS grammaticality judgments. This paper reports on an experiment comparing group oral protocols of native English-speaking subjects enrolled in a 4th semester college French course making grammaticality judgments in English and French. Results indicate that the subjects do not necessarily use the same strategies in rendering grammaticality judgments in L1 and L2, thereby casting doubt on theoretical claims based on data elicited through L2 grammaticality judgments.
1997. "Relational Succession in Kinyarwanda Possessor Ascension," Lingua 101:89 114.
Perlmutter and Postal's (1983) Relational Succession Law (RSL), meant to constrain raising constructions, has been challenged in two ways, both in the area of possessor ascension constructions. On the one hand, it has been proposed that in some languages possessors may raise to indirect object regardless of the grammatical relation of the host nominals; while on the other, it has been proposed that possessors can only raise to direct object. Both types of structures have been proposed for Kinyarwarnda (Bickford 1986). This paper adduces evidence that neither of these challenges to the RSL is empirically motivated. Rather, Kinyarwanda provides compelling evidence for the correctness of the RSL and also provides surprising evidence for possessor ascension from an oblique host. Kinyarwanda thus proves important to the typology of raising constructions.
1996. "Morphological Uniformity and the Null Subject Parameter in SLA," Studies in Second Language Acquisition 17:475-493.
An area of keen interest in applying Chomsky's UG parameter-setting model to SLA has been the Pro-Drop or Null Subject Parameter (Cyrino 1986, Hilles 1986, Phinney 1987, White 1985, 1986). However, the nature of this parameter changes dramatically from the Jaeggli (1982) and Rizzi (1982) conception with Jaeggli and Safir's (1989) proposal linking uniform morphological agreement paradigms with null subjects. Data reported in this paper show a number of L2 learners exhibit knowledge that English is morphologically nonuniform yet still accept English null subject sentences. This is inconsistent with the predictions of the Morphological Uniformity Hypothesis and renders uncertain is applicability to SLA. The results are considered in light of a number of possible positions that can be adopted when faced with data that disconfirm a hypothesis within the UG SLA research program; it is concluded that the Morphological Uniformity Hypothesis is disconfirmed and that any reformulation of the Null Subject Parameter must take the present results into consideration.
1995. "Javanese Adversatives, Passives, and Mapping Theory," Journal of Linguistics 31:15-51.
Relational Grammarians have proposed Union analyses for adversatives in many languages. An odd asymmetry in base predicates contrasts Japanese (Dubinsky 1985), in which adversatives may not be formed on unac cusatives, and Indonesian (Kana 1986), in which adversatives may only be formed on unaccusatives. A close examination of adversatives in Javanese (a language closely relation to Indonesian) resolves this asymmetry, revealing that Javanese adversatives (and by implication Indonesian adversatives) are best analyzed not as Unions but as passives. However, the passive analysis violates Perlmutter and Postal's (1984) 1-Advancement Exclusiveness Law, which figures crucially in Dubinsky's elegant account of the distribution of Japanese adversatives. Gerdts (1993) Mapping Theory, rooted in an RG tradition, provides a solution in which it is possible to capture the similarities of adversatives and other Javanese passives and at the same time preserve the insights of Dubinsky's analysis of Japanese.