Within functionalist approaches, research on the syntax–semantics interface has been aimed at disproving the formalist argument of the autonomy of syntax, by finding instances of semantically determined syntactic structures.1415
Levin and Rappaport Hovav, in their 1995 monograph, reiterated that there are some aspects of verb meaning that are relevant to syntax, and others that are not, as previously noted by Steven Pinker.1617 Levin and Rappaport Hovav isolated such aspects focusing on the phenomenon of unaccusativity that is "semantically determined and syntactically encoded".18
Van Valin and LaPolla, in their 1997 monographic study, found that the more semantically motivated or driven a syntactic phenomenon is, the more it tends to be typologically universal, that is, to show less cross-linguistic variation.19
In formal semantics, semantic interpretation is viewed as a mapping from syntactic structures to denotations. There are several formal views of the syntax–semantics interface which differ in what they take to be the inputs and outputs of this mapping. In the Heim and Kratzer model commonly adopted within generative linguistics, the input is taken to be a special level of syntactic representation called logical form. At logical form, semantic relationships such as scope and binding are represented unambiguously, having been determined by syntactic operations such as quantifier raising. Other formal frameworks take the opposite approach, assuming that such relationships are established by the rules of semantic interpretation themselves. In such systems, the rules include mechanisms such as type shifting and dynamic binding.20212223
Before the 1950s, there was no discussion of a syntax–semantics interface in American linguistics, since neither syntax nor semantics was an active area of research.24 This neglect was due in part to the influence of logical positivism and behaviorism in psychology, that viewed hypotheses about linguistic meaning as untestable.2526
By the 1960s, syntax had become a major area of study, and some researchers began examining semantics as well. In this period, the most prominent view of the interface was the Katz–Postal Hypothesis according to which deep structure was the level of syntactic representation which underwent semantic interpretation. This assumption was upended by data involving quantifiers, which showed that syntactic transformations can affect meaning. During the linguistics wars, a variety of competing notions of the interface were developed, many of which live on in present-day work.2728
Chierchia (1999) ↩
Partee (2014) ↩
Hackl (2013) ↩
Levin & Rappaport Hovav (1995) ↩
Van Valin & LaPolla (1997) ↩
Vendler (1957) ↩
Tenny (1994) ↩
Van Valin (2005) p.67 ↩
Van Valin 2003, p.334 ↩
Since the 1970s, as a response to syntactic-oriented approaches like Chomsky's generativism, the assumption underlying many studies on lexical semantics has been that "syntactic properties of phrases reflect, in large part, the meanings of the words that head them" (Levin& Pinker, 1992) /wiki/Lexical_semantics ↩
Levin & Rappaport Hovav (1995) ch.1 p. 9 ↩
Pinker 1989 ↩
Levin & Rappaport Hovav (1995) ch.5 p.179, Afterword p.279 ↩
Van Valin (2005), ch.5 "Linking syntactic and semantic representations in simple sentences" p.128 ↩
Heim & Kratzer (1998) ↩
Baker (2015) ↩
Partee (2014).pp.2, 6 ↩
Taylor (2017) ↩