Computational models of human sentence comprehension help researchers reason about how grammar might actually be used in the understanding process. Taking a cognitivist approach, this article relates computational psycholinguistics to neighboring fields (such as linguistics), surveys important precedents, and catalogs open problems.
The noun-modifying clause construction (NMCC) in Japanese is a complex noun phrase in which a prenominal clause is dependent on the head noun. Naturally occurring instances of the construction demonstrate that a single structure, schematized as [[… predicate (finite/adnominal)] Noun], represents a wide range of semantic relations between the head noun and the dependent clause, encompassing some that would be expressed by structurally distinct constructions such as relative clauses, noun complement clauses, and other types of complex noun phrases in other languages, such as English. In that way, the Japanese NMCC demonstrates a clear case of the general noun-modifying construction (GNMCC), that is, an NMCC that has structural uniformity across interpretations that extend beyond the range of relative clauses. One of the notable properties of the Japanese NMCC is that the modifying clause may consist only of the predicate, reflecting the fact that referential density is moderate in Japanese—arguments of a predicate are not required to be overtly expressed either in the main clause or in the modifying clause. Another property of the Japanese NMCC is that there is no explicit marking in the construction that indicates the grammatical or semantic relation between the head noun and the modifying clause. The two major constituents are simply juxtaposed to each other. Successful construal of the intended interpretations of instances of such a construction, in the absence of explicit markings, likely relies on an aggregate of structural, semantic, and pragmatic factors, including the semantic content of the linguistic elements, verb valence information, and the interpreter’s real-world knowledge, in addition to the basic structural information. Researchers with different theoretical approaches have studied Japanese NMCCs or subsets thereof. Syntactic approaches, inspired by generative grammar, have focused mostly on relative clauses and aimed to identify universally recognized syntactic principles. Studies that take the descriptive approach have focused on detailed descriptions and the classification of a wide spectrum of naturally occurring instances of the construction in Japanese. The third and most recent group of studies has emphasized the importance of semantics and pragmatics in accounting for a wide variety of naturally occurring instances. The examination of Japanese NMCCs provides information about the nature of clausal noun modification and affords insights into languages beyond Japanese, as similar phenomena have reportedly been observed crosslinguistically to varying degrees.
This paper provides an overview of polarity phenomena in human languages. There are three prominent paradigms of polarity items: negative polarity items (NPIs), positive polarity items (PPIs), and free choice items (FCIs). What they all have in common is that they have limited distribution: they cannot occur just anywhere, but only inside the scope of licenser, which is negation and more broadly a nonveridical licenser, PPIs, conversely, must appear outside the scope of negation. The need to be in the scope of a licenser creates a semantic and syntactic dependency, as the polarity item must be c-commanded by the licenser at some syntactic level. Polarity, therefore, is a true interface phenomenon and raises the question of well-formedness that depends on both semantics and syntax. Nonveridical polarity contexts can be negative, but also non-monotonic such as modal contexts, questions, other non-assertive contexts (imperatives, subjunctives), generic and habitual sentences, and disjunction. Some NPIs and FCIs appear freely in these contexts in many languages, and some NPIs prefer negative contexts. Within negative licensers, we make a distinction between classically and minimally negative contexts. There are no NPIs that appear only in minimally negative contexts. The distributions of NPIs and FCIs crosslinguistically can be understood in terms of general patterns, and there are individual differences due largely to the lexical semantic content of the polarity item paradigms. Three general patterns can be identified as possible lexical sources of polarity. The first is the presence of a dependent variable in the polarity item—a property characterizing NPIs and FCIs in many languages, including Greek, Mandarin, and Korean. Secondly, the polarity item may be scalar: English any and FCIs can be scalar, but Greek, Korean, and Mandarin NPIs are not. Finally, it has been proposed that NPIs can be exhaustive, but exhaustivity is hard to precisely identify in a non-stipulative way, and does not characterize all NPIs. NPIs that are not exhaustive tend to be referentially vague, which means that the speaker uses them only if she is unable to identify a specific referent for them.
Agustín Vicente and Ingrid L. Falkum
Polysemy is characterized as the phenomenon whereby a single word form is associated with two or several related senses. It is distinguished from monosemy, where one word form is associated with a single meaning, and homonymy, where a single word form is associated with two or several unrelated meanings. Although the distinctions between polysemy, monosemy, and homonymy may seem clear at an intuitive level, they have proven difficult to draw in practice. Polysemy proliferates in natural language: Virtually every word is polysemous to some extent. Still, the phenomenon has been largely ignored in the mainstream linguistics literature and in related disciplines such as philosophy of language. However, polysemy is a topic of relevance to linguistic and philosophical debates regarding lexical meaning representation, compositional semantics, and the semantics–pragmatics divide. Early accounts treated polysemy in terms of sense enumeration: each sense of a polysemous expression is represented individually in the lexicon, such that polysemy and homonymy were treated on a par. This approach has been strongly criticized on both theoretical and empirical grounds. Since at least the 1990s, most researchers converge on the hypothesis that the senses of at least many polysemous expressions derive from a single meaning representation, though the status of this representation is a matter of vivid debate: Are the lexical representations of polysemous expressions informationally poor and underspecified with respect to their different senses? Or do they have to be informationally rich in order to store and be able to generate all these polysemous senses? Alternatively, senses might be computed from a literal, primary meaning via semantic or pragmatic mechanisms such as coercion, modulation or ad hoc concept construction (including metaphorical and metonymic extension), mechanisms that apparently play a role also in explaining how polysemy arises and is implicated in lexical semantic change.
Throughout the 20th century, structuralist and generative linguists have argued that the study of the language system (langue, competence) must be separated from the study of language use (parole, performance), but this view of language has been called into question by usage-based linguists who have argued that the structure and organization of a speaker’s linguistic knowledge is the product of language use or performance. On this account, language is seen as a dynamic system of fluid categories and flexible constraints that are constantly restructured and reorganized under the pressure of domain-general cognitive processes that are not only involved in the use of language but also in other cognitive phenomena such as vision and (joint) attention. The general goal of usage-based linguistics is to develop a framework for the analysis of the emergence of linguistic structure and meaning. In order to understand the dynamics of the language system, usage-based linguists study how languages evolve, both in history and language acquisition. One aspect that plays an important role in this approach is frequency of occurrence. As frequency strengthens the representation of linguistic elements in memory, it facilitates the activation and processing of words, categories, and constructions, which in turn can have long-lasting effects on the development and organization of the linguistic system. A second aspect that has been very prominent in the usage-based study of grammar concerns the relationship between lexical and structural knowledge. Since abstract representations of linguistic structure are derived from language users’ experience with concrete linguistic tokens, grammatical patterns are generally associated with particular lexical expressions.
John E. Joseph
Ferdinand de Saussure (1857–1913), the founding figure of modern linguistics, made his mark on the field with a book he published a month after his 21st birthday, in which he proposed a radical rethinking of the original system of vowels in Proto-Indo-European. A year later, he submitted his doctoral thesis on a morpho-syntactic topic, the genitive absolute in Sanskrit, to the University of Leipzig. He went to Paris intending to do a second, French doctorate, but instead he was given responsibility for courses on Gothic and Old High Gerrman at the École Pratique des Hautes Études, and for managing the publications of the Société de Linguistique de Paris. He abandoned more than one large publication project of his own during the decade he spent in Paris. In 1891 he returned to his native Geneva, where the University created a chair in Sanskrit and the history and comparison of languages for him. He produced some significant work on Lithuanian during this period, connected to his early book on the Indo-European vowel system, and yielding Saussure’s Law, concerning the placement of stress in Lithuanian. He undertook writing projects about the general nature of language, but again abandoned them. In 1907, 1908–1909, and 1910–1911, he gave three courses in general linguistics at the University of Geneva, in which he developed an approach to languages as systems of signs, each sign consisting of a signifier (sound pattern) and a signified (concept), both of them mental rather than physical in nature, and conjoined arbitrarily and inseparably. The socially shared language system, or langue, makes possible the production and comprehension of parole, utterances, by individual speakers and hearers. Each signifier and signified is a value generated by its difference from all the other signifiers or signifieds with which it coexists on an associative (or paradigmatic) axis, and affected as well by its syntagmatic axis. Shortly after Saussure’s death at 55, two of his colleagues, Bally and Sechehaye, gathered together students’ notes from the three courses, as well as manuscript notes by Saussure, and from them constructed the Cours de linguistique générale, published in 1916. Over the course of the next several decades, this book became the basis for the structuralist approach, initially within linguistics, and later adapted to other fields. Saussure left behind a large quantity of manuscript material that has gradually been published over the last few decades, and continues to be published, shedding new light on his thought.
Inflection is the systematic relation between words’ morphosyntactic content and their morphological form; as such, the phenomenon of inflection raises fundamental questions about the nature of morphology itself and about its interfaces. Within the domain of morphology proper, it is essential to establish how (or whether) inflection differs from other kinds of morphology and to identify the ways in which morphosyntactic content can be encoded morphologically. A number of different approaches to modeling inflectional morphology have been proposed; these tend to cluster into two main groups, those that are morpheme-based and those that are lexeme-based. Morpheme-based theories tend to treat inflectional morphology as fundamentally concatenative; they tend to represent an inflected word’s morphosyntactic content as a compositional summing of its morphemes’ content; they tend to attribute an inflected word’s internal structure to syntactic principles; and they tend to minimize the theoretical significance of inflectional paradigms. Lexeme-based theories, by contrast, tend to accord concatenative and nonconcatenative morphology essentially equal status as marks of inflection; they tend to represent an inflected word’s morphosyntactic content as a property set intrinsically associated with that word’s paradigm cell; they tend to assume that an inflected word’s internal morphology is neither accessible to nor defined by syntactic principles; and they tend to treat inflection as the morphological realization of a paradigm’s cells. Four important issues for approaches of either sort are the nature of nonconcatenative morphology, the incidence of extended exponence, the underdetermination of a word’s morphosyntactic content by its inflectional form, and the nature of word forms’ internal structure. The structure of a word’s inventory of inflected forms—its paradigm—is the locus of considerable cross-linguistic variation. In particular, the canonical relation of content to form in an inflectional paradigm is subject to a wide array of deviations, including inflection-class distinctions, morphomic properties, defectiveness, deponency, metaconjugation, and syncretism; these deviations pose important challenges for understanding the interfaces of inflectional morphology, and a theory’s resolution of these challenges depends squarely on whether that theory is morpheme-based or lexeme-based.
David R. Mortensen
Hmong-Mien (also known as Miao-Yao) is a bipartite family of minority languages spoken primarily in China and mainland Southeast Asia. The two branches, called Hmongic and Mienic by most Western linguists and Miao and Yao by Chinese linguists, are both compact groups (phylogenetically if not geographically). Although they are uncontroversially distinct from one another, they bear a strong mutual affinity. But while their internal relationships are reasonably well established, there is no unanimity regarding their wider genetic affiliations, with many Chinese scholars insisting on Hmong-Mien membership in the Sino-Tibetan superfamily, some Western scholars suggesting a relationship to Austronesian and/or Tai-Kradai, and still others suggesting a relationship to Mon-Khmer. A plurality view appears to be that Hmong-Mien bears no special relationship to any surviving language family. Hmong-Mien languages are typical—in many respects—of the non-Sino-Tibetan languages of Southern China and mainland Southeast Asia. However, they possess a number of properties that make them stand out. Many neighboring languages are tonal, but Hmong-Mien languages are, on average, more so (in terms of the number of tones). While some other languages in the area have small-to-medium consonant inventories, Hmong-Mien languages (and especially Hmongic languages) often have very large consonant inventories with rare classes of sounds like uvulars and voiceless sonorants. Furthermore, while many of their neighbors are morphologically isolating, few language groups display as little affixation as Hmong-Mien languages. They are largely head-initial, but they deviate from this generalization in their genitive-noun constructions and their relative clauses (which vary in position and structure, sometimes even within the same language).
Matthew J. Gordon
William Labov (b. 1927) is an American linguist who pioneered the study of variationist sociolinguistics. Born and raised in northern New Jersey, Labov studied English and philosophy at Harvard University (BA, 1948) and worked as an industrial chemist for several years before entering graduate school in linguistics at Columbia University in 1961. He completed his PhD in 1964, under the direction of Uriel Weinreich. He worked at Columbia until 1971, when he joined the faculty of the University of Pennsylvania, where he taught until his retirement in 2014. Labov’s influence on the field began with research he conducted in graduate school. His study of changing pronunciations on Martha’s Vineyard, the subject of his master’s thesis, introduced a method for observing sound change in progress and broke with tradition by exploring social motivations for linguistic innovations. For his PhD dissertation, Labov carried out a study of dialect patterns on the Lower East Side of New York City. Using a systematic, quantitative methodology, he demonstrated that linguistic variation is socially stratified, such that the use of pronunciation features (e.g., dropping of post-vocalic /r/) correlates with social class, ethnicity, etc. in regular patterns. Labov’s early research was greatly influential and inspired many scholars to carry out similar projects in other communities. The paradigm came to be known as variationist sociolinguistics. Much of Labov’s scholarship seeks to advance our understanding of language change. Historical linguists traditionally study completed linguistic changes, often long after they occurred, but Labov developed a method for examining active changes through a quantitative comparison of speakers representing several generations. This approach produces a new perspective on the change process by revealing intermediate stages. Labov has brought insights from this research to bear on theoretical debates within historical linguistics and the field more broadly. His work in this area has also documented many active sound changes in American English. Among these changes are innovations underway in particular dialects, such as the vowel changes in Philadelphia, as well as broader regional patterns, such as the Northern Cities Shift heard in the Great Lakes states. Throughout his career, social justice concerns have fueled Labov’s research. He has sought to demonstrate that the speech of stigmatized groups is as systematic and rule-governed as any other. He led a pioneering study in Harlem in the late 1960s that shone new light on African American English, demonstrating, for example, that grammatical usages like the deletion of the copula (e.g., He fast) are subject to regular constraints. Labov has served as an expert witness in court and before the U.S. Congress to share insights from his study of African American English. He has also worked to promote literacy for speakers of non-standard dialects, carrying out research on reading and developing material for the teaching of reading to these populations.
Malka Rappaport Hovav
Words are sensitive to syntactic context. Argument realization is the study of the relation between argument-taking words, the syntactic contexts they appear in and the interpretive properties that constrain the relation between them.