Читать книгу A Companion to Chomsky - Группа авторов - Страница 31

The main idea behind the model of grammar labeled Government and Binding (GB) was developed by Chomsky in several publications (1981, 1982, 1986). According to Chomsky (1981, p. 5ff.), the theory of grammar consists of a rule system and its subcomponents, depicted in (15), and a system of principles and its subcomponents, shown in (16). The overall architecture is provided in (17).

1 (15) Rule system1.lexicon2.syntaxa.categorial componentb.transformational component3.Phonetic Form4.Logical Form

1 (16) Principle systembounding theorytheta‐theorybinding theorygoverment theorycase theorycontrol theory

1 (17) Y‐model

In constructing a sentence (what is typically called the derivation), the grammar machine starts out with lexical and functional items from the lexicon. The lexicon contains all idiosyncratic information about lexical items. In turn, these items feed the syntax by way of assembling a D‐structure where argument structure properties are assigned, i.e. a transitive verb will be in a configuration with an object and a subject, and an intransitive verb with just a subject. Then transformations (called Move α, where α stands for any constituent) map this structure into an S‐structure (e.g. in the case of wh‐movement as in What did Jane read?, which starts out with the structure Jane read what), which fairly closely resembles the surface order of words. At this point, the resulting structure is transferred to the two interpretive interfaces: Phonetic Form (sound) and Logical Form (meaning). Note also that the principal subsystems in effect apply wellformedness constraints on the other levels of representation. Chomsky (1981, pp. 5–6) summarizes the organization of the grammar as follows:

The lexicon specifies the abstract morpho‐phonological structure of each lexical item and its syntactic features, including its categorial features and its contextual features. The rules of the categorial component meet some variety of X‐bar theory. Systems 1.1 and 1.2a constitute the base. Base rules generate D‐structures (deep structures) through insertion of lexical items into structures generated by 1.2a, in accordance with their feature structure. These are mapped to S‐structure by the rule Move‐alpha, leaving traces coindexed with their antecedents; this rule constitutes the transformational component 1.2b, and may also appear in the PF‐ and LF‐components. Thus the syntax generates S‐structures which are assigned PF‐ and LF‐representations by components 3 and 4 of 1. Bounding theory poses locality conditions on certain processes and related items. The central notion of government theory is the relation between the head of a construction and categories dependent on it. Theta‐Theory is concerned with the assignment of thematic roles such as agent‐of‐action, etc…. Binding theory is concerned with relations of anaphors, pronouns, names, and variables to possible antecedents. Case theory deals with assignment of abstract Case and its morphological realization. Control theory determines the potential for reference of the abstract pronominal element PRO.

Much more can be said about each of these components, and Haegeman (1994) provides an excellent and detailed exposition. Many other textbooks also exist, so in the interest of space, we will not provide a more detailed outline of the GB architecture. Some necessary details will be covered in the discussion of the Minimalist Program in the next section, but rather than go through the entire system here, we want to highlight a more general aspect of the model.

The above picture demonstrates the division of the language module into distinct submodules. In addition, it is assumed that the language faculty is a module of the mind distinct from other modules such as for instance the visual system. Chomsky's notion of modularity is very different from Fodorian modules (Chomsky 2018 compares his position to Fodor). For Fodor, human cognition consists of a “central system” and a number of “input systems” (Fodor 1983). The Fodorian modules are cognitive “reflexes” – specialized for particular domains, which operate fast and mandatorily, are hard‐wired for a particular area of the brain, their structure and function largely innately determined, and they are informationally encapsulated. Fodor's view on language is that it is an (i) input system analogous to those devoted to the senses, (ii) the central system, however, is unstructured and uninvestigable.⁸ By contrast, Chomsky assumes that language is a central system used in both input and output (in expressing and communicating thoughts), and there are probably several such central systems: a language faculty; a number “sense,” and possibly others (see e.g. Fodor 2000, Smith and Allott 2016, and Chomsky 2018 on similarities and differences between their views).

While the framework of GB led to very rich empirical generalizations and discoveries (see also the chapters by Baker, Müller, Sheehan, Chapters 10–12, in this volume), there are some issues with its architecture. In particular the proliferation of internal modules and the fact that it postulates four levels of representation can be seen as violating the principle that one should only postulate the “barest essentials” necessary for a theory of grammar. In addition, a new question on the agenda became how to account for the biological or evolutionary origin of language: To what extent is our theory of human language compatible with what we know about language evolution and the biological scaffolding of language? We turn to a discussion of these issues in the next section.