15 February 2011

The Grammar of the World

Filed under: Language, Psychology — ktismatics @ 1:45 pm

17 FEB UPDATE — I should have supplied more context for this post. The main purpose of Tomasello’s book, which I quote extensively here, is to present the empirical evidence supporting a “usage-based” theory of language development versus the two main alternatives: the universal grammar theory of Chomsky et al., and the connectionism models of neurolinguists and cognitive scientists.

Tomasello offers two main objections to universal grammar theory: linguistic diversity and developmental change. Human languages exhibit a wide variety of grammatical structures, which would seem to be maladaptive if all humans are pre-wired to structure their utterances the same ways. Even if, on a highly abstract level and with many exceptions, all human languages can all be made to fit into a universal schema, anyone actually learning to understand and to speak any actually existing language must learn its unique and idiosyncratic structural properties. If grammatical competence is prewired into human brains, then children would be expected to generalize spontaneously from only a few examples of canonical grammatical form into mature adult speech. However, empirical evidence on children’s language acquisition shows that this doesn’t happen. Kids build up competence incrementally and unevenly from the specific examples they hear from other language-users. It seems that competence is achieved not by brain maturation but by categorization and analogical reasoning and practice — the same way human children learn to master other skills prevalent in the societies they grow up in.

Connectionist models are predicated on the assumption that children can learn language without a prewired language module built into the brain’s architecture. Through trial and error, corrected via feedback from other language users, a brain can presumably wire itself to understand and to generate grammatically correct statements. Connectionist simulations via computerized neural network models can build competence in using the basic components of language; e.g., nouns, verbs, modifiers, subject-object relations. However, these simulated learning models are weak where universal grammar theory is strong; i.e., the models don’t generalize across different kinds of examples, and they don’t handle complex sentences very well. Tomasello contends that, to overcome these limitations, connectionist models don’t need grammatical prewiring; rather, they need an understanding of communicative intent. Humans learn grammatical elements and their interconnections not in isolation but in the context of use. People use language with the intent of orienting one other jointly toward some selected features of the world. Absent this understanding of context and intentionality, connectionist models can categorize and analogize from examples only on surface linguistic features like word order and synonyms. The ability to infer similarity in meaning across two statements, both grammatically correct but structurally quite different from each other, requires the learner to infer the speaker’s communicative intent.

*  *  *  *  *

John ran. John went for a run. Both sentences describe the same actor performing the same action in the world. Linguistically though the two sentences are structured differently. The verb in the first sentence is run, whereas in the second it’s go. The word run also appears in the second sentence, but it’s embedded in a prepositional phrase where it functions not as a verb but as a noun. Because of these structural differences, the two sentences carry slightly different connotations.

What is the relationship between language and the world, between the grammar of a sentence and the grammar of the situation it describes? In Constructing a Language (2003), Michael Tomasello proposes that the joint communicative intentions of speaker and hearer determine not just what is said but also how it is said — “that language structure emerges from language use, both historically and ontogenetically.” I’ve written a number of posts about Tomasello’s ideas; here is an extended passage from Chapter 5, “Abstract Syntactic Constructions” (with emphases added by me):

*  *  *  *  *

The prototypical paradigmatic linguistic categories, and the only ones that are even candidates for universal status, are nouns and verbs. The classic notional definitions — nouns indicate person, place, or thing; verbs indicate actions — clearly do not hold, as many nouns indicate actions or events (party, discussion) and many verbs indicate non-actional stats of affairs that are sometime very difficult to distinguish from things indicated by adjectives (as in be noisy, feel good, which in different languages may be indicated by either a verb or an adjective)…

Langacker (1987b) has provided a functionally based account of nouns and verbs that goes much deeper than both simplistic notional definitions and purely formal properties. Langacker stresses that nouns and verbs are used not to refer to specific kinds of things but rather to invite the listener to construe something in a particular way in a particular communicative context. Thus, we may refer to the very same experience as either exploding or an explosion, depending on our communicative purposes. In general, nouns are used to construe experiences as “bounded entities” (like an explosion), whereas verbs are used to construe experiences as processes (like exploding). Hopper and Thompson (1984) contend further that the discourse functions of reference and predication provide the communicative reason for construing something as either a bounded entity, to which one may refer with a noun, or a process, which one may predicate with a verb. Importantly, it is these communicative functions that explain why nouns are associated with such things as determiners, whose primary function is to help the listener to locate a referent in actual or conceptual space, and verbs are associated with such things as tense markers, whose primary function is to help the listener to locate a process in actual or conceptual time (Langacker, 1991; and see Chapter 6). After an individual understands the functional basis of nouns and verbs, formal features such as determiners and tense markers may be used to identify further instances.

Relying on the notion of prototypical categories, Bates and MacWhinney (1979, 1982) proposed that early nouns are anchored in the concept of a concrete object and early verbs are anchored in the concept of concrete action – and these are generalized to other referents only later (very similar to the hypothesis that subjects are originally anchored in agents). The problem is that young children use adult nouns from quite early in development to refer to all kinds of non-object entities (such as breakfast, kitchen, kiss, lunch, light, park, doctor, night, party), and they use many of their verbs to predicate non-actional states of affairs (like, feel, want, stay, be; Nelson, Hampson, and Shaw, 1993). Also problematic for accounts such as these, grounded in the reference of terms, is the fact that early in development young children also learn many words that are used as both nouns and verbs, for example, bite, kiss, drink, brush, walk, hug, help, and call (Nelson, 1995). It is unclear how any theory that does not consider communicative function primary — in the sense of the communicative role a word plays in whole utterances — can account for the acquisition of these so-called dual category words. Instead, the developmental data support the view that children initially understand paradigmatic categories very locally and mosaically, in terms of the particular kinds of things particular words can and cannot do communicatively…

Overall, children’s early paradigmatic categories are best explained in the same theoretical terms as their other cognitive categories… [T]he essence of concepts lies in function; human beings group together things that behave in similar ways in events and activities. In the case of linguistic categories such as nouns and verbs, however, it is important to be clear that these are categories not of entities in the world (that is, not referents) but of pieces of language (words and phrases). When words and phrases are grouped together according to similarities in what they do communicatively — grounded in such functions as reference and predication — cognitively and linguistically coherent categories are the result.


Leave a Comment »

No comments yet.

RSS feed for comments on this post. TrackBack URI

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

Blog at WordPress.com.

%d bloggers like this: