book reviews

No one can seriously undertake a study of linguistics without grappling with the theories and speculations of Noam Chomsky.  From the 1960s on, his influence has been felt in the discipline, and for good reason: Chomsky took nativist philosophy (begun with Plato and expounded by Descartes), dressed it up in scientific language, and succeeded in converting scores of researchers and intellectuals to his cause, all in the heyday of behaviorism.

Reading Chomsky is a chore.  His prose has been described as "dense;" Chomsky packs sentences full of verbiage.  I was able to keep up with him while he discussed philosophy and general issues in linguistics; I got lost when he dove into technical discussions of phrase-markers, sentence structures, diagramming, and rule transformation.  I understood the theoretical underpinnings, but had difficulty following the reasoning.

Chomsky's main presupposition is summed up on the opening page of Aspects:
 

Linguistic theory is concerned primarily with an ideal speaker-listener, in a completely homogeneous speech-community, who knows its language perfectly and is unaffected by such grammatically irrelevant conditions as memory limitations, distractions, shifts of attention and interest, and errors (random or characteristic) in appliying his [sic] knowledge of the language in actual performance (p. 3).

Futhermore, Chomsky asserts that "...no cogent reason for modifying it has been offered" (p. 4).  Chomsky thus distinguishes between competence ("knows its language perfectly") and performance (the actual use of language in concrete situations).

Chomsky defines generative grammar as "a system of rules that in some explicit and well-defined ways assigns a structural definition to sentences" (p. 8).  He is more concerned with syntax than with phonology or semantics.  Generative grammar, unlike traditional grammar, is not prescriptive; it is an attempt to divulge the mental process behind or underneath the speech act.

Chomsky also differentiates between "surface" and "deep" structures.  Surface structures are phonetic; deep structures are semantic.  Thus, the phrases "The boy stole the book" and "The book was stolen by the boy" have differing surface structures (i.e., they do not look or sound the same) but similar deep structures (i.e., the agent and object are the same).  As a counterexample, Chomsky offers two sentences seemingly similar in surface structure, but remarkably different in deep structure: "I persuaded John to leave" versus "I expected John to leave."

Chomsky admits that "there are... very few reliable experimental or data-processing procedures for obtaining significant information concerning the linguistic intuition of the native speaker" (p. 19)  Chomsky confidently asserts, however, that the paucity of the experimental data is no inhibition to theory-building.

Chomsky points out that language acquisition is another problematic aspect of linguistics.  The logical result of Chomsky's train of arguments is that language is innate, in the sense that humans have a "language-acquistion device" through which they discover, within themselves, the universal rules of grammar.

Buried in the thirty-second footnote to the first chapter, Chomsky makes the admission that language acquisition functions are not the sole property of any one section of the mind or brain; in other words, our theory of mind is inadequate to fully explore or understand language acquisition.  This, ultimately, is the strongest ground for criticism of Chomsky's nativist theory.

Educating Eve: The Language Instinct Debate
Geoffrey Sampson, 1997
London: Cassell; 184 pp.

We move from Chomsky to anti-Chomsky.  Sampson carefully dissects the logical and empirical foundations of Chomsky's linguistic nativism, and (in his opinion) finds it lacking.  Premises and refutations include:

1.  Speed of acquisition: infants learn language within only a few years of birth; the speed indicates a strong measure of innateness of a "Universal Grammar."
    Refutation:  Chomsky never proves why 2-3 years is not enough time for a remarkably plastic and learning-oriented brain to acquire a language, or why, logically, we should be "surprised" by a 2-3 year learning time.  In comparison to the speed at which a child can learn to ride a bicycle, 2-3 years might seem like a "long" time.(Ultimately, Sampson dismisses attempts to analogize language learning with any other process; in his mind, language is unique, incomparable to other physical or intellectual tasks.)

2.  Age-dependence: Human ability to acquire a first language "diminishes sharply 'at a relatively fixed age, apparently by puberty or somewhat earlier'" (36).  Lenneberg (a disciple of Chomsky) proved that children's capacity for rapid language learning ended in the early teens.
    Refutation: Lenneberg never distinguished between ability to learn language and general learning ability.  The capacity to intellectually learn any sort of subject matter seems to decrease after puberty, not specific to language.

3.  Poverty of data: The examples of language given to children in their early years are "quantitatively meagre and qualitatively degenerate" and preclude any sort of language learning without strong innate capacities.
    Refutation: The majority of empirical studies support the thesis that Motherese (and other child-directed speech) is "ideal" for language learning, and is in fact highly grammatical.

4.  Linguistic universals: Shared properties of languages suggest an innate language capacity
    Refutation: When it comes down to it, the only truly shared aspects of all languages are nouns and verbs.  Other grammatical structures can vary greatly; no "transformational rules" have been proven to exist in all languages.

Sampson is an empiricist who draws strongly on the theories of Karl Popper, especially in regard to learning as growth through "conjectures and refutations."  Sampson's greatest strength is his continuous reference to counterexamples.  If Chomsky posits a "paucity of data," Sampson is quick to bring up multiple references that suggest otherwise.  If Stephen Pinker (a Chomskyan) points to plurals of words that end in -foot as proof of a "percolation conduit" that distinguishes between "headless" and "headed" compound words, Sampson can readily identify words (Blackfoot, tenderfoot, pinkfoot; commonly pluralized as Blackfeet, tenderfeet, and pinkfeet) that do not seem to fit the rule, as they are used in actual speech.  Sampson rejects Chomsky's (and followers') distinction between competence and performance, instead focusing on the real use of speech by real speakers.  In Sampson's estimation, a theory that cannot account for the exceptions is not viable. The competence / performance dichotomy, as a construct, is an attempt to explain away empirical evidence that contradicts nativist assumptions.
 

The Emergence of Language.
Brian MacWhinney, ed., 1999.
New Jersey: Lawrence Erlbaum Associates; 500 pp.

This volume includes essays covering the emergence of language; its broader purpose is to introduce the reader to theories of "emergentism."  MacWhinney explains that emergentism represents a "middle ground" between strict nativism and empiricism, since each is inadequate to fully explain organismal development.  Emergentism, rather, "...provides accounts in which structures emerge from the interaction of known processes" (x).  In other words, the interaction between environment and organism drives neuronal structure formation and learning, and is key to understanding the genesis or emergence of language in humans.

The essays are fairly technical in parts; familiarity with neuropsychology is assumed. However, I found most of the work accessible after reading other, more fundamental, texts. (This was true of my linguistics experience as a whole; the more I read, the more it started to "click"; I would return to texts that had previously baffled me and finally "understand" what the authors conveyed.)

Example of an essay read:

Elman: "The Emergence of Language: A Conspiracy Theory."

Language is architecturally and chronotopically innate; the sequences of learning are dependent not only on brain plasticity but on developmental and evolutionary time constraints.

Rather than their possession of a language acquisition device, "...it is the children's lack of resources that enables them to learn languages fluently." (Elman, 16)  This stands in direct contrast to Chomsky's assertions about children's innate langugage capacities.

Language is not innate in the sense that a language-specific device exists in the brain.  The process is far more complex; with Hebbsian learning modeling and computerized modeling, we are starting to understand that complexity.
 

Speaking Culturally: Language Diversity in the United States
Fern L. Johnson, 2000
Thousand Oaks: Sage Publications

Johnson’s overall purpose is to dispel myths about American language use.  She looks at language through a lens slightly different from that of standard sociolinguistics.  She focuses on culture as the arbiter of discourse.  Within this framework, Johnson’s tasks are manifold.  First, she outlines the multivariate linguistic environment of the United States, examining census and survey data, and analyzing ethnocultural language use.  Johnson focuses on gendered, African-American, Hispanic, and Asian-American modes of discourse in her analysis of the diverse fabric of American language use.  She carefully delineates the rules of African-American Vernacular English (AAVE), for example, demonstrating its syntactic and semantic complexity that prohibits reduction of AAVE to “slang” or “lingo.”

For me, the most useful portion of Speaking Culturally is the last section on the “consequences and controversies” of cultural linguistic diversity.  Johnson discusses the implications of linguistic diversity in health care, the legal system, schools, and the workplace.  She devotes an entire chapter to the issue of bilingual education
 

further texts skimmed

Reflections on Language
Noam Chomksy, 1975
New York: Pantheon Books; 269 pp.

This book is a collection of lectures given at McMaster University.  Chomsky outlines the basics of his "innateness hypothesis," largely, that "Linguistic theory, the theory of Universal Grammar... is an innate propert of the human mind.  In principle, we should be able to account for it in terms of human biology" (p. 34).

New Horizons in the Study of Language and Mind
Noam Chomsky, 2000
Cambridge University Press; 230 pp.

The ABC's of Languages and Linguistics: A Basic Introduction to Language Science
Curtis Hayes, Jacob Ornstein, & William Gage, 1989
Lincolnwood, Illinois: National Textbook Company; 195 pp.

Encyclopedic Dictionary of the Sciences of Language
Oswald Ducrot & Tzvetan Todorov (Catherine Porter, trans.), 1972
Baltimore: The Johns Hopkins University Press; 380 pp.