On Universal Grammar and Related Matters

Aug 23, 2010 by

Recently, Lera Boroditsky, a psychologist at Stanford and editor in chief of Frontiers in Cultural Psychology, published an interesting article in The Wall Street Journal on the subject of whether language influences the way people see the world. Although much of her own and her colleagues’ research on these issues is fascinating and thought-provoking, some of statements about linguistic theory require me to comment.

For example, Dr. Boroditsky writes:

“the idea [that languages shape the way we think] went out of favor with scientists when Noam Chomsky’s theories of language gained popularity in the 1960s and ’70s. Dr. Chomsky proposed that there is a universal grammar for all human languages-—essentially, that languages don’t really differ from one another in significant ways. And because languages didn’t differ from one another, the theory went, it made no sense to ask whether linguistic differences led to differences in thinking.

This is a misrepresentation or a misunderstanding of the Chomskian theory. Differences across languages are so obvious and have been the focus of linguistic study for so long that for any linguist, Chomskian or otherwise, to say that “languages don’t really differ from one another” is insane. And Universal Grammar is not a “one-size-fits-all” grammatical description that somehow ignores differences across languages. Rather it catalogs both what is universal across languages (i.e., principles) and what is different from one language to the next (i.e., parameters). That both components are important is stressed by the name of the theory: Principles & Parameters Theory. That the differences, i.e. the parameters, are perhaps more worthy of study is emphasized by the more recent name of the theory: the parametric theory of language. In other words, Universal Grammar is not a statement that “languages don’t really differ from one another in significant ways” but rather a catalog of those “significant ways” in which languages differ from each other. Note that what is or isn’t “significant” also depends on one’s approach to language: while researchers like Boroditsky focus mostly on lexical differences, linguists working in the Principles & Parameters framework focus on deeper grammatical differences.

So what are some of these principles and parameters? While there is no exhaustive list of either principles or parameters at the moment, there are many quite well established examples of both. To start with the principles, one such universal principle may be that all languages have such “parts-of-speech” categories as nouns, verbs and adjectives. Although some researchers have proposed that certain languages may lack one of these categories, Mark Baker in his 2002 book Lexical Categories: Verbs, Nouns and Adjectives argues quite convincingly that this is simply not true. According to him, all languages have at least a few nouns, verbs and adjectives although any of the three categories may be quite limited in a given language. For example, there are numerous languages that have only a restricted set of adjectives, but nonetheless all languages have at least a few, as Baker shows.

When it comes to parameters, many have been formulated and reformulated during the last three decades. Indeed, the work focusing on linguistic parameters has proven very fruitful throughout the years. According to Mark Baker, “parameters… play the same foundational role in scientific theories of linguistic discovery that atoms play in chemistry” because “these parameters combine and interact with each other in interesting ways to create the wide variety of languages that we can observe around us”.

One important notion regarding parameters is that they allow to account for the bounds of variation across language. It was noticed that certain properties in language tend to go with certain other properties: for example, if the object typically precedes the verb, then nouns also precede adpositions (i.e., the given language has postpositions rather than prepositions) and auxiliary verbs follow main verbs. Languages of this type include Udmurt (Finno-Ugric), Telugu (Dravidian), Hindi (Indo-European), Basque (isolate), Japanese (Japonic), Turkish (Turkic), Georgian (Kartvelian), Amharic (Afro-Asiatic) and others. Conversely, in languages where the object typically follows the verb, nouns follow adpositions (i.e., the language has prepositions rather than postpositions) and auxiliary verbs precede main verbs. Languages of this type include English, Spanish, Russian, Breton (all four Indo-European), Chichewa (Niger-Congo), Fijian (Austronesian), and others. Note that these typological groupings do not follow the familiar language family divisions but cut across them.

This contrast between the two types of languages has been formulated as the Headedness Parameter, referring to whether the “head” of a phrase (such as the verb, the adposition or the auxiliary) precedes or follows its “complement” (i.e., the object, the noun phrase or the main verb phrase). The important point here is that the we do not get a “mix-and-match” situation across languages where a given order of, say, the verb and its object goes with any order of noun and adposition equally well. In other words, a parameter is like a switch: it can be either “on” or “off”, “yes” or “no”, “right” or “left” (and most parameters described to date are indeed binary in the sense that they offer only two choices, called “settings”). For instance, the Headedness Parameter may be set as either “left” or “right”, referring to whether the head is to the left (i.e., precedes) or to the right (i.e., follows) its complement. But whatever the setting of the parameter is, it must apply across the different categories (verbs, adpositions, auxiliaries). And this is why we do not get Object-Verb languages with prepositions or Verb-Object languages with postpositions (at least not as often as the other combinations).

However, the workings of one parameter may be obscured by the workings of another parameter (or other parameters), thus creating an illusion of exceptions to the clear-cut picture of language variation. For example, German is considered to be a right-headed language: consider Martin hat das Buch gelesen, literally ‘Martin has the book read’. Here, the object das Buch precedes the (lexical) verb gelesen (here, in a participial form). The same is true in Martin kann das Buch lesen, literally ‘Martin can the book to.read’ or Hans sagte dass Martin das Buch gelesen hat (literally, ‘Hans said that Martin the book read has’).

But what about the following sentence: Martin las das Buch (literally, ‘Martin read the book’)? Here, the object das Buch follows rather than precedes the verb las. Is German suddenly not a right-headed language anymore? The answer is no. But the setting (or effect) of the Headedness Parameter is obscured by the setting (or effect) of another parameter, the Verb-Second (or V2, for short) Parameter.

This parameter refers to whether or not the finite/tensed verb must come in the second position in its clause or not. In German (as well as in Dutch, Swedish and Norwegian) this parameter is set as “yes” and in English it is set as “no”. Thus, if a sentence in one of the V2 languages starts with anything other than the grammatical subject — for example, with an adverb like ‘yesterday’ — the finite verb must follow it and precede the subject, as in the German Gestern las ich das Buch or the Swedish Igår läste jag denna bok (both literally translating as ‘Yesterday read I the book’). In a non-V2 language, the verb would follow the subject regardless of whether anything (like an adverb) comes first or not. Hence, in English we say Yesterday I read the book, not Yesterday read I the book.

For more on Verb-Second in Icelandic, read here and on Verb-Second effects in a non-V2 language English, read here.

In the following postings we will look at some other parameters and see how the parametric theory allows us to account not only for variation among human languages, but also for language change and language acquisition by children.

Subscribe For Updates

We would love to have you back on Languages Of The World in the future. If you would like to receive updates of our newest posts, feel free to do so using any of your favorite methods below: