Of Minds and Language (39 page)

There was a theory of language acquisition that went with this framework, more or less explicitly, according to which the language acquisition process is
actually a process of rule induction. That is to say, the child, equipped with the notions of UG, has to figure out on the basis of experience what the properties are of the particular rule system pertaining to the language he is exposed to. So there is a process of rule induction, the determination of a particular rule system on the basis of experience.

There were a number of problems with this approach. One had to do with the difficulty of basing comparative syntax on this way of looking at things. What happened was that linguists would write a formal grammar concerning a particular language, and then when they started analyzing the next language, basically they had to start from scratch and write another system of rules that was in part related to the previous one, but it was truly difficult to pull out the properties that the two systems had in common. That was something that I experienced very directly because my first attempt to do syntactic research was basically to adapt to Italian what Richard Kayne had done about French. I came up with a system of formal rules for certain Italian constructions that had a sort of family resemblance to the rules that Kayne had proposed for French, but it was really hard to factor out the common properties (Kayne 1975).

Then, one major problem with this approach had to do with the acquisition model, because there weren't clear ideas on how rule induction would work.

Things changed around the late 1970s with Chomsky's lectures in Pisa (Chomsky 1981),
¹⁹
which gave rise to his 1981 book
Lectures on Government and Binding
, articulating the principles and parameters approach, based on very different ideas. The key notion really became UG, which was construed as an integral component of particular grammars: UG was conceived of as a system of principles which contain some parameters, some choice points expressing the possible cross-linguistic variation; particular grammars could be seen as UG with parameters fixed or set in particular ways. This went with a particular model of language acquisition. Acquiring a language meant essentially setting the parameters on the basis of experience. This is not a trivial task, as a number of people including Janet Fodor, for instance, have observed. In a number of cases the evidence available to a child may be ambiguous between different parametric values, there are complex interactions between parameters, etc. Still, in spite of such problems, parameter setting is a much more workable concept than the obscure notion of rule induction was. And so language acquisition studies blossomed once this model was introduced, and modern comparative syntax really started. For the first time there was a technical
language that could be used to express in a concise and precise way what languages have in common and where languages differ.

Let me just mention for our non-linguist friends a couple of examples. One fundamental parameter has to do with basic word order properties. In some languages, VO languages, the verb precedes the object, as in English, for example,
love Mary
, or in French
aime Marie
. Other languages have OV, Object Verb order: Latin is one case, Japanese is another. If we are to deal with these properties we need at least a principle and a parameter. The principle is Merge, the fundamental structure-building procedure:

(1)    Merge:…A…B…→ [A B]

It basically says “take two elements, A and B, string them together, and you will have formed a new linguistic entity, [AB] in this case.” But then we need some kind of parameter to account for the difference between, let's say, English and Japanese, having to do with linear order. In some languages the head (the verb) precedes the complement, while in other languages the head follows the complement:

(2)    Head precedes/follows complement

This simple ordering parameter has pervasive consequences in languages which consistently order heads and complements one way or the other. So, two examples like the English sentence (3a) and its Japanese counterpart (3b) differ rather dramatically in order and structure, as illustrated by the two trees (4a) and (4b):

(3)    a. John has said that Mary can meet Bill

         b. John-wa [Mary-ga   Bill-ni     a- eru-     to ]     itte-aru

             John-top [Mary-nom Bill-dat meet-can- that ] said-has

English expressions have a fundamentally right-branching structure, Japanese expressions a fundamentally left-branching structure, not the perfect mirror image because certain ordering properties (such as the order subject–predicate) remain constant, but almost the mirror image.

We have broad parameters of this sort, having to do with the ways in which Merge works, and parameters on the other basic operations. The other fundamental operation is Move, so there are parameters on movement. Some languages have properties like Verb Second having to do with the fact that the inflected verb always occupies the second position (German, for instance, has this property), and the parameter basically amounts to the fact that there are two slots in the left periphery of these languages which must be filled by movement, one by the inflected verb and the other by any constituent. A third
kind of parameter has to do with Spell-out. There are certain elements that can or must be left unpronounced in particular configurations in some languages. One classical case is the Null Subject parameter: subject pronouns can be left unpronounced in languages like Italian, Spanish, etc. You can say things like
parlo italiano (‘(I) speak Italian')
for instance, and this property relates in a non-trivial manner to other properties of the language (Rizzi 1982 and much subsequent work).

So the question that arose at some point, after a few years of development of these ideas, was how to express the format of these parameters. Is it the case that anything can be parameterized, or is there a specific locus for parameters? The first idea on the locus for parameters was that parameters were expressed directly in the structure of principles. This was probably suggested by the fact that the first parameter that was discussed in the late seventies had to do with a particular locality principle, Subjacency, the parameterization involving the choice of the nodes that would count as bounding nodes, or barriers for locality (the S/S′ parameter) (Rizzi 1978). On the basis of this case, it was assumed for some time that maybe parameters were generally expressed in principles, and that could be the general format. Among other things, this assumption gave a certain idea on the important question of how many parameters one should expect in UG. As the UG principles were assumed to be reduced in number, if parameters were expressed in the structure of principles one could expect an equally reduced number of parameters.

This view was abandoned fairly quickly, for a number of reasons. One reason was that some principles turned out not to be parameterized. There are certain things that don't vary at all, certain principles do not allow for any sort of variation. In no language, as far as we know, does a structure like the following

(5)    He thinks that John is crazy

allow for coreference between
He
and
John
(principle C of the Binding Theory). That seems to be a general, invariable property of referential dependencies, and many other principles seemed to work like that.

The second reason was that some macroparameters, big parameters initially assumed to characterize basic cross-linguistic differences, turned out to require reanalysis into clusters of smaller parameters. One case in point was the so-called Configurationality parameter. Some languages have a much freer word order than other languages. Originally it was thought that there was a major parameter dividing languages with free word order vs. languages without free word order, essentially. But it quickly turned out that there are different degrees of free word order: some languages are freer in the positioning of the subjects,
others are freer in the reordering of the complements (scrambling), etc. You have a continuum – not in a technical sense, but in the informal sense that there are different degrees of freedom, so that the big “non-configurationality” parameters really needed to be divided into smaller parameters.

The third reason was that some parametric values turned out to be intimately related to specific lexical items. For instance, consider the Long-Distance Anaphor parameter – the fact that certain reflexives roughly corresponding to English
himself
in some languages allow for an antecedent that is not in the same local clause (in Icelandic, for example). This turned out to be the specific property of certain lexical items: if the language has such special lexical items, that is, anaphors of a certain kind, then these anaphors work long-distance. So, we are not looking at a global property of the grammatical system, but simply at the presence or absence of a certain kind of item in the lexicon. These considerations led to the general view that parameters are not specified in the structure of principles, but rather are properties specified in the lexicon of the language. In fact, assuming the fundamental distinction between the contentive lexicon (nouns, verbs, adjectives, elements endowed with descriptive content), and the functional lexicon (determiners, tense, mood, aspect specifications, auxiliaries, complementizers, etc.), parameters could be seen as specifications in the functional lexicon. So, a reasonable format for parameters would look like the following:

(6)    H has F

where H is a functional head, and F is a feature determining the possibility of one of the major operations, either Merge or Move or Spell-out, essentially. This is the general format of parameters that seems to be justified. This view implies important differences with the view expressing the parameters in the principles. For instance, the order of magnitude of parameters is now related not to the number of principles, but to the size of the functional lexicon.

If you take certain approaches, like the cartographic approach (Belletti 2004; Cinque 1999, 2002; Rizzi 2004), assuming very rich functional structures, the implication is that there can be a very rich system of parameters. Much recent work on the cartography of the left periphery of the clause has led to the identification of a rich system of functional heads corresponding to the C (complementizer) domain, a system delimited by Force and Finiteness and hosting positions for Focus, different kinds of Topics, preposed adverbials, operators for the various A′ constructions, etc. (see various papers in Belletti 2004, Rizzi 2004). And the cartography of the IP structure has uncovered a very detailed functional system for the clausal structure, with dedicated heads of Modality, Mood, Tense, Aspect, and Voice; similar conclusions hold for the
structure of major phrases, DPs, etc. (Cinque 1999 and various references in Belletti 2004 and Rizzi 2004). Putting together the theory of parameters, some minimalist assumptions on linguistic computations, and cartography, we end up with something like the following typology of parameters:

(7)    For H a functional head, H has F, where F is a feature determining H's properties with respect to the major computational processes of Merge, Move, and Spell-out. For instance:

So we have parameters determining the capacity of a functional head to undergo merge: what categories does it select; and does it take complements to the left or to the right?
²⁰
And perhaps even more fundamental properties, such as: does the language use that particular functional head? It may be the case that (certain) heads of the cartographic hierarchy may be “turned on” or “turned off” in particular languages.

Then we have Move parameters. Heads function as attractors: they may attract a lower head which incorporates into the attractor, or a phrase which moves to the attractor's specifier. So, does the tense marker attract the lexical verb, as it does in the Romance languages but not in English or most varieties of Continental Scandinavian? Does a head of the complementizer system attract the inflected verb, as in V-2 languages? And does the head attract some phrase to its specifier position, as the C head in V-2?

And then we have Spell-out parameters, having to do with the phonetic realization of the elements involved. Is a particular head overt or not? For instance, the topic head is realized in some languages (one particular use of Japanese
wa
seems to be analyzable along these lines), but not in others (e.g., in Romance Clitic Left Dislocation). And does a head license null dependents? For instance, does the verbal inflection license a null subject? That is one of a number of possible ways of looking at the null subject parameter in current terms.