Proceedings MTT 2011

There is a certain consensus on some defining features of collocations (they are frequent, recurrent, conventional, institutionalized, etc.), but there is not much agreement on other characteristics, for example whether or not they constitute binary lexical relations, and — crucially— whether they are arbitrary, and must be stipulated, or should instead be deduced or inferred. I will argue that collocations may belong to one of these two groups. Since most collocations are presented as basic or primary pieces of lexical information in most analyses, I will concentrate on those belonging to the second group, which form, I will argue, a rather large paradigm. In MTT, collocations are accounted for through lexical functions (LFs): collocates provide values for keywords, which stand for bases. Although LF values often illustrate semantic groups or classes, these are rarely mentioned in the literature on LFs. Here is an example: if we introduce a LF such a «Magn(increase) = substantially», we will certainly miss a generalization. The adverb substantially typically modifies change of state verbs, most of them belonging to the group of so-called “degree achievement predicates” or “gradual completion verbs”. This adverb modifies a large number of verbs denoting increase (enhance, enlarge, exceed, expand, go up, grow, improve, increase, overshoot, progress, raise, upgrade, etc.), decrease (cut back, decline, degrade, diminish, discount, dwindle, fade, fall, lessen, limit, minimize, moderate, reduce, shrink, slow, tighten, weaken, etc.) and the very notion of change (alter, amend, change, deviate, differ, diverge, fluctuate, modify, redefine, reformulate, revise, rewrite, transform, vary, among many others). To these groups we may add those of comparative adverbs (more, less), comparative adjectives (higher, lower, greater, older and all the other members of this paradigm) and adjectives denoting (un)equality (different, similar, identical). Leaving aside some other minor lexical groups unrelated to these, we may introduce two reasonable hypotheses: a. b. The correspondent LFs should involve these classes, rather than specific lexical items. That is, we would have a formulation such as “Magn(VERBS OF INCREASING) = substantially”, to which we would add the LFs based on the other groups. The semantic features that allow us to make these groups, or some other version of them, are part of the linguistic knowledge of native speakers of English.