The previous post asked about the variation of verbs. Treating verbs, as we do in logic, as predicates, and nouns as arguments, we have intransitive verbs as one-place predicates (one argument), transitive verbs as two-place predicates, and ditransitive verbs as three-place predicates. One of these arguments is the external argument (that is, the argument which is not within the VP), and the rest are internal arguments. In syntax, the internal argument structure (what arguments are allowed to take) is represented by a subcategorization frame, and those frames will fall into classes.
put: [V NP PP] – ditransitive
throw: [V NP (PP)] – transitive optional ditransitive
defeat: [V NP] – transitive
eat: [V (NP)] – optional transitive
sleep: [V] – intransitive
But we do not just want to posit these subcategorization frames with no structure or theory. If this grammar is to fulfill our third feature, psychological possibility, it should consist of a few primitives that can be generalized to other things. After all, it is psychologically unrealistic to say that the human brain has an innate template for each individual verb. And if we take the less modest approach, to simply say that the frames are innate primitives, we still need some explanation of how verbs are sorted into frames (which will require some explanation).
Intuitively, this should be easy: verbs like defeat, for example, require two participants: one to be defeated, and one to do the defeating. These are easily sorted into our previous stipulation of internal and external arguments, and provides a rubric for sorting verbs: if you know the participants involved, you know the argument structure, and vice versa.
Unfortunately, this account hits a snag. Eat and devour or bake and microwave all seem to have similar argument structures: they have a do-er and a do-ee. How then do we explain the following data? The *(X) notation means that the sentence is ungrammatical without X, and the ?(X) notation means the sentence sounds off, if not completely ungrammatical, without X.
(17a) I ate (the fish)
(17b) I devoured *(the fish)
(17c) Stacy baked (bread)
(17d) Stacy microwaved *(the pop-tart)
These verbs, as well as verbs like ‘explode’ (some speakers seem to think it optionally transitive, others seem to see it as intransitive), have been used as evidence that while argument structure may be constrained in some ways by the conceptual participants of the verb, it cannot be forced. I could not disagree more strongly. However, I am going to leave this question unanswered for now, neither because I intend to just ignore it, nor because I have no interest in it. Far from that, I plan on writing my topic on a similar topic. I will write a short blurb up on the topic in the next day or two, but you’ll have to wait until near I graduate to see a full proof, as well as what I think it means. But for future writings, I will assume that the conceptual participants of the verb have a causal relation with the argument structure of the verb (although, there will be other factors as well)
I will end this post with a question: What determines which arguments are internal and which are external?