MegaOrientation

Which aspects of semantic interpretation are due to predicates’ denotations and which are due to the denotations of their arguments? One way to approach this issue is examining clause-embedding verbs in the context of temporal interpretation, which acts as an indicator of underlying syntactic structures and semantic frames. For example, in (1), is it a fact about a verb like regret or a fact about its subordinate clause that explains why the leaving can be yesterday but not tomorrow? Is it the same for want in (2)?

1. a. #Jo regretted leaving tomorrow.
   b. Jo will regret leaving yesterday.

2. a. Jo wanted to leave tomorrow.
   b. #Jo will want to leave yesterday.

One reason this question is difficult to answer is that a change in embedded clause structure sometimes alters the relevant pattern. For example, in (3a), remember patterns like regret, but in (3b), it patterns like want. In both, remember takes a participial (-ing) clause, unlike (1) and (2).

3. a. Jo will remember leaving yesterday.
   b. #Jo will remember to leave yesterday.

Is this an idiosyncratic fact about the verb remember or a general fact about the structure of its complements?

The MegaOrientation dataset consists of ordinal acceptability judgments for 898 clause-embedding verbs of English with a variety of nonfinite subordinate clause structures. For a detailed description of the dataset, the item construction and collection methods, and discussion of how to use a dataset on this scale to address questions in linguistic theory, please see the references below.

Data

References

Researchers

Aaron Steven White

Ellise Moon