This article offers a constructional approach to complement-taking mental predicates (CTMPs), e.g. I think , accommodating a whole class of CTMP types ( i think, i suppose, i imagine etc.) and their variant forms (e.g. I would think , I should have imagined ) in a constructional taxonomy. CTMPs are generally believed to depend on their prototypical simple present form in order to convey an epistemic/evidential meaning. Corpus evidence shows, however, that there exist several variant forms that equally function as interpersonal modifications. Such variation has long presented a stumbling block to studies approaching CTMPs from the point of view of grammaticalization theory, since this framework has traditionally been rather inimical to the idea that a grammaticalized item may encompass a paradigm of variant forms and instead requires internal fixation into an unalterable form. It will be argued that CTMPs should be regarded as constructions constituting a taxonomy characterized by several levels of schematicity. It will be demonstrated that the most frequently used CTMP, i think , has reached the highest degree of entrenchment and schematicity, and consequently sanctions the widest range of variant forms, which are disseminated throughout the taxonomy by virtue of analogization.
No takes yet. Share an insight, caveat, or question.
Julie Van Bogaert (2010) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: