Functions of Language - Volume 8, Issue 2, 2001

Despite the recognition of secondary ‘modal’ resources such as modal adjectives and adverbs, there has been relatively little discussion of the full extent of their contribution to the expression of modal meaning in general. In this corpus-based study, I focus exclusively on the adverb possibly and describe the range of environments that are revealed by a data set of 2000 randomly selected citations. On the basis of the observed data, I argue for a single core sense of possibly that distinguishes it both from modal operators, such as may/might and from closely related adverbs such as perhaps and maybe. I also argue that, beyond the stereotypical function of possibly as a modal adjunct, there is massive evidence to suggest that it functions additionally as a modalising element in units at the lower rank of group. I therefore propose a revision to the structure of these units to incorporate the expression of modal meaning.

This paper attempts to outline a model for the semantic interpretation of mood and modality in constructions of knowledge or belief, as realized in the finite verb in English. It does so in terms of the attitude(s) which the speaker is seen as presenting him/herself to hold towards the proposition expressed in the clause. The account of such attitudes is based on an epistemic logic, extended to take probabilities into account, and is presented in terms of set theory.The approach takes epistemic mood to be realized either by marking (overt or covert) on the stem of the lexical verb or by the choice of an auxiliary verb. Semantic distinctions between different modal auxiliaries of knowledge/belief are analysed in the same terms as the contrast between the epistemic subjunctive and indicative moods of the finite full verb. One effect of this approach is that the semantics of epistemic mood and epistemic modality is treated as an integrated continuum. Some benefits are claimed to derive from this approach in the conclusion, including that of establishing different types, as well as degrees, of certainty.

A broad view of evidentiality is adopted, based on Chafe (1986) and Haviland (1987) which goes beyond the grammatical marking of the speaker’s or writer’s perceived sources of knowledge and reliability of these sources to encode, not only what the speaker knows and how s/he knows it, but also what can be taken to be an addressee’s state of knowledge. According to this view, evidentials are contemplated as interactive devices or resources for redefining common ground between interlocutors. They go beyond referential content to signal such meanings as confrontation and contradictory assumptions. They are necessarily situated in social contexts and have an indexical function. They may also overlap with epistemic stances and with affect, ranging in the case of surely from surprise, disbelief, doubt and disapproval to persuasion and an invitation to share beliefs or to agree on future courses of action. Using data from the British National Corpus, I analyse a sample of concordances of surely with subject personal pronouns, with the aim of providing a preliminary characterisation of the range of interpersonal attitudes expressed by surely and the determining factors which trigger these apparently contradictory stances.

This paper analyzes different types of deverbal -er nominals and factive nominalizations and argues that they can only be fully described and generalized across if, in addition to their ideational properties, the interpersonal categories which they realize are also considered. It is shown that interpersonal functions such as Subject/person deixis, finite/non-finite grounding and the Mood-relation between them are not exclusively clausal categories, but that they are equally operative at word level and in the nominal group. In factive and -er nominalizations, they set us on the track of the systems’ basic grammatico-semantic characteristics: the link which deverbal -er nominalizations establish between an entity and a process turns out to be strikingly similar to that realized by the Subject and the Finite at clause level; the analysis of the internal, interpersonal properties of that-factives, the fact that-constructions and gerundive factives confirms their downranked or ‘nominal’ nature and enables us to define factivity more accurately.