The Acquisition of Modality: Implications for Theories of Semantic Representation

The set of English modal verbs is widely recognized to communicate two broad clusters of meanings: epistemic and root modal meanings. A number of researchers have claimed that root meanings are acquired earlier than epistemic ones; this claim has subsequently been employed in the linguistics literature as an argument for the position that English modal verbs are polysemous (Sweetser, 1990). In this paper I offer an alternative explanation for the later emergence of epistemic interpretations by linking them to the development of the child’s theory of mind (Wellman, 1990); if correct, this hypothesis might have important implications for the shape of the semantics of modal verbs. It is widely acknowledged in the linguistic literature that modal expressions may be used to communicate at least two broad clusters of meanings: epistemic modal meanings, which deal with the degree of speaker commitment to the truth of the proposition that forms the complement of the modal, and deontic modal meanings, concerned with the necessity or possibility of acts performed by morally responsible agents, e.g. obligation and permission (Lyons, 1977; Kratzer, 1981; Coates, 1983; Palmer, 1986, 1990; Sweetser, 1990; Bybee and Fleischman, 1995). The utterances in (1) and (2) (on their preferred interpretations) are examples of epistemic and deontic modality respectively: (1) (a) You must be John’s wife. (b) That may be the postman. (on hearing the doorbell) (c) If you are interested in studying the mind, lectures in Linguistics should prove interesting. (2) (a) I must go on a diet soon. (b) You may leave the room only after having signed these papers. I wish to thank Deirdre Wilson, Neil Smith and an anonymous referee for Mind and Language for comments and suggestions. I am also grateful to the audience of the 1997 Autumn Meeting of the Linguistics Association of Great Britain, where I first presented my thoughts on the acquisition of modality. The research for this paper was partly supported by a grant from the A.G. Leventis Foundation. Address for correspondence: Department of Linguistics, University College London, Gower Street, London WC1E 6BT, UK. Email: annaKling.ucl.ac.uk The Acquisition of Modality 371 (c) Full credit should be given to the city authorities for having done so much for the environment. Apart from the epistemic/deontic distinction, a third main area of modal meaning is often recognized: so-called dynamic modality, which includes the notional categories of real-world ability, possibility and intention/ willingness (von Wright, 1951; Palmer, 1990): (3) (a) My son can speak four languages. (b) I will become the best skier in the world. Normally, deontic and dynamic uses are grouped together under agent-oriented modalities (to be distinguished from speaker-oriented, i.e. epistemic, modalities—cf. Bybee and Pagliuca, 1985; Bybee, Perkins and Pagliuca, 1994; Bybee and Fleischman 1995), or root modalities (Bybee, 1988; Sweetser, 1988, 1990; Traugott, 1989). In the present discussion, I will adopt the root/epistemic distinction, although this does not always correspond to the terms used in the literature on modality.1 An interesting fact about the root and epistemic types of meaning is that they often tend to be expressed by a single class of modal expression in the languages of the world (for cross-linguistic evidence, see Fleischman, 1982; Perkins, 1983; Traugott, 1988; Traugott and König, 1991; Bybee, Perkins and Pagliuca, 1994). Probably the set of items for which this claim has been most extensively illustrated is the set of English modal verbs (for an overview, see Palmer, 1990), mainly must, may, might, can, could, should, shall/will, would, ought to; the rest of my discussion will focus on these as the best-documented area of research on modality. Modal verbs have been semantically analysed in two main ways: one strand considers them ambiguous between root and epistemic meanings (Palmer, 1990; Coates, 1983); another assigns to them a unitary semantics, which is pragmatically developed into epistemic or root interpretations in the process of utterance comprehension (Kratzer, 1977, 1981; Perkins, 1983; Groefsema, 1995; Papafragou, 1998a, 1998c). A purely ambiguous approach to the English modals can quickly be discredited because of its lack of machinery with which to motivate a connection between the root and epistemic clusters of modal meanings. On the other 1 The reader should thus bear in mind that most of the psycholinguistic literature I am going to review uses the term ‘deontic’ loosely to cover dynamic uses as well; I will be consistent in my use of ‘root’ for both cases, unless the arguments bear directly and solely on deontic modality. A second disclaimer is also called for: I am aware that the root/epistemic distinction is a crude one, since I myself have argued elsewhere for the existence of other modal categories which do not fit this dichotomy (e.g. the alethic modality of logicians—see Papafragou, 1998a). What is more, one may distinguish different types of modal meanings even within the root or the epistemic domain. Still, the distinction will do for present purposes: after all, it is difficult enough to design experiments to test the acquisition of these broad categories, let alone more fine-grained modal meanings. I will have something to say on alethic modality in section 2.3.  Blackwell Publishers Ltd. 1998