Modulating grammar trough modality : a discourse approach.

ELIA I, 2000 119 MODULATING GRAMMAR THROUGH MODALITY: A DISCOURSE APPROACH* Francisco Gonzálvez García Universidad de Almería This paper argues for a dynamic, discourse-based approach to modality which departs from the traditional, monolithic account of modality in a number of interesting respects: (i) modality is characterized as conveying the speaker’s involvement in the propositional content of a given utterance (either in the form of agency or subjectivity) (ii) modality can be said to ramify across the whole lexico-grammatical architecture of the language, (iii) many of the meanings usually ascribed to individual modal verbs are in fact derived either from the verb’s sentential environment or from some wider context of utterance, and (iv) modal meaning in discourse can be said to arise out of the interaction of two closely connected layers of meaning: one embracing the inherent linguistic meaning of the modal verb in conjunction with that of other neighbouring modal devices, and another concerned with principles connected with politeness and face-saving strategies. 1. Introduction Modality in English and other languages has been the focus of attention of scholars from distinct disciplines and approaches over the last thirty years. Within linguistics, the study of modality has witnessed a gradual shift from a monolithic, static conception to a more dynamic understanding of modality taking into account the relevance of linguistic and extralinguistic contextual factors in the production and interpretation of modal utterances in discourse (Bybee & Fleischman 1995) or the creation of modal textual * I would like to express my gratitude to Paul J. Thibault (University of Venice) and Judith A. Carini Martínez (University of Almería) for most helpful discussions on the ideas presented in this paper. I would also like to thank the ELIA editorial board for useful comments and suggestions. All usual disclaimers apply. F. Gonzálvez García 120 coherence (Lundquist 1989). This paper seeks to shed some light on a number of conspicuous issues within this relatively new discourse-oriented program of research into modality, which can be briefly summarized as in (1)-(5) below: (1) How can modality be aptly and comprehensively defined on both semantic and pragmatic grounds? (2) What linguistic devices can be considered carriers of modal meaning? (3) What are the means whereby each modal device modulates the speaker’s involvement which is interpreted from the utterance of that sentence in context? (4) How can the modal meaning in general and its dynamic relation to the context of utterance in particular, be adequately accounted for from a theoretical and descriptive standpoint? (5) How can we characterize the enriching (reinforcing or cumulative) effect of contextual factors in the unfolding of “modal” meaning/s in a given piece of discourse? 2. The definition and scope of modality: a question of “attitude” The extensive bulk of literature on modality shows a clear tendency among linguists1 to accept that semantic areas such as possibility, necessity and prediction (knowledge or epistemic “modality”), on the one hand, and “permission”, “obligation” and “volition” (“deontic” or “root” modality), on the other, constitute the domains of modality. A non-concomitant view with the mainstream conception of modality was already advanced by Halliday (1970:349) in the following terms: “Modality....is the speaker’s assessment of probability and predictability. It is external to the content, being part of the 1 See, for instance, Lyons (1977, 1983, 1994), Hermerén (1978), Coates (1983), Perkins (1983) and Palmer (1986, 1990), among others. ELIA I, 2000 121 attitude taken up by the speaker”.2 This “reductionist” view of modality will not be invoked in this paper, since it obscures the fact that deontic modality can nevertheless be regarded as “a form of participation of the speaker in the speech event” (Halliday 1970:335) and plays a significant role in the interpersonal process of negotiation of meaning.3 Instead, the following definition will be adopted, along the lines suggested by Lyons (1977,1983,1994), where the term “attitude” has been expanded into that of “subjectivity”4 understood as “subject/speaker’s involvement” so as to embrace both types of modality: Subjectivity is a matter of speaker’s, or more generally, of the locutionary agent’s involvement of himself in the utterance. In the case of epistemic modality what is involved is his knowledge (or beliefs). In the case of deontic modality it is his will and authority that is involved. But in both cases it is the locutionary agent who is the source of the modality (Lyons 2 Therefore, modality in systemic grammar is restricted to what is called “epistemic” modality, and what is often referred to as “root” or “deontic” modality is regarded as forming a different (though obviously connected) category of its own, namely, “modulation”, the reason being that the latter “...are not speaker’s comments on the process referred to...” (Halliday 1970:338). 3 In this respect, two more specific criticisms can be levelled against Halliday’s restrictive formulation of modality: (i) the fact that a clear-cut distinction between both types of modality can be challenged at an ideational level, since cases of indeterminacy (“merger”) may occur (Leech & Coates:1980, Coates:1983, Bald 1990:354, among others); and (ii) both epistemic and deontic modality may be geared towards the interpersonal component of the language (i.e. politeness strategies in general and face-saving in particular). We shall have more to say about this at a later stage in this paper. 4 Palmer (1986:16) also argues that subjectivity is an essential criterion for modality, which he defines as “the grammaticalization of speaker’s (subjective) attitude and opinions.” In our definition of modality, we shall leave out the “grammaticalization” component as it belongs to the domains of mood rather than modality. F. Gonzálvez García 122 1983:111).5 Therefore, it can be claimed that modality is concerned with the expression of the subject/speaker’s involvement towards the propositional content of an utterance, whether in the form of agency or subjectivity.6 An interesting conclusion ensuing from the above definition is that modality need not, and should by no means, be exclusively restricted to modal auxiliary (or semi-auxiliary) verbs. Over the last fifteen years, linguists have concentrated on surveying a more or less comprehensive inventory of modal elements,7 a representative sampling of which can be said to include (i) adjectives (e.g. “possible”, “necessary”, etc.), (ii) participles (e.g.“alleged”, “demanded”, etc.), (iii) 5 Hoye (1997:43) argues that “there are problems, however, with subjectivity when it comes to an analysis of deontic modality, for here there are varying degrees of speaker involvement”. Even when the modality is clearly subject-oriented, the nonfactual status of the utterance can be questioned, as in: “John can sing in Romanian”). In much the same vein, Hübler (1983) explicitly rejects that semantic notions such as “ability”, “volition” or “futurity” should be treated as involving modality. This is why linguists like Palmer (1986:102ff, 1990:35ff) have posited a third type of modality known as “dynamic modality”. Unlike epistemic and deontic modality, dynamic modality is not subjective and is subjectrather than speaker-oriented: the subject’s ability or willingness is at issue, not the speaker’s attitudes or opinions. This three-fold distinction will be retained in this paper. 6 The term “speaker” should be understood here as a general semantico-pragmatic function, in other words, as the subject of modality, “...the one who invests in the Subject as the modally responsible element in the clause and in relation to whom the Finiteness of the proposition is organized” (Thibault 1995:64). 7 See, for further reference, Bailey (1981), Perkins (1983) and Hermerén (1986), among others. ELIA I, 2000 123 nouns (e.g. “necessity”, “consideration”, etc.),8 (iv) lexical verbs (e.g. “wonder”, “order”, etc.), (v) adverbs (e.g. “perhaps”, “obviously”, etc.), (vi) articles (e.g. “the”/ “a”, “some”/ “any”, etc.),9 (vii) tense (usually in “preterite” or marked forms) (e.g. “I thought you in Paris”, “In 1492, Columbus discovers America”),10 (viii) aspect (marked forms) (e.g. “John is having a headache”),11 (ix) particles (e.g. “if”) and truth-functionally equivalent devices (e.g. subject-verb inversion, putative “should”) (e.g. “If you don’t like it 12/ should you not like it, that’s your hard luck”), (x) the degree of morphosyntactic compression in the encoding of a sentential complement (finite, non-finite or verbless clause) (e.g. “They consider that 8 Jacobson’s (1982) study on modal nouns shows, among other things, that there is a much greater variety of lexical items in the field of modality nouns than in that of modal auxiliaries. This finding points to one of the many inconveniences of treating modal verbs as the exclusive or main carrier of modal meaning. 9 Lyons (1983) provides an interesting outline of the many connections between “demonstrative” and “indicative” and Rauch (1983) also further explores the relationship between modality and deixis in general. Further research will have to address how concepts such as anaphora, generality, individuality (and their grammatical correlates such as “number”) relate to modality. 10 Lyons (1983:104) justifies the modal treatment of this marked form (ie. historic present) on the grounds that “it conveys an impression of vividness and subjective involvement”. Kress (1977), Hutchinson (1985:1), Fleischman (1989:7) and Kvacanovic (1994:67ff), however, contend that all tense forms imply a given degree of speaker’s distance from and attitude towards a given proposition and can therefore be seen as modalized forms. 11 Lyons (1983:101) argues that a dynamic, agentive reading of this sentence is feasible given an adequate supporting context (eg., “John is simulating a headache or doing something to bring it on). It must be borne in mind that, for Lyons, agency is superimposed upon the notion of subjectivity. The exact relationship between subjectivity-agency as the main axes of modality remains to be further explored from a linguistic as well as a socio-semiotic viewpoint. 12 Perkins (1983:110) lumps together all types of “if”-clauses (nominal and adverbial) as modal.Under the view of modality taken here, contextually motivated variants such as indirect speech placed in the past are treated as non-modal. See also Hübler (1983:134) for a position along the same lines. F. Gonzálvez García 124 John is intelligent/John to be intelligent/John intelligent”, etc.),13 (xi) hedging expressions (e.g. “She is a great scholar, I think”), (xii) “emphatic” do (e.g. “Do be getting sleepy!”), (xiii) “get”-passives (e.g. “He got elected Class President”),14 (xiv) question tags (e.g. “You liked the movie, didn’t you?”), (xv) “yes”/ “no” questions (e.g. “Have you ever been to London?”),15 (xvi) (rising) intonation (e.g. “I feel happy”),16 and so forth. The above inventory, by no means complete, is intended to illustrate how varied and sophisticated the modality system in English is, thus fully endorsing Halliday’s (1970:331) statement that “...there is (thus) no single place in the clause where modality is located”. However, the linguistic system of modal resources is further enriched by their numberless combinatory possibilities in a given piece of discourse. In the words of Halliday (1970:331): Nor do the different non-verbal forms of the same lexical item necessarily correspond with each other: “obviously” is not the same as “it is obvious that...”, “surely” as “I am sure that”. But there are discernible groupings, and 13 See Ransom (1977) and Gonzálvez (1997) for an account of the embedded complements as having modal meaning. Differences exist as to where the locus of the modal meaning should be established. Thus, while for Ransom (1977) the embedded complements have a modal meaning of their own and independent of that of their embedding predicates, Gonzálvez (1997) proposes that the modal meaning arises out of a complex, dynamic interaction of the meaning and form properties of the morphosyntactic encoding of the embedded clause with those of the embedding predicate. 14 Bailey (1981:168), quoting Bolinger (p.c.) and Granville Hatcher (1949), argues that “get”- passives are characterized, among other things, by having more saliently affected subjects than their “be”-counterparts, which motivates their being treated as subjectively modalized. 15 Together with questions, markers of negation are likely candidates for a modal treatment. See Palmer (1979) for further discussion on the issue of the relationship between “nonassertion” and modality,. 16 See, for further reference, Stevenson (1967), Crystal (1969, 1975), and Roach (1991:164ff), among others. The latter is particularly insightful for a better and comprehensive understanding of the attitudinal function of intonation in general. ELIA I, 2000 125 a clear distinction can be drawn between pairs which are felt to be equivalent, and thus reinforce each other (“as concord”) when both are present, as in “Perhaps he might have built it”, and those which are not equivalent and are thus cumulative in meaning, as in “Certainly he might have built it” (“I insist that it is possible” or “I grant that it is possible”).17 An interesting conclusion ensuing from the above is that modal meaning can be said to resolve itself into a more or less intricate flow of reinforcing or cumulative modal devices which invades the whole lexicogrammatical architecture of the language, always subject to the communicative requirements of the interlocutors in the modalised space-time if dialogic negotiation.18 3. Towards a “dynamic”, context-based view of modal meaning Traditional (and not so traditional) accounts of modality present in our view two important shortcomings when tackling the issue of modal meaning: (i) they fail to separate the intrinsic linguistic meaning of modal verbs from the linguistic meaning of other neighbouring modal items (of different kinds) in the sentential environment in which they are inserted, and (ii) they also fail to distinguish the meaning of modal verbs from the (endless) number of pragmatic uses or functions these may be put to in specific contexts. 17 See Lyons (1977:Chapter 17) for a similar position in the form of a distinction between modally harmonic and non-harmonic meaning. 18 In addition, it must be borne in mind that modality may be implicated by the semantic structure of the whole utterance without being signalled by any specific (linguistic) markers. Thus, Costa (1975:119-21) (quoted in Brown & Levinson (1978:274), notes that English statements like “One just doesn’t do things like that” can implicate the deontic modality interpretation (“You must not do things like that”). In our view, the discourse approach to modality invoked in this paper can prove useful to help to determine to what extent covert modality may be implicated with the aid of neighbouring (linguistic) devices in a given discourse scenario. F. Gonzálvez García 126 The consequences to be derived from this cannot possibly be expected to be advantageous from either a theoretical or descriptive view. Thus, as Walton (1991:367) rightly argues, “...the meanings that are ascribed to modal verbs multiply arbitrarily as more and more context is added, and any reference grammar aimed at recording these meanings will come to resemble a lexicon and still fail to cover them all”. Let us briefly illustrate the relevance of the above statement with particular relevance to the taxonomy of modal meanings for “might” in three well-known works on modality: Leech (1971), Coates (1983) and Hoye (1997).19 Of the three, the taxonomy proposed by Hoye (1997) is the one coming closer to the line of research into modality invoked here, since it refrains from attributing the meaning of items in the sentential environment of a sentence (i.e. a hypothetical conditional clause, a concessive clause, etc.) to the modal verb itself (“hypothetical might”, “concessive might”), or even lumping together the inherent meaning of the modal with some of the specific pragmatic functions this verb may fulfil in an also specific context of 19 Leech (1971) comes up with something like six different meanings for “might”, namely, (i) past tense of “may”, (ii) a “subjunctive substitute” (in clauses of purpose and concession), (iii) hypothetical past tense, (iv) “polite permission”, (v) possibility and, (vi) “tentative suggestion”. Coates (1983:147-8) makes them eight: (i) epistemic possibility, (ii) past of “may” (“epistemic possibility”), (iii) past of “may” (“root possibility”), (iv) past of “may” (“root permission”), (v) remote of “may” (“epistemic possibility”), (vi) remote of “may” (root possibility) and (vii) remote of “may” (“root permission”). Finally, Hoye (1997:274-5) reduces them to three only: (i) epistemic probability/possibility, (ii) non-epistemic (a blending of possibility and permission) and (iii) collocational combinations with “well” “to convey epistemic probability”, with “just” in “...formulaic expressions of politeness (usually to ask for permission)” and, finally, with “(just) as well” to “make a circumspect or sardonic recommendation” (Hoye 1997:275). ELIA I, 2000 127 utterance (e.g. “polite permission/suggestion”, etc.).20 The question now arises as to how modal meaning can be felicitously systematized in order to account for the dynamic, though nevertheless systematic, relationship between modality, the context of utterance and the interpersonal function of politeness.21 Following Silva Corvalán (1995:73ff), we shall argue for a monosemantic approach to the semantics and pragmatics of modality drawing on intersecting, polysemantic contexts. More specifically, it will be suggested that modal meaning can be shaped and further modulated in terms of (at least) two separate, though closely connected, layers of meaning:22 LAYER 1: (i) SYSTEMATIC MEANING (ie., the meaning present in all uses of a modal) in conjunction with (ii) CONTEXTUAL (LINGUISTIC) MEANING (i.e. the reinforcing or cumulative modal nuances introduced by other satellite elements in the neighbouring sentential (or discourse) environment. LAYER 2: PRAGMATIC (OR INTERPERSONAL) MEANING (i.e. the modulation or fine-tuning on the part of interlocutors of the meaning 20 However, in our opinion, Hoye’s account is not without problems either, since it fails to account for the relationship between the possibility and permission readings of “might”, on the one hand, and the (metalinguistic) function of politeness, on the other, in a neat, systematic way. More specifically, Hoye’s account misses the important fact that the politeness factor is not a mere appendix of modality, which can be traced in specific formulaic expressions with some collocations, but it should rather be seen as an overarching pragmalinguistic function that can be overtly or covertly marked in deontic, epistemic and dyanmic modal utterances, rather than just in the former alone, as Hoye’s taxonomy seems to imply. 21 Politeness is used henceforth in a sense compatible with that of Brown & Levinson (1978), though it is taken to encompass not only linguistic but also those socio-semiotic aspects of communication. 22 The following schema departs from the original presented in Silva Corvalán (1995) in that it replaces the prototypical discourse meaning component with what has been referred to above as pragmatic (or “interpersonal”) meaning. F. Gonzálvez García 134 Inventory of Forms” in G. Tottie (ed.) English in Speech and Writing. Uppsala. Hoye, L. 1997. Adverbs and Modality in English. London: Longman. Hübler, A. 1983. Understatements and Hedges in English. Amsterdam/Philadelphia: John Benjamins. Hutchinson, C. 1985. “The Where of When” Some Arguments for a NonTemporal Interpretation of Tense”, University of East Anglia Papers in Linguistics 21-22:1-31. Jacobson, S. 1982. “Modality Nouns and the Choice between to-infinitive and of +-ing”. Studia Anglica Posnaniensia 15: 61-71. Kavacanovic, N. 1994. “The Preterite in English: Tense or Modality?”. Facta Universitatis 1(1): 65-70. Kress, G.R. 1977. “Tense as Modality”. University of East Anglia Working Papers in Linguistics, 40-52. Leech, G. N. 1971. Meaning and the English Verb. London: Longman. Leech, G. & J. Coates. 1980. “Semantic Indeterminacy and the Modals”, in S. Greenbaum et al. (eds.), Studies in English Linguistics: For Raldolph Quirk. London/New York: Longman. Lundquist, L. 1989. “Modality and Text Constitution” in M.Conte, J.S. Petöfi & E. Sözer (eds.) Text and Discourse Connectedness (Proceedings of the Conference on Connexity and Coherence, Urbino, July 16-21, 1984). Amsterdam/ Philadelphia: John Benjamins. Lyons, J. 1977. Semantics, Vol. II. Cambridge: C.U.P. Lyons, J. 1983. “Deixis and Modality”. Sophia Linguistica 12: 77-117. Lyons, J. 1994. “Subjecthood and Subjectivity” in Proceedings of the Colloquium “The Status of the Subject in Linguistic Theory”. Paris: Institut Francais du Royaume-Uni. Oka, H. 1981. “Function of Modal Auxiliaries in Requests”. Studies in English Language and Literature 31: 77-105. Palmer, F.R. 1979. “Non-Assertion and Modality” in D. J. Allerton, J.Carney, & D. Holeroft (eds.) Function and Context in Linguistic Analysis. Cambridge: C.U.P. Palmer, F.R. 1986. Mood and Modality. Cambridge: C.U.P. Palmer, F.R. 1990. Modality and the English Modals. London: Longman. Perkins, M. 1983. Modal Expressions in English. London: Frances Pinter. Ransom, E. 1977. “On the Representation of Modality”. Linguistics and Philosophy 1: 357-379. ELIA I, 2000 135 Rauch, I. 1983. “On the Modality of the Article”. Monatshefte 75(2): 156162. Roach, P. 1991. English Phonology and Phonetics. Cambridge: C.U.P. Silva Corvalán, C. 1995.“Contextual Conditions for the Interpretation of “Poder” and “Deber” in Spanish” in J. Bybee & S. Fleischman (eds.) Modality in Grammar and Discourse. Amsterdam/Philadelphia: John Benjamins. Stevenson, E.W. 1967. “Intonation Patterns and Modality”. Rocky Mountain (Review of Language and Literature) 21: 94-95. Thibault, P.J. 1995. “Mood and the Ecosocial Dynamics of Semiotic Exchange” in R. Hasan & P. Fries (eds.) On Subject and Theme. Amsterdam/Philadelphia: John Benjamins. Walton, A. 1991. “Modality and the modals in traditional grammars of English” in G. Leitner (ed.) English Traditional Grammars: An International Perspective. Amsterdam: John Benjamin J.M. Martín Martín 136