Although namely different, the three approaches to genre display more similarities than discrepancies.In all the three approaches, genre is explored through activities, ways of expression, membership, roles of participants and content organization, though focus on these aspects may vary in them, sometimes in terms of cognition and at other times with regards to pragmatic needs.As a result of comparison, it can be found that SFL defines genre concisely and comprehensively (as represented by Martin, 1997).However, the other two approaches (i.e.the rhetorical structure approach and the communicative approach）complement the SFL approach in analyzing some specific genres, for instance, academic writings.In short, the three approaches should be seen as a matter of different perspectives or focuses rather than a matter of “to be or not to be”.
Bhatia (1993/2004）proposes the multidimensional and multi-perspective analyses of professional and institutional genres, holding that generic variations are as common as generic integrity on many occasions, as reflections of real-life needs, for example, the promotional nature of some genres like advertisement, book reviews, job applications, book blurbs, and public campaigns.Bhatia (ibid）argues that private intentions are an important cause for generic variations.
It is not hard to find that none of these approaches deviates from field, tenor, and mode proposed by SFL about register, whose super-ordinate is supposed to be genre.Thus, it is proper to study genre under the framework of SFL with complementary contributions from the other two approaches.Also, it should be noted that tenor-the interpersonal role concerned is a core part of genre analysis for each of the approaches.As modality has much to do with the interpersonal metafunction, it is important to think about the relationship between genre and modality.
3.3.1.2 Genre and modality
According to SFL, genre is realized by register which comprises tenor.As modality is a means of expressing attitudes or judgments, naturally it is related to tenor.Dudley-Evans (1994: 220）mentions that genre analysis is not always concerned with the analysis of ‘moves’; it also embraces, for example, studies of tense usage, lexical frequency, classification of reporting verbs and other things alike.Thus, the analysis of modal devices can somehow reflect what a genre is.
Van Leeuwen (2000: 71）points out that three elements are necessary for the discursive construction of purposeful action: (a）the purposeful action, that is, the action whose purpose is to be constructed; (b）the purpose, itself a process, an action or a state; (c）the purpose link, the relation of purposefulness between these two.The social process approach shows that genre is related to goal-oriented social processes and communicative purposes.Since genres are realized by various types of discourse, the analysis of how discourse is purposefully constructed can reflect the nature of genres to a large extent.Hasan (2000: 31-35）analyzes how the choices of transitivity and mood can contribute to the management of discursive control for talks such as those between mothers and children.In SFL, modality is part of the mood structure; thus, the analysis of modality is conducive for the understanding of how discourse is purposefully constructed and in turn what genres are.
Chilton & Sch?ffner (1997: 220-221）explore the relations between discourse and politics, and find that in political discourse modality can appear as devices of promises but conceal the threats concerned.
Coates (2003: 331-347）studies women’s talks and finds that women’s epistemic modal forms have multiple functions, for instance, the expression of doubt and confidence, sensitivity to others’ feelings, searching for the right word, and avoiding playing the expert.Coates (ibid）does not mention that modality supplementing through MSAs can help convey multiple-leveled attitudes or judgments in conversations.However, such situations do exist.Consider the following:
(102）Anna: I bet the milkman couldn’t believe it/
Liz: /yeah.
Sue: /did he know she was going to the funeral?
Liz: probably told /him =
Sue:/probably/ yeah/
Anna: =well she probably told them you know what she’s like/
(103）Karen: well I suppose it is I’ve never really had any worries like that/
Pat: no/ it wouldn’t bother me/ but perhaps-
Karen: mind you as they’re getting younger.I might feel differently/
Pat: yes/
Coates (ibid）mentions that women’s epistemic modal forms mainly satisfy the collaborative floor and the need for open discussions; yet, no statement is made about the likely occurrences of modality supplementing.Through Examples (102）and (103), it can be found that conversation is a genre where one’s talk is embedded in another’s and the sense groups of talk (notice the sign / is the boundary of tonic group as well as sense group）are not confined to one person only.Thus, it is possible for modality supplementing to happen between interlocutors.The modal devices that one woman uses to supplement another one’s modality reinforce the idea about women’s friendship in communication.Hence, MSAs could be adopted in women’s talks, and frequently under some circumstances.
Examples (102）and (103）also indicate that MSAs could modify MVs across tonic groups (or sense groups).This is consistent with the Dutch Functional School’s idea that interpersonal satellites (including attitudinal disjuncts or proposition satellites, and style disjuncts or illocutionary satellites）can have considerable freedom of movement at the clause level (cf.Dik et al., 2005: 169-215).