MSA is a term coined in this dissertation.So far there has not been any systematic or comprehensive research on the linguistic phenomenon of MSAs as a means of modality supplementing.Thus, a direct literature review on the studies of MSAs seems unrealistic.However, as MSAs are parasitic on MVs and similar to MAs, it is necessary to bring together the studies of the two fields (i.e.MVs and MAs）in order to detect the implications for MSAs as a means of modality supplementing in discourse.
In the light of this assumption, this chapter will review the previous studies of modality mainly through MVs and MAs.MVs are auxiliaries in clauses; yet, they are somewhat different from the other auxiliaries, such as do, have or is.In comparison, MAs have not been as widely and profoundly explored as MVs.Like MAs, MSAs are adverbs in nature, though some of them are different from MAs in forms and functions to some extent.Various theories are involved in this area, such as pedagogic grammar, semantics, pragmatics, cognitive linguistics, SFL and corpus linguistics.
The approach of truth-condition correspondence to MVs and MAs blends syntax and semantics, and states that the use of these modal devices arises out of the needs at the clause level.The approach of cognitive linguistics states that force dynamics is an important cause for the use of modal devices at the level of clause or discourse.The approach of pragmatics holds that speech acts and evidentiality play an important role in the adoption of modal devices in communication.According to this approach, power and solidarity are involved in the use of modal devices.The approach of functional representation believes that modal devices are used in relation to such factors as communicative settings, and text types or genres.The approach of SFL stresses that modal devices are employed in discourse as required by the interpersonal metafunction.The fifth approach inherits something of the functional representation approach, but makes an effort in expanding it both theoretically and practically.It should be noted that these approaches hint at the possibility of multi-layered modality in discourse from time to time, and present examples to illustrate it.However, they do not clearly indicate the necessity of modality supplementing through MSAs in discourse.
2.1 Approaches to modality
This part will present the previous studies on modality.The focus is MVs and MAs.There are arguments over modality, which contribute to the formulations of five approaches in this respect of studies, namely, truth-condition correspondence, the approach of cognitive linguistics, the approach of pragmatics, functional representation and SFL.The following part will illustrate how modality is approached by means of these approaches.
2.1.1 Arguments over modality
Modality is interpreted in two ways.On the one hand, modality is regarded as a way of presenting information in an attitudinal way, including both linguistic and nonlinguistic means.Hodge & Kress (1993: 11）state that such resources as gestures, expressions and posture, under the name of ‘paralinguistics,’ greatly enrich and complicate the system of modality.On the other hand, modality is considered as a semantic category which expresses the attitude of the speaker towards the proposition expressed in the clause, realized by specific lexicogrammatical means, such as MVs, MAs and modal nouns.The latter type of modality is what most linguists have focused on.
It is generally agreed that modality is related to subjectivity and is non-assertive.Lyons (1977: 452）argues that modality is concerned with the ‘opinion and attitude’ of the speaker.Modal expressions are not facts but arguments or claims about facts, factuality and truth.Thus, clauses with modal expressions are not assertive.However, some other linguists notice that modality is an evasive term, for it is a complex system.Herslund (2005: 39-48）holds that modality can be either subjective or objective in nature.The subjective-objective distinction concerns the status of the speaker’s evidence for an epistemic evaluation; in another word, it is a distinction between formally reliable evidence and more intuitive guessing.Herslund (ibid）compares such a distinction to that of major and minor keys in terms of music tonality.Nevertheless, Herslund’s argument does not change the overall nature of subjectivity of modality.Instead, it only means that even subjectivity includes the elements of objectivity.Generally speaking, the complexity of the term “modality” lies in three aspects.
First, there are different views on the scope and content of modality.Palmer (1986, 2001, 2003）believes that mood and modality are of hyponymy, mood being a subordinate while modality a super-ordinate.According to Palmer, modality is a general cover term for the modality system with a number of subcategories as Figure 2.1 indicates.
In Palmer’s view, modality as a system of hierarchy is realized by various modal devices, including MVs.This view is similar to that of Halliday & Matthiessen (2004: 616), that is, speakers have indefinitely many ways of expressing their opinions ― or rather, perhaps, of dissimulating the fact that they are expressing their opinions.However, “mood” in Palmer’s view means clause types: indicative or subjunctive, for example.In Halliday (1994), mood is not just clause type but a unit of the analysis of clause as exchange as well; the central structure of mood is subject + finite.In SFL, MVs as finite are in the mood structure.