Throughout this book we have argued for a broad view of stylistics as being concerned with the systematic analysis of style in language in all its forms. This is a wider view of stylistics than some stylisticians might hold; as we have seen, stylistics is often defined more narrowly as the study of literary texts using linguistic techniques. However, as we have also noted at various points throughout this book, the techniques of linguistics are just as applicable in the analysis of non-literary (in the sense of non-fiction) texts as they are in the analysis of prototypically literary works. Furthermore, the problem of defining the concept of literariness (see Chapter 2) lends further weight to the view that stylistics should not be seen as concerned with any one particular text-type. While we cannot dismiss the fact that stylisticians generally have concentrated primarily on the analysis of so-called literary texts, this activity has been motivated by a desire to understand the workings of what is defined socio-culturally as literature rather than by an analytical inability on the part of stylistics to deal with other text-types.
Despite the fact that our broad view of stylistics may not be shared by all practitioners of it, one aspect of our definition of the subject that all stylisticians will be in agreement with concerns its development out of the discipline of linguistics. Stylistics is unremittingly linguistic in orientation. That this is the case can be seen from the fact that stylistics emerged originally out of the discipline of language study, formalised in the early twentieth century as linguistics. Moreover, stylistics proceeds on the principles of linguistics. That is, it aims to be objective (notwithstanding the inherent difficulties of achieving objectivity), rigorous, replicable and falsifiable. Thus, stylistics has much more in common with other sub-disciplines of linguistics than it does with sub-areas of traditional literary criticism. Indeed, stylistics has traditionally not had a good relationship with literary criticism specifically because of this difference of opinion between the two disciplines concerning the methods of criticism. Nowhere is this difference of opinion more vituperatively expressed than in the articles that form the (in)famous Fowler–Bateson controversy (see Fowler 1971), in which the linguist Roger Fowler and the literary critic F. W. Bateson exchanged metaphorical blows concerning the proper method of studying literature. As we have noted, stylistics has always allied itself with linguistics in its view that objectivity and falsifiability are key to any analysis of the workings of a literary text. Traditional literary criticism, on the other hand, has proceeded according to the belief that literary works can only be discussed subjectively. In an article sub-titled ‘a polemic against relativism and fragmentation’, West (2008) exposes another inherent problem with English Studies in its traditional sense, namely the lack of concern with the analysis of texts:
If English is fragmented, then it is so because it is a subject without a soul, and it is a subject without a soul because it lacks a clear object of study and – most scandalously – a clear methodology. What other discipline would delight in these deficits? Do historians not agree, generally, on what they are studying and on the methods that they use? Do mathematicians or astronomers or biologists disagree fundamentally on what their object of study is, or on the methods to study that object? Why is our discipline so fragmented? Why do we shy away from stating explicitly what we are studying, why we are studying it, and how we are going to study it? Why are relativism and fragmentation seen as positive characteristics? A subject without an object of study or a methodology is not a real subject at all, but a playground for amateurs.
(West 2008: 137)
We agree with West's view and it is our contention that literary criticism as a discipline is stagnating because the unfalsifiability of the claims generated by this subjective approach, and the lack of focus on the text, makes critical discussion impossible. It is also our contention that stylistics offers a way out of this impasse.
Leech (2008: 2) describes stylistics as a ‘bridge discipline’ connecting linguistics and literary studies and explains that ‘by undertaking a linguistic analysis as part of the interrelation between the two fields of study, we facilitate and anticipate an interpretative synthesis’ (Leech 2008: 2). On this basis, Leech describes stylistics as an interdiscipline: i.e. a discipline in its own right but one which is informed by the insights of other disciplines. This is a view we subscribe to, though with the caveat that there are other disciplines than literary studies that inform stylistics (we have, for example, discussed the contributions of the visual arts and cognitive science to stylistics; there is no reason why history, politics, sociology, etc. should not also have a contribution to make). Indeed, this is in line with developments in linguistics itself, which as a discipline has progressed far beyond its initial and primary concern with the formal elements of language. Leech explains that
Placing linguistics in a broad humanistic and social science perspective, it no longer seems controversial that when we describe the characteristics of a piece of language, we can (and should) also study its interrelations with those things that lie beyond it but nevertheless give it meaning in the broadest sense. These include the shared knowledge of writer and reader, the social background, and the placing of the text in its cultural and historical context.
(Leech 2008: 3)
To put it succinctly, stylistics is concerned with both form and function, though very much aware of the dangers of interpretative positivism (Simpson 1993).
Finally, as Leech (2008) also notes, stylistics is very much a practical discipline. It has developed and improved its techniques through the application of theoretical frameworks to the practical analysis of texts of all kinds. The success of this approach seems set to continue into the twenty-first century. In this final chapter we will briefly discuss some of the directions that stylistics is currently taking and outline how it might develop in the future.
The capacity of stylistics for taking insights from other disciplines means that it is a subject that is always expanding and developing. In this section we briefly outline just two of the directions that stylistics is currently taking, as a flavour of what is possible within the area.
As we have said at various points throughout this book, stylistics is neither theoretically nor in practice limited to studying the language of literature. Having said that, there are many outside the field who consider stylistics and literary stylistics to be identical, and there is, it must be admitted, a great deal more analysis of literary than of other texts. In this section, we will take it as read that non-literary texts (speeches, gas bills, love letters, committee minutes etc.) can be analysed for their stylistic features in just the same way as poems, plays and novels. In such cases, the same kinds of features may be noted, such as the foregrounding of deviation or the use of parallelism to underline a point.
There is also, however, a use of stylistic analysis which may not originally have been envisaged in either literary or non-literary stylistics, but which is increasingly attractive as an option. That is the use of stylistic analysis in order to establish the ideological basis of a text's meaning. The study of ideology in language has a long and respectable history in feminist linguistics (e.g. Cameron 1998), and in critical discourse analysis (e.g. Fairclough 1989). Not all of this study is text-based, of course, and some is more sociological and political than linguistic. However, insofar as there is a linguistic aspect to the work carried out in these fields, it can be called stylistic. Indeed, some of the founders of the critical discourse analysis (CDA) sub-discipline, such as Roger Fowler (see for example Fowler 1991), are also known as stylisticians (e.g. Fowler 1986), and this overlap in technique, if not in aims, is one that has become increasingly evident in recent years.
Simpson (1993) took a large step forward in demonstrating the usefulness of the tools of analysis commonly used by CDA (e.g. modality and transitivity from Hallidayan functional grammar) in tracing the points of view inherent in literary and non-literary texts. This tendency for readers to be ‘invited’ to view a text world from a particular vantage point is one of the mechanisms by which persuasive, not to say manipulative, texts could influence the ideological outlook of readers, and for this reason, studies combining some of the insights of stylistic analysis with those of CDA are very important. The move towards understanding how readers process text evident in cognitive stylistics, and the development of more ideologically-sensitive tools of analysis, will be vital in this task. Recent examples of the kind of work which seems to combine the text analysis of stylistics with the ideological awareness of CDA include Jeffries (2007b, 2010b) and Davies (2007).
Since stylistics emerged out of linguistics, it is natural that it has traditionally concerned itself with the analysis of language data. However, it is also the case that stylistics has been used predominantly in the analysis of literary texts. Consequently, as literary texts have themselves developed, so too has it become necessary for stylistics to develop new ways of accounting for the effects that such texts can generate. One issue that stylisticians have recently started to grapple with is multimodality. Since a number of different text-types include multimodal elements, a complete analysis of such texts would need to take this into account. Our coverage here of multimodality cannot be comprehensive, so we will instead aim to describe some of the current approaches to the issue that have been applied within stylistics in order to give a sample of work ongoing in this area. We will begin by considering drama, since this is a text-type that is defined in part by its multimodal elements.
Prototypically, dramatic texts are written to be performed and a full stylistic analysis of such texts should be able to account for performance-related effects. That stylisticians of drama have tended not to consider dramatic performances has to do with a methodological problem associated with this, namely that since theatre performances vary from show to show, the object of analysis is unstable and critical discussion is not viable (Short 1998). This, however, is less of an issue in the stylistic analysis of film drama, since, with the exception of remakes, there is only ever one record of a film performance to be taken into account. McIntyre (2008), for instance, examines a film version of the Shakespeare play, Richard III, and presents an analysis that identifies the ways in which the performance emphasises the effects generated by the linguistic elements of the screenplay. The means by which this is achieved is via the application of linguistic analytic frameworks. For example, Brown and Levinson's (1987) politeness theory is predicated on the notion that social distance can be encoded in language and that speakers’ linguistic choices can consequently affect social relations. What McIntyre (2008) suggests is that the Brown and Levinson politeness model can also be used to explain literal movement closer to or further away from an interlocutor. In the case of the Richard III film that McIntyre analyses, it is noted that:
In Brown and Levinson's terms, Richard's movement towards the camera (shot 18), and the camera's consequent backing away from him, represents a paralinguistic threat to the viewer's negative face (i.e. his or her desire to be unimpeded). This is because the eye-line vector established between Richard and the camera suggests that Richard is now looking at the viewer (in effect, the camera creates the illusion that the viewer is somehow in the bathroom with Richard). Richard's movement closer to the camera impinges on what we perceive, as a result of what we know the camera to represent, to be our personal space. The position of the camera creates the illusion of there being a direct connection between the discourse world (Werth, 1997) and the text world of which Richard is a part. The illusion of a paralinguistic threat to the audience's negative face further characterizes Richard as a threatening and potentially dangerous character.
(McIntyre 2008: 325)
McIntyre's analysis also makes use of Kress and van Leeuwen's (2006) approach to multimodal texts, which proposes that transitivity analysis (see section 3.2.1) can be applied to visual images in order to determine the syntactic connections between the constituent elements of an image.
The potential for applying linguistic frameworks to the analysis of visual images remains to be fully explored, though it seems that this may be a way forward for tackling multimodal texts. For example, a New Labour political poster from the 2001 UK General Election campaign was set out like a film advert, with the introductory caption ‘The Tories present’ and the large title ‘ECONOMIC DISASTER II’ across a photo of an apocalyptic landscape. Above the title are the faces of Michael Portillo, who is said to star as ‘Mr Boom’, and William Hague, who is introduced as ‘Mr Bust’. There is a sub-title, ‘Coming to a home, hospital, school near you’. The Labour Party thus used a visual style akin to cinema adverts which has the effect of generating the conceptual metaphor THE CONSERVATIVE PARTY IS A FILM PRODUCTION COMPANY. This in turn gives rise to all the attendant mappings between source and target domain (MPs ARE ACTORS, TORY POLICIES ARE FILM SCRIPTS, etc.) that generate a conception of the Conservative Party as lacking substance and not in touch with the real world. As Forceville (2007: 26) has noted, however, the mappings between source and target domains in a visual metaphor are not triggered in the same way that they are in a verbal metaphor (i.e. by a copula grammatical construction). In multimodal metaphors a different set of mechanisms cues our identification of metaphorical elements. As Forceville explains, at the heart of many of these mechanisms is foregrounding. For example, it is common for the target domain in visual metaphors to deviate from our schematic expectations and to be foregrounded as a consequence. Since foregrounded features invite closer interpretative consideration, we are likely to consider a metaphorical interpretation in order to make sense of the image.
Another example from the 2001 UK General Election campaign, on the other hand, may potentially be analysed using blending theory to explain how the concepts of Thatcherism are seen as intrinsically linked to the policies of the then Conservative Leader, William Hague. The image literally blended together the faces of Margaret Thatcher and William Hague.
It is also becoming possible to use corpus linguistic approaches in the analysis of multimodal discourse, as is reported in Carter and Adolphs (2008) and Dahlmann and Adolphs (2009). The boundaries of stylistics are clearly expanding.
We began this book by noting that stylistics is a subject that is eclectic and open. Evidence of this eclecticism and openness to new ideas can be seen in the wealth of research published in such stylistics journals as Language and Literature and the Journal of Literary Semantics, and in the presentations made at the annual conference of the Poetics and Linguistics Association, an international association of stylisticians.1 We therefore end this book on an optimistic note. Stylistics seems set to continue developing in new and exciting ways. In this chapter we have gestured towards new directions in stylistics. Indeed, new frontiers are opening up all the time. If we speculate on what we might expect to see happening within stylistics over the course of the next few years, we would suggest that one area set to expand substantially is corpus stylistics. This approach to stylistic analysis very obviously continues the rich tradition of objectivity and replicability that is the hallmark of stylistics, and, intriguingly, the advent of corpus linguistics software has led stylistics to return to one of its initial concerns: the analysis of authorial style. The ability to analyse large quantities of data has allowed scholars to overcome the methodological problem associated with studying this aspect of style, and recent research in this area is reported by Hoover (2007 and 2008) and Hardy (2007).
Another area of stylistics that is clearly burgeoning is cognitive stylistics. Over the coming years we would expect to see a strengthening of this approach, as cognitive stylisticians empirically test their theoretical claims and in so doing augment the insights that stylistics is able to provide.
We also expect to see a convergence of CDA and stylistics in the field of critical stylistics, where the object of study may be any text-type or genre, but the interpretative component of the study aims to discover the latent ideological assumptions embedded in the text, rather than the aesthetic or other effect as in more traditional stylistic study.
Finally, we might well expect stylistics to develop in terms of the text-types it is able to deal with. In the twenty-first century multimodality is more of an issue than ever before, and stylisticians have already begun to work on the analysis of texts that incorporate substantial multimodal elements, including film and hypertext, among many others. The common factor in all such new work is that it proceeds according to the core principles of stylistics, and in this way stylistics will continue to expand and shed new light on how texts and readers combine to create meaning.
The value of stylistics for the analysis of how ideologies are manifested in texts is the subject of Simpson (1993) and Jeffries (2010b), both of which are practical introductions to the topic. Jeffries (2007a and 2007b) demonstrates the application of such techniques in the analysis of political and media texts. Multimodal stylistics is discussed in Boeriis and Nørgaard (2008), McIntyre (2008) and Montoro (2006), and Forceville has published extensively on the application of cognitive metaphor theory in the analysis of visual images (see, for example, Forceville 1996, 2002b, 2005a, 2005b, 2007). The concept of applying transitivity analysis to images is covered in Kress and van Leeuwen (2001). Other relevant works on multimodality are Kress and van Leeuwen (2006), Scollon and Wong-Scollon (2003) and Machin (2007). Affect and emotion are covered in Burke (2008) and Oatley (1992 and 2004). Finally, Lambrou and Stockwell (2007) is a collection of chapters dealing with current issues in contemporary stylistics and represents some of the best current work in the subject.