Language, Text and Context

Essays in stylistics

Michael Toolan, Michael Toolan

• Publisher: Routledge
• Year: 2016
• ISBN:
  9781315402369
• Edition: 1
• Topic: Languages & Linguistics
• Subtopic: Linguistics
• Index: Languages & Linguistics

First published in 1992, this wide-ranging collection of essays focuses on the principle of contextualisation as it applies to the interpretation, description, theorising and reading of literary and non-literary texts. The collection aims to reveal the interdependencies between theory, analysis, text and context by challenging the myth that stylistics entails a fundamental separation of text from context, linguistic description from descriptive interpretation, or language from situation. The essays cover a historically diverse set of texts, from Puttenham to Colemanballs, and a number of language-sensitive topics such as post-modernism, irony, newspaper representations, gender and narrative.

Part I Situated fashions of speaking and writing: from nonsense to common sense

DOI: 10.4324/9781315402383-1

1 Making (non)sense of postmodernist poetry

Brian McHale

DOI: 10.4324/9781315402383-3

We do not want more poems about everyday life; there are enough and more than enough poems that do that; but never today enough Dada poems.

(Forrest-Thomson 1978)

TALKING NONSENSE

Accusations of nonsense put literary people on the defensive. ‘This is not nonsense talk’, writes Marjorie Perloff (1987:231), defensively, of a passage from a poem she admires by the postmodernist ‘language’ poet Charles Bernstein. She is right to get defensive, for the passage in question (from a poem called ‘Dysraphism’) certainly looks like nonsense, and a sustainable charge of nonsense is normally fatal to a poem’s claims on our serious readerly attention. A stronger defence, however, would have involved turning the accusation into a description, that is, admitting the charge of nonsense while denying that the label ‘nonsense’ must inevitably be pejorative. ‘Nonsense’ can just as well identify a valuable, and valued, quality. It has functioned that way historically, and not only inalized poetry (‘children’s classics’: Dodgson, Lear), but, more pertinently, in Russian futurist zaum and Dada poetry. Many postmodernist poems might appropriately be described as ‘neo-Dada’ or ‘nonsense’, and part of the process by which we might come to understand why such poems could be worth writing and reading involves coming to understand the possible uses and value of nonsense.¹ In recognition of this, the present essay uses the term ‘nonsense’ neither pejoratively nor dismissively, but as a neutral descriptive category.

Why might one value nonsense? First, nonsense, far from being only too easy to fall into, as one might infer from the pejorative contexts of the term’s use, proves to be quite difficult to make. This is because readers or hearers of sentences are such resourceful sense-makers, able to extract sense from the least tractable materials. Anecdotal evidence of such resourcefulness is to be found, for instance, in Stanley Fish’s by now notorious experiment in which well-drilled students of religious poetry were able to develop a plausible interpretation for a cryptic (to all appearances nonsensical) pseudo-poem – in fact a list of linguists’ names left on the blackboard from a previous class (Fish 1980: 322–37).²

Philosophers of language and philosophically oriented linguists often assume that certain grammatically well-formed expressions are ‘inherently nonsensical’, e.g. ‘Colorless green ideas sleep furiously’. But J. F. Ross has persuasively argued that, on the contrary, ‘there is no grammatically well-formed string of words that is in all environments semantically impossible or semantically unacceptable. … Something is nonsense only relative to an environment’ (Ross 1981: 55). Consequently, nonsense arises only when extraordinary efforts are made to render an environment semantically inhospitable to sense: ‘Meaninglessness [i.e. nonsense] occurs only when meaning is environmentally prevented’ (172).

If this is so, and nonsense really is as difficult to produce as Ross contends, and as Fish’s experiment seems to corroborate, then it might be valued precisely for this quality of difficulty surmounted. Of course, the value attached to difficulty surmounted is not by any means a universal; it is, we might suppose, a modernist value, but not necessarily a postmodernist value. In any case, the evidence of nonsense’s difficulty would lead us, at the very least, to assume that nonsense must be motivated; in other words, it would lead us to ask why, if nonsense is so difficult to achieve, would someone have bothered to produce it?

Second, nonsense might be valued precisely for the light it throws on its antithesis: sense-making. Nonsense yields valuable insight into how sense is made, giving us access to the sense-making process in a way, perhaps, that nothing else can. This is because of the intimate relationship between nonsense and common sense: nonsense depends on common sense. ‘Our ways of making nonsense will depend upon our ways of making common sense’, writes Susan Stewart (1979: viii); ‘the nature of nonsense will always be contingent upon the nature of its corresponding common sense’ (51). ‘Where there is a common sense, there will be a common nonsense’ (52); consequently, ‘There will be as many varieties of nonsense as there are varieties of sense’ (16). Now this is, in effect, a negative value: nonsense is to be valued for what it tells us about what it is not. Whether it can acquire some positive value in its own right, and if so what, is a question to which we shall return later.

The best and most economical way to investigate how nonsense is made and common sense resisted in postmodernist poetry is to analyse specific texts. Thus, the bulk of this chapter will be devoted to readings of three poems: John Ashbery’s ‘Metamorphosis’ (1979); J. H. Prynne’s ‘Of Movement Towards a Natural Place’ (1974); and Charles Bernstein’s ‘Live Acts’ (1986). But before we can undertake these readings, we will need to equip ourselves with an appropriate descriptive apparatus, one designed to capture Stewart’s insight into the dependency of nonsense on common sense. We will need, in other words, to give some account of how we typically make sense of difficult or obscure poetry, as a preliminary to accounting for how we fail to make sense of postmodernist nonsense poetry.

MAKING SENSE

The most attractive and persuasive account of sense-making I know of, for all its incompleteness and eccentricities, is the one proposed by Veronica Forrest-Thomson (1978; see also Forrest-Thomson 1971, 1972, and 1973). Her approach to literary intelligibility might be characterized as a ‘strong misreading’ of William Empson, especially the Empson of Seven Types of Ambiguity.³ What in particular Forrest-Thomson retains from Empson is his emphasis on the reader’s resourcefulness in rationalizing (Forrest-Thomson says ‘naturalizing’) the text’s semantic anomalies, cruxes, and ‘ambiguities’ (in Empson’s extended sense of the term).

How do we make poems, especially ‘difficult’ or ‘obscure’ or apparently ‘nonsensical’ poems, intelligible?⁴ We do so, according to Forrest-Thomson, by identifying pertinent levels (we might just as properly say frames) of coherence or integration. Identifying a level of coherence or integration (or what Forrest-Thomson somewhat anomalously calls an ‘image-complex’) enables us ‘to assimilate features of various kinds, to distinguish the relevant from the irrelevant, and to control the importation of external contexts’ (1978: xii). In other words, it enables us to integrate a range of features, both semantic and non-semantic,⁵ under the same explanatory rubric; it enables us to establish, relative to this rubric, a hierarchy or priority of features, some of them judged to be relevant, others irrelevant; and, finally, it enables us to decide which, if any, external frames of reference (in the sense of Hrushovski 1979, 1984a, and 1984b) might relevantly be referred to.

The actual frames or levels of coherence which might be pertinent to specific poems are, of course, very various, but we can propose three basic frame types or conventional levels of integration which have served readers well in their naturalizations of (at least) western poetry since (at least) the Renaissance.

1. The level of world, which Forrest-Thomson calls the ‘empirical image-complex’. This involves the reader’s reconstructing a situation, scene, event, etc., at the extreme limit an entire cosmology.
2. The level of voice, Forrest-Thomson’s ‘discursive image-complex’. At this level the reader reconstructs for the poem a ‘speaker’ or source persona, in some cases a more or less fully personified ‘character’, in others a supra-personal or conventional source, a register, discourse, or level of style keyed to a specific genre or topic.
3. The level of theme, Forrest-Thomson’s ‘thematic synthesis’.⁶ This involves identifying an ‘idea’ sufficiently abstract to allow for the assimilation of other local ‘ideas’ identified in the text.⁷

In reading most poems, all three frames come into play and interact; in some types of poetry, one frame is clearly dominant (e.g. ‘world’ in topographical-descriptive poetry, or ‘voice’ in dramatic monologue poems), the others subordinate (or inapplicable). In the process of naturalizing or rationalizing poems, ‘feedback loops’ typically function among these three levels: thus, the identification of a world (situation, scene, event) helps us to integrate a speaker at the level of voice, and vice versa; identification of a speaker helps us to integrate a situation at the level of world; while identification of a theme may guide us in reconstructing a world and/or a voice, and vice versa; reconstruction of a world, a voice, and/or the interaction between them, may guide us in identifying a theme.

Forrest-Thomson’s approach to sense-making in terms of the assimilation of textual features at different levels of coherence finds interesting corroboration in the work of others who have investigated these processes from other perspectives. For example, J. F. Ross has proposed an account of disambiguation and semantic coherence whereby the relatively intransigent words in a sentence (the ones whose range of meanings is most narrowly prescribed) coerce (or ‘dominate’, to use Ross’s own term) the less intransigent words, weeding out their irrelevant meanings (‘differentiating’ them, as Ross puts it) and assimilating these words to the semantic context (Ross 1981; cf. Thompson and Thompson 1987). This, it seems to me, amounts to a version of Forrest-Thomson’s account of assimilation and hierarchies of relevance and irrelevance, but pitched at the sentence-level rather than, like Forrest-Thomson’s, at the text-level. Similarly, Alex McHoul’s (1982) ‘Cumulex’ exercise tends to corroborate not only Forrest-Thomson’s general approach, but even the levels of coherence which she specifies. In grappling with a semantically diffuse and enigmatic text (in fact, a synthetic ‘pseudo-poem’), McHoul’s student readers construct ‘scenes’ or fragments of a world (1982: 22, 30), often importing external frames of reference to ‘flesh out’ these world-fragments (32); they try to identify voices or personas, and routinely assume that the text has been ‘authored’, that is, that it emanates from a single source and intention (18, 29); and above all, they strive for thematic synthesis, easily the most conspicuous and dynamic of the sense-making operations they apply.

If Forrest-Thomson is, as we have noted, a faithful Empsonian in her general orientation towards sense-making, she nevertheless parts company with him at one important juncture. For Empson, the resolvability of semantic anomalies is a positive value; texts which successfully resist rational resolution are described as ‘decadent’, and are said to ‘misuse’ ambiguity (1947: 165, 160). Empson’s negative valuation of unresolvable ambiguities is explained by his anxiety about control: texts which do not contextually limit or constrain ambiguity give the reader too much interpretative freedom. Forrest-Thomson exactly reverses Empson’s valuation. Where he regards the resolvability of semantic cruxes as the mark of poetic success, she regards it as a measure of overly facile sense-making, premature or ‘bad’ naturalization; while the texts which in Empson’s view ‘misuse’ ambiguity Forrest-Thomson regards as encouraging suspended or ‘good’ naturalization.

The poet Charles Bernstein, who has read Forrest-Thomson as carefully and sympathetically as she has read Empson, proposes a concept of ‘absorption’ to cover much the same ground as Forrest-Thomson’s assimilation and levels of coherence. Texts may be ‘absorptive’ in two interrelated senses: they absorb diverse materials, that is, they integrate or assimilate them and make them cohere, in roughly Forrest-Thomson’s sense; and in the process of absorbing materials they may also absorb the reader, in the sense of engrossing or fascinating him or her. Bernstein describes approvingly a range of ‘antiabsorptive’ strategies by which his postmodernist contemporaries’ poems counteract the absorptive tendencies of sense-making.⁸ These strategies might just as readily be called nonsense strategies, and it is their operation, and the resistance they present to the operations of sense-making, that we will observe in action in the analyses which follow.⁹

BUILDING WORLDS

Some working hypothesis about the reconstructed world of a poetic text is often essential for distinguishing between the two frames of reference whose interaction constitutes a metaphor (see Hrushovski 1984a; McHale 1987: 133–47). Metaphor is not always or inevitably signalled grammatically, so in many cases we must operate with a semantic hypothesis about which frame is ‘present’ and ‘literal’ (i.e. t...