Language And Literature Research Paper

1. Literary Linguistics

Linguistics looks for generalizations and so treats any text, any instance of language, just as an example of general processes at work and hence of interest only to the extent that it demonstrates those general processes. In contrast, literary criticism, at least traditionally, is interested in the text in itself, its characteristics considered unique and special, carrying traces of the historical moment of the text’s production and of the individual genius of the author who produced it. This is a fundamental distinction between linguistics and literary studies, abstraction as opposed to concreteness, generalization as opposed to specificity, and means that any combination of linguistics and literary studies must compromise in one direction or the other.

Linguistics might put itself at the service of literary studies by offering to analyze the underlying linguistic system which is instantiated in a specific text; this is an approach often associated with more functionalist orientations in linguistics. Functionalist literary linguistics characteristically uses linguistics to illuminate the workings of language in a particular text, and in particular to establish an interpretation of the text. This ‘critical linguistics’ spreads beyond literary studies, into the analysis of media texts (in ‘critical discourse analysis’). It has also been influential in language teaching, with linguist-educationalists arguing that the student better understands a language by analyzing its use in literature (see Widdowson 1975).

Alternatively, literature might serve linguistics by offering the literary text as evidence for generalizations about a particular kind of language: the language of literature. This is particularly associated with formalist literary linguistics, such as Generative Metrics, closely allied to Generative Phonology. Formalist literary linguistics looks for generalizations specific to language in literature, and ways in which literary practice develops from the abstract principles underlying all language. Formalist literary linguistics has its roots in Russian Formalism and related movements, and in its early stages was particularly associated with the work of Jakobson (1960). Through Jakobson, it influenced the structuralist anthropology of Claude Levi-Strauss, and then the structuralist literary analysis of Roland Barthes, in which literary studies briefly adopted the generalizing and universalizing tendencies of linguistics. The history of formalist literary linguistics can be traced in a series of anthologies and conference proceedings: Sebeok (1960), Freeman (1970, 1981), Fabb et al. (1987), Kiparsky and Youmans (1989); see also Fabb (1997).

2. What Makes ‘Literature’ Different?

The language of literary texts may sometimes differ from the language of nonliterary texts, but this is not a defining characteristic of literature. What makes a text literary is not form, but function. Jakobson (1960) proposed that verbal behavior might perform any (and any combination) of six distinct functions, one of which was the poetic function. In literature, the poetic function dominates. For Jakobson, a text functions poetically when it draws attention to its own form— when our attention is drawn to the text itself and not just to what it means. Bauman (1975) suggests that the poetic function itself has a further purpose, which is to enable a literary text to be performed; for Bauman it is performance, understood as a type of social behavior requiring evaluation by an audience, which defines a text as literary. The poetic function draws attention to the form of the text in order to permit evaluation by the audience according to the text’s adherence to rules of form. For both Jakobson and Bauman a literary text can differ from a nonliterary text only in terms of function; in principle the same text could be made literary or nonliterary depending on the extent to which it is made to draw attention to its own form, by being presented or performed in different ways.

Though it is function rather than form which makes a text literary, the poetic function can nevertheless give rise to distinctive kinds of literary form. A literary text tends to differ from a nonliterary text in two fundamental ways relating to form: in having specialized kinds of constituent structure, and in having ‘difficult form,’ form which is not straightforward to ascertain. Specialized kinds of constituent structure include lines of verse, metrical constituents such as the foot, and narrative constituents such as ‘orientation,’ ‘coda,’ or narrative episodes. These specialized kinds of form are regulated by rules which the text can overtly adhere to, thus drawing attention to form. Difficulties for interpretation arise from formal ambiguities, as well as from exploitations of the ways in which verbal communication works to produce effects such as metaphor and irony. By making the text difficult, attention is drawn to the workings of the text and thus again the poetic function is performed. These formal complexities and difficulties may also be responsible for some of the aesthetic effects of literary texts, if the experience of complexity constitutes one of the kinds of aesthetic experience.

‘Literature’ can exist either in the medium of writing or speech. (Because the term implies writing, ‘literature’ is sometimes substituted in literary linguistics by the medium-neutral term ‘verbal art.’) Language in written literature and language in oral literature differ, just as writing and speech differ. Any medium both constrains and enables certain kinds of aesthetic possibility in its own medium-specific way. Thus we might ask whether spoken literature makes special use of the characteristics of speech in order to carry out literature-specific functions; the same question arises for written literature. Some kinds of literary–linguistic form are restricted to speech. Metrical verse is the most obvious example, because it organizes the components of sound: loudness, pitch, and duration. There have been proposals that certain kinds of linguistic form are restricted to writing; thus Banfield (1982) proposes that certain kinds of sentence, expressing a character’s inner experience but at the same time describing that experience from another’s point of view and so paradoxically combining two centers of perspective, are ‘unspeakable,’ found only in writing (Fludernik 1993 and other authors have suggested that complexities of this kind can also be understood as exploitation of the ‘looseness’ of communication, see Sect. 7).

3. Metrical And Parametrical Form

Verse is a kind of text defined by its having a particular constituent structure: it is divided into lines. The line is marked out as a constituent by different means in different types of verse: in free verse it is marked out as a constituent by sentence structure, layout on the page, and performance; in canonically parallelistic verse the line is characterised by parallelism with other lines (see Sect. 4); in metrical verse the line is regulated by metrical rules which fix the length of the line, measured in phonological units such as the syllable or the mora (a ‘weight unit’ contained with the syllable). The metrical rules can also determine a rhythm within the line, in terms of the sequencing of relatively stressed and unstressed syllables, or relatively heavy and light syllables.

One of the key questions about language in literature is whether literary language ever has its own distinctive kind of linguistic form, not just a way of using general linguistic form, but a separate rulesystem which governs literary language. Metrical verse perhaps offers the best opportunity of finding specifically literary-linguistic form of this kind, because in metrical verse the rules which govern the meter appear to be similar to linguistic rules, while also differing from linguistic rules. The similarities involve phonological rules, the rules which govern how sounds are combined into words. First, metrical rules build ‘tree structures’ (hierarchical constituent structures) which resemble the tree structures of phonology. Second, the metrical rules form a set of interacting rules so that apparent exceptions can be understood in terms of rule interaction and even appear to operate on abstract structures, like phonological rules. Third, the metrical rules differentiate strong and weak phonological constituents, just like the phonological rules which organize the rhythmic structure of speech in general. Formalist literary linguistics has always seen metrical verse as the most important kind of text for understanding the relation between linguistic rules in general and specifically literary linguistic rules, and there has often been a close relation between work in phonology (the sound-structure of the language) and metrics; as new theories of phonology have emerged, so new insights become available into metrical structure. (Representative samples of linguistic work on meter include Halle and Keyser 1971, Attridge 1982, Hayes 1989, Hanson and Kiparsky 1996. See Language and Poetic Structure.)

Metrical rules control small phonological units. These include the morae which are counted in Japanese verse; the syllables counted in Welsh verse, organized into stress patterns in English verse, and weight patterns in Latin verse; and the phonological feet (combinations of syllables grouped around a strong syllable) counted in older Germanic verse like Beowulf. But it is also fairly common to find another kind of rule operating simultaneously with the metrical rules, and governing larger phonological units such as the word. These word-boundary rules—caesura and bridge rules—govern where words can begin or end within the line, and may contradict the constituent structures built by the meter. Thus Homer’s line is divided into six constituents (‘feet’), and thus can be divided into two halves with three feet in each half. But word-boundary rules operate to ensure that the middle of the line does not coincide with the end of a word; instead a word must end just before or just after the middle boundary as defined by the foot constituent structure. The fact that word-boundary rules sometimes contradict the metrical rules with which they cooccur suggests that contradiction and complexity are design features of literary linguistic rules, perhaps because contradiction and complexity are a source of aesthetic experience. Poetic word-boundary rules are of interest to linguists partly because of their interaction with metrical rules, but also because they are sensitive to different levels of linguistic structure; Devine and Stephens (1984) have shown for Greek that in tragedies, word-boundary rules govern the lexical word, while in comedies the rules govern the ‘phonological word’ (which contains a lexical word plus neighboring grammatical words). This shows that literature-specific rules are able to differentiate abstract linguistic structure.

4. Parallelism

Most verse traditions are governed by one of two basic organizatory systems: meter or canonic parallelism (free verse, a third organizatory system, has become increasingly common in the twentieth century). Canonic parallelism (the term is Jakobson’s) is parallelism as a structuring principle in verse holding throughout a text. Parallelism draws on the linguistic structure of the text: parallel pieces of text (usually lines, or half-lines) differ in certain respects but have some fairly salient linguistic component in common. In phonological parallelism, a sequence of sounds is repeated: the Welsh system of cynghanedd (harmony) requires the consonant sequence in the first half of a line to be repeated in the second half of the line; in some Gaelic songs the sequence of stressed vowels is the same in every line; and in Efik tonal riddles the sequence of tone changes is the same in both halves of the riddle. In syntactic parallelism, phrase structure is repeated; the words are different but they belong to the same parts of speech organized in the same manner. Syntactic parallelism is often tied to the third type of parallelism: lexical parallelism, in which words with related meanings are paired in adjacent sections of the text (see Fox 1988).

Parallelism is another instance of a literary practice where linguistic form is controlled in a literary text. Canonic parallelism has some similarities to meter, and functions in a similar way; both, for example, draw attention to form (hence define the text as verbal art) through ostensive repetition of form. However, parallelism and meter clearly operate by different principles; in parallelism the only general rule is ‘imitate the preceding line half-line,’ while in meter the line has its structure because it is governed by a complex set of metrical rules. Thus, parallelism does not appear to involve a distinctive kind of linguistic form, but instead is a way of using linguistic form to build literary form. It is worth noting that parallelism has emerged apparently independently in widely dispersed literatures, for example, Mayan, Russian, Tswana, Egyptian, and the isolated cultures of New Guinea. In contrast, almost all known metrical traditions can be historically connected, and traced back to Indo-European or Arabic origins (and these may themselves be connected). It is relatively uncommon for metrical traditions to emerge apparently spontaneously elsewhere (rare exceptions include Somali and the East Australian Dyirbal). This correlates with a hypothesized difference between parallelism as a development of possibilities inherent in language, and meter as a rule system separate from language in general and not implied by language in general.

5. Narrative Structure

Narrative is a type of discourse, including as a subtype ‘literary narrative.’ Questions about language in literary narratives are thus to some extent the same as questions about language in narrative as a general discourse genre. One of the most influential linguistic works on narrative told to interviewers, Labov and Waletzky (1967), describes true stories in which the narrator was in danger. Though these are not literary narratives, they nevertheless have characteristics which literary narratives share. Labov and Waletzky drew particular attention to three characteristics. First, sentences in the narrative can be separated into (‘storyline’) sentences whose sequence mimics the sequence of events and (‘non-storyline’) sentences whose role is to provide contextual information. Subsequent research has shown that storyline and nonstoryline sentences typically differ in their linguistic form (storyline clauses tend to be higher in transitivity; may have the verb in a more prominent position, etc.). Second, the narrative as a discourse has a clearly structured trajectory; it often begins with an ‘orientation’ constituent, and ends with a ‘coda’ constituent, and there is a movement from a complication to a resolution. Third, the point at which the complication is resolved is often made salient by particular linguistic devices; this practice is called the ‘evaluation’ by Labov and Waletzky, and may involve an explicit judgment on the narrative by the narrator; more generally it draws the hearer’s attention to the point of the narrative. This strand of work on narrative offers a way of thinking about literary narratives as developments of narrative as a general discourse genre, but does not involve a specifically literary kind of form. Nevertheless, it is possible to see how literary narratives exploit these basic possibilities to generate effects such as suspense, to create complex and ambiguous structures, and to draw attention to narrative form (thus performing the poetic function).

Focus on the language of specifically literary narrative comes from another tradition, particularly associated with Dell Hymes’s work on American Indian narratives (Hymes 1981). In this tradition, the narrative is seen to have a constituent structure analogous to verse constituent structure (typically nonmetrical, but sometimes with canonic parallelism), in which the text’s constituent structure is signaled by the salient use of grammatical material—connective words (‘and,’ ‘so,’ ‘then’), hearsay elements (‘it is said’), shifts in tense, and so on. As Tedlock (1983) argues, components of speech, such as intonational contours or the placement of pauses, can also have an important role in marking out constituency. This research tradition also looks at the organization of content into episodes within the narrative, and ways in which episode boundaries are signalled by grammatical means. This research, sometimes involving the reanalysis and reconstruction of texts gathered in the early twentieth century from now lost languages, has as one of its goals to assert the literariness of the texts, by revealing their complexity. This clarifies one of the ideological goals which is characteristic of literary linguistics in its more ethnographic and anthropological modes: to promote the artistry of traditional verbal arts, demonstrating that traditional literatures from marginalized cultures can be as complex as the canonical literatures of the ‘first world.’

6. Access To Abstract Linguistic Form

With Transformational Generative Grammar, Noam Chomsky invented a way of thinking about a sentence or a word as the ‘surface’ manifestation of a set of interconnected representations. Each instance of spoken or written language is in this sense like the tip of an iceberg, with multiple representations underlying each sentence and each word. Users of the language have no direct access to those underlying representations, in which a sentence might have its phrases ordered differently from its surface representation, or a word might have sounds which are different from those which appear on the surface; this is an ‘encapsulated’ type of knowledge which is hidden away from introspection.

In this tradition, one of the things which has interested linguists about literary texts is the possibility that underlying sentence or word structures might be manipulated directly in the construction of some literary texts. This would be an extraordinary access to linguistic material which is normally hidden. Along these lines, Austin (1984) argues that instances of distorted phrasing in poetry by Shelley might involve the violation of specific constraints on transformational rules. Thus the line ‘And the swift boat the little waves which bore’ sounds odd because there is leftward movement of ‘the swift boat’ which is the object of the verb ‘bore,’ out of a relative clause in a noun phrase; this is a violation of the ‘complex noun phrase constraint,’ which was proposed in 1960s transformational grammar to regulate abstract syntactic structure. Austin shows that Shelley violates constraints as though he has access to the underlying workings of the language, which ought to be hidden to introspection or intervention. Similar kinds of argument have been made about sound structure. Thus Malone (1988) argues that rhymes in Turkish poetry seem to be somewhat dissonant on the surface but can be seen to be underlying perfect rhymes, if the linguistic process of ‘vowel harmony’ is factored out. Vowel harmony alters the abstract vowels which make up a word to ensure that surface vowels in the same word ‘harmonize’ to have similar features. Thus in these poems the rhyme shows more regularity at an abstract level than on the surface of the text.

The possibility that abstract and supposedly ‘inaccessible’ form is manipulated in literary texts might interestingly suggest that authors (and audiences) can have privileged though presumably unconscious access to kinds of linguistic form which are normally not available to introspection. This suggestion that literature is special in its relation to language would find favor with some traditional (specifically Romantic) views of literature. But in some cases, it might also be suggested that authors use a slightly different ‘language’ from the ordinary language, which operates according to slightly different rules. In some cases, these different literary languages might correspond to archaic versions of the contemporary language.

7. The Communication Of Literary Meaning

The communication of meaning is a primary function of most ways of using language. Communication has two components. The encoding of meaning into form (performed by the vocabulary and syntax of the language, such that a sentence encodes a proposition) is one component which provides evidence for the informative intention of the communicator. But other kinds of evidence can be added, and may override the evidence offered by the encoded meaning; these kinds of evidence can come from the way in which the form is used, and rely on the context in which the communicative act occurs (including the cognitive context of the interlocutors—what they know). This second component involves inferencing, and falls under the scope of pragmatics. It is here that most of the interesting questions about literary meaning have been asked.

Metaphor is a characteristic ‘meaning effect’ in literary texts. In a metaphor, the encoded meaning (what is actually said) is not the communicated meaning (what is intended and can be inferred). Two lines of enquiry into metaphor have proved particularly fruitful for linguistics; both take the position that metaphor is not specific to literature but is characteristic of the way language works in general. In the approach particularly associated with the work of George Lakoff and co-workers (Lakoff and Johnson 1980), the vocabulary of a language is claimed to be structured by metaphor, collected into sets of relations between words and meanings which develop basic metaphors, the ‘metaphors we live by.’ Thus, for example, the use of the term ‘journey’ to stand for ‘life’ is a metaphorical substitution which underpins many other substitution (we get a start in life, reach the end of our life, choose a path, etc). Metaphors in literature draw upon the ‘metaphorical’ structure of the vocabulary in general, using this as raw material and potentially innovating on it, as when Robert Frost writes a poem on The Road Not Taken; thus language in literature makes special use of the potential of language in general. A second approach is taken by Sperber and Wilson (1995), who see the possibility of metaphor as a development of the way in which communication works. In verbal communication, a literal meaning is encoded into the linguistic form of a sentence, but that literal meaning need not itself be the intended meaning; instead the literal meaning can constitute indirect evidence for an intended meaning which is not actually stated. This is true of all verbal communication; it is also the basis of metaphor, where the literal meaning is not the intended meaning, but instead offers evidence by which the intended meaning can be inferred.

A second characteristic ‘meaning effect’ in literary texts involves the attribution of a source for any thought which is communicated, and the attribution of an attitude or a set of attitudes towards that thought. Again Sperber and Wilson (1995) have proposed an influential way of seeing the literary uses of attribution and attitude—in effects such as ‘free indirect speech’ or irony—as developments of the possibilities of communication in general. The simplest case in verbal communication is for the thoughts expressed by the communicative act to be attributed to the communicator, along with the assumption that the communicator believes those thoughts. These attributions are part of the communicative process; they are worked out by the hearer or reader alongside the working out of the referential content of the communicative act. The attribution of source and attitude is affected by the same ‘looseness’ which gives rise to metaphor. In literature, the possibility of looseness in the attribution of speaker is exploited in techniques such as the use of hearsay particles (which displace the source of a thought onto persons unknown, with various literary consequences), or free indirect speech (which blur the source of a thought between narrator and character). Literature also exploits the possibility of ‘looseness of attitude’ in modes such as irony; in irony, the communicator does not have the ‘unmarked’ attitude of belief towards the thoughts which she expresses. Instead, she typically communicates her own disbelief and at the same time re-assigns the source of the thoughts to another person or set of persons who are assumed to believe those thoughts.

8. The Communication Of Literary Form

A fundamental element of linguistic form is linguistic constituent structure. The interpretation of speech involves the assignment of a phonological constituent structure to the sequence of sounds, and the interpretation of a sentence involves the assignment of a syntactic constituent structure to the sequence of words. Literary texts have their own distinctive kinds of constituent structure, examples of literary form (rather than linguistic form). Examples include lines and stanzas in verse, or the orientation and coda and division into episodes in narrative. Each of these kinds of constituent structure is specific to literary texts, each adding noticeably to the formal complexity of the text and thus drawing attention to form and performing the poetic function. An interesting question about these kinds of constituent is whether they are similar in kind to types of syntactic or phonological constituent.

As a way of thinking about this question, it is helpful to consider two distinct ways in which a text can have a specific constituent structure. A text has phonological and syntactic constituent structure because it is processed by phonological and syntactic rules; these structures enable words to exist, to be spoken and to relate to one another in an interpretable manner. These kinds of constituent structure are in principle determinate; they can be discovered by linguistic investigation though they may not be obvious on the surface. In contrast, constituent structure might also exist as a description of the text only to the extent that it is interpreted as being present in the text, by analogy with the interpretation of implied meanings. The constituent structure in this sense is a self-description which is evidenced by the text, basically a kind of meaning, rather than being like linguistic form. Evidenced constituent structure of this kind might sometimes be vague or indeterminate, ambiguous or contradictory, and there is some reason to think that many kinds of literary constituent structure are evidenced in this manner. This includes the division of verse into lines, or the division of a narrative into functional constituents like ‘evaluation’ or structural constituents like ‘episode.’ In all cases, the text contains evidence that these constituents exist—usually some element at the beginning or end of the constituent (such as a connective like ‘so’ at the beginning of a narrative, or a rhyme at the end of the line). Of the various kinds of literature-specific constituent structure, only metrical constituent structure seems to be ‘inherent’ and determinately present in the manner of phonological structure. If these speculations are right, then most kinds of literary constituent are best understood under a theory of pragmatics, which looks at interpretation—in this case, the interpretation of form.

Sperber and Wilson (1995) suggest that the experience of weak implicature—a large number of weakly evidenced interpretations stimulated by a text—is a ‘poetic effect’; perhaps this is the cognitive component of a certain kind of aesthetic experience. This might be linked with a position held by some within literary studies and particularly postmodern literary studies. This is that textual complexity is a source of aesthetic experience, with complexity manifested in ambiguity, uncertainty, and vagueness, contradictoriness, looseness, and so on. ‘Difficulties’ in meaning and in form, both of which are literary exploitations of the pragmatic aspects of language, might thus be seen as ways of generating aesthetic experience. When we consider also that even the most regulated kind of literary linguistic form—metrical form—seems to be designed to give rise to complexity and contradiction, we might conclude that whereas linguistics seeks the underlying order in language, a literary linguistics might look at how the ordering principles of language give rise to the kinds of productive disorder which we experience as aesthetic.

Bibliography:

1. Attridge D 1982 The Rhythms of English Poetry. Longman, Harlow, UK
2. Austin T R 1984 Language Crafted. A Linguistic Theory of Poetic Syntax. Indiana University Press, Bloomington, IN
3. Banfield A 1982 Unspeakable Sentences. Narration and Representation in the Language of Fiction. Routledge & Kegan Paul, Boston
4. Bauman R 1975 Verbal art as performance. American Anthropologist 77: 290–311
5. Devine A M, Stephens L D 1984 Language and Metre. Resolution, Porson’s Bridge and their Prosodic Basis. Scholar’s Press, Chico, CA
6. Fabb N 1997 Linguistics and Literature: Language in the Verbal Arts of the World. Blackwell, Oxford, UK
7. Fabb N, Attridge D, Durant A, MacCabe C (eds.) 1987 The Linguistics of Writing: Arguments Between Language and Literature. Manchester University Press, Manchester, UK
8. Fludernik M 1993 The Fictions of Language and the Languages of Fiction. Routledge, London
9. Fox J J (ed.) 1988 To Speak in Pairs. Essays on the Ritual Languages of Eastern Indonesia. Cambridge University Press, Cambridge, UK
10. Freeman D C (ed.) 1970 Linguistics and Literary Style. Holt, Rinehart and Winston, New York
11. Freeman D C (ed.) 1981 Essays in Modern Stylistics. Methuen, London
12. Halle M, Keyser S J 1971 English Stress: Its Form, its Growth, and its Role in Verse. Harper and Row, New York
13. Hanson K, Kiparsky P 1996 A parametric theory of poetic meter. Language 72: 287–335
14. Hayes B 1989 The prosodic hierarchy in meter. In: Kiparsky P, Youmans G (eds.) Phonetics and Phonology 1. Rhythm and Meter. Academic Press, San Diego, CA
15. Hymes D 1981 ‘In Vain I Tried to Tell You.’ Essays in Native American Ethnopoetics. University of Pennsylvania Press, Philadelphia, PA
16. Jakobson R 1960 Linguistics and poetics. In: Sebeok T (ed.) Style in Language. MIT Press, Cambridge, MA
17. Kiparsky P, Youmans G (eds.) 1989 Phonetics and Phonology 1: Rhythm and Meter. Academic Press, San Diego, CA
18. Labov W, Waletzky J 1967 Narrative analysis oral versions of personal experience. In: Helm J (ed.) Essays on the Verbal and Visual Arts. Proceedings of the 1966 Annual Spring Meeting of the American Ethnological Society. University of Washington Press, Seattle, WA
19. Lakoff G, Johnson M 1980 Metaphors We Li e By. Chicago University Press, Chicago
20. Malone J L 1988 Underspecification theory and Turkish rhyme. Phonology 5: 293–7
21. Sebeok T (ed.) 1960 Style in Language. MIT Press, Cambridge, MA
22. Sperber D, Wilson D 1995 Relevance: Communication and Cognition. Blackwell, Oxford, UK
23. Tedlock D 1983 The Spoken Word and the Work of Interpretation. University of Pennsylvania Press, Philadelphia, PA
24. Widdowson H G 1975 Stylistics and the Teaching of Literature. Longman, London