by Donald Freeman

Research on the language of literature has a long tradition under the rubric of philology. This unity of concerns has lapsed as linguistics has become independent and linguistic training for literary scholars has been virtually eliminated. Still, the linguistic revolution of the last four decades has produced substantial work by linguists in prose style, narrative, metrics, poetic syntax, and metaphor.

Style and Textual Function

Early work in prose style related syntactic patterning to the author's world-view, and, as speech-act theory developed, to the social conventions surrounding literary 'utterances.' The world-view of Ernest Hemingway, action-oriented, stoic, is depicted in the unadorned, unmodified, verb-centered syntax of his prose. As this paradigm crested in North America, functional linguists in Britain and Europe focused on the textual function in literary prose: voice; patterns of related words called 'collocations'; cohesion, or patterns of pronoun reference and specialized language that unify a text; deixis, or the 'pointing' function in language (in English, how using 'this' and 'that' depends on the relative positions of speaker and hearer) and information flow. Functional linguists drew on the sustained collaboration between linguists and literary theorists in the Linguistic Circle of Prague. Functionalists sought to explain such phenomena as our intuitions that while the collocation 'bailiff', 'prosecutor', 'your honor', and 'voir dire' would be unremarkable in an account of a trial, in a scene between lovers, the collocation 'defendant' and 'cross-examine' would display a motivated prominence, or 'foregrounding'. The text itself, foregrounding some elements, 'automatizing' others, making them seem natural or 'automatic', becomes a meaningful linguistic unit.

Approaches to Narrative

This view of text underlies contemporary research in narratology—how fictional narrative represents speech and thought. Narratology blends text linguistics (linguistic structures beyond the sentence) and the literary-critical 'unreliable narrator' (where the narrative voice itself is an engaged character). Recent work has focused on 'free indirect discourse,' employed, for example, by Jane Smiley in her recent novel, Moo. Smiley fuses her narrator's voice at one point with a character—a hog named Earl Butz—recalling his piglet-hood: 'it had happened that around the time of Earl's birth the farmer who bred him found his complex rather lightly booked...and so he had amused himself by letting the animals out in the yard every day, they were so lively and cute.' (p. 270). Free indirect discourse is marked by such properties as different tense requirements (because general narrative is usually past tense, free indirect discourse usually is past perfect ('had'), and the broader play of consciousness (the narrator is Earl the hog, but Earl cannot know that the farmer's complex was 'lightly booked'; moreover, the perception that the piglets were 'lively and cute' is the farmer's, not Earl's).

Poetic Language and Meter

Linguistic research on poetic language has centered on meter, syntax, and metaphor. Researchers in generative metrics sought to establish the line as the unit of metrical analysis rather than the traditional poetic foot. On this account, a line of iambic pentameter, for example, would be analyzed not as five iambic feet but as 10 metrical positions, the odd-numbered weak and the even-numbered strong, e.g.:

When to the sessions of sweet silent thought
w s w s w s w s w s
1 2 3 4 5 6 7 8 9 10
[Shakespeare, Sonnet 30]

Traditional theory could not explain why the syllables '-sions' and 'of' constituted an iambic 'foot,' a sequence of an unstressed syllable followed by a stressed syllable. Generative metrics claimed that the only stress unit that 'counted' was an entity they called the stress maximum, sequences like 'the sess-,' where a syllable to which the stress rules of English assign primary stress is on an even-numbered position and flanked on both sides by syllables assigned lesser or no stress. The only 'rule,' generativists claimed, was proscriptive. In iambic pentameter, stress maxima could not appear in odd-numbered metrical positions. This 'rule' explains phenomena like the problematic 'foot' 'swéet sí-,' where both syllables of the 'iamb' are stressed. This common 'exception' is no problem for generative metrics. 'Sweet', though stressed, is not a stress maximum because it has an unstressed syllable only to its left.

The stress maximum also explains another common 'exception' in English poetry, the reversed first foot, 'When to'. 'When' does not count as a stress maximum because there is no unstressed syllable but only a line boundary to its left.

This body of research, which is by no means settled, posited natural relationships among a previously random set of 'permissible licenses' to 'strict' iambic pentameter—a phenomenon that indeed rarely occurs. 'Licenses' like pyrrhic, trochaic, and spondaic substitution, headless lines, synaloepha, and diaeresis—which occur so often in poetry that the 'exceptions' nearly outnumber the 'norm'—were shown to be consistent with our knowledge of English phonology and phonetics. Generative metrists synthesized these unrelated 'exceptions' into a perspicuous method for analyzing metrical tension, the extent to which poets depart from the abstract metrical norm. An account of this tension can play a significant role in the analysis of poetic style.

Poetic Language and Syntax

Linguists have used the more powerful analytical tools of generative syntax in close readings of poetry. A syntactic analysis of Coleridge's "Kubla Khan", for example, contests the traditional view of this poem as a fragment, demonstrating that the dislocated syntax at the poem's start, and again as its closing vision takes shape, conclusively resolves into strongly canonical syntactic patterns as the poet finds his voice and subject, poetry and the process of creation. Another study shows that the syntactic patterns of Wallace Stevens' "The Snow Man" require a reanalysis of the poem's sense that constitutes the poem's statement itself. The poem's mind of winter is just the kind of mind that constantly reanalyzes and readjusts its view of reality. In this research, linguists have argued that a poem's syntactic patterns are central to its form.

Literature and Cognitive Metaphor

The most recent linguistic approach to literature is that of cognitive metaphor, which claims that metaphor is not a mode of language, but a mode of thought. Metaphors project structures from source domains of schematized bodily or enculturated experience into abstract target domains. We conceive the abstract idea of life in terms of our experiences of a journey, a year, or a day. We do not understand Robert Frost's "Stopping by Woods on a Snowy Evening" to be about a horse-and-wagon journey but about life. We understand Emily Dickinson's "Because I Could Not Stop for Death" as a poem about the end of the human life span, not a trip in a carriage. This work is redefining the critical notion of imagery. Perhaps for this reason, cognitive metaphor has significant promise for some kind of rapprochement between linguistics and literary study.

Suggested Readings

Fludernik, M. 1996. Towards a 'natural' narratology. London and New York: Routledge.

Halle, M., and Keyser, S. J. 1972. "English III. The iambic pentameter." Versification: Major language types, ed. by W. K. Wimsatt. New York: Modern Language Association.

Halliday, M. A. K. 1971."Linguistic function and literary style: An inquiry into the language of William Golding's The Inheritors." Literary style: A symposium, ed. by S. Chatman. London and New York: Oxford University Press.

Turner, M. 1996. The Literary Mind. New York: Oxford University Press.