Descriptive Rhetoric

It may be clear by now that what I am advocating, as one of the best services linguistics can at present pay to literary studies, is a ' descriptive rhetoric'. By this I mean a body of theory and technique devoted to the analysis of the characteristic features of literary language, and to the explanation of terms in the critic's vocabulary, where this can be done, using the linguist's insights at a level where they become useful to the student of literature. The present book, limited as it is in breadth of scope and depth of detail, will be, I hope, a step in this direction.

It may be helpful, in this light, to discuss two much criticized aspects of the traditional handbook of rhetoric. The first of these is its preservation of, and seeming reverence for, a vocabulary of unnecessarily difficult technical terms. Beside such well-known words as 'metaphor' and 'irony', as names for rhetorical figures, are many more forbidding Greek labels like 'epana-lepsis ', ' homoiotelcuton and ' antistrophe '. It would be foolish to lay any store by the mastery of this cumbersome terminology in an age when the classical languages and cultures are little studied. However, because such terms have a certain currency in literary scholarship, and serve a real communicative purpose, they cannot be altogether discarded. It would be even more foolish, in the present age, to try to replace the classical terms by a completely new terminology, as George Puttenham, the Elizabethan literary theorist, did in his Arte of English Poesie.3 As a considerable part of the present book is concerned with what are traditionally known as 'rhetorical figures' or 'figures of speech',4 it is as well to bear in mind from the start that the technical names for these figures are not sacrosanct, nor have their definitions been laid down once and for all time. In fact, the definitions of rhetorical terms have always been notorious for vagueness and inconsistency. My main preoccupation will be not how to define these terms, but how to get at the realities behind them - that is, the basic characteristics of poetic language.

Connected with this is a second weakness of traditional rhetoric - its cultivation of what I am tempted to call the' train-spotting' or' butterfly-collecting' attitude to style. This is the frame of mind in which the identification, classification, and labelling of specimens of given stylistic devices becomes an end in itself, divorced from the higher goal of enriching one's appreciation and critical understanding of literature. The response conveyed by 'Aha, there's an instance of hystcron proteron' is one of satisfaction without enlightenment. This train-spotting mentality was particularly prevalent in Elizabethan times,5 but its persistence to the present day is shown in the survival in modem textbooks of figures like hendiadys, which we can value only as curiosities. Hendiadys (Greek for 'one-by-two') consists in the use of a co-ordinating construction where a structure of modification would be strictly appropriate: 'charmed by bright eyes and a woman' instead of'charmed by the bright eyes of a woman'. It is so rare that I have found no certain instance of it in English literature.

There is danger of train-spotting whenever anyone tries, as I do in this book, to deal with the general properties of poetic language, without particular attention to a given text, a given writer, or a given period. With such a programme, one cannot help (except by avoiding illustrations altogether) quoting short passages, lifted from their contexts, simply as instances of this or that stylistic feature. The corrective to this use of labelled specimens lies in the opposite approach, whereby a student considers a characteristic of language only within the context of the poem to which it belongs, as a contribution to its total communicative effect. This is the method of'practical criticism'.

However, both these approaches, the isolating and the synthesizing of stylistic effects, are necessary roads to the understanding of language in literature. We cannot appreciate how a poem fits together, unless we have first found the means to take it to pieces. Detailed exegesis of poems uses up more space than this book can accommodate, so I cannot avoid a certain bias towards specimen-collecting. But in the section called 'Examples for Discussion' at the end of each chapter, the student is invited to redress the balance for himself, by putting the content of that chapter and previous chapters to work on the explication of lengthier passages of poetry, sometimes of whole poems. I therefore stress at this point the importance of these exercises, which are indispensable to the plan of the book.


The investigation of poetic language cannot proceed very far unless we have some notion of the relation between the kind of language which occurs in poetry, and other kinds of language. Here, if anywhere, we would expect linguistics, as the study of language in general, to help; for the subject matter of linguistics is all language - language as used not only in literary composition, but in everyday gossip, in scientific reports, in commercial or political persuasion, and in a multitude of other more or less mundane functions. The literary critic, on the other hand, concentrates on that relatively minute, but inordinately precious body of texts which are thought worthy of preservation as 'literature', to be studied for their own sake, rather than for their extrinsic value as (say) guide books or political tracts. Both the critic and the linguist are to some extent involved in the same task of describing and explaining linguistic communications: but in comparison with that of the critic, the linguist's perspective is broad and unspecialized. His approach to literature may be in many ways a crude one, but it results in generalizations and particular observations which could not easily be made from the critic's point of view.

As the position of poetic language with respect to 'ordinary' language is the subject for discussion in the first and second chapters, I shall merely anticipate here themes important to this book as a whole by observing that the relation between the two is not a simple one, and has at least three aspects:

1. Poetic language may violate or deviate from the generally observed rules of the language in many different ways, some obvious, some subtle. Both the means of and motives for deviation are worth careful study.

2. The creative writer, and more particularly the poet, enjoys a unique freedom, amongst users of the language, to range over all its communicative resources, without respect to the social or historical contexts to which they belong. This means, amongst other things, that the poet can draw on the language of past ages, or can borrow features belonging to other, non-literary uses of language, as Ezra Pound and T. S. Eliot, for example, have made use of the English of banal, prosy conversation in some of their poems.

3. Most of what is considered characteristic of literary language (for example, the use of tropes like irony and metaphor) nevertheless has its roots in everyday uses of language, and can best be studied with some reference to these uses.

Just as there is no firm dividing line between 'poetic' and 'ordinary' language, so it would be artificial to enforce a clear division between the language of poetry, considered as verse literature, and that of other literary kinds. I shall not hesitate to make use of prose illustrations where they are apposite, but in general the topics to be discussed can be more strikingly exemplified by verse extracts.

0 0


  • Sabrina
    What is descriptive rhetoric?
    6 years ago

Post a comment