TRANSLATION THEORY(2)

What is a translation theory?

The increasingly interdisciplinary nature of translation studies has multiplied

theories of translation. A shared interest in a topic, however, is no guarantee

that what is acceptable as a theory in one field or approach will satisfy the

conceptual requirements of a theory in others. In the West, from antiquity to the

late nineteenth century, theoretical statements about translation fell into

traditionally defined areas of thinking about language and culture: literary theory

and criticism, rhetoric, grammar, philosophy. And the most frequently cited

theorists comprised a fairly limited group. One such catalogue might include:

Cicero, Horace, Quintilian, Augustine, Jerome, Dryden, Goethe, Schleiermacher,

Arnold, Nietzsche. Twentieth-century translation theory reveals a much

expanded range of fields and approaches reflecting the differentiation of modern

culture: not only varieties of linguistics, literary criticism, philosophical

speculation, and cultural theory, but experimental studies and anthropological

fieldwork, as well as translator training and translation practice. Any account of

theoretical concepts and trends must acknowledge the disciplinary sites in

which they emerged in order to understand and evaluate them. At the same

time, it is possible to locate recurrent themes and celebrated topoi, if not broad

areas of agreement.

Louis Kelly has argued that a “complete” theory of translation “has three

components: specification of function and goal; description and analysis of

operations; and critical comment on relationships between goal and operations”

(Kelly 1979:1). Kelly is careful to observe that throughout history theorists have

tended to emphasize one of these components at the expense of others. The

component that receives the greatest emphasis, I would add, often devolves into

a recommendation or prescription for good translating.

The Latin poet Horace asserted in his Ars Poetica (c. 10 BC) that the poet

who resorts to translation should avoid a certain operation—namely, word-forword

rendering—in order to write distinctive poetry. Here the function of

translating is to construct poetic authorship. In a lecture entitled “On the

Different Methods of Translating” (1813), the German philosopher and theologian

Friedrich Schleiermacher advocated word-for-word literalism in elevated

language (“not colloquial”) to produce an effect of foreignness in the translation:

“the more closely the translation follows the turns taken by the original, the

more foreign it will seem to the reader” (Lefevere 1992a:155). For

Schleiermacher, textual operations produced cognitive effects and served

cultural and political functions. These operations, effects and functions were

described and judged according to values that were literary and nationalist,

according to whether the translation helped to build a German language and

literature during the Napoleonic wars. Even with modern approaches that are

based on linguistics and tend to assume a scientific or value-free treatment of

language, the emphasis on one theoretical component might be linked to

prescription. During the 1960s and 1970s, linguistics-oriented theorists

emphasized the description and analysis of translation operations, producing

typologies of equivalence that acted as normative principles to guide translator

training.

The surveys of theoretical trends in the section introductions have both

benefited from and revised Kelly’s useful scheme. To my mind, however, the key

concept in any translation research and commentary is what I shall call the

relative autonomy of translation, the textual features and operations or

strategies that distinguish it from the foreign text and from texts initially written

in the translating language. These complicated features and strategies are what

prevent translating from being unmediated or transparent communication; they

both enable and set up obstacles to cross-cultural understanding by working

over the foreign text. They substantiate the arguments for the impossibility of

translation that recur throughout this century. Yet without some sense of

distinctive features and strategies, translation never emerges as an object of

study in its own right.

The history of translation theory can in fact be imagined as a set of changing

relationships between the relative autonomy of the translated text, or the translator’s

actions, and two other concepts: equivalence and function. Equivalence has

been understood as “accuracy,” “adequacy,” “correctness,” “correspondence,”

“fidelity,” or “identity”; it is a variable notion of how the translation is connected to

the foreign text. Function has been understood as the potentiality of the translated

text to release diverse effects, beginning with the communication of information

and the production of a response comparable to the one produced by the foreign

text in its own culture. Yet the effects of translation are also social, and they have

been harnessed to cultural, economic, and political agendas: evangelical programs,

commercial ventures, and colonial projects, as well as the development of

languages, national literatures, and avant-garde literary movements. Function is a

variable notion of how the translated text is connected to the receiving language

and culture. In some periods, such as the 1960s and 1970s, the autonomy of

translation is limited by the dominance of thinking about equivalence, and

functionalism becomes a solution to a theoretical impasse; in other periods, such

as the 1980s and 1990s, autonomy is limited by the dominance of functionalisms,

and equivalence is rethought to embrace what were previously treated as shifts or

deviations from the foreign text.

The changing importance of a particular theoretical concept, whether autonomy,

equivalence or function, may be determined by various factors, linguistic and

literary, cultural and social. Yet the most decisive determination is a particular

theory of language or textuality. George Steiner has argued that a translation theory

“presumes a systematic theory of language with which it overlaps completely or

from which it derives as a special case according to demonstrable rules of

deduction and application” (Steiner 1975:2801). He doubted whether any such

theory of language existed. But he nevertheless proceeded to outline his own

“conviction” before offering his reflections on translation.

A translation theory always rests on particular assumptions about language

use, even if they are no more than fragmentary hypotheses that remain implicit or

unacknowledged. For centuries the assumptions seem to have fallen into two

large categories: instrumental and hermeneutic (cf. Kelly 1979: chap. 1). Some

translation theories have assumed an instrumental concept of language as

communication, expressive of thought and meaning, where meanings are either

based on reference to an empirical reality or derived from a context that is primarily linguistic, but may also encompass a pragmatic situation. Other theories have

assumed a hermeneutic concept of language as interpretation, constitutive of

thought and meaning, where meanings shape reality and are inscribed according

to changing cultural and social situations. An instrumental concept of language

leads to translation theories that privilege the communication of objective

information and formulate typologies of equivalence, minimizing and sometimes

excluding altogether any question of function beyond communication. A

hermeneutic concept of language leads to translation theories that privilege the

interpretation of creative values and therefore describe the target-language

inscription in the foreign text, often explaining it on the basis of social functions

and effects.

These concepts of language and translation are obviously no more than

abstractions. Before they can contribute to any explanation or interrogation of

translation theories and practices, they require analysis in specific historical

contexts.

In the section introductions they have been used as heuristic devices to describe

and distinguish among different theoretical texts and trends.