What Practitioners Can Bring to Theory – The Good and the Bad
©Brian Mossop 2003
The relationship between translational theory and practice is usually phrased in the form: Can theory help practice? In this article, however, I shall be asking the opposite question: What can practitioners bring to theory? My answer will be that translators can both help and hinder theory.
The expression ‘translation theory’ is used in several different ways. Here reference will be made to two of these. The primary sense with which I will be concerned (theory sense 1 or Theory1) will be theory as description and explanation. The purpose of Theory1 is to understand translation. In order to achieve such understanding, theorists describe what is happening now or has happened in the past in the world of translation. They also attempt to explain their findings—to identify the factors that lead translations to have the features which they have observed and described.
I shall also have occasion to refer to a very different sense of ‘theory’, Theory2. Translators with theoretical knowledge in sense 2 can verbalize a rationale for what they are doing. For example such a translator might say: “I have translated this passage in this way rather than in that way because the readership of the translation consists of non-experts in the field while the readership of the source text consisted of experts.” In other words, such a translator is able to invoke a general communicative scheme, in this instance one which distinguishes two types of reader. Theory2 can be seen as a summary of the practice of experienced translators. Translation teachers can pass on Theory2 to students in order to speed up the learning process.
There has been a tendency, within theory sense 1, to focus on the social functioning of translations in target societies and on the ideological goals of translation commissioners as the factors explaining why translations have certain features. Theoreticians have also looked at text-linguistic factors, genre-related factors and cognitive factors. However another set of potential explanatory factors has received little attention, namely factors related to the translator’s workplace, the point of production.
Translation theory has not really registered the fact that much of the translation going on in the world today is commodity production; that is, translations are often done not simply to convey some message but in order to make a living, or a profit. Yet only rarely does one find theorists invoking, as causal factors, the conditions that prevail in translation workplaces, such as:
- the division of labour among translators, terminologists, documentalists, proofreaders, quality controllers, webpage designers, etc
- the deadlines imposed by clients and the demands for speed made by employers
- the functioning of the government or company for which the translators works as either an employee or freelance supplier (conditions of employment and contracting; structure of decision-making, and the like).
If one is seeking the reason why a passage was translated in a certain way, the answer may be not some ideological factor but simply the fact that the translator ran out of time. If one wants to know why a certain translation is not very consistent from one section to another, the answer may be not some cognitive factor but rather the fact that the text was chunked—it was divided up among three translators and no one ensured consistency. If one wants to know why a translation is uneven in style and why the inter-sentence transitions are not very smooth, the answer may be that it was created by cutting-and-pasting paragraphs, sentences or parts of sentences from previous translations or from original target-language documents.
Chunking is a common workplace procedure, as is the re-use of old target-language material. If theorists are not aware of such practices, they will sometimes not succeed in explaining various features of translations. As it happens, almost all empirical studies of translation take place in vitro—at a university campus rather than in a workplace. It seems that there are only one or two published Think-Aloud studies based on research conducted in a translator’s workplace (Riitta Jääskeläinen - personal communication), and no in vivo studies at all using Translog, a program which records the translator’s keystrokes (Arnt Lykke Jakobsen – personal communication).
The lack of descriptions of workplace procedures represents a huge gap in our knowledge about translation. Practitioners could assist theory either by writing about the workplace themselves or by inviting theorists to come and observe them in the workplace.
When practitioners do concern themselves with theory, they have the bad habit of demanding that theory be useful, that it help improve practice. Theory2 can help practice in that formulations invoking abstract schemes (such as the previously mentioned communicative scheme involving reader types) can make translators, and translation students, more aware of the options available to them. Awareness can then lead to self-criticism, and self-criticism opens the door to improvement. However the demand for usefulness is inimical to Theory1, for it implies rejection of any ideas whose utility is not apparent—ideas which may in fact prove to be of explanatory value.
With the advent of Descriptive Translation Studies in the late 1970s, theory finally began to escape the demand for utility. Theory before then was intellectually limited by overconcern with applicability (whether to translator training, machine translation, or—in the case of Nida—missionary work). But the demand for utility continues to be strong and this, in my view, slows the process of understanding. Why? To put the matter in a nutshell, an increase in practical success is not necessarily based on an increase in understanding. Translator training can improve without us having an iota more understanding of what happens when translations are produced.
This is well known from the history of such practical arts as medicine and architecture. For most of history, no one had the faintest idea why certain treatments kept people alive, or why a structure had to be built in a certain way if it was not to fall down. There were improvements in medicine and architecture as time passed, but only in the past 350 years has there been any increase in understanding. Medicine and architecture have became arts which draw on independently established bodies of scientific knowledge.
To see how improvements can occur in practice without any increase in understanding, consider a scenario in which certain members of a group of translators are seen to produce higher quality translations faster than other members of the group. Why is this? A theorist is called in, who proceeds to make careful observations, perhaps using keystroke recordings, think-aloud protocols, video recordings of the translators at work, or interviews. After a while, some factors are found that distinguish the successful from the less successful translators. A correlation has been obtained between observed behaviour during translation production and commercial success.
This is very useful, but has it advanced our understanding of either the mental or the social processes that occur when translations are produced or received? No, for two reasons. First, these research results tell us nothing at all about why behaving in a certain way produces a certain result. Second, the study was slanted at the outset to a predefined notion of translation, namely ‘good’ (i.e. successful) translation. The purpose of Theory1 is not to figure out how commercially successful translations can best be produced but to discover what happens, in general, when translations are produced, whether they are successful or not.
The theory of translation is too often the theory of good translation. Many of the commonest concepts in the literature we think of as theoretical are more or less direct reflections of practical agendas promoting a particular approach to translation. This is the case, for example, with Nida’s distinction between formal and dynamic translation, and Venuti’s distinction between domesticating and foreignizing translation. But a valid Theory1 of translation will not define translation in terms of a particular desired approach or social norm.
Theory1 does not confuse utility with enlightenment. An example of a concept that is useful but not necessarily enlightening is ‘deverbalization’. Everyone in the field of Translation Studies is familiar with those box diagrams which show some sort of translinguistic meaning being extracted from the source text and then re-expressed in the target language. Translation teachers have found it useful to talk about ‘translating ideas, not words’. But the fact that the notion of deverbalization is pedagogically useful does not mean that anything like deverbalization actually takes place in the brain.
Now consider the opposite case: a phenomenon whose investigation would be enlightening but not useful. Interference is a much noted feature of translation. An interesting thing about interference is that it does not occur with every syntactic structure. For example, in translations from French by native speakers of English, one might find the ungrammatical ‘Mary sees often John’ (for Marie voit souvent Jean), but one would not, I think, ever find ‘Mary him sees’ (for Marie le voit). The question is: why not? This is an interesting theoretical question if one is interested in knowing what happens in the mind when people are translating. However the answer would not be of any practical value, since even the worst students do not produce sentences like ‘Mary him sees’, and hence do not need help to avoid producing them.
A further bad effect of the demand for utility is that it parcelizes translation theory, in the sense that every translation practice will tend to ‘call for’ its own theory that can serve it (both by helping it to improve and by enhancing its status through the aura of respectability created by theory).
Take the distinction between translating and interpreting. Certainly there are differences between these modes, reflecting differences between speech and writing, but then there are also differences within the oral mode between simultaneous interpreting and dialogue interpreting; and there are differences within the written mode between translating the script of a play and translating a Web page. Why is any special status given to the oral/written distinction, leading to the recent creation of the separate (sub?)discipline of Interpretation Studies, with its own apparatus of journals and conferences? The reason such status is accorded stems from practice: it reflects the separate professional organization of translators on the one hand and interpreters on the other. One might perhaps similarly explain the still largely separate sphere of literary translation theory as a reflection of the distinctive economic status and mode of employment of literary translators. Indeed, one might predict that if the translators of Web pages come to be organized into a separate profession, then a theory of Web translation will arise. This would doubtless be a good thing from the point of view of Theory2, but for Theory1 it would not.
Understanding translation means searching for truths about translation in general, between all language pairs, at all times and places, and in all genres and modes. There cannot be a Theory1 of Web page translation any more than there can be a Theory1 of birds: Web translation is translation, just as birds are living things that fall within the scope of biology.
What then is the proper relationship between Theory1 and practice? I suggest a three-part answer. First, practice provides data to suggest or test hypotheses. (Example: observations of a translator’s workplace might be used in this way). Second, while theorists do not aim to improve practice, they do need to know about practice in order to be able to identify those things that need explaining. (An example of theorizing uninformed by practice would be the largely vacuous discussions about ‘untranslatability’ in the 1960s and 1970s.) Third, a theory may turn out to be useful, even if utility is not its purpose.
On this third point, consider Ernst Gutt’s work on translation and relevance (Gutt 2000). Whatever reservations one might have about his specific claims, his approach to theory is the right one from the point of view of advancing our understanding of translation. He uses the concept of relevance (achieving maximum contextual effects with minimum processing effort) to explain observations about translations. The concept of relevance is drawn from general considerations about language; it is not a Theory2-type summary of some aspect of translation practice. Nevertheless teachers might find the idea useful. Andrew Chesterman (2003:43) suggests that the the idea of achieving maximum contextual effects can perhaps be seen as providing a theoretical justification for taking into account the reader’s background and expectations, while the idea of minimum processing effort provides a justification for writing clearly in the target language even if the source text is not clearly written. Thus a translator might use a formulation phrased in terms of reader’s background and ease of reading in order to provide a rationale for a particular translation; and a teacher might use it to help students translate better. But from the point of view of theory sense 1, it does not matter whether this is possible. It does not matter whether teachers or translators find relevance or any other theoretical concept useful. The understanding sought by theory sense 1 is an end in itself.
Chesterman, Andrew & Emma Wagner, Can Theory Help Translators? Manchester: St Jerome 2002.
Gutt, Ernst, Translation and Relevance 2nd edition. Manchester: St Jerome 2000.
Mossop, Brian, “What is a Translating Translator Doing?” in Target 10:2, 231-266.
Peeters, Jean, La médiation de l’étranger, Artois Presses Universitaires, 1999.
 For an attempt at a definition which is neutral with respect to differing social norms of translation, see Mossop 1998 as well as ‘A socially neutral definition of translation’, available at http://www.geocities.com/brmossop/mypage.html. A very different but still neutral approach, presenting translation as a special case of intersubjective exchange, is offered in Peeters 1999.
 I do not think that it does, but this article is not the place to set out an alternative; see ‘An Alternative to ‘Deverbalization’, available at http://www.geocities.com/brmossop/mypage.html. Here is a simple illustration of the point from another field, the pedagogy of singing. Singing teachers frequently suggest placing a sound behind the forehead, or imagining a high note as going out the top of the head. These suggestions have proven to be extremely helpful, but from an explanatory point of view they are absurd: there are no resonating chambers behind the forehead, or holes in the top of the skull. Just as singing students are told to imagine that they are pushing the sound out the top of the head, so translation students can be told to imagine that they are translating ideas. In actual fact, their brains may well be making direct links between SL wordings and TL wordings.