We actually do not know precisely what takes place in the translator's mind when he translates, for psychologists and neurologists do not know the manner in which language data are stored in the brain. The fact that in some brain injuries the capacity of bilingual persons has been reduced to the speaking of one language and not the other has led to the belief that there may be compartmentalization in the brain. However, this hypothesis has been generally rejected.

One argument against any strict compartmentalization is the fact that a person who knows several languages is not always aware of the language in which certain concepts were communicated to him. However, a person who has, for example, studied mathematics in German, music in French, and science in English seems to retain a good deal of this compartmentalization in his participation in such activities. At the same time, merely having a knowledge of two languages is no guarantee that a person can function as a translator; for in certain individuals there seems to be no connection between the two sets of experiences which provide the cultural contexts for the use of the two languages. If a person is to serve as a translator, and especially if he is to be an interpreter of continuous discourse (simultaneous interpreting), he must have had a good deal of experience in language switching. Moreover, some bilingual persons are good in translating from one of their languages into the other, but cannot easily reverse the roles. Before we can actually understand the psychological basis of translating, a great deal of research must be done.

Fortunately, however, for the purpose of our study of the translator's role, we need not have a comprehensive understanding of the psychological processes involved. These we shall simply have to take for granted. More important for us is a careful analysis of the larger cultural context into which the translator's activity fits.

In contrast with the relatively simple model of one-language communication, the translator's activity is obviously complex, as the diagram in Figure 1 attempts to show.

In this model a message in language A is decoded by the receptor into a different form of language A. It is then transformed by a "transfer mechanism” into language B, and the translator then becomes a source for the encoding of the message into language B. If we understood more precisely what happens in this transfer mechanism, we should be better able to pinpoint some sources of the difficulty persons have in interpreting from one language to another. One thing we do know, however--- that the translator must not only discover corresponding symbols with which to communicate the message in B, but must also organize these symbols in the form required by language B. Basically we may describe translating as a process in which the concept is transferred, possibly in essentially "kernel" form, and then the corresponding utterance in language B is generated. Some descriptions of the translational process would seem to imply that the procedure consists merely in matching successive corresponding symbols and grammatical structures of the two languages. In some artificially simple situations this may be true, but it is much more likely that the message of language A is decoded into a concept, and that this concept then provides the basis for the generation of an utterance in language B.

In terms of a psychology employing the image-plan model (Miller, Galanter, and Pribram, 1960), we could say that the transfer takes place at the level of the image, not at some stage of the plans. If the transfer occurs on the level of plans, especially on the lower hierarchical levels, the resulting translations will inevitably be awkward and artificial. Persons with a good deal of experience in simultaneous translating may become so expert in "matching" that they tend to short-circuit this transfer on the image level by selecting certain frequently recurring correspondences on the level of plans (before the conceptual image is complete). Careful literary translating must be concerned with "meaningful mouthfuls" (i.e. with reasonably whole images) and with translating entire concepts, not merely series of words or isolated fragments.

Anyone experienced in translating from one particular language to another does tend to speed up the process of transfer by anticipating in the decoding something of the encoding process. But if there is too much automatic or mechanical adjustment, there is likely to be severe distortion, for each language must be treated as a system in and of itself.

责任编辑：admin