In the previous three units, we focused mainly on two mental activities ' reading and writing ' that are part of the translation process, and we tried to describe some of their phases and the applicable implications with reference to mental functioning. We observed that ' even within the framework of one code, that is to say without shifting to another language ' we have to accomplish more than one translation process, involving nonverbal processing. Moreover, we observed that there is an intermediate stage during which words, or word combinations, are translated into an idiosyncratic mental nonverbal language that is understandable (and hence translatable into words) only by the individual accomplishing such effort in her mind.

  Such analytic exam was necessary in order to identify the single mental processes involved in the mentioned activities; we know, however, that such activities are actually carried out in a minor span. During this mental work, there is a constant focusing shift between microanalysis and microanalysis, between micro-expression and macro-expression, i.e. a constant comparison between the meaning of the single utterances and the meaning of the text as a whole. Or, on a larger scale, a constant comparison between the sense of a single text and the comprehensive sense of the corpus that, consciously or unconsciously, forms the "intertext". In this context, "intertext" should be understood as the complex of intertextual links in which a text is located, with, or without, the authors acknowledgement.

  After said analytic exam, we must bear in mind that the mental processing of verbal data undergoes many simultaneous, interdependent, and holistic processes¹. In order to describe the mental process occurring while translating, it is necessary to temporarily put aside the individual mechanisms of the microactivities and analyze the translation process in the whole, with a systemic approach.

  An important translation-studies researcher, James S. Holmes, has proposed a mental approach to translation processes, the so-called «mapping theory». He presents a synthesis of his approach in this paragraph:

  The translation process should, therefore, be considered a complex system in which understanding, processing, and projection of the translated text are interdependent portions of one structure. We can therefore put forward, as does Hönig, the existence of a sort of "central processing unit" supervising the coordination of the different mental processes (those connected to reading, interpretation, and writing) and at the same time projecting a map of the text to be.

  Let us follow Hönig's passages. The original, in order to be translated, is "moved out" of its natural context and projected onto the translator's mental reality. The translator does not work on the original text, consequently, but on its mental projection. There are two kinds of processing, the controlled workspace, and the uncontrolled workspace. In the uncontrolled workspace, the first understanding of the text takes place, consisting of the application of frames and schemes, an assortment of semantic patterns based on the perceptive experience of the translator. Such semantic schemes are not very different, from a conceptual point of view, from the cognitive types we dealt with in the unit about the reading process.

  As it happens in reading (it is possible to read fractions of words or sentences and construct the unread parts), using the semantic schemes our minds tend to postulate the affinity of the utterances in the original to utterances already read or heard and assimilated.

  Semantic schemes are long-term memory structures reflecting the reader's expectations, her meaning conjectures, and in part are already oriented towards a translated text that ' although existing only within the translator's mind ' is taking shape in her mental map.

  Translation micro-strategies are composed of the interaction between the original text, the hypotheses on the translated text and the uncontrolled workspace. For the experienced professional a nearly automated process can become more conscious owing to the translation-oriented analysis of the text.

  Researchers postulated the existence of an uncontrolled workspace using a thinking aloud protocol. Some translators were asked to say aloud what they were doing or thinking to do while they were intent in their work. Mental processes described by these protocols are those that are called "controlled workspace". Uncontrolled workspace contains, in contrast, mental activities different from those described in thinking aloud protocols. In the controlled workspace, mental processing is conscious: the translator knows that given mechanisms take place but, at the same time, she is usually unaware of them because she performs them automatically.

  A translator using only the uncontrolled workspace does not have any comprehensive strategy, which takes into consideration the translated text as a whole. Such a hypothetical translator is guided only by her linguistic reflexes originating from her perception of the original text. If one wants to achieve a complete translation competence, it is necessary to adopt a rational macro-textual strategy as well.

BIBLIOGRAPHICAL REFERENCES:

HOLMES J. S. Translated! Papers on Literary Translation and Translation Studies. Amsterdam, Rodopi, 1988. ISBN 90-6203-739-9.

HÖNIG H. G. Holmes' "Mapping Theory" and the Landscape of Mental Translation Processes, in Leuven-Zwart and Naaijkens (ed.) Translations Studies: The State of the Art. Proceedings of the first James S. Holmes Symposium on Translation Studies, Amsterdam, Rodopi, 1991. ISBN 90-5183-257-5, p. 77-89.