«el indiferente anteayer se convierte de golpe en 'los últimos años', según la fórmula de las crónicas y biografías, que a menudo dicen eso del muerto, 'durante sus últimos años...', como si hubiera podido anticiparlo nadie»¹.

"what was no more than the day before yesterday is suddenly transformed into 'the final years', in the standard phrase of articles and biographies, which often speak of the deceased 'during his final years', as if anyone could have anticipated that"².

One of the main points studies of translation processes revolve around is the description of translation choices or, if you want to consider them from the point of view of subsequent process, (see previous unit), of the shifts existing between prototext and metatext.

1. Models analyzing micro-structural shifts (those concerning small units of text, words for example) independent of the textual context and draw inferences concerning the whole text, or at least produce a list of micro-shifts that interest a given metatext.
2. Then there are chronotopical models that aim at creating a typology of (meta)texts, considered as whole texts, rather than as groups of micro-shifts, taken as whole macro-shifts. In this case, the shift categories are not absolute as above, but are related to the poetological and culturological context of the text in question.

An example of the former kind is Leuven-Zwart's model, an example of the latter type is Torop's chronotopical model, that we have already seen in unit 21 of the first part and reproduced here:

Let us start from the model of the Netherlands researcher van Leuven-Zwart, who was among the first to study translation criticism from a pragmatic point of view, concretely trying to solve the question of the comparative analysis of two texts in general, of prototext and metatext in particular. Her method is applicable to intralingual translation as well.

Aim of the model is to describe microstructural shifts deriving from a conscious or unconscious choice by the translator, on a semantic, syntactic or pragmatic level, and classify important shifts as Lefevere had described them³. After the comparative-model analysis, a descriptive model is used, analyzing the effects of micro-structural shifts on macrostructure, i.e. on characters, events, time, space and other textual elements. As you can see, this model proceeds from the bottom up, because no preventive analysis is made of the text's poetics from which one can arrive at the categories to be put to comparative analysis.

In order to understand this model, it would be necessary to get familiar with Leuven-Zwart's vocabulary: by "transeme" the researcher means an understandable text unit with or without a predicate. Any speech act can be disassembled into transemes.

By "relationship" similarities or dissimilarities between the texts are meant and, in the given case, between transemes of a text and transemes of the other texts that is compared to it. There is, moreover, another complex concept, "architranseme": it is the common denominator between prototranseme and metatranseme, expressed in simplified terms.

I think that the introduction of these new terms (with the exception of the word "relationship", altogether understandable) in this course would be a mistake, and would create some confusion, without necessarily giving the reader a more precise or more articulate vision. Therefore, having named the terms to give the readers the opportunity to recognize them in an eventual reading, I take on the duty of translating Leuven-Zwart's view into a form that I consider more easily understandable.

Three kinds of relations are identified as follows:

• contrast relation cases in which an element of the text is modified rendering it unrecognizable. Contrast can be recognized as

• an omission
• an addition or
• a radical change of sense.

Here is an example:

PT Gertrude stood up

MT Camilla thought

• There is a modulation relationship (binary change) when the change occurring between a prototext's and a metatext's element follows the logics of a dichotomy, of a bifurcation along the continuum generalization versus specification or along another bipolar continuum.

For example:

PT Gertrude walked

MT Gertrude ran (= walk + fast) ? modulation, specification

In Leuven-Zwart's model, modulation relationships (specification or generalization) can be further divided depending on the fact they concern: culture (nation, region, ethnic group, community, profession etc.) non verbal time (historization, modernization, achronization), space (localization, exotization, universalization) text style (accentuation, reconstruction, leveling), register (accentuation, reconstruction, leveling), rhetoric figures (preservation, adaptation, paraphrase), author's idiolect (accentuation, reconstruction, leveling), character's idiolect (accentuation, reconstruction, leveling).

• There is a non-binary shift when between the two elements of prototext and metatext there is a difference that, however, not ascribable to any dichotomy, it is not along any continuum between poles, rather it is a difference of a non-binary type. It refers, instead, to a text element to which there is not one single alternative but many.

For example

PT Yesterday

MT Today

is a non-binary change, because possible alternatives to "Yesterday" on the paradigmatic ax are manifold (for example tomorrow, one day, in a week, last month etc.)

This relationship considers also the categories, mostly morphological or grammatical or syntactic, that, owing to their nature, could not consist of a binary modulation. Here are some of such categories:

• tense
• number
• mood
• gram. person
• gram. class
• verb form (passive/active)
• verb aspect (perfective/imperfective)
• word order
• explicitation/implicitation (number of informative elements)
• deixis/anaphora (intratextual and extratextual references)
• assoncance, alliteration, rhyme

After this first part of the essay, in which she analyzes various micro-shifts, in the second van Leuven-Zwart examines the potential effects of such shifts, especially on a large scale, on the macrostructure of the text. Specification, depending on the elements involved, can modify the text in an emotional, picturesque, evocative, suggestive, aggressive, or stereotyped way⁴. In specification, the fact that the reader's attention is directed toward specific details can involve some distraction from other aspects of the text. Moreover

Another emphasized element is that some characters have different idiolects according to the situations they are in: the homogenization of the expressive modes on a single idiolect in the metatext has an unavoidable impact on its reception. The verbal shift of the person modifies the emotional involvement of the reader, or of the inner narrator, with respect to the characters. Syntactic structure can cause differences in the text perception, particularly with right and left text displacement.

Deixis and anaphora deserve separate coverage. First, using a pronoun instead of a noun implies a given degree of implicitness: "If, in the narrator's opinion, sufficient pragmatic information is given in order to adequately interpret a referring expression, the cross-reference will be achieved through deictic/anaphoric elements"⁶. If a translator substitutes a deictic expression for its reference name, the attitude toward the reader is changed, therefore the author's strategy is changed.

Van Leuven-Zwart makes some general conclusions on translations in the second half of the 20th century. There is an abundance of specification and explicitation (a trend that, as we have seen, closes interpretive perspectives left open in the prototext), as if "the translator tries to make things 'logical' and comprehensible for the reader"⁷. This general trend, this unstated, but documentable 'norm', of modern translations "seems to be in accordance with Toury's initial norm of acceptability: passages which, in the translator's opinion, are difficult, strange or 'illogical' are adapted to the assumed taste of the reader"⁸. If it is true that sometimes adaptations are necessary because otherwise the utterance would sound unnatural and clumsy, it is true that often they are voluntary as well: when using them, a subjective element is introduced.

In the next unit we will further examine possible models of shift relations.

Bibliographical references

LEFEVERE A. Translating Poetry. Seven Strategies and a Blueprint. Amsterdam, Van Gorcum, 1975, ISBN 90-232-1263-0.

LEUVEN ZWART K. van Translation and original. Similarities and dissimilarities. In Target, n. 1:2 (1989) e n. 2:1 (1990).

MARÍAS J. Negra espalda del tiempo, Punto de lectura, 2000 (original edition 1998), ISBN 84-663-0007-7.

MARÍAS J. Dark Back of Time, New York, New Directions, 2001 (translated by Esther Allen), ISBN 0-8112-1466-4.

TOROP P. La traduzione totale, edited by Bruno Osimo, Modena, Logos, 2000, ISBN 88-8049-195-4.