shifts1

Task A4.3
➤ What means of comparison did you use when assessing shifts in Task A4.2
above? How objective do you feel this comparison was?
Attempts at objectifying the comparison have included van Leuven-Zwart’s
Architranseme concept where the dictionary meaning of the ST term was taken
as a comparator and used independently to evaluate the closeness of the ST and TT
term (van Leuven-Zwart 1989, 1990). However, the success of the Architranseme
rests upon the absolute objective dependability of the decontextualized dictionary
meaning and the analyst’s ability to accurately and repeatedly decide whether a
shift has occurred in the translation context. In view of the difficulty, not to say
impossibility, of achieving this, many theorists have moved away from the tertium
comparationis (see Snell-Hornby 1990). Gideon Toury is the Israeli scholar who
has been the prime proponent of Descriptive Translation Studies, a branch of the
discipline that sets out to describe translation by comparing and analysing ST–TT
pairs. In his work, Toury initially used a supposed ‘invariant’ as a form of comparison
(Toury 1980), but in his major work Descriptive Translation Studies – and
Beyond (Toury 1995) he drops this in favour of a more flexible ‘ad-hoc’ approach
to the selection of features, dependent on the characteristics of the specific texts
under consideration. Importantly, he warns against ‘the totally negative kind of
reasoning required by the search for shifts’ (Toury 1995:84) in which error and
failure and loss in translation are highlighted. Instead, for Toury translation shift
analysis is most valuable as a form of ‘discovery’, ‘a step towards the formulation of
explanatory hypotheses’ about the practice of translation (1995:85). The relevance
and applications of translation shifts are issues which we shall explore further in
Sections B and C

This unit describes a theoretical position that promotes the systematic analysis of
the changes that take place in moving from ST to TT. A change, known technically
as a ‘shift’, is generally any translation that moves away from formal correspondence.
Analysis normally first requires identification of the translation unit. The bestknown
work in this area is by Catford, who first used the term shift, and by Vinay
and Darbelnet, whose detailed taxonomy has influenced many theorists. But, as we
shall see in Unit 11, shifts also occur on the higher levels of text, genre and discourse..

Tertium comparationis1

A non-linguistic, intermediate form of the meaning of a ST and TT. The idea
is that an invariant meaning exists, independent of both texts, which can be
used to gauge or assist transfer of meaning between ST and TT

This can be described graphically as follows:
ST chunk TT chunk
comparator
This has long been a thorny issue in Translation Studies and no one measure
has ever been accepted by all..

Compensation, loss and gain

A translation technique used to compensate for translation loss. The translator
offsets an inevitable loss at one point in the text by adding a suitable
element at another point, achieving a compensatory translation gain.
For example, an informal text in French using the second personal pronoun
tu might be rendered in English by informal lexis or use of the first name
or nickname. Compensation in an interpretive sense, restoring life to the TT,
is the fourth ‘movement’ of Steiner’s hermeneutic process (Steiner 1998:39,
see Part A, Unit 13).

These translation procedures have influenced later taxonomies by, amongst others,
van Leuven-Zwart (1989, 1990), who attempts a very complex analysis of extracts
from translations of Latin American fiction. However, despite a systematic means
of analysis based on the denotative meaning of each word, the decision as to whether
a shift has occurred is inevitably subjective since an evaluation of the equivalence
of the ST and TT units is required. Some kind of evaluator, known in translation
as a tertium comparationis, is necessary.

shiffts

Task A4.2
➤ The Eurostar ST has been reproduced below together with the German
translation. Look at the translation units that are matched up and, using the
back-translation to help you, note any ‘mismatches’, denoting shifts.
Example A4.3a
Eurostar
Please note that smoke detectors will be fitted on-board. Any misconduct will result in
necessary action being taken by rail staff and/or police.
Example A4.3b German
Beachten Sie bitte, daß die Züge mit Rauchdetektoren ausgestattet werden. Jeder
Verstoß wird mit den erforderlichen Maßnahmen durch das Bahnpersonal und/oder die
Polizei geahndet.
[Note please that the trains with smokedetectors fitted will be. Each violation will-be
with the necessary measures through the railstaff and/or the police punished.]

ST TT
1. please bitte
2. note beachten Sie
3. that daß
die Züge
4. smoke detectors mit Rauchdetektoren
5. will be fitted ausgestattet werden
6. on-board
7. Any misconduct Jeder Verstoß
8. will result in wird . . . mit
9. necessary action den erforderlichen Maßnahmen
10. being taken geahndet
11. by rail staff durch das Bahnpersonal
12. and/or und/oder
13. police die Polizei

Clear shifts in the second sentence can be seen in translation unit 7, where any
misconduct becomes the more specific and stronger ‘each violation’ in the German,
and in units 8 and 10, where will result in . . . being taken is altered to ‘will . . . with
. . . be punished’.Yet numerous issues arise when this type of analysis is undertaken,
not least what the translation unit is. This is illustrated by the term smoke detectors
in this and other versions (the leaflet also contained French and Flemish versions):
the German and Flemish have a one word equivalent (Rauchdetektoren and rookdetectoren
respectively) but the French needs the multi-word unit détecteurs de fumée
[‘detectors of smoke’]. Few would argue that the translations are correct and close
equivalents, but should the number of word forms used be taken into consideration
when deciding if a shift has taken place? Similarly, has a shift occurred in the
German because of the obligatory placing of the passive werden at the end of sentence
1? And how are we to decide if Verstoß, infraction (in the French) or overtreding
(Flemish) involves a shift from ST misconduct?
Identifying that a shift has taken place leads to questions such as what kind of
shift, what form of classification we can use and what the importance of the shifts
is. As will begin to become clear in Section B Text B3.1, Vinay and Darbelnet’s
categorization of translation procedures is very detailed. They name two ‘methods’
covering seven procedures:
1. direct translation, which covers
borrowing, calque and literal translation, and
2. oblique translation, which is transposition, modulation, equivalence and
adaptation.
30
I n t r o d u c t i o n
A
SECTION
ST TT
1. please bitte
2. note beachten Sie
3. that daß
die Züge
4. smoke detectors mit Rauchdetektoren
5. will be fitted ausgestattet werden
6. on-board
7. Any misconduct Jeder Verstoß
8. will result in wird . . . mit
9. necessary action den erforderlichen Maßnahmen
10. being taken geahndet
11. by rail staff durch das Bahnpersonal
12. and/or und/oder
13. police die Polizei

These procedures are applied on three levels of language:
i. the lexicon
ii. the grammatical structures and
iii. the ‘message’,which is used to refer to the situational utterance and some of the
higher text elements such as sentence and paragraphs.
At the level of message, Vinay and Darbelnet discuss such strategies as compensation,
an important term in translation which is linked to the notion of loss and
gain.

shifts translaton11

Task A4.1
➤ Look at these two examples. How many departures from formal correspondence
can you detect? How do you decide what a departure is?
Analysing these examples, it is clear that there are many formal correspondences at
lexical and grammatical levels:
please – bitte
beachten – note
that – daß
smoke detectors – Rauchdetektoren
will be – werden

Systemic differences between the languages must be accepted. These include wordorder
changes and the construction of the German imperative with the addition
of the pronoun Sie [‘you’]. However, there is a clear departure from formal
correspondence in the translation of the ST on-board and the restructuring of the
second clause. In this text, the only possible textual equivalent for on-board is die
Züge (‘the trains’) which is added with a change of grammatical subject (ST smoke
detectors to TT die Züge). The analyst then has to decide whether ausgestattet is a
formal correspondent of fitted. A dictionary definition is not enough since some
dictionaries may give ausstatten as a translation of to equip or fit out but not fit.
However, the role occupied by ausgestattet and fitted in the two languages is very
similar, so it is highly unlikely that we would class this as a shift.
Catford was the first to use the term shift, but the most comprehensive taxonomy
of translation shifts, based on their ‘translation procedures’, was set out by the
Canadians Jean-Paul Vinay and Jean Darbelnet in their A Comparative Stylistics
of French and English (1958/1995).While it is true that they approach the subject
from the point of view of comparative or contrastive stylistics, using parallel
non-translated as well as translated texts, they describe a detailed and systematic
model for the analysis and comparison of a ST–TT pair. The first step involves
identification and numbering of the ST units and the units of translation (see
Section A, Unit 3). This is followed by a matching of the two.

shfits

A shift is said to occur if, in a given TT, a translation equivalent other than
the formal correspondent occurs for a specific SL element. This is what has
occurred between the French and English texts in Example A4.1

The following example, from a leaflet distributed on board Eurostar trains explaining
the measures being taken to detect smoking, can illustrate these differences.
Example A4.2a English
Please note that smoke detectors will be fitted on-board.
Example A4.2b German
Beachten Sie bitte, daß die Züge mit Rauchdetektoren ausgestattet werden.
[Note you please, that the trains with smokedetectors fitted will-be.].

translation shifts6

Translation shifts
Unit 3 looked at the unit of translation, whether word, phrase or higher level.
The present unit will now discuss models or taxonomies that have been proposed
for examining the small changes or ‘shifts’ that occur between units in a ST–TT pair.
A connecting theme of the examples is rail travel, perhaps a symbolic counterpoint
to the best known taxonomy of translation shifts, devised by Vinay and Darbelnet
and initially inspired by the study of bilingual road signs in Canada.
TRANSLATION SHIFTS
On some international trains in Europe, there is, or used to be, a multilingual
warning notice displayed next to the windows:

Example A4.1
Ne pas se pencher au dehors
Nicht hinauslehnen
È pericoloso sporgersi
Do not lean out of the window
The warning is clear, even if the formis different in each language. The English, the
only one to actually mention the window, is a negative imperative,while the French
and German use a negative infinitive construction (‘not to lean outside’) and the
Italian is a statement (‘[it] is dangerous to lean out’). Of course, these kinds of
differences are typical of translation in general. It is not at all the most common for
the exact structure of the words to be repeated across languages and, even when the
grammatical structure is the same (as in the French and German examples above),
the number of word forms varies from six (ne pas se pencher au dehors) to two (nicht
hinauslehnen).
The small linguistic changes that occur between ST and TT are known as translation
shifts. John Catford was the first scholar to use the term in his A Linguistic Theory
of Translation (1965, see Section B Text B4.1).His definition of shifts is ‘departures
from formal correspondence in the process of going from the SL to the TL’ (Catford  1965:73). The distinction drawn between formal correspondence and textual
equivalence will be crucial and relates to Saussure’s distinction between langue and
parole:

THE UNIT OF TRANSLATION AS A PRELUDE TO ANALYSIS

THE UNIT OF TRANSLATION AS A PRELUDE TO ANALYSIS
Division of ST and TT into the units of translation is of particular importance in
Vinay and Darbelnet’s work as a prelude to analysis of changes in translation, the
translation shifts that will occupy us in Unit 4.As an illustration of how this division
works, and how it might illuminate the process of translation, look at Example
A3.4a, a poster located by the underground ticket office at Heathrow airport,
London:
Example A3.4a
Travelling from Heathrow?
There are easy to follow instructions on the larger self-service touch screen ticket
machines.

Task A3.3
➤ Before reading further, imagine you have been asked to translate this poster
into your first language (or main foreign language).Write down your translation
and make a note of the translation units you use when dividing up
the ST.

A translator approaching this short text will most probably break it down into the
title (Travelling from Heathrow?) and the instructions in the second sentence.While
that sentence will be taken as a whole, it might also in turn be sub-divided more or
less as follows:
There are/
[easy to follow/instructions]/
[on the/larger/self-service/touch screen/ticket machines]
Here, the slashes (/) indicate small word groups with a distinct semantic meaning
that might be considered separately, while the brackets ([. . .]) enclose larger units
that a practised translator is likely to translate as a whole.
The actual French TT on the poster indicates how this operates in real life:

THE TRANSLATION SHIFT APPROACH1

THE TRANSLATION SHIFT APPROACH
2
4
I 1 Note the similarity with recommendations by Nida and Newmark which were
discussed in chapter 3.
2 This forms the basis of the discourse analysis models discussed in chapter 6.
3 These functions originate in Riihler (1939165) and are later developed by
Halliday. See chapters 5 and 6 for a more detailed explanation.
4 See, for example, Fish (1981) or van Peer (1989).
5 In The Royal River Thames: Westminster to Greenwich Cruise and Sail and Rail
Guide (1997), London: Paton Walker, pp. 7 and 14.

system theories3

For Toury (1995: 13)' translations first and foremost occupy a position in
the social and literary systems of the target culture, and this position determines
the translation strategies that are employed. With this approach, he is
continuing and building on the polysystem work of Even-Zohar and on
earlier versions of his own work (Toury 1978, 1980, 1985, 1991). Toury
(1995: 36-9 and 102) proposes the following three-phase methodology for
systematic DTS, incorporating a description of the product and the wider
role of the sociocultural system:
1 Situate the text within the target culture system, looking at its significance
or acceptability.
2 Compare the ST and the TT for shifts, identifying relationships between
'coupled pairs' of ST and TT segments, and attempting generalizations
about the underlying concept of translation.
3 Draw implications for decision-making in future translating.

An important additional step is the possibility of repeating phases (1) and (2)
for other pairs of similar texts in order to widen the corpus and to build up a
descriptive profile of translations according to genre, period, author, etc. In
this way, the norms pertaining to each kind of translation can be identified
with the ultimate aim (as more descriptive studies are performed) of stating
j laws of behaviour for translation in general. The concepts of norms and laws
are further discussed in sections 7.2.1 and 7.2.2 below.
I The second step of Toury's methodology is one of the most controversial
areas. The decisions on which ST and TT segments to examine and what the
1 relationships are betureen them is an apparatus which Toury (1995: 85) states
I should be supplied by translation theory. Yet, as we have seen in chapters 4 l
and 5, linguistic translation theory is far from reaching a consensus as to
what that apparatus should be. Most controversially, in earlier papers (1978:
93, 1985: 32), Toury still holds to the use of a hypothetical intermediate
invariant or tertium comparationis (see page 49 for a discussion of this term) as
an 'Adequate Translation' (AT) against which to gauge translation shifts.
However, at the same time he also admits (1978: 88-9) that, in practice, no
translation is ever fully 'adequate'; for this contradiction, and for considerl
i ing the hypothetical invariant to be a universal given, he has been roundly
I ~ criticized (see, e.g., Gentzler 1993: 131-2, Hermans 1999: 56-7).