Lecture 5.

Translation shift types and translation equivalence types

Changes or ‘shifts’ often occur in translation. By ‘shifts’, we mean departures from

formal correspondence in the process of going from the source language to the target
language. These shifts are actually the changes that occur due to the translation of
text from one language to another, also known as class shift in translation.

But why do these class shift in translation take place? Well, there could be many
possible reasons for it. For instance, they appear due to differences in vocabulary,
grammar, or syntax between the original or source language and the target language.
Other possible reasons include differences in cultural nuances and in the intended
audience for which the text is being translated. Moreover, translators and linguists
may also need to make certain changes to the start text so that it can comply with the
target language convention or meet other requirements of the target audience.

“Equivalence” is one of the most traditional and critical concepts in translation

theory, being considered by some “constitutive for translation” (Koller 1997: 189)
and the “nucleus of all translation theory” (Albrecht 1987: 13).
Adopted by translation theory in the fifties as a key term in machine
translation, the notion of “Umkehrbarkeit” (equality) was its focus; later within
human translation theory the focus shifted to “Gleichwertigkeit” (equatability) (Reiß
1971: 12). From a communicative point of view, Nida (1964: 159f) in the sixties
distinguished between “formal equivalence” (referring to the relationship of ST and
TT form and content elements) and “dynamic equivalence” (orienting itself towards
the reader). Similar distinctions are made from a linguistic point of view by Catford
(1978: 27ff) who distinguishes between “textual equivalence” and “formal correspon-
dence” indicating two levels on which “equivalence” can be considered: a systematic,
virtual level (“formal correspondence”) and a parole-related, actual level (“textual
equivalence”). Jäger (1975: 107) differentiates between “communicative equivalence”
and “functional equivalence,” already including theme-rheme patterns as an equiva-
lence parameter. The most detailed concept of equivalence is presented by Koller
(1997: 214ff), specifying five “equivalence frameworks”: denotative, connotative,
textnormative, pragmatic and formal equivalence. In the eighties and nineties, the
equivalence concept was severely criticized for being too linguistically oriented
(Snell-Hornby 1986: 13ff and others, cf. most recently Gallagher 1998) and even
totally discarded as a valid concept for translation theory on the grounds that the
concept suggested a symmetry between languages that did not exist. Halliday’s register-
specific categories (House 1997) or which may vary by so-called “Invarianzforderungen”
(invariance postulates) to be
negotiated between translator and the author of the original (Albrecht 1990).
(https://www.erudit.org/en/journals/meta/2001-v46-n2-meta159/002886ar.pdf )

Shifts result from attempts to deal with systemic differences. Translation involves the
transfer of certain values of expression or content across a semiotic border; shifts are
concomitant with this transfer. The relation between any two systems confronted in the
process of trans-lation is asymmetric, and the way the transfer is carried out is not
determined a priori. It is because the translation operation performed on an initial
semiotic entity can lead to different resultant entities that ‘shift’ is not a category of
competence (see also semiotics). The description and explanation of shifts in trans-lation
as performance therefore ought to be concerned with the dynamics of culture rather
than the static contrastive description of languages and cultures which takes place within
the framework of various comparative disci-plines (Toury 1980a: 11–18).

Shifts and invariance

The transformation which is occasioned by the translation process can be specified in

terms of changes with respect to the source texts, changes which are termed ‘shifts’. The
two concepts of invariant and shift are therefore interdependent, such that any classifi-
cation or definition of shifts entails a definition of the invariant (Bakker and Naaijkens
1991: 204–5).

Definitions of the concept of invariant (i.e. those elements which remain unchanged in
the process of translation) necessarily serve a certain theoretical purpose, while presup-
posing a certain point of view. A very rough and schematic division can be made
between those conceptions of invariance in which the point of view lies ‘before’
translation (be it actual or ideal), and those in which it lies ‘after’ the fact. In accordance
with this division, two classes of definitions of the invariant can be distinguished. The
first consists of those defini-tions in which the invariant is postulated as a necessary
condition to be met before the transfer operation can qualify as translation; here, the
invariant coincides with the tertium compa-rationis of translation (cf. Steiner 1975: 319;
Lefevere and Bassnett 1990: 3). In definitions of the second type, the invariant is meant
for use as a descriptive, purely heuristic construct; here, the tertium comparationis is a
device in the methodology of the description.

Invariance defined ‘prior’ to translation

When a certain type of invariance is considered a requirement for appropriate translation

behaviour, the corresponding notion of shift is likely to be a normative or prescriptive
one. The directive statements in which this notion is found can be cast either in an
affirmative form as do, or in a negative form as don’t (van Leuven-Zwart 1990b). The
choice of either the positive or the negative formulation depends on the way the initial
differences between source and target codes or systems are taken into account. In both
types of statement the concept of shift is especially relevant to the applied branches of
translation studies: training and education and criticism (see reviewing and criticism).
In negative formulations, shifts are looked upon as unwelcome results of the translation
act, as something to be avoided: the implied performance instruction is don’t. The term,
then, refers to transformations of certain source text values or properties which ought to
remain, or have remained, unaltered; the result is described as an error or mistranslation.
Since shifts are thus seen as unnecessary devia-tions from the due course of the
translation process, the concept could be said to operate within a restricted theory of
translatability (cf. Toury 1980a: 26–8). While being derived from the source text, this
theory already allows for systemic differences between the source and target languages to
a certain, variable extent: the source text-based theory is modified to accommodate
target language possibilities and impossibilities, whether only linguistic, or textual and
cultural as well. Consequently, shifts are shifts with respect to a specific trans-lation ideal
and some postulated concept of equivalence. If, for instance, it is stipulated as an
invariance condition that the translation be (at least) the maximal reconstruction of the
conceptual semantic meaning of the source text, any deviation from this potential
reconstruction will be marked as a shift.

In positive formulations, on the other hand, shifts are seen as required, indispen-sable
changes at specific semiotic levels, with regard to specific aspects of the source text.
Their supposedly necessary, or desirable, occur-rence is a consequence of systemic
differences. Shifts are the means which allow the trans-lator to overcome such
differences. In other words, changes at a certain semiotic level with respect to a certain
aspect of the source text benefit the invariance at other levels and with respect to other
aspects. With this notion of shift, the focus is not on departures from a given normative
concept of translatability but on the systemic differences which, in the projected
translatability model, remain to be provided for. It is these systemic differences which
are rewritten in terms of performance instructions (do). The concept of shift, then, is
instrumental within a set of translation procedures. Examples of shifts postulated as do’s
include changes at the level of formal linguistic means which are brought about in favour
of functional or text-pragmatic equivalence (see functionalist approaches; pragmatics).
For instance, Nida’s notion of dynamic equivalence, where ‘the focus of attention is
directed, not so much toward the source message, as toward the receptor response’
(Nida 1964: 166), adopts a functional-pragmatic concept of invariance and presupposes
shifts away from static, or formal hierarchies of source text properties. Two of the
translation procedures discussed in Vinay and Darbelnet (1958) provide further
examples of positive performance instructions: transposition, where an SL word is
rendered by a TL word of a different word class, and modulation, ‘[a] translation
method consisting of changing a point of view, an evocation, and often a category of
thought’ (Vinay and Darbelnet 1995: 346). Chesterman (1997: 87–116) elaborates on the
categories of Vinay and Darbelnet in his classification of translation strategies (which
conceptually can be viewed as the methodo-logical mirror image of shifts) on syntactic,
semantic and pragmatic levels.
Invariance defined ‘after’ translation

As a descriptive category, shifts are defined and identified retrospectively. They are
reconstructed or established during the description of actual, existing translations. The
descriptive focus may be on the reconstruction of the translation process or on the
product, particularly with respect to its relation to the source. However, the distinction
between process-oriented and product-oriented description is not clear cut. Process-
related elements may play a role within the description of translation as a product, and
the study of the product is the principal means for describing translation as a
process.When the focus is on the process, typologies of shifts generally attempt to
account for the nature of translation operations and the considerations underlying
certain decisions taken during the course of translation. Because the translation process
is essentially a ‘black box’ (Holmes 1972: 72), any classification of shifts at this level has
to be based on translation competence, that is, on the possible relationships and differ-
ences between systems or codes. But, since the empirical testing of cognitive processes
involved in translation is problematical (see think-aloud protocols; psycholinguistic and
cognitive approaches), process-oriented typologies tend to reduce theoretical, general
translation compe-tence to a specific translation ideal. A distinction is often made
between obligatory and optional shifts (van den Broeck and Lefevere 1979; Toury 1980a;
van Leuven-Zwart 1989).

Obligatory shifts are dictated by differences between linguistic systems, for example a
lack of corre-spondence between related lexical items in the source and target languages
(Kade 1968: 79 ff.). Optional shifts are those opted for by the trans-lator for stylistic,
ideological or cultural reasons. This distinction is similar to one made by Popovič
between constitutive and individual shifts (see below), but according to Popovič
constitutive shifts are not exclusively linguistic. As far as the product-oriented view of
shifts is concerned, the following definition by Popovič (1970: 79) may serve as a
starting point: ‘All that appears as new with respect to the original, or fails to appear
where it might have been expected, may be interpreted as a shift’. In this definition,
three elements can be discerned: (a) a relationship between the source and target texts
(‘new with respect to the original’); (b) a relationship between the target text and its
reception in the target system (‘where it might have been expected’); and (c) a descriptive
point of view (‘may be interpreted’). The descriptive focus can be either on (a) or on (b).
For example, a zero-shift established at specific textual or linguistic levels in the
source/target text relationship (i.e. an instance of invar-iance, where nothing new
appears) may still be interpreted as a shift in terms of (b): by violating the expectations
of the target system, a target text may acquire a function other than that fulfilled by the
source text in the source system. This double point of view implies that there is always
the possibility of a description in which shifts are shown to occur in translation. For this
reason, shifts are sometimes called a categorial quality (van den Broeck 1984–5: 117)
of the class of translation. This quality can be causally linked to the double status of the
translation as a reconstruction of another text and a text functioning in its own right in
the target culture (see, for instance, Levý 1969: 72).

Definition and classification of shifts in product-oriented descriptions

Any typology of shifts presupposes a descriptive point of view (see descriptive vs.
committed approaches). This point of view can be made explicit in terms of criteria or
parameters for comparative analysis. One might suggest that the relationship between a
translation and its original may be defined as a ‘network of correspondences and shifts’
(Koster 2002: 34).For any given parameter, the degree of corre-spondence that will be
taken as invariance has to be established.
Catford (1965) discusses shifts within the framework of a linguistic theory of
translation (see linguistic approaches). Within this framework, shifts occur on the
grammatical and lexical levels, and their investigation is therefore pursued within the
boundaries of the sentence as an upper rank. Catford distinguishes between a textual
equivalent, ‘any TL text or portion of text which is observed on a particular occasion . .
. to be the equivalent of a given SL text or portion of text’, and a formal
correspondent, ‘any TL category (unit, class, structure, element of structure, etc.) which
can be said to occupy, as nearly as possible, the “same” place in the “economy” of the
TL as the given SL category occupies in the SL’ (Catford 1965: 27). He limits his theory
of shifts to instances of translation which satisfy the condition that the relationship
between source and target utterances can be identified by a competent bilingual as
textual equivalence. The invariant of comparison Catford employs is formal
correspondence. Shifts, in his definition, are ‘departures from formal correspondence in
the process of going from the SL to the TL’ (ibid.: 73). If, from a descriptive point of
view, a given TL instance is observed to be a textual equivalent of a given SL form, this
does not entail that formal correspondence exists between the units under comparison,
since the TL categories cannot necessarily ‘be said to occupy, as nearly as possible the
“same” place in the economy of the TL as the given SL category occupies in the SL’
(ibid.: 32). The type and degree of divergence between formal correspondence and
translation equivalence can be detailed in terms of shifts. Catford distinguishes two
major types, level shifts (where an SL item at one linguistic level, for example grammar,
has a TL equivalent at a different level, for instance lexis) and category shifts, which
involve (a) changes of structure (structure shifts, for example a subject–predicate–
object structure may be trans-lated as a predicate–subject–object structure), (b) changes
of rank (unit shifts, for example a word may be translated by a morpheme or a group by
a clause), (c) changes of class (class shifts, for example an adjective may be trans-lated by
a noun or a verb), or (d) changes of term (intra-system shifts, shifts which occur inter-
nally, within a system, when source and target language systems have the same formal
consti-tution but translation involves the selection of a non-corresponding term in the
TL system) (Catford 1965: 73 ff.).
 In Popovič (1970), the main concern is with literary translation, and shifts are
therefore defined as a stylistic category and termed ‘shifts of expression’. For Popovič, ‘a
systematic evalu-ation of the shifts of expression that occur in a translation’, and hence
‘the objective classi-fication of differences between the translation and its original’ (ibid.:
84), should be based on a theory of expression, such as can be found in Miko (1970).
The linguistic means employed in the source and target texts cannot be compared in
isolation, but only ‘in relation to the entire system of expression’ (Popovič 1970: 84). It is
this system of expression which allows us to determine the expressive values of the
respective linguistic devices, a necessary precondition for the establishment of shifts ‘in
the sphere of style’ (ibid.: 83). Style, for Popovič, is a multi-layered and hierarchically
organized concept. It is because it covers abstract and general categories and qualities, as
well as more specific stylistic means, that it can be used as an invariant for the
comparison of source and target texts.
For the evaluation of shifts, it is necessary to examine the respective
differentiation of stylistic qualitities in the source and target languages and texts. Popovič
distinguishes between constitutive shifts and individual shifts. Constitutive shifts are
system-bound, but the concept is wider than that of obligatory shifts. Popovič defines a
constitutive shift as ‘[a]n inevitable shift that takes place in the translation as a
consequence of differences between the two languages, the two poetics and the two
styles of original and trans-lation’ (1976: 16). One might also conceive of these shifts as
constitutive in the sense that they are constitutive for the style of the translation (cf. the
notion of shifts as a categorial quality of translation above). According to Popovič, the
style of the translation, conceived of as the ‘integrative principle’ in the development of
its structure (1970: 79), is necessarily determined by shifts because of its ‘dual character’
(ibid.: 82): it has to comply both with the norms of the original and with a given target
‘translation ideal’. Individual shifts differ from constitutive shifts in that they are
prompted by the stylistic propensities and the subjective idiolect of the individual
translator. When changes at the level of macro-stylistics cause the translation to fit a
literary genre different from that of the original (see adaptation), Popovič speaks of a
generic shift (see also van den Broeck 1986).
Within the methodology of Toury (1980a: 89–121, 1985: 32), the invariant of the
comparison is the adequate translation (AT) and the unit of comparison is the texteme.
An adequate translation is a reconstruction of source text textemes and consists of an
explicitation of the textual relations and functions of the source text. As such, it is not an
actual text but a hypothetical construct, serving only methodo-logical purposes (see
Hermans 1995: 218–20 for a critical assessment of this concept). The degree of
correspondence taken as invariance within this method is adequacy at the textemic level,
and shifts are defined as deviations from adequacy. The purpose of comparison is to
determine the distance between the ‘actual equivalence’ found between source and target
texts and the maximal norm of adequate translation, inasmuch as this distance can be
attributed to norm-governed translational behaviour.
 Since obligatory shifts are rule-governed, they cannot be taken to reflect
translational norms and are therefore not taken into consideration; methodologically,
they are accounted for in the invariant itself (the weak version of adequacy, see Toury
1980a: 69). The comparative procedure starts by assuming equiv-alence at the textual-
functional level, hence the parameter of comparison is textual-functional. When the
dominant relationship between target text texteme and source text unit is found at that
level, the translational relationship is one of adequacy. When there is no textual-
functional correspondence, the procedure is to look for correspondence at lower textual
and linguistic levels. Norms determine the position of the actual translation equivalence
between adequacy and acceptability, and the establishment of individual shifts ultimately
leads to the establishment of the translational norms governing the text in question.
When, after further generalization and expansion of the investigated corpus, shifts show
a certain pattern or statistical regularities, they can be explained by the existence of a
historically and culturally determined poetics of translation or translation ideal. In later
stages of Toury’s thinking (1985, 1990), the above procedure became part of a larger one
in which an additional unit of comparison was introduced: the ‘coupled pair of “problem
solution” ’ (see unit of trans-lation). The notion of shift gradually became less central in
his method of description (see, especially, Toury 1995).
Within the methodology of van Leuven-Zwart (1984, 1989, 1990a), a distinction is
made between shifts at the level of a text’s microstructure (comparative model) and the
effects of these shifts at the macrostructural level (descriptive model). At the
microstructural level, the invariant of the comparison is the architranseme (ATR),
which expresses the common denominator(s) in the relation between specific textual
units of the source and target texts; these textual units are called transemes. In as much
as the descriptive model is comparative, it works with an invariant at the macrostructural
level as well. The invariant in this case is based on a theory of the genre to which the
texts under comparison belong. Van Leuven-Zwart limits her methodology to the
domain of narrative texts, and the invariant is therefore derived from specific
narratological concepts such as ‘story level’ and ‘focalization’. The ATR has to be estab-
lished separately for each pair of transemes, the invariant at the macrostructural level will
be established a priori.
Esssential to the method of van Leuven-Zwart is the priority given to the concept
of relation. Any comparative description involves establishing the relation between
elements as well as attributing certain features to those elements. According to Van
Leuven-Zwart, a comparison based on the prior attribution of features is only a ‘second
degree comparison’, since it departs from a descriptive operation, and the relationship
between the elements is established afterwards. In a direct comparison, the order is
reversed. According to this scheme, a texteme, for instance, would be a unit of
description rather than a unit of comparison. At the microstructural level, a relation of
complete conjunction between the transemes and archi-transemes (in which case there is
a relation of synonymy between transemes) is assumed as a starting point, and shifts
occur when there are aspects of disjunction between transemes and the ATR. Van
Leuven-Zwart distinguishes three main categories: modulation (where a source or target
transeme shows one or more aspects of disjunction with the ATR; a relation of
hyponymy between transemes), modification (where a source and target transeme show
one or more aspects of disjunction with the ATR; a relation of contrast between
transemes), and mutation (where there are no aspects of conjunction, and therefore no
ATR can be established; no relation between transemes). The purpose of this method is
to arrive at hypotheses about the interpretation and the strategy underlying the
translation involved in the comparison. As a consequence, shifts that do not reflect a
translator’s interpretation or strategy are not taken into account: only optional shifts and
substantial shifts are considered. As far as the distinction between obligatory and
optional shifts is concerned, van Leuven-Zwart expresses a reservation as to its
applicability. In the first instance, the decision whether shifts are to be considered
optional or obligatory is suspended. Not until the effects of the microstructural shifts on
the macrostructural level have been established will it be possible to determine to what
extent the shifts are due to other than purely linguistic factors. Initially, all substantial
shifts are noted; that is, all the shifts that have some bearing on one of the substantial
levels, namely the semantic, stylistic or pragmatic level. As to syntactic shifts, only those
that affect these substantial levels are taken into account. Purely formal shifts are
disregarded. On the distinction between formal and substantial shifts, see also van den
Broeck and Lefevere (1979). In more relativistic terms, Koster (2000: 131–2, 153) relates
substantiality to the purpose of a descriptive procedure. Substantial shifts, from that
perspective, are those shifts which are relevant to what the procedure sets out to
describe (norms, strategies or the transla-tional interpretation).

See also: equivalence; explicitation; linguistic approaches; norms; translatability; unit of translation.

Further reading

Catford 1965; Popovič 1970, 1976; Holmes 1972, 1978; Toury 1980a; van den Broeck
1986; Van Leuven-Zwart 1989, 1990a; Koster 2000.

Written by

MATTHIJS BAKKER, CEES KOSTER AND KITTY VAN LEUVEN-ZWART

Source
https://www.academia.edu/40889257/ROUTLEDGE_ENCYCLOPEDIA_OF_TRA
NSLATION_STUDIES_2nd_Edition_ed_by_Mona_Baker_and_Gabriela_Saldanah_?e
mail_work_card=view-paper