Journal Across Languages and Cultures, A Multidisciplinary Journal

Across Languages and Cultures 11 (1), pp.
1–28 (2010)
DOI: 10.1556/Acr.11.2010.1.1
ABANDONING THE NOTION OF

“TRANSLATION-INHERENT” EXPLICITATION:
AGAINST A DOGMA OF TRANSLATION STUDIES
VIKTOR BECHER
University of Hamburg, Research Center on Multilingualism
Max-Brauer-Allee 60, 22765 Hamburg
Phone: +49 (0)40/428386427; Fax: +49 (0)40/428386116
E-mail: viktor.becher@uni-hamburg.de
Abstract: One of the best-known hypotheses of translation studies, the Explicitation

Hypothesis, postulates that explicitation is “inherent” in the process of translation and may
therefore be regarded as a “universal of translation”. In recent years, a number of corpus-
based studies on explicitation have been produced, most of which purport to offer evidence
in favor of this hypothesis. As a consequence, the alleged universality of explicitation has
achieved the status of dogma in translation studies. The aim of the present article is to show
that the dogma of translation-inherent explicitation rests on fallacious theoretical considera-
tions and premature interpretations of empirical data. In the first place, it will be argued that
the Explicitation Hypothesis strictly speaking does not even qualify as a scientific hypothe-
sis, since it is unmotivated, unparsimonious and vaguely formulated. In the second place, it
will be shown that previous studies on explicitation fail to provide conclusive evidence for
the translation-inherent nature of explicitation due to a number of methodological shortcom-
ings.
Keywords: explicitation, implicitation, translation universals, translation theory
1. INTRODUCTION
The assumption that translations are “inherently” more explicit than (1) their
corresponding source texts and (2) comparable, non-translated texts written in
the target language has become a dogma in translation studies. This is largely
due to the tremendous influence that Blum-Kulka’s (1986) Explicitation Hy-
pothesis has exerted on the development of this young discipline. The Explicita-
tion Hypothesis claims that “explicitation is a universal strategy inherent in the
process of language mediation” (Blum-Kulka 1986:21).
1585-1923/$ 20.00 © 2010 Akadémiai Kiadó, Budapest

2 VIKTOR BECHER
The present article is a plea to abandon the dogma of translation-inherent

explicitation. It will be shown that previous research on the Explicitation Hy-
pothesis has suffered from severe theoretical and methodological problems that
we need to avoid in the future if we want to find out what role explicitation
really plays in translation. First, I am going to argue that the assumption of the
translation-inherence of explicitation is unmotivated and should thus be re-
placed by a better and more useful hypothesis. Second, I will discuss two fa-
mous empirical studies that are often cited as proof of the existence of transla-
tion-inherent explicitation (Olohan and Baker 2000, Øverås 1998), showing that
they in fact fail to provide conclusive evidence in favor of the Explicitation Hy-
pothesis due to a number of methodological shortcomings (additional studies
are discussed in Becher forthcoming a).
This article is structured as follows. Section 2 will begin by introducing
some important definitions. In Section 3, Klaudy’s (2008) classification of ex-
plicitation phenomena into four different types will be briefly presented. Section
4 will then outline some grave theoretical problems in connection with the con-
cept of translation-inherent explicitation, starting with a short exegesis of the
famous passage from Blum-Kulka (1986) where the Explicitation Hypothesis
was first formulated. In Section 5, the two arguably most important empirical
studies on explicitation and their methodological problems will be discussed,
leading us to the conclusion that they (as well as all other relevant studies I
know of) fail to provide conclusive evidence for the translation-inherent status
of this phenomenon. Section 6 will then conclude that due to the severe theo-
retical and methodological problems outlined, the Explicitation Hypothesis
should be abandoned altogether and replaced by Klaudy’s (2009) Asymmetry
Hypothesis.
2. DEFINITIONS
Many studies on explicitation suffer from the problem that the term “explicita-
tion” is not properly defined, most commonly because terms such as “explicit-
ness” and/or “implicitness” appear in the definiens which are themselves in
need of a definition. In this section, I would like to propose proper definitions of
all three terms. Let us start with implicitness:
Implicitness is the non-verbalization of information that the addressee might

be able to infer.
The reader is asked to note the following important points about this defini-
tion. First, there is the epistemic modal might: it does not matter whether the
addressee is actually able to infer the non-verbalized information or whether the
Across Languages and Cultures 11 (1) (2010)

AGAINST A DOGMA OF TRANSLATION STUDIES 3
inference fails – which obviously happens occasionally in conversation. For the

definition above, it is sufficient when the piece of information in question is in-
ferable in a “theoretically motivated sense” (cf. Steiner’s 2005:11 definition of
explicitation).
Second, there is the (deliberately) vague term “information”. This may be
syntactic, semantic/pragmatic, or even phonological information. Any aspect of
a linguistic message may be left implicit by the sender.
Third, the definition deliberately avoids spelling out from where the ad-
dressee might infer the non-verbalized information. From the previous dis-
course? From the extralinguistic context? From her world knowledge? It is of
course legitimate and highly relevant to ask for the inferential sources that are
available to the addressee. But we do not want this question to make our defini-
tion unnecessarily complicated.
We can now define explicitness, which is nothing more than the absence of
implicitness, as follows:
Explicitness is the verbalization of information that the addressee might be

able to infer if it were not verbalized.
To put it somewhat informally, explicitness means saying something that the

addressee might have understood anyway. From this definition, it also becomes
clear that explicitness often (but not necessarily) entails redundancy, i.e. the en-
coding of information by means of more linguistic material than is necessary.
Now that we have defined explicitness, it is easy to provide a proper defini-
tion of explicitation:
Explicitation is observed where a given target text is more explicit than the
corresponding source text.
It should be pointed out that the definitions of implicitness, explicitness

and explicitation provided in this section are purely product-based; they delib-
erately ignore processing considerations. Thus, the definitions of implicitness
and explicitness discussed above are agnostic as to whether relatively explicit
texts are easier to process than comparatively implicit texts. And the given defi-
nition of explicitation does not say how the target text’s higher degree of explic-
itness is related to the translation process. This latter point is particularly impor-
tant. When we investigate translation corpora, we do not see an increase in ex-
plicitness in the process of translation, strictly speaking. We merely observe the
product, i.e. a higher degree of explicitness in the target text (TT) as compared

4 VIKTOR BECHER
to the source text (ST), so we need to be careful with our conclusions. This is
why the definition of explicitation provided does not say anything about the
translation process.
3. TYPES OF EXPLICITNESS
In an often cited encyclopedia article, Klaudy (2008) distinguishes between the
following four kinds of explicitation in translation:1
1. Obligatory explicitation. Caused by lexicogrammatical differences be-
tween the source language (SL) and the target language (TL).
■ e.g. En. to be → Sp. ser/estar
2. Optional explicitation. Motivated by differences in stylistic prefer-
ences between SL and TL.
■ e.g. En. group → Ger. Unternehmen (‘company’)
3. Pragmatic explicitation. Motivated by differences in cultural and/or
world knowledge shared by members of SL and TL communities.
■ e.g. Ger. die Alster → En. the lake Alster
4. Translation-inherent explicitation. Caused by “the nature of the trans-
lation process itself” (Klaudy 2008:107).
■ e.g. ???
Two important points deserve to be mentioned in connection with Klaudy’s

classification.2 First, Klaudy’s Types 1 to 3 are (obligatorily) caused or (non-
obligatorily) motivated by certain differences between SL and TL (in the case
of pragmatic explicitation: between the source and target language communi-
ties). This means that these types are predicted to exist; they simply have to ex-
ist. When we translate from English to Spanish, for example, we know in ad-
vance that at some point explicitation will become necessary because Spanish
has two copulas with different meanings (cf. e.g. Maienborn 2005), while Eng-
lish has only one. In general, we know from linguistics that lexicogrammatical,
stylistic and cultural differences exist for every conceivable language pair.
These will inevitably cause or motivate instances of explicitation in translation,
and we would be very surprised if this prediction were not supported by empiri-
cal data. Type 4, on the other hand, is a very different beast: the translation-
inherent type of explicitation is not predicted, but rather postulated to exist,
namely by Blum-Kulka’s (1986) Explicitation Hypothesis (see next section). It
would therefore not be surprising if we were not to find evidence of this type of
explicitation in a given corpus.
Second, while the list above provides typical examples of obligatory, op-
tional and pragmatic explicitations, it fails to give an example of a translation-
inherent explicitation. This is because it is not clear to me at all what an in-

stance of this type of explicitation is supposed to look like. In this connection it

is interesting to note that Klaudy (2008) herself provides many examples for her
Types 1 to 3 (the examples given above are my own), but does not provide a
single example of a translation-inherent explicitation. She probably had the
same problem as the author of the present article.
4. THEORETICAL ISSUES
Klaudy’s fourth explicitation type seems to be shrouded in mystery. In order to
find out more about this postulated type of explicitation, we travel back in time
to witness the birth of Blum-Kulka’s Explicitation Hypothesis.
4.1. The Birth of the Explicitation Hypothesis
Blum-Kulka proposed her famous Explicitation Hypothesis in a seminal and

highly insightful paper from 1986. It is instructive to quote the relevant passage
in full:
The process of translation [...] necessitates a complex text and discourse
processing. The process of interpretation performed by the translator on the
source text might lead to a TL text which is more redundant than the SL
text. This redundancy can be expressed by a rise in the level of cohesive
explicitness in the TL text. This argument may be stated as ‘the
explicitation hypothesis’, which postulates an observed [increase in, VB]
cohesive explicitness from SL to TL texts regardless of the increase
traceable to differences between the two linguistic and textual systems
involved. (1986:19; my emphases)
Blum-Kulka tells us that the process of interpretation that invariably occurs in

translation “might” lead to a TT which is more redundant, or explicit, than the
corresponding ST. However, we do not learn how the interpretation process is
supposed to produce this effect. The author then goes on to call her claims an
“argument”, which is somewhat surprising: where is the “argument”? If there is
an argument at all, neither its premises nor its conclusion are discernible. While
we might be inclined to agree that a given TT might come out as more explicit
than its ST, Blum-Kulka provides no reason for putting forward her Explicita-
tion Hypothesis, which claims that this has to be the case.
The postulated increase in cohesive explicitness is supposed to occur “re-
gardless of the increase traceable to differences between the two linguistic and
textual systems involved.” This additional stipulation makes clear that Blum-
Kulka does acknowledge the existence of obligatory explicitations (due to dif-

6 VIKTOR BECHER
ferences between “linguistic” systems) and optional explicitations3 (due to dif-

ferences between “textual” systems); however, the Explicitation Hypothesis
postulates an additional, translation-inherent type of explicitation which is sup-
posed to be caused by the “process of interpretation performed by the translator
on the source text”.
Note that this famous passage from Blum-Kulka (1986) definitely assumes
the existence of a separate type of explicitation qualitatively different from the
other, language-pair specific types. The passage cannot be taken in a different
way. If there are explicitations that are caused or motivated by differences be-
tween SL and TL and if there are explicitations that occur regardless of these
differences, then there has to be an additional type of “translation-inherent” ex-
plicitation – Klaudy’s (2008) mysterious Type 4.
4.2. Problems with the Explicitation Hypothesis
There are three serious problems to be noted in connection with Blum-Kulka’s

Explicitation Hypothesis:
Problem 1. Why do I criticize that it is not clear where the “argument” is

behind the Explicitation Hypothesis? Does a hypothesis have to rely on an ar-
gument? Yes, it has to. More precisely, a scientific hypothesis has to be moti-
vated, i.e. its postulation has to be justified, and Blum-Kulka’s Explicitation
Hypothesis is not motivated. This is of crucial importance, as non-motivated
hypotheses entail the danger of producing what I like to call pseudo-significant
findings, i.e. statistically significant but otherwise meaningless results. In his
online statistics textbook, Dallal (2007) provides interesting examples of
pseudo-significant findings:
[I]n the early part of the twentieth century, it was noticed that, when
viewed over time, the number of crimes increased with membership in the
Church of England. This had nothing to do with criminals finding religion.
Rather, both crimes and [...] Church membership increased as the popula-
tion increased. Association does not imply causation! Should opposition
increase or decrease accuracy? During WWII it was noticed that bombers
were more accurate when there was more opposition from enemy fighters.
The reason was that fighter opposition was less when the weather was
cloudy. The fighters couldn’t see the bombers, but the bombers couldn’t
see their targets! Association does not imply causation, at least not neces-
sarily in the way it appears on the surface! (Dallal 2007; emphasis re-
moved)

Clearly, we would not want to accept the hypothesis that, for example, the pres-
ence of many enemy fighters leads to bomber pilots aiming better, even if the
correlation mentioned by Dallal is significant. Statistical association in this con-
text clearly does not imply real-world causation. The reason is that this hy-
pothesis is not motivated; the theoretical assumptions underlying it (if there are
any) just does not make much sense.
On the other hand, we do want to accept the hypothesis that, for example,
smoking causes lung cancer when we find a significant correlation, e.g. in a
large-scale epidemiological study. The reason is that this hypothesis is well-
motivated; we are justified in postulating it, as small-scale laboratory studies
make the assumption of a causal link seem plausible. In other words, in this
case the potential connection between association and causation is backed by
evidence obtained on independent grounds.
It will have become clear that the Explicitation Hypothesis is much like the
fighter–bomber hypothesis mentioned above. It is not motivated on independent
grounds and therefore does not qualify as a scientific hypothesis. Thus, even if
we did find that translations are significantly more explicit than their source
texts and comparable TL texts, we still could not accept the Explicitation Hy-
pothesis. The higher degree of explicitation observed could have causes other
than “[t]he process of interpretation performed by the translator on the source
text” (cf. the weather conditions in the bomber example). An example of an al-
ternative cause would be a (hypothesized) universal tendency of translators to
simplify (Baker 1993:244, 1996:181ff), which potentially “raises the level of
explicitness by resolving ambiguity” (1996:182) and thus may also result in tar-
get texts that are more explicit across the board. It is a general problem of cor-
pus-based translation studies that “the same surface expression may point to dif-
ferent features or tendencies” (1996:180). If we want to accept Blum-Kulka’s
Explicitation Hypothesis, we need to demonstrate that it is better motivated than
the ‘Explicitation through Simplification Hypothesis’ just sketched. But the big
problem is that it is not motivated at all.4
Problem 2. Blum-Kulka’s Explicitation Hypothesis clashes head-on with

Occam’s Razor: entia non sunt multiplicanda praeter necessitatem (‘entities
must not be multiplied beyond necessity’). This principle, which has become
one of the cornerstones of scientific research, calls for hypotheses to be parsi-
monious in their assumptions and thus not only easier to handle, but also more
likely to be true. Blum-Kulka assumes the existence of a new entity, namely a
new, translation-inherent type of explicitation. This means that any other hy-
pothesis that might explain an observed tendency of explicitation in translation
without assuming a new type of explicitation would be more compatible with
Occam’s Razor and thus preferable to the Explicitation Hypothesis.

8 VIKTOR BECHER
For example, the Explicitation through Simplification Hypothesis sketched

above could explain a universal tendency of explicitation without postulating a
new type of explicitation. We would assume that the (hypothesized) urge of
translators to simplify leads them to resort to optional and pragmatic explicita-
tions more often than appropriate and/or necessary, et voilà, we have explained
the data without assuming any mysterious, translation-inherent kind of explici-
tation. In this way, this hypothesis is more compatible with Occam’s Razor and
should thus be preferred over Blum-Kulka’s Explicitation Hypothesis. (How-
ever, the ‘Explicitation through Simplification Hypothesis’ is similar to the Ex-
plicitation Hypothesis in that it is not motivated; cf. Problem 1).5
Problem 3. In a later passage of her paper, Blum-Kulka paraphrases the

Explicitation Hypothesis as postulating that “explicitation is a universal strategy
inherent in the process of language mediation” (1986:21), where our attention is
particularly called to the vague term “strategy”. It is not clear whether Blum-
Kulka means a conscious or a subconscious strategy. Olohan and Baker (2000)
seem to interpret her as meaning the latter, while Øverås (1998) seems to as-
sume that she means the former. In other words, although both Olohan and
Baker as well as Øverås invoke Blum-Kulka’s considerations as the basis for
their studies (which will be discussed in the following section), it is not even
clear whether they are investigating the same thing. The vagueness with which
Blum-Kulka has formulated her hypothesis has led to much confusion in the lit-
erature on explicitation right from the outset.
The three problems pointed out above are not merely minor shortcomings,
but rather fundamental issues that seriously question the usefulness of Blum-
Kulka’s Explicitation Hypothesis. In summary, the hypothesis is unmotivated,
unparsimonious and vaguely formulated. Since there is a much better hypothe-
sis that can be motivated on independent grounds and is compatible with Oc-
cam’s Razor (namely Klaudy’s Asymmetry Hypothesis; see Section 6), the up-
shot from the above discussion is that Blum-Kulka’s Explicitation Hypothesis
should be abandoned entirely and no longer investigated, at least not in its pre-
sent form.
5. METHODOLOGICAL ISSUES
This section discusses the two arguably most often cited empirical studies on
explicitation, carried out by Olohan and Baker (Section 5.1) and Øverås (Sec-
tion 5.2), both of which fail to provide conclusive evidence in favor of Blum-
Kulka’s Explicitation Hypothesis. I have not chosen these two particular studies
as the focus of my criticism because they are particularly good or bad, but rather
because they (and their shortcomings) are representative of previous empirical
research on explicitation as a whole (see Note 10 for some references), in part

because they have served as models for other studies. It should be stressed that
it is not the aim of this section to malign the importance or quality of the studies
to be discussed. Both studies are highly interesting and offer intriguing results,
which, however, are difficult to interpret due to the problems that will be
pointed out in the following. Due to lack of space, I am forced to focus on the
shortcomings of these studies and cannot dwell on their merits, such as the ex-
emplary detail with which Olohan and Baker (2000) discuss differences in the
occurrence of the complementizer that with different verb forms.
5.1. Subconscious Processes of Syntactic Explicitation?

The Case of Reporting that
Olohan and Baker (2000) investigated the optional use of the complementizer
that in combination with the reporting verbs say and tell in translated vs. non-
translated English texts (“reporting that”). It is laudable that the authors must be
given credit because they “have tried to be as explicit as possible concerning
[their] methodology in order precisely to allow future studies to confirm or
challenge [their] results” (2000:158). In the following, I hope to do just that,
namely challenge Olohan and Baker’s results. I am going to argue that their
study design is problematic in several respects and that their findings can also
be explained as the (combined?) effect of other alleged universals of translation.
Olohan and Baker carried out their research using the Translational English
Corpus (TEC) and a comparable sample from the British National Corpus
(BNC). The TEC consists of English target texts from four different genres
translated from “a range of source languages” (2000:151), and the BNC sample,
containing non-translated English texts, was chosen so as to mirror the makeup
of the TEC. Both corpora contain approximately 3.5 million words. I have two
main points of criticism regarding the TEC as employed by Olohan and Baker
(2000):
1. Olohan and Baker conducted their research using a preliminary, work-

in-progress version of the TEC. As they themselves point out, this
version of the TEC was very imbalanced, with each of the four genres
being represented very differently. Most notably (and problematically),
82% of the corpus consisted of fiction texts, while newspaper texts
represented only 1% of the corpus material (2000:152). This means that
the corpus used by Olohan and Baker could essentially be described as
a single-genre corpus and not as a representative sample of translated
English.

10 VIKTOR BECHER
2. The authors fail to disclose the source languages of the texts contained
in their preliminary version of the TEC. The current version of the TEC
contains translations from 24 different source languages: Arabic,
Brazilian Portuguese, Chinese, Czech, Danish, Dutch, Finnish, French,
German, Hebrew, Hopi, Hungarian, Italian, Modern Greek, Norwegian,
Polish, Portuguese, Russian, Serbo-Croatian, Spanish, Swedish, Tamil,
Thai and Welsh.6 It is not clear which source languages were
represented in the version of the TEC used by Olohan and Baker
(2000).
While the first problem is not grave, as single-genre analyses can yield interest-
ing results concerning explicitation, the second problem will have to be ad-
dressed in more detail later on.
Let us briefly review how Olohan and Baker conducted their investigation.
In a first step, the authors searched the TEC and the BNC sample for occur-
rences of the reporting verbs say and tell. In a second step, occurrences where
these verbs do not occur with a clausal complement (as e.g. in to tell a lie or to
tell someone to go away) were excluded so that only cases remained where that
could potentially be used as an optional complementizer (as e.g. in She told me
[that] she’s happy). In a third step, the two corpora were compared with respect
to the frequency with which say and tell were used with or without the optional
complementizer. The rationale behind this approach was that:
a higher incidence of the optional that in translated English would provide

evidence of inherent, subliminal processes of explicitation in translation.
Translators clearly do not adopt a conscious strategy of spelling out
optional syntactic elements such as that in reporting clauses more often
than writers producing original texts in the same language. (Olohan and
Baker 2000:143)
In other words, Olohan and Baker set out to test a version of Blum-Kulka’s Ex-
plicitation Hypothesis. Although it is probably overly optimistic to say that
translators “clearly” do not have any conscious control over their use of the
complementizer that after reporting verbs, Olohan and Baker nonetheless found
a very interesting object of investigation in reporting that, since translators ar-
guably put less thought into using or not using this purely syntactic – i.e. se-
mantically empty – element than is the case for semantically laden explicitating
shifts.
The main results of Olohan and Baker’s study can be summarized as fol-
lows:

z In the TEC, occurrences of say with and without reporting that are
essentially equally frequent (50.2% vs. 49.8%, respectively). In the
BNC sample, on the other hand, the picture is very different. Here, only
23.7% of all occurrences of say occur with the complementizer that,
while 76.3% occur without.
z In the case of tell, similar differences can be observed between the TEC
and the BNC sample. In the TEC, the optional complementizer was
used in 62.7% of all cases and omitted in 37.3%. In the BNC sample,
the figures showed nearly the reverse: 41.5% for that vs. 58.5% for its
omission.
These results indicate that reporting that is indeed used more frequently in
translated than in non-translated English. But unlike Olohan and Baker suggest,
their findings do not represent evidence for the Explicitation Hypothesis; there
are alternative explanations that do without the dubious assumption of “sub-
liminal processes of explicitation in translation” and are thus more plausible.
Two explanations that come to mind are:
z Source language interference.7 As has been mentioned above, the

texts contained in the TEC were translated from a large variety of
different source languages. Now the question is: how many of these
languages allow for the use of a complementizer with reporting verbs,
and how many require it? (For example, Spanish and Portuguese favor
the use of a complementizer, but also allow complementizer-free
infinitive constructions [Vanderschueren, personal communication],
while French and Italian have an obligatory complementizer after
reporting verbs.) Without a full answer to this question, Olohan and
Baker’s results cannot be interpreted properly, because the greater the
number of source languages represented in the TEC stipulating a
complementizer after reporting verbs, the more likely it becomes that
the higher occurrence of reporting that in this corpus is the result of
source language interference (cf. Saldanha 2008).8
z Conservatism. It has been suggested that translators tend to use more

conservative language than authors of non-translated texts (Baker
1993:244, 1996:183ff). The related effort to employ more formal means
of expression should make translators choose reporting that more often
than authors of originally English texts, since that is typically omitted
when writing “in an informal style [...]. After more formal and less
common verbs, that cannot be left out” (Swan 1980, cited in Olohan
and Baker 2000:144).

12 VIKTOR BECHER
Both of these approaches, which draw on proposed translation universals other

than explicitation, are more attractive than the Explicitation Hypothesis, since
they (1) offer real, more plausible explanations for the observations made and
(2) do not presuppose a subconscious tendency to explicitate on the part of
translators.
5.2. Explicitation in English–Norwegian and Norwegian–English

Literary Translations
Øverås (1998) reports the results of a study whose aim was to test Blum-
Kulka’s Explicitation Hypothesis through the investigation of English–
Norwegian and Norwegian–English literary translations. Her corpus consisted
of 1000 sentences taken from 40 novel fragments for each translation direction.
She manually identified and counted all explicitations and implicitations occur-
ring in these sentences, with the exception of obligatory shifts, i.e. shifts due to
lexicogrammatical differences between English and Norwegian. As with Olo-
han and Baker (2000), Øverås deserves praise for presenting her methodology
with exemplary transparency, making it possible to properly evaluate – and
criticize – her results. Again, due to lack of space I will have to focus on the
problematic points of the study and neglect its merits such as the elaborate and
highly useful typology of explicitating shifts that Øverås proposes.
The first problem with Øverås’ study is the improper and inconsistently
applied definition of explicitation on which it is based. Øverås defines explicita-
tion in passing as “the kind of translation process where implicit, co-textually
recoverable ST material is rendered explicit in TT” (1998:4). Although this
definition is quite vague, it seems to be compatible with the definition proposed
here (in Section 2). However, as we will see in the following, Øverås does not
adhere to this definition, counting cases as explicitations in which information
is verbalized that is definitely not “co-textually recoverable”.
The second issue is of a theoretical nature and is related to the general
weaknesses of the Explicitation Hypothesis that we noted in Section 4. As we
have seen, Blum-Kulka (1986) leaves us wondering about the exact nature of
translation-inherent explicitation. Is it a conscious or a subconscious phenome-
non? What are its causes? Øverås offers a simple answer to these questions:
translation-inherent explicitation is the result of an operational norm9 in the
sense of Toury, i.e. a norm on the lowest and most concrete level that directly
governs “the decisions made during the act of translation” (1995:58ff). It re-
mains completely mysterious to me how Øverås has come to conceive of trans-
lation-inherent explicitation as the effect of a translational norm. Toury
(1995:61f) goes to great lengths to emphasize “two features inherent in the very
notion of norm [...]: the socio-cultural specificity of norms and their basic insta-

bility.” If translation-inherent explicitation were indeed the result of a transla-

tional norm, it should be highly language-pair specific (contrary to the claims of
the Explicitation Hypothesis). Since norms are (1) culture-specific and (2) un-
stable, they epitomize the opposite of universality. A conception of translation-
inherent explicitation as the result of an operational norm thus has to be re-
jected.
A third problem is connected with an important suggestion given by Blum-
Kulka. In her 1986 paper, Blum-Kulka concludes that “it should be possible to
ascertain by empirical research to what extent explicitation is indeed a norm
that cuts across translations from various languages and to what extent it is a
language pair specific phenomenon” (1986:23). In other words, she recognizes
the problem of distinguishing between optional and translation-inherent explici-
tations. Optional explicitations of course need to be identified and excluded
when investigating the Explicitation Hypothesis. Blum-Kulka thus cautions her
readers that “it would be necessary to first carry out a large scale contrastive
stylistic study (in a given register) [...] and then to examine translations to and
from both languages to investigate shifts [...] that occur in translation” (Blum-
Kulka 1986:33).
Øverås quotes this important advice given by Blum-Kulka but then goes on
to say that “such investigation into contrastive features was not possible”
(1998:9). It is understandable that a full-blown investigation into the stylistic
norms of English and Norwegian was not possible within the scope of Øverås’
study. But the lack of this contrastive foundation implies a big problem: igno-
rant of many stylistic contrasts between English and Norwegian, Øverås of
course encounters a number of cases where it is not clear whether they are op-
tional or potentially translation-inherent explicitations, which she “included on
the assumption that, while not part of the present survey, the investigation of
initial norms may benefit from research that includes all occurrences” (1998:9).
I cannot see how the investigation of initial norms (Toury 1995:56f), which is at
best only peripherically related to the aim of Øverås’ study, could justify such a
methodologically fatal step. Doubtful cases should never be regarded as evi-
dence for or against anything.
Speaking of doubtful cases, let us have a look at three examples that
Øverås included in her study as potentially translation-inherent explicitations:
(1) NorOrig: Den hvite mannen knipser.
‘The white man clicks.’
EngTrans: The white man clicks his camera. (Øverås 1998:8)
In (1), the author of the Norwegian original uses the verb knipse, which means
‘to click’, or, in this context, ‘to photograph, to take a snapshot’. Since the di-
rect English equivalent of knipse, to click, does not share the idiomatic meaning

14 VIKTOR BECHER
of its Norwegian counterpart, the translator decides to expand the verb to the
collocation to click one’s camera (which, for obvious reasons, is more explicit
than knipse). Since English does not have an expression comparable to knipse in
terms of implicitness, the translator is forced to perform this explicitating shift
given her prior decision to translate knipse as to click. We are therefore dealing
with what appears to be an instance of obligatory explicitation possibly trig-
gered by an instance of source language interference. Thus, it is incorrect to
count this example as evidence in a study aiming to test the Explicitation Hy-
pothesis (and purporting to exclude obligatory shifts). The instance of explicita-
tion in question is clearly the result of a lexical contrast between English and
Norwegian and is thus specific to this particular language pair.
(2) NorOrig: Jeg lente meg fram over bordet og fisket ut en Hobby.
‘...and fished out a Hobby.’
EngTrans: I leaned forward over the table and fished out a Hobby
cigarette. (Øverås 1998:11)
In (2), the translator explicitates the word cigarette. The motivation behind this
minor shift seems to be the fact that while most Norwegian readers probably
know that Hobby is a cigarette brand, English readers might have difficulty in
drawing this inference. This example therefore appears to be a paradigm case of
pragmatic explicitation, which should of course be excluded from a study on
translation-inherent explicitation.
Let us look at a final example:
(3) NorOrig: Nå er St. Patric den største helgenen i hele Irland.
‘Now St. Patric is the greatest saint in all of Ireland.’
EngTrans: Now Saint Patric is regarded as the greatest saint in all of
Ireland. (Øverås 1998:10)
Øverås’ inclusion of (3) as evidence for the Explicitation Hypothesis is particu-

larly troubling, as this example does not even qualify as an instance of explicita-
tion, according to either the definitions provided in Section 2 or to Øverås’ own
definition quoted above. In the Norwegian ST of (3), the author expresses his
belief in the proposition that St. Patric is the greatest saint in all of Ireland. In
the English TT, on the other hand, things are very different. Here, the translator
has expanded the verb phrase to is regarded as, entailing a considerable change
in meaning: the belief in the proposition is no longer attributed to the author, but
rather to an unspecified person or group of persons. The translator has funda-
mentally changed the truth-conditional meaning of the TT vis-à-vis the ST.

So we are not dealing with a shift from implicit to explicit meaning here,
but with an (ideologically motivated?) change in meaning brought about by the
translator; the TT encodes a different state of affairs from the ST, so the ques-
tion of whether the expansion of the verb phrase performed by the translator is
to be counted as a case of explicitation does not even arise. (If anything, the ex-
pansion should be counted as an implicitation rather than an explicitation, since
the passive verb form is regarded as leaves implicit to whom the belief of the
proposition expressed is attributed.)
Øverås justifies her decision to include (3) as an instance of explicitation
by informing us that “it often proved difficult to determine the extent to which a
shift affects meaning” and that “all instances perceived to explicitate have there-
fore been included” (1998:11). It should go without saying that counting data as
evidence for a hypothesis should not rely on the “perception” of the researcher
but on objective criteria such as those proposed in Section 2.
Let us turn to Øverås’ results. Table 1 (taken from Øverås 1998:15) pre-
sents an overview of the explicitating and implicitating shifts that she counted
in her data:
Table 1
Explicitations and implicitations counted by Øverås (1998) in her translation corpus
(containing 1000 sentences per translation direction)
English–Norwegian Norwegian–English
Explicitation 347 248
Implicitation 149 76
We can see that there are roughly 100 more explicitating shifts in the English-
Norwegian translations than in the Norwegian-English translations investigated
by Øverås. As for the implicitating shifts, they show a similarly skewed distri-
bution over the two translation directions which in this case is even more pro-
nounced: there are almost twice as many shifts from English into Norwegian
than in the opposite translation direction. Most interestingly, explicitating shifts
in both translation directions are consistently more frequent than implicitating
ones.
Despite the lopsided distribution of explicitations across the two translation
directions, Øverås optimistically tells us that “one may safely conclude that [...]
Blum-Kulka’s explicitation hypothesis is confirmed”. However, she adds the
proviso that “[c]onfirmation was stronger in translations from English into
Norwegian than in the opposite direction” (1998:16). I find this conclusion
highly implausible. How can the hypothesis that “explicitation is a universal

16 VIKTOR BECHER
strategy inherent in the process of language mediation” (Blum-Kulka 1986:21)

find “stronger” confirmation in one translation direction than in the other?
There is clearly something wrong here.
I would like to propose an alternative conclusion that seems much more
plausible: the explicitations identified by Øverås are not of the translation-
inherent type; rather, they go back to a mixture of obligatory, optional and
pragmatic explicitations (cf. the examples discussed above). This would explain
the imbalanced distribution of explicitations across the two translation direc-
tions quite nicely: we would expect English and Norwegian to differ in terms of
the degree of explicitness they favor lexicogrammatically and stylistically, and
these differences probably account for the skewing observed by Øverås. Over-
all, the lexicogrammatical properties and/or stylistic preferences of English
seem to favor a higher degree of explicitness than those of Norwegian.
Since the alternative interpretation of Øverås’ findings offered in the previ-
ous paragraph does not require the assumption that a number of translation-
inherent explicitations are ‘hidden’ among her data, it is more in line with Oc-
cam’s Razor and thus to be preferred over Øverås’ interpretation of her results
as evidence for the Explicitation Hypothesis. Still, her finding that there are
more explicitations than implicitations in both translation directions is remark-
able. We will have to explain it in the following section.
6. ABANDONING THE EXPLICITATION HYPOTHESIS
In sum, this paper has pointed out four main problems with the Explicitation
Hypothesis. The first two problems are of a theoretical nature (see Section 4 on
Blum-Kulka 1986), while the other two are of the methodological kind (see
Section 5 on Baker and Olohan 2000 and Øverås 1998):
1. The assumption of a separate, translation-inherent type of explicitation
is (1) unmotivated and (2) collides with Occam’s Razor. Instead of
explaining anything, the Explicitation Hypothesis only creates the need
for further explanation. Moreover, its investigation entails the danger of
producing what I have called pseudo-significant findings.
2. The nature of translation-inherent explicitation is not clear; is it
supposed to be a subconscious or a conscious phenomenon?
3. All studies on translation-inherent explicitation (that I know of10) fail to
control for interfering factors such as other types of explicitation,
source language interference, the effect of other (potential) translation
universals, etc.
4. Many studies either do not provide a definition of explicitation at all, or
they provide one but do not adhere to it (cf. Becher forthcoming a).

This of course entails the danger of counting pseudo-explicitations such

as (3) above.
The four problems summarized above permit the following two, somewhat
radical conclusions. First, the Explicitation Hypothesis should be abandoned11
because it is not a useful12 hypothesis. Second, previous studies have failed to
provide conclusive evidence for the hypothesis anyway.
In the remainder of this section, I am going to argue that a slightly ex-
tended and motivated version of Klaudy’s (2009) Asymmetry Hypothesis13 can
serve as a more useful and plausible guide for further research on explicita-
tion.14 The hypothesis postulates that
explicitations in the L1→L2 direction are not always counterbalanced by
implicitations in the L2→L1 direction because translators – if they have a
choice – prefer to use operations involving explicitation, and often fail to
perform optional implicitation. (Klaudy and Károly 2005:14)
I find this formulation somewhat problematic,15 as (1) it does not specify which
kinds of explicitation are covered (Klaudy’s Types 1 to 3, or only optional ex-
plicitations?), (2) the term prefer evokes the impression that a conscious deci-
sion is being made on the part of the translator (I would like to admit the possi-
bility of subconscious explicitation in my version of the hypothesis) and (3) the
term fail has a prescriptive flavor to it. As will become clear in the following, I
do not think we can blame translators for being more explicit than authors of
non-translated texts. I would thus like to propose a slightly modified version of
Klaudy’s hypothesis:
The Asymmetry Hypothesis (modified version):

Obligatory, optional and pragmatic explicitations tend to be more frequent
than the corresponding implicitations regardless of the SL/TL constellation at
hand.
The Asymmetry Hypothesis claims that translators display a tendency to

explicitate, but it avoids positing a separate, translation-inherent type of explici-
tation. It makes do with Klaudy’s explicitation Types 1 to 3, which are unprob-
lematic and uncontroversial. The only remaining problem is that the Asymmetry
Hypothesis still needs to be properly motivated (see Klaudy 2009 for a first
sketch). This is what I am going to do in the following.

18 VIKTOR BECHER
6.1. Motivating the Asymmetry Hypothesis
It is well known that human communication is driven by two competing forces,

or principles (see e.g. Atlas and Levinson 1981, Horn 1984, Fabricius-Hansen
2005; cf. also Grice 1975):
1. The Q Principle: “Say as much as you can!” (→ explicitness)
2. The R Principle: “Say no more than you must!” (→ implicitness)
(adapted from Horn 1984:13)
It is obvious that strictly speaking, the two principles contradict each other: “[a]
speaker obeying only Q would tend to say everything she knows on the off-
chance that it might prove informative, while a speaker obeying only R would
probably, to be on the safe side, not open her mouth” (Horn 1984:15). It is im-
possible for a speaker to stick to just one of the two principles (which would not
be very smart anyway); rather, when preparing her message for formulation, the
speaker has to decide which principle to follow to which degree. In other words,
the speaker has to determine the most favorable trade-off between the two prin-
ciples. The Q and R principles can thus be regarded as the two (virtual) end
points of an explicitness–implicitness scale inherent to linguistic communica-
tion.
It will be apparent that the specific communication situation at hand deter-
mines where a favorable trade-off between the two principles might be, i.e.
which point on the explicitness–implicitness scale should be chosen for the
message to be formulated. In face-to-face communication, the trade-off will
tend towards the implicit end of the scale: if the hearer signals that my message
turned out to be too implicit “Huh, what do you mean?”, I can elaborate, i.e.
make it more explicit ex post. In written text, on the other hand, the trade-off
will tend towards the explicit end of the scale: since I do not have access to di-
rect hearer feedback in this case,16 I will tend to be too explicit rather than too
implicit when in doubt (cf. von Hahn 1997).
What is important to see here is that in terms of the explicitness–
implicitness scale spanned by the Q and R Principles, translations are written
texts par excellence. In other words, translations should tend to be located even
further towards the explicit end of the scale than non-translated texts (cf.
Klaudy 2009). This is due to two properties of the communicative situation
typically underlying translation:
Property 1: The communicative situation underlying translation is typi-
cally characterized by cultural distance between (SL) author and (TL) reader
(House 1997).
Konrad Ehlich has insightfully described written discourse as a “dilated
speech situation” (1984). Writers and readers communicate as they would in

face-to-face communication, albeit in a spatially and temporally “dilated” man-

ner. (For example: an author produces a book and the reader answers by writing
the author a letter.) This is why it was claimed above that written communica-
tion tends to be relatively explicit: increasing the explicitness of the message ex
post is not very practical in this communication situation.
Now in translation, the speech situation is even more dilated than in mono-
lingual written discourse: the distance between writer and reader is not only
spatial and temporal, but also cultural. Recognizing this, responsible translators
will move even further than authors of non-translated texts towards the explicit
end of the Q–R scale in order to compensate for the perceived cultural distance.
Viewed from this perspective, the tendency of translators to explicitate is noth-
ing mysterious; it is not due to “subconscious processes” (Olohan and Baker
2000), but rather the result of a number of – conscious or subconscious (it does
not matter in the present context) – attempts of translators to compensate for the
cultural ‘dilatedness’ characteristic of translation as a specific communication
scenario (cf. Saldanha 2008:28, who stresses the dependence of explicitation on
“translators’ assumptions about their readership and about their role as literary
and cultural mediators”).
Note that the perceived need to deal with a dilated speech situation is not
translation-specific. Thus, explicitations resulting from translators’ preoccupa-
tion with reducing cultural distance cannot be called “translation-inherent”.
First, as I have argued above, translators follow the same communicative prin-
ciples as authors of non-translated texts: they choose a certain point on the ex-
plicitness–implicitness scale in accordance with the perceived dilatedness of the
communication situation at hand. Second, the need to bridge cultural gaps also
arises in monolingual communication. Imagine, for example, the case of an
immigrant author writing about the culture of his country of origin. Here, the
same cultural bridging takes place as in translation. We should thus expect the
author to pick a point on the Q–R scale near its explicit end – as would a trans-
lator.
Finally, it is worth pointing out that compensatory strategies are not neces-
sarily applied only where appropriate. It may well be that translators, driven by
an unspecific desire to bridge the cultural gap between SL author and TL
reader, explicitate even in contexts where cultural differences are not directly
relevant. For example, translators might compensate for the (possible) incoher-
ence of text passages expressing source-culture specific concepts by increasing
explicitness in other (related) text passages. (Another possibility would be that
translators simply overuse compensatory, explicitating strategies.) Explicita-
tions of this type might look puzzling to the researcher as to their origin. How-
ever, as the above line of argumentation has shown, we do not need to appeal to
a mysterious notion of “translation-inherence” to explain their occurrence.

20 VIKTOR BECHER
Property 2: The communicative situation underlying translation is typi-

cally characterized by a great deal of communicative risk (cf. Pym 2005, whom
the line of argument below follows).
Authors of non-translated texts are paid for content, whereas translators are
paid for communication.17 The main task of the author is to make her thoughts
available to the reader by putting them on paper. The main task of the translator,
in contrast, is to ensure understanding between SL author and TL reader, avoid-
ing misunderstanding at all costs. If the reader – for whatever reason – has diffi-
culties in understanding the translated text, she will be quick to blame the trans-
lator. Since translators are paid for linguistic mediation, i.e. for achieving un-
derstanding, they are always the prime suspects when communication problems
occur. There is even the unpleasant case where a translator is blamed for the dif-
ficulty of a text that was already hard to understand in its SL version.
The upshot is that translators have to cope with a certain kind of risk – the
risk of not being understood. Accordingly, it is not surprising that translators
will go to great lengths to ensure understanding, and this is where explicitation
comes into play. As we have noted in Section 2, implicitness is a viable option
when the reader might be able to infer the implicit information. Even in cases
where the inference is an easy and obvious one, it may still happen that a given
reader fails to draw it. The result may be a failure to understand – and thus a
problem for the poor translator, who had merely attempted to stay true to the SL
text. Therefore, it seems plausible to assume that translators will move up on the
Q–R scale, i.e. they will tend to be too explicit rather than too implicit when in
doubt (and maybe even when not in doubt).
What is the risk of being too explicit? A waste of energy and paper – not
too bad. What is the risk of being too implicit? Communicative breakdown –
very bad. This shows that it is perfectly natural and justifiable for translators to
adopt a strategy of avoiding implicitness, even where it is not licensed by the
source text.
Again, note that the tendency of translators to avoid risk cannot be called
“translation-inherent”. First, it depends on the individual translator how much
risk she is willing to take. Translators may either be rather confident or averse
to risk-taking. Moreover, risk differs across translation scenarios (cf. e.g. the
translation of a law vs. the translation of a cooking recipe). Second, authors of
non-translated texts have to deal with the same kind of risk that translators face
(cf. von Hahn 1997), albeit to a lesser degree. Imagine, for example, the case of
a scientist writing for a lay public, who has to deal with the very same kind of
communicative risk as a translator.

6.2. What Explicitational Asymmetry Looks Like in Practice
We have now provided a motivation for postulating the Asymmetry Hypothesis:

explicitations tend to occur more frequently than implicitations in translation
due to two typical properties of the communication scenario underlying this
kind of discourse. In this way, we have also offered an explanation for Øverås’
(1998) finding that explicitations outnumber implicitations in her corpus. Let us
see how the explicitation–implicitation imbalance posited by the Asymmetry
Hypothesis works out in practice by looking at a concrete example (taken from
Becher forthcoming a):
(4) EngOrig: The virus is one of the major causes of chronic liver disease,
probably accounting for even more cases than excessive alcohol use.
GerTrans: HCV verursacht wahrscheinlich mehr chronische
Lebererkrankungen als Alkoholmißbrauch. Das Virus steht damit in
vorderster Reihe der Faktoren, die zu Leberleiden führen.
‘HCV probably causes more cases of chronic liver disease than alcohol
abuse. The virus thus is among the prime factors that cause liver
disease.’
The example consists of an excerpt from an article published in the popular sci-
entific magazine Scientific American alongside its German translation, which
appeared in the German magazine Spektrum der Wissenschaft. The English ST
sentence consists of a main clause to which an ing-adjunct has been attached, a
construction which is highly underspecified semantically (cf. Blühdorn 2009 on
semantic underspecificity in clause linkage). Ing-adjuncts may express a variety
of relations ranging from temporality to causality (Behrens 1999). The transla-
tion problem that we witness here is that German does not have a construction
syntactically and semantically equivalent to the English ing-adjunct. The trans-
lator has thus decided to split the ST sentence into two TT sentences, a common
translation choice in such cases (cf. Fabricius-Hansen 1998, 1999). This leaves
the translator with an interesting choice: if she uses a connective such as damit
‘thus’ to clarify the semantic connection between the two sentences, as was
done in (4), the TT comes out as more explicit that the ST; if she does not use a
connective, the TT will be more implicit than the ST.
We see here a confirmation of Toury’s claim that “the need [...] to deviate
from source-text patterns can always18 be realized in more than one way”
(1995:57). In this case, the necessary deviation from the ST pattern may either
be an explicitation or an implicitation – both would be justifiable choices given
the semantic unspecificity of the English ing-adjunct. We do not know the exact
reason why the translator of (4) chose the explicit over the implicit variant.

22 VIKTOR BECHER
Maybe risk avoidance? Since German discourse is generally characterized by a

higher degree of explicitness in the encoding of semantic relations than English
discourse (see e.g. House 2004, Fabricius-Hansen 2005, Behrens 2004, Becher
forthcoming b), it is also possible that the explicitation performed by the trans-
lator of (4) is simply due to the application of a “cultural filter” (House 1997),
i.e. adjustment of the ST to the stylistic norms of the TL. But even if this is not
the case, we have no fundamental problem in explaining this instance of explici-
tation, as the two typical properties of translation qua communication scenario
outlined above allow us to predict the predominance of explicitating shifts in
cases where translators could have chosen an impliciating shift instead.
7. CONCLUSION
Since the main points of this paper have already been summarized in Section 1
and at the beginning of the previous section, I will confine myself to drawing
some more general conclusions in this final section.
In the previous section, I have argued that translation is not fundamentally
different from monolingual discourse as far as the balance between explicitness
and implicitness is concerned. When choosing a point on the explicitness–
implicitness scale, translators are guided by the very same considerations as
monolingual authors. At times, the latter also have to deal with cultural distance
and/or communicative risk. This means that explicitation, insofar as it is caused
by the tendencies of translators to compensate for cultural distance and to avoid
risk, is neither “translation-inherent” (translators do not do anything translation-
specific, they only do what authors of non-translated texts do) nor “universal” in
a strict sense (there will always be situations in which translators do not display
the mentioned tendencies).19
As far as the explicitness–implicitness dimension of language use is con-
cerned, translation is not a “third code”;20 it is not an exceptional or anomalous
kind of discourse governed by other constraints than normal language use. As
stated by House, “Translation is no more and no less than a practical activity. It
can be described as an act of performance, of parole, not of langue or compe-
tence” (2008:11). Rather, the relative explicitness of translated discourse is a
straightforward result of the communicative circumstances under which it is
typically produced. The Asymmetry Hypothesis that I have argued for above
does justice to this insight, as it merely claims that explicitations tend to be
more frequent that implicitations in translation, thus allowing for exceptional
cases where cultural distance is insignificant and/or communicative risk is low.
In these cases, we do not expect explicitations to outnumber implicitations.21
One might be tempted to say that explicitation is universal in a wider sense,
namely in the sense that the tendency to explicitate is not characteristic, but

typical of translated discourse. This, however, would be misleading, since it

contradicts the normal usage of the term “universal”. In linguistics, for example,
universals are usually defined as “those properties that are necessarily common
to all human languages” (Comrie 2003:195). Given this normal use of the term,
a predominance of explicitating vis-à-vis implicitating shifts would have to be a
property that is necessarily common to all translations, or translation scenarios,
in order to qualify as a translation universal. However, as the above considera-
tions have shown, this is highly unlikely. It is therefore misleading to call ex-
plicitation a (possible) “universal of translation”, as e.g. Baker (1993, 1996)
does.
Finally, I would like to take up the “important theoretical question” raised
by House “of how useful or indeed possible and thus justifiable the positing of
translation universals such as the ones mentioned above are, [which] has not
been touched let alone recognized by all researchers in the field of translation
studies” (2008:10f). House’s own view is that “the quest for translation univer-
sals is in essence futile, i.e. that there are no, and there can be no, translation
universals.”22 One of her five arguments for this stance is that much of what
may look like a translation universal can in fact be attributed to general “univer-
sals of language also applying to translation.” (2008:11) This is quite a radical
view, but it gains some support from the considerations made in the present pa-
per: the Q and R principles and the explicitness–implicitness scale they span
form a genuine universal of language use, and it is this linguistic universal
which accounts for the observation that translators tend to explicitate rather than
to implicitate – in certain contexts.
Acknowledgements. The research for this paper has been carried out within the
project Covert Translation led by Juliane House and located at the University of
Hamburg’s Research Center on Multilingualism. The research center is funded
by the Deutsche Forschungsgemeinschaft (German Research Foundation),
whom I would like to thank for their generous financial support. I am indebted
to Silvia Bernardini, Andrew Chesterman, Juliane House, Svenja Kranich,
Stella Neumann, Sonia Vandepitte and an anonymous referee for their critical
comments on an earlier version of this paper.
Notes
1
As the labels chosen by Klaudy can be a bit misleading, it might be desirable to replace
them with better terms in the future. For example, optional explicitation is ‘pragmatic’ in the
sense that it is dependent on the communicative preferences of the target language community,
i.e. on its pragmatic norms; and pragmatic explicitation is ‘optional’ in the sense that it normally
does not have to be performed.

24 VIKTOR BECHER
2
Due to lack of space, I cannot discuss Klaudy’s (2008) classification in due detail, nor can
I comment on the given examples of explicitation. The reader is referred to Klaudy’s interesting
paper.
3
Blum-Kulka (1986) also seems to accept the existence of Klaudy’s Type no. 3, viz. prag-
matic explicitations.
4
It is sometimes claimed that the Explicitation Hypothesis can be motivated as follows.
When translators interpret the ST, they enrich their interpretation with inferential meaning (e.g.
by interpreting temporal sequence as causal sequence), as is normal in text comprehension (cf.
e.g. Graesser et al. 1994, Carston 2009). This pragmatically enriched interpretation is of course
more explicit than the ST itself. (So far I agree.) It may thus lead to a more explicit TT. This con-
clusion, however, is a fallacy. It depends on the assumption that translators directly verbalize their
(more explicit) mental representation of the ST without applying operations that might render it
more implicit, such as politeness strategies, omission of contextually inferable material, etc. There
is no reason why translators – in contrast to authors of non-translated texts – should skip the ap-
plication of such operations.
5
Another problem with respect to Occam’s Razor is that Blum-Kulka applies her Explicita-
tion Hypothesis not only to translations, but to all kinds of linguistic mediation, under which she
also subsumes speech production by foreign language learners (1986:19–21). It is easy to see that
the latter assumption, which presupposes that translators rely on similar cognitive processes as
foreign language learners, strongly conflicts with Occam’s Razor.
6
See http://ronaldo.cs.tcd.ie/tec2/jnlp/, where the TEC may be queried online free of
charge. Again, Olohan and Baker deserve credit for making their data openly available, giving
other researchers the chance to challenge and/or expand upon their findings.
7
The problem pointed out here is not specific to Olohan and Baker’s study. In general, one
has to be very cautious when doing research on monolingual translation corpora, i.e. corpora con-
taining translations only. Corpora of this type should only be used for hypothesis formation, not
for hypothesis testing (cf. Bernardini 2010).
8
Kenny (2005) attempted to validate Olohan and Baker’s (2000) results by investigating
the (non)occurrence of optional that in a German–English parallel corpus. However, her study has
its own problems and thus cannot be taken as a confirmation of Olohan and Baker’s results. Due
to lack of space, I cannot discuss Kenny’s study in detail. Suffice it to say that Kenny herself ad-
mits that in order to properly interpret her results, “one would first have to ascertain whether
translators insert that in places where the German [author, VB] did not use [a complementizer],
but did use a subjunctive form to make the reported nature of a clause explicit. If this were the
case, it might be difficult to argue that the English structure including that was actually more ex-
plicit than the German structure with no [complementizer].” (2005:161)
9
Strangely, nowhere in Øverås (1998) is it directly said that she views translation-inherent
explicitation as a translational norm. We have to infer this from some vague remarks on page 3 of
her article, as well as from its subtitle, “An investigation of norms in literary translation”.
10
See e.g. Baker and Olohan (2000), Øverås (1998), Olohan (2002), Chen (2004), Pápai
(2004), Kenny (2005), Konšalová (2007), Kamenická (2008), as well as Becher (forthcoming a)
for critical discussion. There are also studies that provide direct evidence against the Explicitation
Hypothesis (e.g. Baumgarten et al. 2008, Puurtinen 2004, Musacchio and Palumbo 2010), the dis-
cussion of which unfortunately lies outside the scope of the present paper.
11
If in the future we should find out that more conservative hypotheses cannot explain the
occurrence of explicitation phenomena in translation, we can still go back to Blum-Kulka’s as-
sumption of a translation-inherent type of explicitation, which, however, would have to be made a
lot more precise before it could serve as a useful guide for research.
12
The Explicitation Hypothesis is not useful in the sense that it is dangerous to investigate.
On the other hand, it has sparked a great deal of interest in the important phenomenon of explici-
tation (and impliciation) in translation. In this respect, the Explicitation Hypothesis has been ex-

tremely useful for translation studies. I would like to stress in this connection that the aim of the
present article is not to denigrate Shoshana Blum-Kulka’s seminal and still very readable 1986
paper, but to show that after an initial phase of pioneering explicitation research, the time has
come to abandon the Explicitation Hypothesis and to look for a better alternative.
13
The hypothesis was originally proposed in a 2001 conference paper (Klaudy 2001).
14
Another promising approach to the study of explicitation in translation is not to depart
from a hypothesis at all, but to approach the data quantitatively from a bottom-up, hypothesis-
generating perspective. Such a data-driven approach has been pursued in recent years by Hansen-
Schirra, Kunz, Neumann and Steiner at the University of Saarbrücken. One of their goals is to es-
tablish a more precise operationalization of the concept of explicitness/explicitation, which they
are attempting to achieve “by defining explicitness and explicitation, by stratifying it in terms of
different linguistic levels, by tightening its boundaries, and by modularizing it in a multifunctional
perspective” (Steiner 2005:19). Initial results from this promising approach are presented in Han-
sen-Schirra et al. (2007), Steiner (2008) and Kunz (2009).
15
What I like about Klaudy and Károly’s formulation of the Asymmetry Hypothesis is the
“not always” part (corresponding to “tend to” in my version of the hypothesis): they do not claim
that explicitations outnumber implicitations in each and every case. As will become clear in the
following, this weak formulation is fully justified.
16
To ease exposition, I will ignore hybrid scenarios such as real-time written communica-
tion (as it takes place in Internet chatrooms, for example), which on the explicitness–implicitness
scale would fall somewhere between face-to-face communication and ‘traditional’ written dis-
course. Also, it should be borne in mind that, as Biber (1988) has pointed out, there is no single
linguistic dimension that neatly corresponds to the distinction ‘spoken vs. written’, and the di-
mension ‘explicit vs. implicit’ is no exception. The relative explicitness of written as compared to
spoken discourse is just a tendency.
17
This sweeping claim is of course an overgeneralization, but I think it nicely highlights an
important difference between the task of the author and that of the translator (see the following
remarks).
18
The always part of this claim might be too strong, but that does not need to concern us
here.
19
Moreover, there seem to be strong individual differences between translators with respect
to their use of explicitating shifts (Saldanha 2008:30ff).
20
As far as less controversial phenomena such as source language interference are con-
cerned, it may well be justified to call translation a “third code”. The term goes back to Frawley
(1984).
21
Denturck provides results from the investigation of a corpus consisting of translations be-
tween French and Dutch in which implicitations (of connectives) were found to be more frequent
than explicitations in one of the two translation directions (Dutch→French), whereas the opposite
was the case in the other direction (Denturck 2009, Denturck and Niemegeers 2010). While it re-
mains to ascertain in how far stylistic differences between French and Dutch are responsible for
this remarkable finding, Denturck’s results show that explicitations do not invariably outnumber
implicitations.
22
Cf. Tymoczko (1998, 2005), who makes a similar point as House in saying that “[n]ot all
conclusions of research are applicable to all translation types or all translation contexts. It follows
[...] that [translation studies] should give up the search for universals.” (2005:1095). Unfortu-
nately, warning voices such as the ones by Tymoczko and House have gone largely unnoticed in
the development of corpus-based translation studies.

26 VIKTOR BECHER
References
Atlas, J. & Levinson, S. C. 1981. It-Clefts, Informativeness, and Logical Form: Radical Pragmat-
ics (revised standard version). In: Cole, P. (ed.) Radical Pragmatics. New York: Academic
Press. 1–61.
Baker, M. 1993. Corpus Linguistics and Translation Studies: Implications and Applications. In:
Baker, M., Francis, G. & Tognini-Bonelli, E. (eds) Text and Technology: In Honour of John
Sinclair. Amsterdam & Philadelphia: John Benjamins. 233–250.
Baker, M. 1996. Corpus-Based Translation Studies: The Challenges That Lie Ahead. In: Somers,
H. (ed.) Terminology, LSP and Translation. Studies in Language Engineering in Honour of
Juan C. Sager. Amsterdam & Philadelphia: John Benjamins. 175–186.
Becher, V. Forthcoming a. Towards a More Rigorous Treatment of the Explicitation Hypothesis
in Translation Studies. trans-kom.
Becher, V. Forthcoming b. Differences in the Use of Deictic Expressions in English and German
Texts. Linguistics.
Behrens, B. 1999. A Dynamic Semantic Approach to Translation Assessment. ING-Participial
Adjuncts and Their Translation in Norwegian. In: Doherty, M. (ed.) Sprachspezifische As-
pekte der Informationsverteilung. Berlin: Akademie-Verlag. 90–112.
Behrens, B. 2004. Cohesive Ties in Translation: A Contrastive Study of the Norwegian Connec-
tive dermed. Languages in Contrast Vol. 5. No. 1. 3–32.
Baumgarten, N., Meyer, B. & Özçetin, D. 2008. Explicitness in Translation and Interpreting. A
Review and Some Empirical Evidence (of an Elusive Concept). Across Languages and Cul-
tures Vol. 9. No. 2. 177–203.
Bernardini, S. 2010. Parallel Corpora and the Search for Translation Norms/Universals. Paper
presented at the conference Methodological Advances in Translation Studies, January 8–9,
Gent. http://liquid.hogent.be/~disgram/MATS2010/bernardini.ppt [last accessed 2010-01-
30].
Biber, D. 1988. Variation Across Speech and Writing. Cambridge: Cambridge University Press.
Blühdorn, H. 2009. Semantische Unterbestimmtheit bei Konnektoren. In: Pohl, I. (ed.) Seman-
tische Unbestimmtheit im Lexikon. Frankfurt: Peter Lang.
Blum-Kulka, S. 1986. Shifts of Cohesion and Coherence in Translation. In: House, J. & Blum-
Kulka, S. (eds) Interlingual and Intercultural Communication. Tübingen: Narr. 17–35.
Carston, R. 2009. The Explicit/Implicit Distinction in Pragmatics and the Limits of Explicit
Communication. International Review of Pragmatics Vol. 1. 35–62.
Chen, W. 2004. Investigating Explicitation of Conjunctions in Translated Chinese: A Corpus-
Based Study. Language Matters Vol. 35. No. 1. 295–312.
Comrie, B. 2003. On Explaining Language Universals. In: Tomasello, M. (ed.) The New Psychol-
ogy of Language, vol. 2. Mahwah: Erlbaum. 195–209.
Dallal, G. E. 2007. The Little Handbook of Statistical Practice. Chapter on Cause and Effect.
http://www.jerrydallal.com/LHSP/cause.htm [last accessed 2010-01-30].
Denturck, K. 2009. Explicitation of Causality in Translation: a Corpus-Based Study on French
and Dutch. Paper presented at the 42nd Annual Meeting of the Societas Linguistica Eu-
ropaea, workshop Connectives across languages: Explicitation and grammaticalization of
contingency relations, September 9–12, Lisbon. http://www.francais.ugent.be/in-
dex.php?id=28&type=file [last accessed 2010-01-30].
Denturck, K. & Niemegeers, S. 2010. Explicitation – Implicitation – Unique Items: the Case of
Modal Particles and Causal Connectives. Paper presented at the conference Methodological
Advances in Translation Studies, January 8–9, Gent. http://liquid.hogent.be/~disgram/
MATS2010/denturck.ppt [last accessed 2010-01-30].
Ehlich, K. 1984. Zum Textbegriff. In: Rothkegel, A. & Sandig, B. (eds) Text – Textsorten – Se-
mantik. Linguistische Modelle und maschinelle Verfahren. Hamburg: Helmut Buske. 9–25.

Fabricius-Hansen, C. 1998. Informational Density and Translation, With Special Reference to

German – Norwegian – English. In: Johansson, S. & Oksefjell, S. (eds) Corpora and Cross-
Linguistic Research. Theory, Method, and Case Studies. Amsterdam: Rodopi. 197–234.
Fabricius-Hansen, C. 1999. Information Packaging and Translation: Aspects of Translational Sen-
tence Splitting. In: Doherty, M. (ed.) Sprachspezifische Aspekte der Informationsverteilung.
Berlin: Akademie-Verlag. 175–214.
Fabricius-Hansen, C. 2005. Elusive Connectives. A Case Study on the Explicitness Dimension of
Discourse Coherence. Linguistics Vol. 43. No. 1. 17–48.
Frawley, W. 1984. Prolegomenon to a Theory of Translation. In: Frawley, W. (ed.) Translation:
Literary, Linguistic and Philosophical Perspectives. London & Toronto: Associated Uni-
versity Presses. 159–175.
Graesser, A. C., Singer, M. & Trabasso, T. 1994. Constructing Inferences During Narrative Text
Comprehension. Psychological Review Vol. 101. No. 3. 371–395.
Grice, H. P. 1975. Logic and Conversation. In: Cole, P. & Morgan, J. (eds) Syntax and Semantics,
vol. 3. New York: Academic Press. 41–58.
Hahn, W. von. 1997. Das Postulat der Explizitheit für den Fachsprachengebrauch. In: Hoffmann,
L., Kalverkämper, H. & Wiegand, H. E. (eds) Ein internationales Handbuch zur Fach-
sprachenforschung und Terminologiewissenschaft / Languages for Special Purposes. An In-
ternational Handbook of Special-Language and Terminology Research. Berlin & New
York: Walter de Gruyter. 383–389.
Hansen-Schirra, S., Neumann, S. & Steiner, E. 2007. Cohesive Explicitness and Explicitation in
an English-German Translation Corpus. Languages in Contrast Vol. 7. No. 2. 241–266.
Horn, L. 1984. Toward a New Taxonomy for Pragmatic Inference: Q-Based and R-Based Impli-
cature. In: Schiffrin, D. (ed.) Meaning, Form, and Use in Context: Linguistic Applications.
Washington: Georgetown University Press. 11–42.
House, J. 1997. Translation Quality Assessment. A Model Revisited. Tübingen: Narr.
House, J. 2004. Explicitness in Discourse Across Languages. In: House, J., Koller, W. & Schu-
bert, K. (eds) Neue Perspektiven in der Übersetzungs- und Dolmetschwissenschaft. Bo-
chum: AKS. 185–207.
House, J. 2008. Beyond Intervention: Universals in Translation? trans-kom Vol. 1. No. 1. 6–19.
Kamenická, R. 2008. Explicitation Profile and Translator Style. In: Pym, A. & Perekrestenko, A.
(eds) Translation Research Projects 1. Tarragona: Intercultural Studies Group. 117–130.
Kenny, D. 2005. Parallel Corpora and Tanslation Studies: Old Questions, New Perspectives? Re-
porting that in GEPCOLT: A Case Study. In: Barnbrook, G., Danielsson, P. & Mahlberg,
M. (eds) Meaningful Texts: The Extraction of Semantic Information from Monolingual and
Multilingual Corpora. London & New York: Continuum. 154–165.
Klaudy, K. 2001. The Asymmetry Hypothesis. Testing the Asymmetric Relationship Between
Explicitations and Implicitations. Paper presented at the Third International Congress of the
European Society for Translation Studies: Claims, Changes and Challenges in Translation
Studies. August 30 – September 01, Copenhagen.
Klaudy, K. 2008. Explicitation. In: Baker, M. & Saldanha, G. (eds) Routledge Encyclopedia of
Translation Studies. London: Routledge. 80–85.
Klaudy, K. 2009. The Asymmetry Hypothesis in Translation Research. In: Dimitriu, R. & M.
Shlesinger (eds) Translators and Their Readers. In Homage to Eugene A. Nida. Brussels:
Les Editions du Hazard. 283–303.
Klaudy, K. & Károly, K. 2005. Implicitation in Translation: Empirical Evidence for Operational
Asymmetry in Translation. Across Languages and Cultures Vol. 6. No. 1. 13–28.
Konšalová, P. 2007. Explicitation as a Universal in Syntactic De/condensation. Across Languages
and Cultures Vol. 8. No. 1. 17–32.
Kunz, K. 2009. Variation in English and German Nominal Coreference. A Study of Political Es-
says. Universität des Saarlandes (Saarbrücken): unpublished PhD dissertation.

28 VIKTOR BECHER
Maienborn, C. 2005. A Discourse-Based Account of Spanish ser/estar. Linguistics Vol. 43. No. 1.
155–180.
Musacchio, M. T. & Palumbo, G. 2010. Missing Links? A Corpus-Driven Approach to the Study
of Connectives in English-Italian Translation. Paper presented at the conference Methodo-
logical Advances in Translation Studies, January 8–9, Gent. http://liquid.hogent.be/~dis-
gram/MATS2010/musacchio.ppt [last accessed 2010-01-30].
Olohan, M. & Baker, M. 2000. Reporting that in Translated English: Evidence for Subconscious
Processes of Explicitation? Across Languages and Cultures Vol. 1. No. 2. 141–158.
Olohan, M. 2002. Leave It Out! Using a Comparable Corpus to Investigate Aspects of Explicita-
tion in Translation. Cadernos de Tradução Vol. 9. 153–169.
Øverås, L. 1998. In Search of the Third Code: An Investigation of Norms in Literary Translation.
Meta Vol. 43. No. 4. 557–570.
Pápai, V. 2004. Explicitation: A Universal of Translated Text? In: Mauranen, A. & Kujamäki, P.
(eds) Translation Universals. Do They Exist? Amsterdam & Philadelphia: John Benjamins.
143–164.
Puurtinen, T. 2004. Explicitation of Clausal Relations: A Corpus-Based Analysis of Clause Con-
nectives in Translated and Non-Translated Finnish Children’s Literature. In: Mauranen, A.
& Kujamäki, P. (eds) Translation Universals. Do they exist? Amsterdam & Philadelphia:
John Benjamins. 165–176.
Pym, A. 2005. Explaining Explicitation. In: Karoly, K. & Fóris, A. (eds) New Trends in Transla-
tion Studies. In Honour of Kinga Klaudy. Budapest: Akadémiai Kiadó. 29–34.
Saldanha, G. 2008. Explicitation Revisited: Bringing the Reader Into the Picture. trans-kom Vol.
1. No. 1. 20–35.
Steiner, E. 2005. Explicitation, its Lexicogrammatical Realization, and its Determining (Inde-
pendent) Variables – Towards an Empirical and Corpus-Based Methodology. SPRIKreports
36.
Steiner, E. 2008. Empirical Studies of Translations as a Mode of Language Contact – “Explicit-
ness” of Lexicogrammatical Encoding as a Relevant Dimension. In: Siemund, P. & Kintana,
N. (eds) Language Contact and Contact Languages. Amsterdam & Philadelphia: John Ben-
jamins. 317–345.
Swan, M. 1980. Practical English Usage. Oxford: Oxford University Press.
Toury, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam & Philadelphia: John
Benjamins.
Tymoczko, M. 1998. Computerized Corpora and the Future of Translation Studies. Meta Vol. 43.
No. 4. 652–660.
Tymoczko, M. 2005. Trajectories of Research in Translation Studies. Meta Vol. 50. No. 4.
1082–1097.

Across Languages and Cultures 11 (1), pp. 29–49 (2010)
DOI: 10.1556/Acr.11.2010.1.2
REPROCESSING TEXTS. THE FINE LINE BETWEEN

RETRANSLATING AND REVISING
OUTI PALOPOSKI1AND KAISA KOSKINEN2

1
Department of Modern Languages, P. O. Box 24,
FI – 00014 University of Helsinki,
Phone: +358-9-191 40068; Fax +358-9-191 23072,
E-mail: outi.paloposki@helsinki.fi
2
School of Modern Languages and Translation Studies,
FI – 33014 University of Tampere,
Phone: +358-3-3551 6119,
E-mail: kaisa.a.koskinen@uta.fi
Abstract: Retranslations are a frequent object of study in Translation Studies. They

can be used as data for a number of research problems, or retranslation can be studied as a
phenomenon on its own. There are no large-scale surveys on retranslation, however, let
alone surveys that would be coupled with in-depth case studies, no doubt due to the labori-
ousness and complexity of the task. Our own interest in the issue started from a small-scale
project testing the so-called Retranslation Hypothesis, but gradually our research expanded
into a wider range of questions. We have addressed three main areas: the extent and propor-
tion of retranslation in Finland; the motives for and reception of retranslations (publishers,
critics); and finally, what happens to a text when it is either retranslated or revised (textual
analysis). For this purpose, we have compiled three different sets of data from the Finnish
context. These sets consist of synchronic data (retranslations and their reviews from the year
2000), diachronic data (charting the retranslation history of classics shortlisted in 1999 and
1887) and case studies (by e.g. Victor Hugo, Nikolai Gogol, Astrid Lindgren and Juan Va-
lera translated into Finnish). This paper presents an overview of the results of our investiga-
tion, argues for a need for a comprehensive treatment of retranslation as a phenomenon, and
discusses the implications of textual analysis for the understanding of the fuzzy area be-
tween retranslation and revision. The cases presented include Hugo’s Les Misérables,
Gogol’s Dead Souls and Lagerlöf's Gösta Berling.
Keywords: retranslation, revision, hybrid texts, categorization, textual scholarship
1. INTRODUCTION
What happens in retranslation seems fairly simple: a text that has previously
been translated is translated again into the same language.1 The reasons for this
retranslating may seem simple. Translations are said to “age”: their language
becomes obsolete or they do not conform to prevailing standards of faithfulness

30 OUTI PALOPOSKI AND KAISA KOSKINEN
or accuracy; therefore, new translations are needed. The ageing of translations is

one of the most common arguments in reviews and media discourse in favour of
new translations, at least in Finland, whereas within academia, probably the
best-known claim about the nature of retranslation is expressed in what is often
called the Retranslation Hypothesis: the idea that the first translations’ inherent
assimilating qualities create a need for source-oriented translations. If all first
translations indeed were assimilating, “ageing” could also be understood as
their becoming old-fashioned; hence the two explanations could be seen to coa-
lesce into one.
The question of retranslations, however, is far more complex than this.
First, many intuitive assumptions about assimilating first translations and more
accurate second versions do not seem to hold true against empirical data: con-
trary examples abound. This is evident both from our own data and from many
other studies in which second translations are shown to be less source-oriented
than first translations (see below). Our earlier studies have shown that domesti-
cating translations may be a feature of a certain phase in translated literature
rather than a property of all retranslations. Source- or target-orientedness may
also be dependent on the observer’s viewpoint: at some other time in history (or
in the future) or with a different audience there may be a demand for target-
oriented (second) translations. The second crux is methodological: measuring
concepts such as improvement, closeness or accuracy in translations is singu-
larly difficult. It does not come as a surprise then that theoretical writings on re-
translations show a variety of starting points and methods of inquiry, and em-
ploy different units of comparison with which to study texts. This makes it hard
to compare the results of existing studies.
The theoretical body of literature on retranslations includes detailed and in-
sightful studies, but they are often single case studies, or they address other is-
sues than retranslation itself (such as norms, strategies or audience expecta-
tions). Şebnem Susam-Sarajeva who studied texts with several retranslations,
has pointed out:
Currently, there is no detailed or systematic study on retranslations per se.
Although the practice itself is common, theoretical discussions on the sub-
ject are rather rare. Retranslations often serve as case studies illuminating
other aspects of translational research rather than drawing attention onto
themselves as a topic in its own right. (2006:135)
Retranslation as a theoretical issue and as a widespread practice is still far

from being understood and certainly merits more comprehensive treatment than
what it has been given hitherto. Therefore Siobhan Brownlie’s (2006) article on
Narrative Theory and Retranslation was a welcome and timely contribution to a
field of study that has been touched from many angles but not properly mapped

REPROCESSING TEXTS 31
out, and in which there exist a number of intuitive assumptions which have not
been thoroughly studied.
Not just presenting a case study but also trying to understand the nature of
the phenomenon in question, Brownlie draws on insights from two sets of theo-
ries: Narrative Theory and Retranslation Theory.2 While showing the usefulness
of an interdisciplinary approach, she also provides a wealth of material for
studying retranslation. Her article charts some of what has been written about
retranslation so far; it also sums up empirical studies, discusses the explanatory
potential of suggested factors behind retranslation, and presents her own study
on the five different English translations of Zola’s Nana. Some of the most im-
portant issues at stake in retranslation surface through the discussion: the com-
plicated relationship between first and subsequent translations and the effect of
time on translations. In the present article, we would like to take issue with
some of the points Brownlie has made and sum up our findings, focusing on
textual analysis. As will be shown, the results of this analysis not only address
the issue of the closeness of first and later translations to their source texts (the
point where we originally started out) but will also raise meta-methodological
questions, which, to our understanding, are central to the study of retranslation.
First, however, we would like to present a short overview of what has been
written on retranslation in Translation Studies.
2. RE:TRANSLATION
The term Retranslation Hypothesis (RH) has often been used to encapsulate An-
toine Berman’s (1990, 1995) ideas: we, too, have used it previously as a short-
cut (Koskinen and Paloposki 2003; Paloposki and Koskinen 2004), following
writers such as Andrew Chesterman (2000:22–23).3 RH may be seen as an in-
terpretive hypothesis as far as it proclaims that only later translations can be
“great” translations, or it may be seen as a descriptive hypothesis, measuring the
distance between source and target texts and describing later translations as
more source-oriented (ibid). That Berman’s ideas (and Goethe’s before him) are
seen as forming a basis for a hypothesis to be tested is evident in writings such
as Vanderschelden (2000:13) and Ballard (2000:19, 20). Recently, the use of
the term Retranslation Hypothesis has increased in TS, judging from various
conference presentations we have listened to recently; it has also often (confus-
ingly) been called “Chesterman’s Retranslation Hypothesis”. This is no doubt
due to Chesterman’s work in defining different hypotheses with the help of RH.
Berman’s idea was that a first translation tends to be more assimilating and
tends to reduce the text’s otherness in the name of cultural or editorial require-
ments, whereas a retranslation would mark a return to the source-text. Contem-
poraneous with Berman’s writings was Yves Gambier’s (1994) article on re-

translations, where Gambier suggested several factors that might be at stake in

retranslation, including issues such as the marketing potential of a new transla-
tion. Despite Gambier’s explicit call for research and the many questions he
posed for a potential research programme, it seems that not many scholars (not
even Gambier himself) have been interested in empirical research or even theo-
retical development of the topic before the turn of the century. The “second
wave” of interest in retranslations was pioneered by Isabelle Vanderschelden’s
(2000) article discussing potential reasons behind retranslation and Michel Bal-
lard’s (2000) case study on retranslating Camus’ L’Etranger into English; these
two articles appeared in the same volume. In both cases, the starting point was
the vague but very widespread idea of classics (or “great books”) needing or
calling for retranslation. Vanderschelden (ibid:1) states that “it is just a matter
of time before a literary translation is challenged or replaced by another”; Bal-
lard (ibid:19) cites J. M. Cohen: “Every great book demands to be re-translated
once in a century”.4 Further on, Vanderschelden is inspired by the writings of
Berman, while Ballard draws on those of Goethe and Berman.
Vanderschelden presents a number of remarks and observations made by
translators and researchers on retranslation that point towards a more variegated
understanding of the reasons behind retranslation. Among other things, she dis-
cusses the difference between “hot” and “cold” translations (terms coined by
Claude Demanuelli; see Vanderschelden 2000:9): translations are “hot” when
they have to be done on the spot, with the translation appearing right after the
original work and with no research knowledge available yet on the work in
question, and “cold” if enough time has passed for the translator to resort to re-
search and audience responses when preparing her/his translation. The relevant
body of knowledge accessible to the translator would explain the increased “ac-
curacy” which may appear in a later translation. Probably due to a lack of
wholesale empirical studies on the issue, however, Vanderschelden’s observa-
tions mainly focus on improvement and quality, along the lines of Berman’s ar-
guments.
Ballard, for his part, presents a detailed case study focusing on textual
markers and their translation (proper names, titles and designators, units of
measurements) and the structural and stylistic organization of the text. Even if
Ballard evokes the often quoted assumption about the “need” for new transla-
tions, his results show that it is not merely a question of time that differentiates
between translations: stylistic preferences of each translator, for example, can
have a role to play. Isabelle Collombat (2004), who makes an overview of the
“retranslation boom” (in Canada?), also places stress on the translators: she
claims that their increased awareness of their task and of the pertinent circum-
stances around the translation make for new, better translations.
At this point, a “boom” in this issue became evident in Translation Studies.
Cadernos published a special issue on retranslation in 2003 and Palimpsestes in

2004 (this journal had been among the first to concentrate on retranslation: its
1990 special issued included Berman’s aforementioned article); more articles
appeared on the topic in Translation Studies journals, and Brownlie (ibid.) made
an overview of retranslation. In this overview, she discusses two published arti-
cles, Du Nour (1995) and Kujamäki (2001), which point towards quite different
reasons behind retranslation than improvement (the articles are among the
many, noted earlier, that have not been written with retranslation hypothesis in
mind but which had other objectives). She also makes the link to Narrative
Theory in explaining the variety of retranslations with the help of the concept of
versions: each translation is but one version that can be made of one single
original. It is obviously useful to make comparisons with other theories and find
similar cases; this observation, however, would seem to arise out of the empiri-
cal data so clearly that a Narrative Theory is probably not even needed to “jus-
tify” it. In Translation Studies literature, there have been other cases where the
borderline between versions/translations has been discussed (see Bassnett 2000
and Pym 1998:68–70).
Brownlie’s own study of Nana also shows that different translations can
exist more or less simultaneously, and explores the contextual factors behind
them. In the case of Nana, different audience expectations and restrictions re-
sulted in two different versions of the text in English. There are many similar
cases, among them the two almost simultaneous translations into Finnish of
Lewis Carroll’s Alice in Wonderland in our own data. Şebnem Susam-
Sarajeva’s work on the translations of French Structuralism into Turkish (2003,
2006) and Tiina Puurtinen’s (1995) results from her study of two Finnish trans-
lations of the American children’s classic The Wizard of Oz in the same year
(1977) are other examples of contextual variation. In all these cases, research
looking at translations that appear close to each other in time is a useful re-
minder of time not being the only affecting factor.
The evidence available from various case studies internationally would suf-
fice by itself to show that the strong version of the retranslation hypothesis, im-
plying that later translations are closer to the original or better than an earlier
translation (and/or substituting earlier translations), is not sufficient alone to
cover the field of retranslations. These studies include, for example, Tymoczko
1999, which shows how Irish epic poetry has been afforded very different
treatment at different periods in time and in the hands of different translators.
The same is true for Lopes 2006 on several translations into Portuguese of Har-
riet Beecher-Stowe’s Uncle Tom’s Cabin, Brownlie’s Nana, and the aforemen-
tioned Kujamäki 2001 on the German translations of the Finnish writer Aleksis
Kivi’s The Seven Brothers. Our own Finnish data abounds with such cases: for
example the two translations into Finnish of Juan Valera’s Pepita Jiménez and
the four translations of Bernardin St. Pierre’s Paul et Virginie underline the im-
portance of paying attention to translatorial styles and to audiences. Time and

order of appearance cannot be seen as a single monolithic entity or causal factor

behind retranslations – there are always different tendencies and multiple orien-
tations at work at any one specific time, just as there are different audiences and
translators.
3. RETRANSLATIONS IN FINLAND
We have been working for several years now – as a kind of a side-project along
our other tasks – on retranslation. We started out with a certain feeling of un-
easiness connected with the Retranslation Hypothesis, which, to our mind, did
not seem to sufficiently cover the specific cases we had been working on in our
previous research (the translation of the Vicar of Wakefield into Finnish, the
long history of translations of a Thousand and One Nights, and the literary re-
translation of the Gospel according to Matthew in Finland in the 1970s). These
were just a few single individual cases, but since they so clearly testified against
the Retranslation Hypothesis, we decided to continue with additional data (we
published the results of the first case studies as Paloposki and Koskinen 2004).
We then chose one specific year in Finland, the year 2000, and studied the sta-
tistics of retranslation for that year as well as combining it with a study of re-
prints, which started to attract our attention since they seemed the obvious first
alternative for retranslation. We also contacted publishers to unearth the reasons
behind the retranslations that year, and looked up translation criticism to study
the reception of retranslations (Koskinen and Paloposki 2003).
In the year 2000, the number of retranslations was 9 out of a total produc-
tion of 359 translations (261 of these were new and 89 reprints). Nine retransla-
tions is a fairly large number for a small language like Finnish; however, what
astonished us most was the number of reprints, which can be interpreted, not
negatively as a lack of the will to retranslate, but positively as a desire to keep a
stock of works available for the readers. The actual lists of titles – reprints and
retranslations – seem to indicate, rather unsurprisingly, that in order to be resur-
rected from the past, the work typically needs to have acquired the status of a
“classic”. But why are some classics reprinted while others are retranslated?
Some books such as Alice or Robinson Crusoe existed in a number of previous
versions and yet a retranslation was commissioned (or offered). Nor can one
conclude that the need for retranslation arises when older versions get dated:
even though there is variation, both retranslations and reprints tend to be of
books the previous versions of which date from the same period, the mid-19th
century. Not everything can be retranslated, as that would be too costly; a
choice has to be made regarding what to reprint and what to retranslate, and the
choice at times seems to be random. One possible answer is the different pro-
files of the publishers at different times: while some seem to favour retransla-
tions (especially young publishing houses which naturally do not have the stock

to recycle), others focus on reprints (and here it is the older publishing compa-
nies that do have the required stock). There is also a potential positive charisma
attached to retranslations and their marketing potential (translation reviews for
the year 2000 indicate that retranslations attract much greater publicity than new
translations and reprints).
One year is a very short period on which to make generalizations. More-
over, in the study of a complex question such as that of retranslations, even syn-
chronic data were not enough: we needed a diachronic view of what had been
retranslated over time. For this purpose, we next turned to a list of 100 classics
that was compiled in Finland at the turn of the millennium, and went through
the statistics of the translated classics included in the list (52 in all, the rest was
Finnish literature and non-fiction) (Koskinen and Paloposki 2005). This study
helped to contextualize the interest in retranslations today and see the impor-
tance and build-up of what is considered worth retranslating today (classics,
mainly, but how are they defined?). In other words, it gave us today’s perspec-
tive. Another angle was offered in the form of the translation programme of the
Finnish Literature Society, which drafted a classics list in 1887. Comparisons
between these two lists show differences and similarities in the understanding of
a classic and also raise a number of forgotten works that, however, were consid-
ered worth retranslating during the early years. A detailed study of some early
retranslations also widened our perspective on the different contextual reasons
why retranslations are made: there may be far more mundane reasons behind re-
translation than “dated” translations or the glory attached to retranslation. For
example, prior to Finland’s signing the Bern agreement in 1928, authorial rights
were not always respected and many translations which were made without the
author’s or his/her publisher’s consent were “collisions”: two or even three ver-
sions of the same book may have appeared more or less simultaneously, due to
a lack of coordination between Finnish translators and publishers.
It also became obvious from our data that classics lists generate retransla-
tion: they keep up the image of a classic. A classic is normally a book that has
been retranslated often, but it also appears to work conversely: a retranslation
becomes a classic more easily than a one-off translation. This is an expectation
that can be put to work to boost the sales of an older book.
In addition to our lists and statistics, we have also looked more closely at a
number of translated classics: Les Misérables by Victor Hugo, Dead Souls by
Nikolai Gogol, The Saga of Gösta Berling by the Swedish author Selma Lager-
löf (the 1909 Nobel prize winner), Pepita Jiménez by Juan Valera, and Three
Men in a Boat by Jerome K. Jerome. To compare the results with revisions of
translated work, various editions of Astrid Lindgren’s Pippi Långstrump, Wal-
ter Scott’s Ivanhoe and L. M. Montgomery’s Anne of Green Gables have been
examined. It is these case studies that bring out some of the most interesting
findings of our research: first, the huge variety of different textual and editorial

practices involved in what is called retranslation; and second, the difficulty of

marking any clear boundaries between retranslations and edited versions (con-
sequently, the whole class of “retranslations” is called into question). It is these
two issues we will turn to now.
4. DEFINITIONS AND CATEGORIES: THE PROBLEM

OF IDENTIFYING AND CLASSIFYING RETRANSLATIONS
The question of retranslation acquires another dimension when we start to pick
out individual translations for closer study. Identifying retranslations itself has,
in the first place, been a complicated process: we all know a fair number of re-
translations and can safely guess which classics are most likely to have been re-
translated, so there is always material to start with. But a systematic study of the
numbers of retranslations and the compilation of a bibliography of retranslated
works is more time-consuming. Retranslations cannot be picked out from bib-
liographical databases the way authors, translators or source languages can, as
there is no search word or bibliographical field for the crucial piece of informa-
tion that a translation is in fact a retranslation. To find a retranslation, one needs
to compare bibliographical entries and look for one source text with at least two
target texts with different translators.5 The work thus needs to be done manually
and needs to be cut down to sectors, say, on certain authors, classics, or source
languages. An all-inclusive list for any one target language is nearly impossible
(exceptions are of course those source languages from which very few transla-
tions are made into the target language in question); even a near approximation
is almost necessarily the result of the combined efforts of several scholars work-
ing together, and it also relies on scholarly literature on specific areas of interest
(on the translation of certain source language classics, for example).6 This is one
of the reasons we decided to base our work partly on lists compiled by other
people: it would widen our perspective, and we would not rely only on our own
intuition and knowledge. Our work was also facilitated by the fact that along-
side our work with retranslations, one of us (OP) was a member of the editorial
team for the history of Finnish translation, a 1,300 page volume that came out in
2007 and that provided a wealth of scattered information on retranslations from
several languages, among others (Riikonen et al. 2007), thus helping us form
the overall picture.
In Finland, literature has been translated for a little less than two hundred
years. The first bibliographies were drafted towards the mid-19th century, and
the predecessor of today’s on-line database, Valfrid Vasenius’ bibliography,
dates from 1878. Bibliographical information today differs from what was con-
sidered important at some other points in history, so that, for example, transla-
tors and writers were often not indicated in the printed books themselves, mak-

ing it more difficult to classify books. In addition, a number of books are miss-
ing from bibliographical records. In other words, bibliographical standards in
the early decades were not the same as today’s standards, which makes it more
difficult to distinguish between different versions and identify what a retransla-
tion is, and what a reprint is.
When the first problem of compiling a list of retranslations has been over-
come and a body of translations has been identified that, in the bibliographical
references, turns out to be translations of the same source text, it becomes pos-
sible to carry out comparative analyses on individual texts, first translations and
retranslations. It was at this stage of our research that we became aware of the
fact that the actual categorizing of translations into first and subsequent transla-
tions, which has formed the basis for almost all theorizing about retranslations,
is ultimately misleading – unless we accept the claim that retranslation can be
anything, from a slight editing of a previous translation to a completely different
text. What we needed to ask now was: even if two separate translators were
mentioned in bibliographies as having translated the same source text, was it
really a question of two different translations? The actual textual practices and
their study were the key to the next phase of our quest.
5. HYBRID VERSIONS: VICTOR HUGO’S LES MISÉRABLES

IN FINNISH
Les Misérables from 1862 is probably Victor Hugo’s best-known work. It is the
story of an ex-convict, Jean Valjean, whose life is a constant struggle to find
and defend his place under a ruthless, enslaving system. The novel was a cry
against social injustice and misery and has been read as such by many an op-
pressed prisoner. The novel is divided into five parts, which in turn are divided
into books and further into chapters. The epic story has been immensely popu-
lar: it was translated into several languages during the year of its original publi-
cation; later, numerous versions including film adaptations, animations and mu-
sicals have been made all over the world, including Mexico, Turkey and Japan.
The musical was the longest-running show in the history of Broadway; in film
and television, actors such as Gérard Depardieu and Liam Neeson have played
Valjean. The book has been important for soldiers and prisoners, and intertex-
tual links referring to the characters of the story still appear in modern popular
culture.
For all practical purposes, in comparing the Finnish translations of the
work they are referred to here by the name of the translator for the sake of clar-
ity, even if the point of comparison is not to study individual translatorial
strategies or styles. It is our aim to compare the texts to see how they have been
categorized and whether this categorization reflects the textual profiles of the

work and also to see whether these textual profiles lend support to Retranslation
Hypothesis.
The book has appeared in a number of versions and under different transla-
tors’ names throughout its Finnish history. The first time Jean Valjean set foot
in Finland was in a Swedish translation, printed in the town of Tampere in
1895, to be followed a year later by a Finnish translation published in the same
town. The textual make-up of these two texts was syntactically and even mor-
phologically very similar, but instead of the Finnish version being based on the
Swedish text printed in Tampere, both texts can be shown to have had an earlier
precedent in Sweden; the second Swedish translation of the work from the year
1888–1889 (this “genealogy” is evident from the translator’s footnotes, which
in the Finnish version are mostly direct translations of the Swedish 1888–1889
version but which have been omitted in the later Swedish version printed in
Tampere). Both translations printed in Finland appeared in parts (periodical
leaflets) of about 50 pages each as was the custom in those days, for if the first
leaflets did not sell well or there was some other mishap, the publication could
easily be called off. As there are no documents left of the publication process, it
is not known what happened, but both the Finnish and the Swedish text printed
in Tampere were ceased abruptly. Of the first Finnish translation, only 442
pages appeared.
A decade later the whole work was commissioned from the young aspiring
translator (later professor of Romance literatures) Vihtori Lehtonen. The pub-
lishing company (Kansa) was also fresh and ambitious, with a prestigious advi-
sory board made up of young literati, the to-be elite of Finland’s literary life.
However, Kansa was not successful with the marketing efforts for any of its
books, and the company went bankrupt soon after the first and second parts of
Les Misérables (Kurjat in Finnish) had appeared in 1908–09.
The third effort at publishing the work in Finnish was made in the 1920s
when WSOY, the largest publishing house in Finland, had acquired the rights to
all of Kansa’s translations. Les Misérables was now designed as part of a new
series of classics called “Valiokirjasto” (“select library”), the title of the series
indicating the serious and select nature of the books to be published. The first of
the five parts appeared in 1927 in Vihtori Lehtonen’s earlier translation, not re-
printed as such, but with extensive language corrections (corrections were ac-
knowledged in the book). This was not uncommon in the 1920s, with the writ-
ten language standardization process still going strong in Finland: several older
translations were “corrected” linguistically to suit the evolving standards. After
the first part, the idea of editing was dropped and the remaining four parts were
translated by one of the most prolific translators of French literature at the time,
Eino Voionmaa. The editor of the first part may have been the translator, now
better known as J. V. Lehtonen, who had meanwhile become a professor of

Romance literature, but it would seem more likely that the task of editing was
given to Voionmaa, who then continued with translating the second part.
Thus, the first part of the five-part novel had been translated twice (1896
and 1908) and reprinted in an edited version once (1927); the second book had
also been translated twice, but in a different cycle than the first book (1909 and
1928), and the last three books had only been translated once (1929–1931). It is
here that the neat (theoretical) division into first and retranslations becomes dif-
ficult to maintain. The discussions on retranslations thus far have not taken into
account cases where parts of the text have been retranslated, perhaps more than
once, whilst other parts have only been translated once, and some parts have
been edited, reprinted or abridged. Les Misérables in Finnish seems a hybrid
text par excellence. But to see whether there is evidence for the Retranslation
Hypothesis here, we will now turn to the textual profiles of the translations.
Every translator has his own distinguishable style: the first one, J. J.
[Aulén], followed the Swedish translation very closely, and for all practical
purposes it served as his original, as he simply copied the word order and in-
serted foreign loan words where there were no Finnish equivalents. This was
very much in the style of Aulén, whose translations had been criticized for be-
ing too slavishly bound to their source texts. Incidentally, the Swedish transla-
tion was syntactically very close to the original French work. Both Lehtonen
and Voionmaa translated more freely and fluently. Voionmaa shows no traces
of copying Lehtonen: his syntax and lexicon differ considerably from those of
Lehtonen but he does not seem to attempt a closer translation to the original
than what Lehtonen had done. At times, one of them is closer as to the syntax
or, for example, the rendering of units of measurement, but in other places the
roles are reversed. In comparing the two translations, we have only checked
small sections: these random checks reveal that differences exist, but that there
is no discernible pattern of increased closeness in the second one. Since there is
no data on why the decision to retranslate the second part was taken, we can
only suggest some potential factors behind the decision. It may be reasonable to
assume that revision may be problematic because of the translators’ personal
choices and styles, and the revising translator may have felt frustrated working
with an earlier translation (such complaints can be found in translators’ corre-
spondence). The Voionmaa retranslation of the second part of the book may
thus have been borne out of a desire to use his own language and own words,
not merely to polish what Lehtonen had done previously, and the rationale be-
hind the retranslation may thus not have been increased accuracy. It stands to
reason that translators may actually feel much more constrained working with
an earlier translation than when translating anew, and this fact – their personal
involvement – might well have contributed to at least some cases of retransla-
tion.

In the 1940s, a reprint was planned, and the first part appeared in 1941 (un-
changed from the 1927 version). The remaining four parts, however, were never
republished – probably due to the war effort, which curbed the purchase of pa-
per, among other things. After the war, plans went ahead again to publish Les
Misérables, but what came out in 1945–1947 were not the four remaining parts
but an abridged version of the whole work in two volumes. The abridging had
been done by Reino Rauanheimo at WSOY, which was acknowledged in the
new edition: his name, together with the two translators’ names, Lehtonen and
Voionmaa, appeared in the book.7 Roughly half of the whole work was cut out.
Basically, there are two ways of abridging: either whole chunks are omitted and
the remaining bits are left as they are, or the work is paraphrased. Such para-
phrasing/abridging takes place, for example, with many children’s classics or
the Readers’ Digest abridged versions of novels. For example, there exist at
least two paraphrased abridged versions of the work in Swedish, by Hugo Gyl-
lander in 1906–07 and by Gemma Funtek in 1947. In the Finnish 1945–47 ver-
sion the abridging was done by cutting out whole chunks of text (sentences,
paragraphs, chapters) without paraphrasing the rest. This was made possible by
the structure of the work itself: full of side-stepping comments, anecdotes and
separate histories, Les Misérables lends itself easily to abridging without overtly
appearing to be bowdlerized and even without the need to paraphrase in order to
maintain the coherence.
This shortened version is the only one published during the last 70 years;
all the reprints are from this version (the latest edition is from 1999; it has sold
out in the bookshops). There is one copy of the full version in the storeroom of
the city library of Helsinki but all shelf copies are of the shortened version. The
latest reprint is unchanged from the 1940s version except for the fact that the
translators’ and the editor’s names have been left out – they do not appear any-
where any more, effectively disguising the complex translation history of the
book.
This history shows that, in the first place, the trajectory of a classic is not
always in line with the Retranslation Hypothesis: with time, readers do not al-
ways get a more accurate or closer-to-the-original version. Rather, the timeline
shows a curve here: from the first, incomplete translation made via a third lan-
guage slowly towards a full-length version and then to a shortened version. The
only full-length volume of Les Misérables in Finland is the one from the 1920s
and is thus eighty years old (obviously, this fact alone does not say anything
about the quality of the translation, nor about its future fate). Completeness
(versus abridging) is one facet of closeness to the original; there are other as-
pects, which are more difficult to gauge. Syntactically the first translation, al-
though made through a mediating version, is the closest and retains the structure
of the text. This closeness, however, entails the fact that it is cumbersome to
read and at points unintelligible, whereas the latter translations avoid problems

of incomprehensibility and therefore seem to carry over the meaning better.

Neither accuracy nor completeness follow the schema assumed by the Retrans-
lation Theory here. The reasons behind each translation and revision have var-
ied and include variable factors such as the war effort, international popularity,
publisher’s profiling and the pocket book market. Another lesson learned from
the Finnish Les Misérables is that categorization and labeling may be mislead-
ing: extensive archival work is needed to piece together the history of a transla-
tion.
6. REVISION & RETRANSLATION:

DEAD SOULS, DISAPPEARED TRANSLATORS
In the case of the translation of Les Misérables into Finnish, there has been no
linear progress towards a closer rendering. Rather, the textual history of the
book in Finnish has been a meandering road. With Gogol’s Mertvyje Dusi
(Dead Souls), the situation seems different from the start and is more in line
with what is normally expected of retranslations: three translations with even in-
tervals (1882, 1939, 1970), identifiable by the translator’s names, and with no
discernible differences in the size/completeness of the books. A closer textual
study reveals some interesting facts, however.
The three translations are given in the bibliographies as follows: 1882 (by
Samuli S.), 1939 (by J. K., later reprints in 1945, 1951, 1962, 1968, 1977, 1984
and 1992 with the translator names Jalo Kalima and Juhani Konkka) and 1970
(by Juhani Konkka). The Finnish title is the same throughout, Kuolleet sielut,
Dead Souls. A textual analysis shows that there is extensive usage in the latter
two translations of the first translator’s singular style and expressions. Idiosyn-
cratic lexicon (words such as lallukka, tallukka) remained constant; alliterated
expressions and tender diminutives (pyöreä pönttöseni, pallea papuseni, pullea
pupuseni used to tenderly describe a person) and sayings (viljainen vissi) were
used in all three translations. And it was not just isolated expressions and words
but whole passages and chapters which were kept the same or nearly the same.
The differences between the first two translations are minor; the third transla-
tion is more independent but keeps nevertheless many of the inventive creations
of the first translation.
The first translation by Samuli S. obviously served as the basis on which
several grammatical changes were made in 1939, but the wording in much of
the work is exactly the same as in the 1882 edition. The alias Samuli S. stands
for Samuli Suomalainen, turn-of-the-century journalist, writer, translator and
teacher, well known for his mastery of the Finnish language at a time when the
literary language was not yet fully exploited and when many translations were
word-for-word renderings of their originals (or of Swedish translations of the

original, as in the case of Les Misérables). An especially distinctive feature of

Samuli S.’s style was the many descriptive words he picked up from popular
usage and used in his translations (and also in his own writing). The Finnish
language is rich in onomatopoetic and descriptive usages, in alliteration and in
multiple dialectal expressions, and these were the cornerstone of Samuli S.’s
style, rendering it witty and humorous. Samuli S. is also considered to have ex-
erted a considerable influence on one of Finland’s foremost (and first) novelists,
Juhani Aho, who was familiar with Suomalainen’s translations (Sarajas
1968:60; Kalima 1939:82–83).
This first translation of Dead Souls was awarded the Finnish Literature So-
ciety’s prize in 1882; it was widely praised in the newspapers and it was still
remembered and revered at the time when the 1939 translation appeared, more
than fifty years later, awakening reminiscences of the old translation in the
minds of the reviewers of the new one. Suomalainen’s Dead Souls translation
was first republished in 1977, almost a hundred years after its first appearance,
by the publishing house Otava and was praised as recently as 2003 by professor
of Russian literature Pekka Pesonen. In 2008 it was republished again by the
BTJ publishing house in their classics series.
There is a mystery behind the 1939 edition regarding the identity of the
persons in charge of the translation, the solving of which may help in under-
standing the evolution of the different versions. The paratextual information in
the 1939 book gives the translator’s initials, J. K. The identity of this J. K. was
not general knowledge (newspaper reviewers were puzzled by it). Later reprints,
however, were under the names of Jalo Kalima and Juhani Konkka, which are
both abbreviable as J. K. The question is, who did what? A partial answer to
this question can be found in the archives of the Finnish Literature Society,
where Juhani Konkka’s personal archives are stored: his correspondence, trans-
lation contracts, etc. In a letter to his publisher Otava (20.2.1969), where
Konkka consents to do a new translation of Dead Souls (the one that appeared
the following year, 1970), Konkka also mentions that he had “corrected Ka-
lima’s old translation more than thirty years ago”. Both Kalima and Konkka had
worked for the WSOY publishing house in the 1930s, and Kalima had previous
experience of editing old translations. Kalima may have first been given
Suomalainen’s translation to edit, and Konkka may have been asked to go
through this version for some reason. This is intelligent guessing since nowhere
is there any information about why such double revision might have been
needed. Kalima was a renowned Slavist, he enjoyed general esteem and had
done both translating and revising earlier; Konkka, on the other hand, was fairly
young and new on the market at this time. But if this indeed was the case,
Konkka may not have been aware that the version he was working with may not
have been Kalima’s translation but Suomalainen’s edited text (although he must
undoubtedly have been aware of the first translation). It may thus possibly be

just a strange mishap or coincidence and not intentional that Suomalainen’s

name does not appear in the 1939 version of the work in Finnish.
The assumption that the original idea behind the 1939 edition was rather to
produce a revised version of the old translation than to publish a whole new
translation thus gains support from three things: first, the person chosen to be in
charge of the new version was Jalo Kalima, whose previous work for the
WSOY publishing house included correcting the language of translations pub-
lished earlier (The Christian and The Eternal City, by Hall Caine) and who thus
was a trusted and experienced person for such work; second, the amount of un-
changed text in the 1939 version is very large, and thirdly, the nature of the
changes calls to mind a retouched older translation rather than a new translation.
These changes, for the most part, consisted of such orthographic and morpho-
logical corrections that had been made during the standardization process of the
Finnish language after the publication of the first translation (and which were
not current at the time Samuli S. was writing his translation). These included,
among others, the orthography of compound nouns, syntactic changes favouring
non-finite clauses instead of subordinate clauses, or inversion.
The third version of the book, Juhani Konkka’s translation of 1970, is syn-
tactically clearly a new translation, but even this version keeps much of Samuli
Suomalainen’s descriptive vocabulary and expressions (lallukka, vonkero, etc.),
and several sentences and paragraphs in the book are either exactly the same or
closely reminiscent of Suomalainen’s version.
Thus, despite the fact that bibliographies claim there are three translations
of the work into Finnish, there is a strong dependency between these transla-
tions. It can be argued that the 1939 translation is in fact a revised version of the
earlier, 1882 translation; even the latest translation owes much to the first one.
Pekka Kujamäki (1998:285) has (drawing on the work done in Göttingen) dis-
cussed the concept of the translators’s source text, not from the point of view of
the original, source-language text but of all the potential mediating texts the
translator may have had at his/her disposal. His data of the German translations
of the Finnish author Aleksis Kivi’s The Seven Brothers lend support to the un-
derstanding that there is a complex interrelationship and dependency between
the different texts at hand, similarly as in our analysis of Dead Souls, captured
by the German expression “Starkes Abhängigkeit” (see the Glossary in Frank,
ed., 1989).
7. THE MANY FACES OF REVISING

From the cases discussed in previous chapters it has become evident that texts
placed in the category of ‘retranslation’ can have a varied and hybridized trans-
lation history. What has also become evident during our research project is that
‘revision’ is an equally − if not more − confused category. In contrast to retrans-

lations, revisions have, however, attracted few researchers, and there is not a
wealth of case studies from various linguistic and cultural areas and eras to
build on. In this chapter we will briefly sketch some of our own findings of dif-
ferent revised translations. Vanderschelden (2000:1–2) calls revision “often the
first step towards retranslation”, involving “making changes to an existing TT
whilst retaining the major part, including the overall structure and tone of the
former version”. According to her, revision may be resorted to if the existing
translation contains “a limited number of problems or errors”, but the alterations
may be anything from “simple copy-editing” to extensive rewriting. Distin-
guishing between retranslation and revision thus becomes very difficult. The
problem, of course, is this: how much change can there be in the revision proc-
ess for the translation still to be the same, i.e. under the name of the previous
translator, and where is the line to be drawn to a new translation? And what
about the different kinds of revising? Do “orthographic” corrections go under
the process of revising, while “stylistic” corrections would merit the title of re-
translation? To complicate things further, several Finnish words denote ‘revi-
sion’, all being used to describe revised translations: tarkistaa, tarkastaa, kor-
jata, uudistaa – to check, to revise, to correct, to modernize/update. Both their
usage and the practice they refer to are not straightforward or consistent.
A minimalist revision might only entail few orthographic improvements; at
the other end of the continuum the text is entirely reworked so that it blurs the
dividing line between revision and retranslation. As readers, we might assume
that a ‘checked’ or ‘corrected’ version would also be somehow closer to the
source text, with potential mistranslations and omissions corrected. However, it
seems that revisions can be either passive or active in their relation to the source
text (cf. Pym 1998:82). In some cases revisions can be done without any com-
parison to the source text; in other cases a changed source text may actively call
for a revised translation to accommodate the changes. One case where the
source text was given an entirely passive role is the (‘modernized’ and
‘checked’) revision of the Finnish translation of L. M. Montgomery’s Anne of
Avonlea (1909; first trans. Hilja Vesala, 1921) by an anonymous reviser in 1961
(see Karonen 2007). The first translation had been abridged (not mentioned in
the book), and had not been translated from the English original but was based
on a Swedish translation8. One might thus assume that the revised version
would restore the omitted sections and check how the old translation corre-
sponds to the original text. Surprisingly, this is not the case: in fact, the revised
version introduces some new omissions. According to Karonen the source text
for the revised version was the first translation, and the source text of the first
translation was the Swedish translation, and thus the (only) Finnish versions do
not have any direct link to Montgomery’s original text at all (ibid. 68).
An illustrative case of varying reviser styles can be found in looking at the
two revised versions of Astrid Lindgren’s Pippi Långstrump. It was first trans-

lated into Finnish in 1946 (trans. Laila Järvinen), immediately after its publica-
tion in Sweden. Before a recent retranslation (2007 by Kristiina Rikman), it was
revised twice (1970 by Inka Makkonen and 2005 by Päivö Taubert). A detailed
comparison of both revised versions reveals that the two revisers have ap-
proached their task in totally different ways: Makkonen makes extensive and
liberal revisions, Taubert is extremely conservative. Within the analysed section
of two chapters Makkonen made close to 350 major and smaller revisions,
whereas Taubert only made 21, most of them related to a recurrent need to
avoid the Finnish word ‘neekeri’ (‘negro’) that was in 2005 considered inappro-
priate usage (norms like these are a major explanatory force in retranslations
and revisions alike, but space does not allow us to elaborate on this here). Most
of Makkonen’s revisions are best described as either personal preferences or
stylistic changes to bring the text into line with the aesthetic values of the time.
The characteristics of the translation were so extensively reworked by
Makkonen that it has been argued that the original translator’s voice was lost in
the process (Taubert in Heikkinen 2006:6). Comparing Makkonen’s revision to
the “retranslation” of Dead Souls makes one wonder about the status of differ-
ent genres and different revisers: why is an entirely reprocessed children’s clas-
sic by a female copy editor still a “revision” and why does a revised version of a
world classic by a male copy editor end up being classified as a “retranslation”?
The translation history of Dead Souls already indicated that the categories
may also be historically unstable: “assumed retranslations” (cf. Toury 1995)
may turn out to be revisions and vice versa. The translation history of Selma
Lagerlöf’s Gösta Berling into Finnish provides evidence of another type of revi-
sion than the grammatical and orthographic changes evident in the second ver-
sion of Dead Souls, but it also sheds light on the problem of identifying retrans-
lations. Here again, we seem to have two different translations of the same
source text: according to the bibliographies, there was a translation of this work
in 1902 by Auk. Andberg, a revised version in 1912, and a translation by Joel
Lehtonen as late as 1952. A study of the different versions proves that the first
translation, made in 1902 and reprinted in 1912 (revised by the writer Joel
Lehtonen), is the same translation that later appeared under Joel Lehtonen’s
name only. Since the 1912 edition clearly states on the title page that it is
Andberg’s translation, corrected by Lehtonen, it may again be pure coincidence
or an accident that the original translator’s name disappeared from the book’s
later editions, giving readers (and scholars) the idea that Lehtonen had made a
completely new translation. Lehtonen’s reviser style seems similar to
Makkonen’s: Lehtonen changed the style of the Andberg version, creating a
more lively and descriptive version of the book in the process. This he did by
using words that were either less common or more literary than the ones used
by Andberg, by reducing repetition, and by making the syntax more fluent.
These changes were, however, written down as corrections at the time, and the

text was not given the status of a new translation. Later this edited version was
then marked as a new translation.
The Finnish data we compiled revealed various ways of revising earlier
translations, and the revisions have not been uniformly signaled paratextually (if
at all). Case studies reveal that ‘to revise’ often stands for orthographic mod-
ernization and not, for example, comparison with the original with the view of
correcting mistakes or minor errors (even if the verb korjata, “to correct”, is
used), but sometimes it may also encompass comparison. In other cases a work
signaled as retranslation might be better called a revised translation, when it
comes to retaining the style of the previous translation (as in Vanderschelden).
However, revising cannot be seen as “a first step” towards retranslation, as most
revised works have not been retranslated, and retranslation does not often pre-
suppose revising, on the contrary.
8. CONCLUSIONS
Our study as a whole supports Siobhan Brownlie’s findings. She criticizes both
the norms and ideologies approach in the study of retranslation and the assump-
tion that time is the single influential factor in the make-up of retranslations;
both of these critiques are leveled against the assumed monocausality in retrans-
lations. She refers to multiple relations of many kinds and a “rhizomatic” rela-
tion between different factors (Brownlie 2006:155). We fully agree with her
that it is the local context that is often conclusive in the final make-up of the re-
translation and that it is the individual commissioners and actors, i.e. translators
and other agents, who should be given more emphasis in the study of retransla-
tion. Our results also point in the direction of multiple causation. What lies be-
hind a phenomenon as complicated as retranslation necessarily seems to be
caused by a multiplicity of different factors in different combinations: retransla-
tion cannot be encapsulated by a simplistic cause-and-effect formula. Case stud-
ies of existing retranslations and revisions also raise a number of ethical consid-
erations. On the one hand, reprocessed texts that rely heavily on a previous
translator’s work bring to the fore issues of plagiarism and copyright (in par-
ticular, paternity and integrity) − or even “trans-piracy”, to use a term coined by
Ljuba Tarvi (2005:137). On the other hand, publishers may see it as their moral
duty to take care of the “maintenance” of aging translations (see Heikkinen
2006; also Sillman 1996:44).
On a metatheoretical level a central finding of our study is that we cannot
blindly trust the existing categorizations in the study of retranslations. First, bib-
liographies do not give us the basic facts to start with. Two texts may have been
printed at different times, probably by different publishing houses, and there
appear different translators’ names, but are they indeed two different texts, and

if so what way? The scale of difference between two translations may vary: an
edited or corrected earlier translation is sometimes passed on as a new transla-
tion; a completely renewed and changed text may still appear under the earlier
translator’s name. You can decide to draw the line between retranslations and
corrected/edited second versions following the previously established metatex-
tual (bibliographical) practices, taking as a retranslation only the ones that the
bibliographies list under different translators, and in this way you can do statis-
tics and say that there were so and so many retranslations of certain texts. How-
ever, you need to be aware that you may be leaving out the real distinctions (or
non-distinctions) – the label of retranslation (likewise that of revising) covers a
great many different kinds of reworkings.
A purported revision may consist of minor linguistic amendments to keep
up with the standardization process of a language without changing the idiosyn-
cratic expressions of the first translator, but this kind of a process may also end
up being called a new translation, as in the case of Dead Souls. Then again, a
reviser may end up enhancing the style of the whole text, as in the case of Gösta
Berling. If we simply look at what has happened to the texts and forget what
they have purportedly been through – revision, modernization or retranslation –
we end up with a multi-layered schema where a previous translation may have
been used by later translators in very different ways: orthographic correction,
stylistic correction, correction against the source text, or, at the other end of the
scale, a completely new translation, not indicating any closeness to the previous
translation (the different translations of Les Misérables into Finnish are all fairly
idiosyncratic and do not seem to depend on each other).
Researchers often want neat categorizations, but our extensive data show
these categorizations do not arise out of reality. Binary categorization into first
and retranslations is not always helpful; neither is the categorization into revi-
sions and retranslations. It is more a question of a continuum where different
versions seamlessly slide together or even coalesce. Even the idea of a contin-
uum might be too simplified, as revision and changes may be brought about at
various levels of the text. To borrow Brownlie’s term, the textual relations be-
tween different versions, whether they are called retranslations or revisions,
seem to form a “rhizomatic” network of influences, ideologies and value judg-
ments. For a researcher they offer a rich and varied field of study, but as this is
still a largely uncharted terrain, we are well advised to walk lightly and to avoid
jumping to hasty conclusions.
Notes
1
Some scholars include simultaneous or near-simultaneous translations in the category of
retranslations (Susam-Sarajeva 2006:138). Anthony Pym (1998:82) calls simultaneous transla-
tions for different markets ‘passive retranslations’, while ‘active retranslations’ are those compet-
ing for the same audiences.

2
Brownlie (2006:145, n. 1) calls theoretical discussions and observations concerning the
phenomenon of retranslation by the name of Retranslation Theory. Under this label she discusses
Antoine Berman’s (1990) “theory of retranslation” (Brownlie 2006:147) and several articles,
some of them initially geared towards studying the phenomenon of retranslation itself, others to-
wards studying other topics (such as the changing of translation norms).
3
Brownlie (2006:148) observes that the term ‘hypothesis’ was not there in Berman’s writ-
ings and is actually part of a different theoretical framework than what Berman would have em-
braced. However, Berman’s ideas can be seen to figure behind the Retranslation Hypothesis in the
way hypotheses are understood in Chesterman (2000:22–23).
4
In other sources, this distance in time is often claimed to be 50 years, sometimes 30 or 20
years. See e g. Helin 2005:145 and authors cited in Collombat 2004:4.
5
There are cases where one translator has translated the same text twice with an interval of
time (e.g. some of the Seven Brothers translations into German, see Kujamäki 1998).
6
Obviously, bibliographies may be differently organized and in some national databases
there may well be a relevant field marking some translations as retranslations (this is not done in
Finland), which would greatly facilitate the study of retranslations. However, unless one is pre-
pared to execute a textual comparison of each version, one needs to accept a certain margin of er-
ror in these labels – and the categories are bound to be shifting, as can be seen later.
7
As can be seen, language correction and editing are sometimes attributed to a specific per-
son in the paratexts, sometimes not.
8
Indirect or mediated translations (that are a common phenomenon in translation history)
are also sometimes called ‘retranslations’. As mediated translations merit independent research
(which is currently being carried out, see e.g. Ringmar 2008), we have found it less confusing to
disregard them here.
References
Ballard, M. 2000. In Search of the Foreign: A Study of the Three English Translations of Camus’s
L’Étranger. In: Salama-Carr, M. (ed.), On Translating French Literature and Film II, Am-
sterdam/Atlanta: Rodopi, 19–38.
Bassnett, S. 2000. Adventures Across Time: Translational Transformations. In: Salama-Carr, M.
(ed.) On Translating French Literature and Film II, Amsterdam/Atlanta: Rodopi, 155–170.
Berman, A. 1990. La Retraduction comme espace de traduction. Palimpsestes Vol. 13. No. 4,
1–7.
Berman, A. 1995. Pour une critique des traductions: John Donne. Paris: Gallimard.
Brownlie, S. 2006. Narrative Theory and Retranslation Theory. Across Languages and Cultures
Vol. 7. No. 2. 145–170.
Chesterman, A. 2000. A Causal Model for Translation Studies. In: Olohan, M. (ed.) Intercultural
Faultlines. Manchester: St. Jerome, 15–27.
Collombat, I. 2004. Le XXIe siècle: l’âge de la retraduction. Translation Studies in the New Mil-
lennium Vol. 2. 1–15.
Du Nour, M. 1995. Retranslation of Children’s Books as Evidence of Changes in Norms. Target
Vol. 7. No. 2. 327–346.
Frank, A-P. (Hrsg.) 1989. Die literarische Übersetzung. Der lange Schatten kurzer Geschichten.
Amerikanische Kurzprosa in deutschen Übersetzungen. Erich Schmidt, Berlin.
Gambier, Y. 1994. La Retraduction, retour et détour. Meta Vol. 39. No. 3. 413–417.
Heikkinen, H. 2006. Palmikkotukkaisesta neekeriprinsessasta saparopäiseksi alkuasukaskunin-
kaan tyttäreksi. Ajan vaikutus Peppi pitkätossu–suomennoksiin 1946−2005. Unpublished
proseminar thesis. University of Tampere.

Helin, I. 2005. Kääntäjä ja yhteiskunta – uudelleenkääntämisen yhteiskunnallinen viitekehys. In:

Yli-Jokipii, H. (ed.) Kielen matkassa multimediaan. Näkökulmia kääntämisen tutkimiseen ja
opiskelemiseen. Helsinki: Helsingin yliopiston käännöstieteen laitos.
Jianzhong, Xu. 2003. Retranslation: Necessary or Unnecessary. Babel Vol. 49. No. 3. 193–202.
Kalima, J. 1939. Juhani Ahon lastu ”Pienehkö, pyöreä, pehmeähkö”. Virittäjä Vols. 1–2. 82.
Karonen, T. 2007. Vihervaaran Anna ajan hampaissa. Unpublished pro gradu thesis, University of
Tampere.
Koskinen, K., Paloposki, O. 2005. Mapping Retranslation. Proceedings of the XVII World Con-
gress. FIT (Federation Internationale des Traducteurs.) Paris/Tampere: FIT. 194–195.
Koskinen, K., Paloposki, O. 2003. Retranslations in the Age of Digital Reproduction. Cadernos
Vol. 1. 19–38.
Kujamäki, P. 2001. Finnish Comet in German Skies: Translation, Retranslation and Norms. Tar-
get Vol. 13. No.1. 45–70.
Kujamäki, P. 1998. Deutsche Stimmen der Sieben Brüder. Frankfurt am Main: Peter Lang.
Lopes, A. 2006. Landscaping Emotion(s) – Translating Harriet Beecher-Stowe in Portugal. In:
Michaela Wolf (ed.) Übersetzen – Translating – Traduire. Towards a “Social Turn”? Band
I. Wien & Berlin: LIT Verlag. S. 199–208.
Paloposki, O., Koskinen, K. 2004. Thousand and One Translations. Retranslation Revisited. In:
Hansen, G., Malmkjaer, K., Gile, D. (eds) Claims, Changes and Challenges. Amster-
dam/Philadelphia: John Benjamins, 27–38.
Puurtinen, T. 1995. Linguistic Acceptability in Translated Children’s Literature. Joensuu: Joen-
suun yliopisto.
Pym, A. 1998. Method in Translation History. Manchester: St. Jerome.
Riikonen, H. K., Kujamäki, P., Kovala, U., Paloposki, O. (eds) 2007. Suomennoskirjallisuuden
historia I–II. Helsinki: SKS.
Ringmar, M. 2008. Von indirekten zu direkten Beziehungen im Finnisch-isländischen Litera-
turaustausch. Trans-kom Vol. 1. No. 2. 164–179.
Sarajas, A. 1968. Tunnuskuvia. Suomen ja Venäjän kirjallisen realismin kosketuskohtia. Porvoo–
Helsinki: WSOY.
Sillman, P. 1996. Uudelleenkääntäminen kustantajan, kääntäjän ja lukijan näkökulmasta. Unpub-
lished pro gradu thesis. University of Turku.
Susam-Sarajeva, Ş. 2003. Multiple-entry visa to Travelling Theory: Retranslations of Literary and
Cultural Theories. Target Vol. 15. No. 1–36.
Susam-Sarajeva, Ş. 2006. Theories on the Move. Translation’s Role in the Travels of Literary
Theories. Amsterdam & New York: Rodopi.
Tarvi, L. 2005. The Problems of Managing the ‘Translation Stock’: Freelancing vs. Freebooting.
In: Reintjes, M., Tåqvist, M. (eds) Ethics and Politics of Translation. Norwich Papers 13.
Norwich: University of East Anglia, 125–140.
Toury, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam: John Benjamins.
Tymoczko, M. 1999. Translation in a Postcolonial Context. Manchester: St. Jerome.
Vanderschelden, I. 2000. Why Retranslate the French Classics? The Impact of Retranslation on
Quality. In: Salama-Carr, M. (ed.) On Translating French Literature and Film II. Amster-
dam/Atlanta: Rodopi. 1–18.

DOI: 10.1556/Acr.11.2010.1.3
ON COMPOSITE INTERRELATIONSHIP
IN LITERARY TRANSLATION
CHUANMAO TIAN
School of Foreign Studies, Yangtze University,

Hubei, 434023 P. R. China
Intercultural Studies Group, Universitat Rovira i Virgili,
Tarragona, 43002 Spain
E-mail: tcm_316@ 163.com
Abstract: Complex relations, such as those between and within intertextuality, inter-
subjectivity and interculturality, are inherent in all kinds of translation, especially in literary
translation. These relations themselves interact with each other to form higher-level rela-
tions which may be labelled “composite interrelationship” (CI). Research on composite in-
terrelationship can help traditional Translation Studies break away from the research para-
digms of mono-centrism and dualism, thus allowing translation phenomena to be seen as a
multi-dimensional and multi-directional network of interrelationships. More importantly, it
can help us better understand fidelity as copying of inherent relations in the source text and
creation as conscious or unconscious establishment of new relations in the target text in lit-
erary translation from a new angle. Research on composite interrelationship may possibly
open up a new area of “inter-” studies for translation research.
Keywords: literary translation, composite interrelationship, implication, copying,

creation
1. INTRODUCTION
Inspired and influenced by Western theories on intersubjectivity and intertextu-
ality, research on “interrelationships” has become a major topic of discussion in
Chinese academic circles, especially in the field of cultural studies. The bounda-
ries of interrelationship have been expanding, resulting in the appearance of in-
terculturality, interliterariness, interlinguisticality, and so forth. The theory of
intersubjectivity was formulated first by the German philosopher Edmund
Husserl as a reaction to the philosophical framework of subject-object binary
opposition, then developed by Heidegger, Sartre, Merleau-Ponty, and others,
with a focus on equal dialogue and communication between subjects. The con-

52 CHUANMAO TIAN
cept of intertextuality was first expressed in the work of the Russian philoso-
pher and literary scholar Mikhail Bakhtin. In articulating Bakhtin's work on dia-
logism, Julia Kristeva (1980:66) coined the term “intertextuality” to describe
the idea that “any text is constructed as a mosaic of quotations; any text is the
absorption and transformation of another”. In other words, various linkages ex-
ist between texts, or between different discourse events. Relations exist not only
between subjects or texts but also between a subject and a text. Bakhtin holds
that as a kind of semantic relation between parallel texts or statements, intertex-
tuality has the whole discourse as its component, behind which is the real or po-
tential speech subject, namely the speaker of the uttered speech (quoted in Liu
2005:65). Therefore, the agent of the speech act serves as the bridge between in-
tertextuality and intersubjectivity. That is to say, the horizontal movement of in-
tertextuality towards context forms intercontextuality (Liu 2005:65).
In both Chinese and Western translation research, a few scholars have dis-
cussed interrelationships in translation. Western authors give more attention to
causality (Chesterman 2002; Brownlie 2003) and power relations (Fawcett
1995; Zauberga 2000; Susam-Sarajeva 2001) in translation, while Chinese writ-
ers seem to be more interested in intersubjectivity (Xu 2003a/2003b; Chen
2004/2005). However, all these studies generally restrict themselves to just one
kind of interrelationship and pay no attention to the relation between all kinds of
interrelationships. In the following paragraphs, we will borrow one concept
from literary studies to explore the interrelationships underlying translation so
as to seek new perspectives and methods for Translation Studies.
2. “COMPOSITE INTERRELATIONSHIP” (CI)

AS A UNIVERSAL IN TRANSLATION
The term “composite interrelationship” (my translation of the original
“复合间性”) is a relatively new concept formulated by Liu Yuedi (2005:64), a
pioneer in Chinese literary studies, in his exploration of the relation between in-
tersubjectivity and intertextuality. He claims that “literary ‘composite interrela-
tionship’ means an ‘interrelationship’ between ‘intertextuality’ and ‘intersubjec-
tivity’. It has the feature of ‘intertextuality’ and ‘intersubjectivity’ but it is an
‘interrelationship structure’ higher than either of them.” Because translation in-
volves more subjects (i.e. translator and target-language reader) and more texts
(i.e. target text and other texts of the target culture) than literature, composite
interrelationship is inherent in translation. In terms of subject, author vs. transla-
tor, translator vs. target-language reader and author vs. target-language reader
constitute a multi-dimensional structure of intersubjectivity. In fact, transla-
tional intersubjectivity is far more than a three-dimensional structure because
there are other subjects involved in translation. Such subjects may be source-

ON COMPOSITE INTERRELATIONSHIP IN LITERARY TRANSLATION 53
language reader and critic, translation project initiator, sponsor, publisher, tar-
get-language critic, and so forth. In terms of text, there are various kinds of in-
teractions between source/target text and a potential constellation of previous
texts in addition to the objectively existing relation between the source and tar-
get texts. Furthermore, as translation concerns at least two language-cultures,
the source/target language and culture develop a relation of dialogue and ex-
change through the intermediary of translation and translator. Within and be-
tween different dimensions, various interrelationships interact with each other
to form a highly complex network of interrelationships which may be labelled
“translational composite interrelationship” as illustrated in the following dia-
grams:
intersubjectivity intertextuality interculturality
author SL culture
translator ST TL culture
TT
previous texts other cultures

others
composite interrelationship
intersubjectivity
intertextuality
interculturality
From the diagrams we can see that translational composite interrelationship

is a kind of heterogeneous interrelationship consisting of homogeneous interre-
lationships including intersubjectivity, intertextuality, interculturality, and so
on. It is actually an “interrelationship within another interrelationship”, a con-
glomerate, or complex whole of interrelationships. There are more dimensions
and orientations in literary translation than in literary writing because the former
has more components, namely target text and target-language reader, which
give rise to more interrelationships, such as more intersubjectivities. Literary

54 CHUANMAO TIAN
creation concerns the author, previous authors and source-language readers

while literary translation relates to the author, previous authors, source-language
readers (mainly critics), the translator and target-language readers. In transla-
tion, the “previous authors” include more authors than those with whom the
source-text author interacted in the overall process of creative writing, namely
those authors to whom the translator relates but the author did not. Similarly,
translational intertextuality includes the interrelationship between the target text
and the new constellation of previous texts which cannot be found in literary in-
tertextuality. And translational interculturality contains the interrelationship be-
tween languages, aesthetic traditions, power relations, and so on, which may be
absent from literary writing proper. Some of these interrelationships are hidden
in translation practices and they have not yet been discerned and detected by re-
searchers; some of them have already been exposed, discussed and conceptual-
ized as various kinds of translation theories and techniques. All of them consti-
tute the main content of translation.
Previous translation research, of course, took into consideration and dis-
cussed, at least indirectly, interrelationships in translation. Western discourse
describes the translator as a servant with two masters. This is repeated in the
Chinese female translator, Yang Jiang’s famous metaphor “yi pu er zhu” (liter-
ally, a servant serves two masters at the same time). Liu Miqing (1999:125–
132), an original writer, discusses cross-reference of texts in his major work, An
Outline of Cultural Translation. All of these relate to intersubjectivity or inter-
textuality. However, these studies focused on just one interrelationship. They
did not speculate on the composite interrelationship in translation, such as the
relation between intertextuality and intersubjectivity. Moreover, they were
strongly influenced by the traditional philosophy of subject-object opposition.
Dualistic opposites such as form and content, source and target language, author
and translator, translator and receptor, source and target text, source and target
culture, were widely present in previous studies.
3. THE IMPLICATIONS OF CI FOR TRANSLATION STUDIES

A brief examination of Chinese and Western history of translation theory and
practice shows that the paradigms of research on translation, though shifting
from one focus to another, have not yet broken away from the model of a single
centre, such as the text-centred, author-centred, translator-centred, reader-
centred models, and so on. Take Chinese translation history, for example. The
monk-translator, Dao An (312–385 AD) of the Eastern Jin Dynasty, stressed
that “the translator should render the original text as it is without any omission
or addition” (see Luo 1984:26). As a result, many unidiomatic expressions ap-
peared in the Chinese translation of Buddhist Scriptures. This is a typical case

of character-for-character translation which represents text-centrism. St.

Jerome’s word-for-word Biblical translation was also executed after this model
(see Bassnett and Lefevere 2001:2). As for text-centrism and author-centrism,
there are both similarities and differences between them. In some sense, being
faithful to the source text means fidelity to the author. But for any text, there is
the matter of unity and contradiction between content and form. In cases where
there is unity of content and form, fidelity to both of them can be achieved.
Conversely, in cases where there is a contradiction between content and form,
fidelity to only one of them can be maintained. Therefore, the text-centred para-
digm seeks fidelity to the form of the source text and technically manifests itself
in literal translation while the author-centred paradigm pursues the author’s
thoughts, feelings, aesthetic orientation and values between the lines and tech-
nically externalizes as liberal translation. Bian Zhilin et al. (1959) gave a vivid
description of the author-centred translation model as follows:
The original author creates his text freely while we translators must render
his text faithfully, faithful to his free creation. If he beats around the bush
we must follow him closely. If he goes up high into the heavens or down
deep into the earth we have to chase him very closely. If he is far-sighted,
we should not be short-sighted. (Luo 1984:655, my translation)
In the late Qing Dynasty, Yan Fu and Lin Shu freely added plots and made
comments on the source text in their translations; this broke away from the
text/author-centred translation model and developed into a translator-centred
paradigm. This translation-practice paradigm found its theoretical echo and ex-
planation in China in the 1990s, with the dominant propositions being “transla-
tion is a kind of creative treason” (Xie 1992), “the translator is the translational
subject” (Chen 2004), among others. Thereafter, Chinese Translation Studies
turned its attention to the translator, trying to construct a translator-centred re-
search paradigm which was related to the rise of Western hermeneutics and de-
construction, especially to Foucault’s and Barthes’ claim that the author is dead.
The appearance of Western hermeneutics and reception aesthetics shifted the
focus of Translation Studies to the target-language reader, thus giving rise to a
reader-centred paradigm. In China, a few writer-translators, such as Lu Xun and
Mao Dun, focused their attention on the target-language reader at an earlier
stage. For example, Lu Xun classified readers as Groups A, B and C; Mao Dun
strongly opposed translationese and advocated the use of language the broad
masses would like to read and hear (see Luo 1984). In Taiwan, the famous poet-
translator, Yu Guangzhong, attached vital importance to his readers. For exam-
ple, he adopted the translation principle of “being pleasant to the reader’s eye,
to the audience’s ear and to the player’s mouth” when he translated Oscar
Wilde’s The Importance of Being Earnest, giving all his attention to the recep-

56 CHUANMAO TIAN
tor (see Jin & Huang 2001:3). Mu Lei is one of the earliest scholars in China
who applied the theory of reception aesthetics to translation research.
There are similarities and differences between the target-text-centred para-
digm and the reader-centred paradigm. Both paradigms are biased towards the
target language-culture. Fu Donghua, the translator of Gone with the Wind by
Margaret Mitchell, followed the guideline of “sparing some effort for the
reader” (see Luo 1984:442), and sinicized all the names of persons and places in
the source text, which clearly follows the reader-centred paradigm. However,
the reader-centred paradigm bases all the specific translation operations on the
reader’s expectations while the target-text-centred paradigm aims to create the
perfect target text. Xu Yuanchong’s three-beauty theory of poetry translation
represents the latter. According to him, the target text can surpass the source
text so as to produce a perfect translation which is characterized by beauty, first
in meaning/image, then in sound and lastly in form. He also claims that literary
translation, like creative writing, is a kind of competition between two lan-
guage-cultures, with the common goal of producing high-quality intellectual
food for readers.
On the other hand, the binary tendency has always existed in the history of
discourse on translation, especially on ways to translate. The categorization of
translation ways or attitudes as literalism/liberalism or faithfulness/freedom can
be traced back to Marcus Tullius Cicero’s dichotomy of translating as a literal
translator (ut interpres) and translating as an orator (ut orator), which has found
its modern versions in Eugene A. Nida’s classification of formal equivalence
and dynamic equivalence and Peter Newmark’s categories of semantic transla-
tion and communicative translation (Robinson 2006:9). The dualistic perspec-
tive was also adopted by the English author and translator John Dryden in his
observation of the English translation practice of his day. He observed two ex-
treme translation biases in the translations of his contemporaries, which he la-
belled “metaphrase”, a word-for-word literal rendering, and “imitation”, an ex-
cessively free rendering which departs radically from the original text and in-
cludes additions and re-interpretations, as in Cowley’s and Denham’s transla-
tions (see Dryden’s Preface to Ovid’s Epistles). He satirized those metaphrasts
by saying “That servile path thou nobly dost decline / Of tracing word by word
and line by line”, and wanted to make a compromise between the two extremes
to follow what he called “paraphrase” in his own translating. In today’s transla-
tion research, we can see that this kind of binary polarity dies hard and is still
embraced by many scholars, such as Lawrence Venuti, who re-introduced the
dichotomy under the name of foreignizing and domesticating translation strate-
gies (see Baker 240–244), Gideon Toury who coined his own terms “acceptabil-
ity” and “adequacy” in his discussion of translation norms (see Hermans 2004:
76), André Lefevere (1992) who classified translation patronage as “differenti-
ated” and “undifferentiated”, and Anthony Pym (1998:82–83) who created

“passive retranslation” and “active retranslation” in his analysis of reasons for

retranslating.
Obviously, there are weaknesses inherent in single-centre, single-relation
or dualistic research. For example, the text-centred model places too much em-
phasis on the text itself, thus ignoring the other translational elements such as
author, translator and reader. The translation outcome of this model will inevi-
tably have imperfections and biases in expressing the author’s thoughts and aes-
thetic values and in treating the relation between the two language-cultures. The
result is unfaithful translations, unidiomatic expressions or extreme foreignizing
translation. If we should cling to one interrelationship, say, intersubjectivity, we
would ignore the intertextual and intercultural relations. As for dualistic re-
search, it will easily lead to opposition between two extremes, such as literal
and liberal or foreignizing and domesticating translation. The six pairs of trans-
lation principles listed by T. H. Savory are most representative of the dualistic
model (see Liao et al. 2001:55–56).
Research on composite interrelationship in translation does not look at
translational issues from one point, one line or one plane. Neither does it study
translation phenomenon from two points, two lines or two planes, but it sees
translation activity as a multi-intersectional network, looking at interrelation-
ships between points, lines and planes along a multi-dimensional and multi-
directional perspective. This research paradigm which sees translation as a liv-
ing whole may reformulate the real picture of translation as it is.
4. TRANSLATION PRACTICE FROM THE

PERSPECTIVE OF CI
From the angle of interrelationship, translation activity may be viewed as a kind
of copying and creation of interrelationship. By copying of interrelationship, I
mean that the composite interrelationship in a literary text should be exactly and
completely reproduced in the target text. By creation of interrelationship, I
mean that a literary translation will unavoidably produce various kinds of inter-
relationships which do not exist in the source text. While the copying of interre-
lationship guarantees maximum fidelity of the target text to the source text,
creation of interrelationship denies that literary translation is a type of mechani-
cal reduplication of the source text. Rather, the latter affirms that it is something
like original writing, a kind of humanistic activity conducive to the promotion
of human civilization. In the following sections, we will take David Pollard’s
translation of Yu Guangzhong’s Thus Friends Absent Speak 《 ( 尺素寸心》, see
Yang 2003:96–100) as an example to look at various kinds of interactive rela-
tions in translation practice.

58 CHUANMAO TIAN
5. COPYING OF INHERENT RELATIONS

A text is a network of relations. In terms of the text, the author is at the central
point of the network, while for the reader – including the translator – the centre
is the text. The author creates their own text by interacting with a galaxy of pre-
vious authors and texts, namely their predecessors, contemporaries and their
texts. The previous texts, in a broad sense, also include the current and histori-
cal context in which the author is situated. The previous authors and texts are
not necessarily restricted to the author’s own language-culture, because literary
creation is a kind of unbridled imagination which can transcend any temporo-
spatial boundaries and which is delimited only by the author’s pre-compre-
hension (Heidegger 1962; Gadamer 1988). Besides the subjective and textual
“other”, literary writing involves the cultural “other”. Different and sometimes
contradictory cultural elements may be integrated into one literary work. There-
fore, in translating the translator should not only pay attention to the under-
standing of specific words and expressions at the micro-level but also of all
kinds of interrelationships at the macro-level. A good command of all the inter-
relationships in the source text is helpful to the comprehension of the whole text
as well as to the solution of micro-level linguistic, pragmatic and cultural diffi-
culties. It can be argued that the copying of interrelationship is a global faithful-
ness to the source text.
The author’s pre-comprehension is filled with a world view, life experi-
ence, and real people and events, and fictitious people and things, all of which
will be activated to a lesser or greater extent and occupy a specific mental and
spiritual presence in the overall process of creative writing. That is to say, these
real and fictitious people and things interact with the author. Let us take quota-
tion, for example. Quotation, whether direct or indirect, is a common device in
literary writing and is characterized by a composite interrelationship because it
embodies a kind of dialogic relation between author, text and “other”. The in-
troduction of a quotation by the author is the result of a series of communicative
events. For instance, an author first comes up with a piece of utterance, then the
text in which it appears, or the person who produced it, or the person (s) who
commented on it. Or, an author may first think of a certain previous author or
text, then some statement relevant to their current artistic creation and finally di-
rectly or indirectly quote it in their own text. The translator should see clearly
the lines of such interaction and then, bearing in mind the reader’s capacity for
reception, determine the proper translation strategy. In Thus Friends Absent
Speak, David Pollard rendered three quotations using three different strategies.
(1) Strategy: literal translation
Original Chinese version:
英国诗人奥登曾说, 他常常搁下重要的信件不回, 躲在家里看他的侦探
小说。

English version:
W. H. Auden once admitted that he was in the habit of shelving important
letters, preferring instead to curl up with a detective novel…
(2) Strategy: restitution (i.e. finding the original words)

王尔德有一次对韩黎说：“我认得不少人, 满怀光明的远景来到伦敦,
但是几个月后就整个崩溃了, 因为他们有回信的习惯。 ”
English version:
… while Oscar Wilde remarked to Henley: “I have known men come to
London full of bright prospects and seen them complete wrecks in a few
months through a habit of answering letters.”
(3) Strategy: addition

（七零八落的新简旧信, 漫无规则地充塞在书架上、抽屉里, 有的回过,
有的未回,）“只在此山中，云深不知处，”（要找到你决心要回的那一封,
耗费的时间和精力，往往数倍于回信本身。）
English version:
As the poet was told about the recluse he was looking for: “I know he’s in
these mountains, but in this mist I can’t tell where.”
Example (1) is an indirect quotation embodying a subject–subject relation,

namely the relation between the author and a previous foreign author. The trans-
lator uses a literal translation and does not point out the source of the quotation
because the original meaning is very clear. Example (2) is a direct quotation
containing a subject–subject–text relation, namely the relation between the au-
thor, two previous foreign authors and the text in which the quotation appears.
The translator directly moves the original words of the quotation into the trans-
lation. Here the back-translation method cannot be used. David Pollard specifies
the reason. He says, quotations are generally uniquely and wonderfully phrased,
and the only solution is to find the quoted author’s words.
Let us look at some more examples. In Qian Zhongshu and His Fortress
Besieged, Yang Jiang quoted Gustave Flaubert’s well-known words: “Madame
Bovary, c’est moi [it is I].” The translator, LinYuan, used the original French
words in his translation, followed by an English paraphrase in brackets (see
Qian 2003:257). If a quotation is a famous utterance, the translator’s work is
relatively easy. If a quotation is relatively unknown, the translator must have ei-
ther erudite knowledge or seek help from relevant resources or colleagues or

60 CHUANMAO TIAN
experts in order to solve the problem. In translating Yu Dafu’s On Marriage,

Zhang Peiji, a veteran translator, encountered such a sentence:
“同兰姆所说的一样, 儿子们去上了断头台.” His translation is: “As Lamb
said, the sons were sent up to the gallows.” He did not use the fully equivalent
“guillotine” for “断头台.” Instead, he used “gallows” because Charles Lamb
used gallows in his own essay (see Zhang 2003:135). This is a typical example
representing the influence of a composite interrelationship (specifically subject-
text relation) on the translator’s diction. The quotation of verse lines and allu-
sions is quite common in literary writing, the translation of which is far from
easy work. What is self-evident cultural information for the author and the
source-language reader is often hidden between the lines and may be invisible
to the target-language reader. As a result, the translator is expected to use addi-
tions properly to bring to the surface the implicit message which the average
source-language reader can obtain so as to achieve an equivalent effect for the
target-language reader. The words added in example (3) are not enough for the
English reader to understand the sentence. They must be puzzled, asking, “Who
is the poet?” However, with no effort, a common Chinese reader will get the
implicit message behind the verse line “只在此山中，云深不知处”: it comes
from a poem by Jia Dao of the Tang Dynasty.
What is true in literary translation is that a photographic copy of interrela-
tionship by means of translation is generally impossible and that semantic mi-
gration and deviation will inevitably take place in translation. Moreover, the in-
terrelationships in the source text may be either overt or covert. Overt interrela-
tionships are relatively easy to grasp. Covert interrelationships, which can be
found in some ancient Chinese classics as well as in the works by George Eliot,
James Joyce, and other writers, are really a great challenge to translators who
must make major efforts to verify them.
6. CREATION OF NEW RELATIONS
The value of translation is represented more in the creation of interrelationships.

This value demands that literary translation not be a parrot-like imitation but
rather a creative value-added artistic work. Moreover, such value is the key to
the rebirth of the source text in a new language-culture. Thus, translating does
not only mean that the translator interacts with the author and their text but that
the translator, by means of the author and their text, communicates with readers,
other authors, other texts and characters in texts. With their own life experience,
values, aesthetics, indelible mother-tongue marks, as well as readers’ expecta-
tions and their own translational intention, the translator agrees, disagrees or

even fights with the “other” authors, texts and cultures. Translators are one side
of the dialogue and also the “third party”, namely the intermediary. Without the
translator, it would be impossible for the copying and creation of interrelation-
ship to be achieved. Creation of interrelationship is necessary for overcoming
various kinds of intersubjective and intercultural contradictions in translation, a
buffer zone for the translator to survive and mediate between author and reader.
In this zone, the translator deals with the author, the reader and the “other” (in-
cluding other authors and texts).
David Pollard’s translation of “《尺素寸心》” as Thus Friends Absent
Speak is actually a borrowing from the English writer John Donne (specifically,
from his book To Sir Henry Wotton), which preserves the original style of for-
mality, solemnity and euphemism of the Chinese title. We can find many exam-
ples of this type. In translating Li Dazhao’s Today (“《今》”), Zhang Peiji re-
ferred to the famous line in “A Psalm of Life” by the American poet H. W.
Longfellow: “Act, –act in the living Present!” and rendered “《今》” as “The
Living Present” (see Zhang 2003:3). Feng Huazhan rendered
“《王若飞在狱中》”, a Chinese book title (literally, Wang Ruofei in Prison), as
Iron Bars but Not a Cage which actually is an adaptation from the verse line
“Nor iron bars a cage” of the poem “To Althea from Prison” by the English poet
Richard Lovelace, so as to avoid any possible misunderstanding of an otherwise
literal translation by the target reader who can well understand that the person in
prison was a noble revolutionary, not a criminal (see Luo 1984:1000–1001).
This kind of borrowing is not restricted to the translation of titles; it can also be
employed in the overall process of translating. To be more exact, in all cases
where the translator thinks that their own words fail to express the original
meaning, they can make full use of other people’s words to express their inter-
pretation of the source text. For example, David Pollard borrowed William
Wordsworth’s well-known expression, “surprised by joy” (from the poem “Sur-
prised by Joy – Impatient as the Wind”), to render the Chinese “喜出望外”, so
as to achieve a forceful and effective reproduction of the original tone and feel-
ing. The famous translator, Xu Yuanchong, once said, in translating poetry from
Chinese to English, “one can fully employ the well-known expressions and
lines of foreign poets, making them well serve our translation” (1984:53).
Likewise, in translating poetry from English (actually, any foreign language)
into Chinese one can borrow those widely known expressions and lines in Chi-
nese poetry. The following is Feng Huazhan’s translation of the English poet
Thomas Gray’s first stanza of “Elegy Written in a Country Churchyard”:
晚钟殷殷响，夕阳已西沉。

62 CHUANMAO TIAN
群牛呼叫归，迂回走草径。
农夫荷锄犁，倦倦归家门。
惟我立旷野，独自对黄昏。
(see Weng 1983:135)
Back translation (my translation):
The evening bell tolls constantly; the sun’s already setting.
Herds of cattle are crying home, winding along a grassy path.
The farmer carries his hoe and plough, going home wearily.
I stand in the wilderness all by myself, facing the dusk alone.
Weng Xianliang (1983:136) made the following comment on the transla-

tion:
“农夫荷锄犁” (The farmer carries his hoe and plough) reminds us of Tao
Yuanming’s verse line “戴月荷锄归” (The farmer carries his hoe home
under the moonlight). This association cannot be avoided unless we ex-
clude the use of classical Chinese. Now that a lot of classical Chinese ele-
ments have been incorporated into contemporary Mandarin Chinese, no-
body can exclude them. The problem is not what we can do to avoid asso-
ciations but what we can do to avoid producing an association which is not
in harmony with the original poetic image. If the association produced is in
perfect harmony with the original image and what is added in the transla-
tion can enhance the artistic effect very well, why can we not do it? (my
translation)
From Weng’s remarks we can see that literary translation allows for the
creation of interrelationship, but this kind of creation should follow the princi-
ple of “harmony”. In addition, creation of interrelationship and copying of inter-
relationship are mutually convertible. In other words, copying of interrelation-
ship produced by the translator in the course of their rendering of the source text
can convert to creation of interrelationship by the target-language reader in the
course of their comprehension of the target text. For example, the above transla-
tion will make readers who are familiar with ancient Chinese poetry, immedi-
ately think of the famous poet Tao Yuanming (365–427 AD) and his poem Re-
turn to Nature.
Copying and creation of interrelationship can be viewed as two equal strata
of author–translator intersubjectivity. In the entire process of communication
with the author the translator may sometimes identify themselves with the au-
thor’s values, aesthetics and cultural outlook, and they will accept their attitudes

and viewpoints with no hesitation in translating, which results in a copy of in-

terrelationship. However, sometimes the translator may not agree with the au-
thor’s views. That is to say, the translator will be affected and restricted by the
target culture, reader’s expectation and translation purpose when they interact
with the author and source culture. They may employ strategies such as addi-
tion, omission and revision to impose drastic changes on the source text. This
kind of translation practice is where creation of interrelationship is. Creation of
interrelationship has advantages, such as upgrading the value of the source text;
it also has disadvantages, such as the intentional misreading and mistranslation
of the source text by the major translators in the late Qing Dynasty.
7. CONCLUSION
Composite interrelationship in literary writing and translation comprises inter-

textuality, intersubjectivity, interculturality, and especially the relation between
them. The nature of freedom in literary creation determines that composite in-
terrelationship is inherent in a literary work. It includes the interrelationship be-
tween the author and other authors, their text and prior texts, their native culture
and other cultures. This kind of composite interrelationship is, ideally, repro-
duced in the translation of literary works. Moreover, literary translation in-
volves a kind of composite interrelationship which does not exist in the source
text. It is created by literary translators as a result of the very use of the target
language.
As we know, culture-loaded words usually associate with a myth, a prov-
erb, a folk tale, a literary classic, an utterance of a famous person, an account of
a historical event or other textual materials. When the target-language reader
encounters such words in a literary translation, they will naturally think of the
myth, the proverb, the folk tale etc. of the target culture. Therefore, new inter-
textual relations are inevitably introduced consciously or subconsciously into
the translation by the translator and then realized by the reader in their interpre-
tation of the translated text. When the values, aesthetics, writing norms, and ide-
ologies of the two languages or cultures contradict each other, the translator
may lean towards the target side to incorporate target cultural elements into the
translation, where new intercultural relations are created. In a word, a deep and
systematic study of the composite interrelationship created by the translator
may open up new perspectives for Translation Studies.

64 CHUANMAO TIAN
References
Bassnett, S., Lefevere, A. 2001. Constructing Cultures. Shanghai: Shanghai Waiyu Jiaoyu Chu-
banshe.
Brownlie, S. 2003. Investigating explanations of translational phenomena: A case for multiple
causality. Target Vol. 1. No.1. 111–152.
Chen, Daliang. 2004. “Shui shi fanyi zhuti?” [Who is the translational subject?]. Zhongguo Fanyi
Vol. 2. 3–7.
Chen, Daliang. 2005. Fanyi yanjiu: Cong zhutixing zouxiang zhuti jianxing [Translation research:
From subjectivity to intersubjectivity]. Zhongguo Fanyi Vol. 2. 3–9.
Chesterman, A. 2002. Semiotic modalities in translation causality. Across Languages and Cul-
tures Vol. 2. No.2. 145–158.
Fawcett, P. 1995. Translation and power play. The Translator Vol. 2. No. 2. 177–192.
Gadamer, H. G. 1988. Truth and Method, trans. by J. Weinsheimer & D.G. Marshall. New York:
Crossroad.
Heidegger, M. 1962. Being and Time, trans. by John Macquarrie & Edward Robinson. London:
SCM Press.
Hermans, Th. 2004. Translation in Systems: Descriptive and System-oriented Approaches Ex-
plained. Shanghai: Shanghai Waiyu Jiaoyu Chubanshe.
Jin, Shenghua & Huang, Guobin. 2001. Yin Nan Jian Qiao [Ingenious Solutions in Translation].
Beijing: Zhongguo Duiwai Fanyi Chuban Gongsi.
Kristeva, J. 1980. Desire in Language: A Semiotic Approach to Literature and Art, trans. by
Thomas Gora, Alice Jardine and Leon S. Roudiez. New York: Columbia University Press.
Lefevere, A. 1992. Translation, Rewriting and the Manipulation of Literary Fame. London and
New York: Routledge.
Liao, Qiyi et al. 2001. Dangdai Yingguo Fanyi Lilun [Contemporary Translation Theories in Eng-
land]. Wuhan: Hubei Jiaoyu Chubanshe.
Liu, Miqing. 1999. Wenhua Fanyi Lungang [An Outline of Cultural Translation]. Wuhan: Hubei
Jiaoyu Chubanshe.
Liu, Yuedi. 2005. “Zai ‘wenben jianxing’ yu ‘zhuti jianxing’ zhijian – Shilun wenxue huodong
zhong de ‘fuhe jianxing’” [Between intertextuality and intersubjectivity: On composite in-
terrelationship in literary activity]. Wenyi Lilun Yanjiu Vol. 4. 64–69.
Luo, Xinzhang. 1984. Fanyi Lunji [Selected Papers on Translation]. Beijing: Shangwu Yinshu-
guan.
Pym, A. 1998. Method in Translation History. Manchester: St. Jerome.
Qian, Zhongshu. 2003. Wei Cheng [Fortress Besieged]. Beijing: Waiyu Jiaoxue yu Yanjiu Chu-
banshe.
Robinson, D. 2006. Western Translation Theories from Herodotus to Nietzesche. Beijing: Waiyu
Jiaoxue yu Yanjiu Chubanshe.
Susam-Sarajeva, S. 2001. Thinking about power relations once again: Response to Ieva
Zauberga’s paper. Across Languages and Cultures Vol. 2. No. 1. 127–132.
Tian, Chuanmao, Chen, Yifang. 2005. Shilun wenxue fanyi de ‘fuhe jianxing’ [A tentative study
of composite interrelationship in literary translation]. Waiyu Jiaoxue No. 2. 83–86.
Weng, Xianliang. 1983. Yitai Youlai Huabucheng? [Can Image Never Be Drawn?]. Beijing:
Zhongguo Duiwai Fanyi Chuban Gongsi.
Xie, Tianzhen, 1992. “Wenxue fanyi de chuangzaoxing panni” [Creative treason in literary trans-
lation]. Waiguoyu No. 30–37.
Xu, Jun. 2003a. “Chuangzaoxing panni yu fanyi zhutixing de queli” [Creative betrayal and estab-
lishment of translational subjectivity]. Zhongguo Fanyi No. 1. 6–11.
Xu, Jun. 2003b. “Zhuti jianxing yu shiye ronghe” [Intersubjectivity and fusion of horizons].
Waiyu Jiaoxue yu Yanjiu Vol. 4. 51–56.

Xu, Yuanchong. 1984. Fanyi de Yishu [The Art of Translation]. Beijing: Zhongguo Duiwai Fanyi
Chuban Gongsi.
Yang, Ping. 2003. Mingzuo Jingyi [Quality Translation of Classics]. Qingdao: Qingdao Chuban-
she.
Zauberga, I. 2000. Rethinking power relations in translation. Across Languages and Cultures Vol.
1. No. 1. 49–56.
Zhang, Peiji. 2003. Yingyi Zhongguo Xiandai Sanwenxuan [English Rendition of Modern Chinese
Classical Essays]. Shanghai: Shanghai Waiyu Jiaoyu Chubanshe.

DOI: 10.1556/Acr.11.2010.1.4
COMPARISON OF PARALLEL TEXTS

OF PETITIONS IN POLISH–HUNGARIAN AND
POLISH–ENGLISH TRANSLATION
KAROLINA KACZMAREK1 AND ALEKSANDRA MATULEWSKA2

Institute of Linguistics, Adam Mickiewicz University
Al. Niepodległości 4, 61-874 Poznań, Poland
Phone/Fax: +48-61-829-36-63
1
E-mail: karkacz@amu.edu.pl
2
E-mail: aleksandra.matulewska@gmail.com
Abstract: This article explores the possibilities of using parallel texts in Polish, Hun-
garian and English legal translation and legal translation teaching. The authors focus on pe-
titions, a special genre of the legal register, and show that the macrostructure of Polish,
Hungarian and English are different: they are composed of the same elements, but these are
ordered differently. Polish petitions may be written by lawyers as well as lay persons, thus
the language used in them may be both legal and colloquial depending on the author. Eng-
lish petitions are usually executed by lawyers, or they are simply forms which are filled in
by claimants. In Hungary a claimant also fills in a form. What may pose serious translation
problems is the part called the grounds of the petition, as it may be written in a special-
purpose register or in colloquial language. Despite the differences, parallel texts of petitions
in Polish, Hungarian and English provide the translator and trainee translator with more re-
liable equivalents than dictionaries, and not only for technical terms, but also for phrase-
ological units and formulaic sequences.
Keywords: parallel text, translation, legal translation, petition
1. INTRODUCTION
The purpose of this article is to demonstrate the usefulness of parallel texts in
the translation process.
Parallel texts are texts produced by users of different languages under near-
identical communicative conditions. [...] Parallel text files [...] are part and
parcel of the material and mental equipment of the competent translator.
This equipment is a vast database storing enormous experience. It is the
key to an extensive knowledge of how texts are structured in the (text)
world of different (communicative) cultures. (Neubert 1996:101)

68 KAROLINA KACZMAREK AND ALEKSANDRA MATULEWSKA
The analysis and comparison of the content and structure of parallel legal
texts in two languages provides a translator not only with legal terminology but
also with phraseology and formulaic sequences which may be considered dy-
namic equivalents. In the case of language pairs with similar legal systems (e.g.
two languages with civil law systems based on the Napoleonic Code), parallel
texts are often surprisingly similar. But even in the case of documents from two
different legal systems (e.g. civil law and common law), such documents are
very helpful. As Smith (1995:187) states,
recognizing a case of absent terminology requires constant comparison be-
tween the legal systems of the source and the target languages, as well as
being familiar with up-to-date legal literature. (…) «equivalency» refers to
equal value between source and target text, and «adequacy» concerns par-
tial dimensions of the text. In the global legal world, where legal traditions
and cultures differ so much that one system uses a legal concept com-
pletely unknown or even alien to another system, equivalency must also in-
clude cultural equivalency. From a linguistic point of view, the ideal trans-
lation is one that does not resemble one. Applied to legal texts, a successful
translation should communicate the content of a document, all the while
employing equivalent accurate syntax, semantics and pragmatics. This is
particularly true for legal texts. (Smith 1995:187)
Moreover, “finding equivalent and adequate terminology requires transla-

tors to recognize differences between the cultures of the legal texts with which
they work. Legal culture represents the status and behaviour of a judicial class.”
(Smith 1995:187) Parallel texts help establish such equivalent pairs on the level
of terms, phraseological units and formulaic sentences.
2. TRANSLATION PRODUCT AND ITS RECIPIENT
In 1978 Vermeer formulated the skopos theory which is now generally consid-
ered to be a part of the theory of translational action based on the aim of a text,
where great importance is assigned to the pragmatic aspects of the source text
and target text. The word skopos stands here for the aim or objective of a trans-
lation product. The function of the source text may be the same as or different
from the function of the target text. Usually clients (called by Vermeer commis-
sioners) supply translators with sufficient information to determine the function1
(that is the audience) of a target text. According to Vermeer (2001), in order to
identify the skopos of a text, translators should be experts in translational action
and therefore should know more about the field than ‘outsiders’ and conse-
quently they should be able to determine what is the role of the source text in

COMPARISON OF PARALLEL TEXTS OF PETITIONS 69
the translational action they perform. The fact that the source text is addressed
to the source language reality and the target text to the target language reality
does not mean that the function of those two texts will be different. Every legal
translation has its skopos, which probably cannot be said about literary texts,
and it is essential for the translator to determine it.2 The theory of skopos has
been modified by Kierzkowska (2002) to serve the purpose of translation of le-
gal terms. She has described the pragmatic model of translation of legal texts.
The skopos of translation determines the orientation which is to be applied
by a translator. In general two opposing orientations are mentioned: source-
language oriented translation3 and target-language oriented translation4. The
comparison of parallel texts provides translators with dynamic (connotative)
equivalents which serve the purpose of target-language oriented translation for
so-called distant recipients (cf. Kierzkowska 2002:87–95).
3. COMPARISON OF PARALLEL TEXTS

There are two types of parallel texts. A parallel text is defined as “a text that
represents the same text type as the source text” (Delisle et al. 1999:166) or
a text that treats the same or a closely related topic in the same subject field
and that serves as a source for the mots justes and terms that should ideally
be incorporated into the target text to ensure collocational cohesion. (De-
lisle et al. 1999:166)
Sometimes, parallel texts are defined as source-language texts accompa-

nied with their translations into a target language. However, in this paper it is
the former type that will be of interest to the authors. Within that type, we will
regard as parallel texts only those which belong to the same genre within the le-
gal register. It should be borne in mind that texts may be classified in accor-
dance with the following criteria: (i) the subject matter of the text, (ii) the pur-
pose of the text, (iii) types of structure elements constituting a given text, and
(iv) communicative orientation. Texts do not exist independently but they al-
ways remain in relations with other texts (de Beaugrande and Dresler 1990).
Parallel documents in both source and target languages are very useful.
They help solve problems with terminology and phraseology. Well-matched
parallel texts are better than dictionaries as they are more detailed and give us
terminology and phraseology in a wider context of a sentence or a text. Of
course they cannot solve the problem of source-language concepts with no
equivalents in the target legal reality. Though laborious and time-consuming,
the analysis of such texts is worth the trouble.
Teachers of languages for special purposes (LSP) and legal translation
should make students aware of the benefits which may be reaped from the use

of parallel texts as compared to the use of dictionaries, whose deficiencies be-

come even more apparent from analyses of parallel texts. In comparing parallel
texts, students and translators should pay attention to four aspects of the docu-
ments, namely (i) their macrostructure, (ii) the terms used, (iii) the phraseologi-
cal units, and last but not least (iv) the frozen or formulaic sentences or expres-
sions found in them. Let us analyse those four aspects in more detail.
4. ANALYSIS
4.1. The Macrostructure of Petitions
Detailed guidelines specifying obligatory elements of applications submitted to

courts may be found in Article 126 § 1 of the Polish Code of Civil Procedure.
According to the provisions of that article, each application shall be composed
of the following elements:
(i) the court addressed,
(ii) the parties to the proceedings or their representatives,
(iii)the type of the petition,
(iv) the subject matter and statement of fact as well as evidence supporting
the claim,
(v) the signatures of the petitioner (claimant) or his/her representative;
(vi) a list of enclosures.
In addition to the above, the first application submitted to the court shall in-
clude:
(i) the place of residence or principal place of business of the parties to
the proceedings (and/or their representatives),
(ii) the subject matter and its value.
A petition is
a formal, written application to a court requesting judicial action on a cer-
tain matter. An application made to a court ex parte, or where there are no
parties in opposition, praying for the exercise of the judicial powers of the
court in relation to some matter which is not the subject for a suit or action,
or for authority to do act which requires the sanction of the court (…).
(Black 1979:1031)
Sometimes the term claim is used instead of petition in English civil ac-
tions. A claim is ‘a demand for money, property, or a legal remedy to which
one asserts a right; esp., the part of a complaint in a civil action specifying what

relief the plaintiff asks for.’ (Black 2004:264) Some courts even introduce the
so-called ‘claim forms’ which are to be filled in with relevant data by claimants.
The term petition may be translated into Polish as pozew when the parties
to the contentious proceedings are in opposition or wniosek when the parties are
participating in non-contentious proceedings. Petition in civil law is an applica-
tion which serves the purpose of instigating civil proceedings (cf. Siedlecki
Świeboda 2003:186, Jędrzejewska in Ereciński Gudonski 2003:187n). The pro-
visions of the Polish Code of Civil Proceedings (Article 187 § 1) are very spe-
cific as the following components shall be included in a petition:
(i) a clearly specified claim and the relief sought (if the case concerns
property rights, the monetary value of the relief sought must be given
in a petition),
(ii) the statement of fact justifying the claim and if need be also the choice
of court.
The petition as a special kind of application in Poland has the following

macrostructure:
(ii) the parties (petitioner/claimant/plaintiff and the defendant), their ad-
dresses and representatives if any,
(iii) the title of the petition,
(iv) a relief sought,
(v) some additional claims (optional),
(vi) grounds,
(vii) signatures,
(viii) enclosures.
The macrostructure of English petitions is the following:

(i) the title of a petition,
(ii) the court addressed,
(iii) the parties and subject matter,
(iv) the main body of the petition (the statement of fact, grounds and the
relief sought),
(v) the date,
(vi) the statement of a solicitor, and
(vii) the details concerning filing the petition and the date of hearing
(added by the court).
On the other hand, Hungarian petitions are composed of the following elements
(Bakó et al. 2000:24):


(ii) a courtesy expression,
(iii) the petitioner and subject matter,
(iv) the statement of fact,
(v) a formulaic expression that a petition is justified,
(vi) the relief sought,
(vii) the justification of the relief sought value,
(viii) legal grounds,
(ix) other conclusions,
(x) legal grounds for the court addressed,
(xi) a courtesy expression,
(xii) the signature.
In general, it may be stated that the contents of the petitions in Polish,

Hungarian and English legal realities are more or less the same. The flow of in-
formation and the structure differ. However, the content is alike.
4.2. Key-words
Translators of Hungarian must cope with numerous problems which are not
faced by the translators of more widespread languages. First of all, the fact that
there are no bilingual dictionaries is very troublesome. As far as the language
pair Polish and Hungarian is concerned, there is only one bilingual dictionary
which was written in the 1950s and was reprinted in 1980s, and a supplemen-
tary thematic dictionary which was written at the Institute of Linguistics of
Adam Mickiewicz University by the staff of the Department of Hungarian Stud-
ies. These two dictionaries contain a limited number of legal terms and thus
they are not sufficient. On the one hand, it stems from the fact that they are not
specialized legal dictionaries, and on the other hand many new legal terms have
been coined for the last fifteen years or so due to the social and economic
changes. Therefore, translators have to resort to monolingual Hungarian dic-
tionaries or to Hungarian–English ones which is a very hazardous practice be-
cause of the differences in legal systems (common law and civil law countries
have dissimilar legal realities). It should also be remembered that
translation is generally possible because in everyday language words, their
component parts, their order, and most importantly their meanings are not
like fixed points, but fuzzy blots at best. While the blots of L1 may rarely
correspond exactly to those of L2, chances are that their fuzz will overlap
enough for most translation purposes. Literary translation, especially the
translation of poetry, exploits those fuzzy areas to the fullest. But law, or at
least some areas of it, requires those blots to shed their fuzz and become

point-like; and indeterminacy resides in the difficulty we have in making

language do this with any consistency. (Joseph 1995:23)
The tables below feature the terms which may be found in the dictionaries listed
in the bibliography and the parallel texts under scrutiny.
Table 1
The terms included in dictionaries and parallel text equivalents
Parallel text Dictionary Parallel text

Polish term Dictionary equivalent
equivalent equivalent equivalent
pozwany defendant, respondent defendant beidézett alperes
powód 1. plaintiff, claimant, petitioner, petitioner/ – felperes
demandant, 2. reason, cause, plaintiff/
occasion, grounds (two mean- claimant
ings of the Polish polysemous
term are sometimes even not
distinguished)
wartość value, worth value érték, ár, becs érték
wnosić o coś file, submit, petition, motion petition for, – kér
(no preposition given in dic- pray for, ask to
tionaries)
nakazać coś order, summon, command, pre- order sth to sb megparancsol elren- kötelez vkit
komuś scribe, enjoin (no preposition del, intézkedik, uta-
given in dictionaries) sít, parancsol
rozprawa hearing, trial hearing tárgyalás, bírósági tárgyalás
tárgyalás
przesłuchać interrogate, question hear kihallgat, kivallat kihallgat
świadek witness witness tanú tanú
uzasadnienie reasons, statement of reasons, grounds megokolás, indoko- indoklás
grounds, substantiation, motives lás, megalapozás
należność duties, fees, payments, debts, fee, payment illetmény, járandó- összegszerűség
(należności) dues, claims, receivables ság, kinnlevőség,
összeg, illeték, adós-
ság, tartozás
dowód proof, evidence, receipt evidence, proof bizonyíték, bizony- bizonyíték
ság, jel
żądanie request, demand, claim relief (asked) követelés, kívánság, –
kwota sum, amount the sum of kvóta, összeg, há- összeg
nyad, fizetési arány
zasądzić coś adjudge, grant, concede, award, to award sth to elítél, ítélkezik ‘kérem a T.
od kogoś adjudicate sb Bíróságot,
hogy szíves-
kedjen köte-
lezni az alpe-
rest arra, hogy
fizessen meg
részemre...’*

To sum up, dictionaries provide translators with a much wider array of po-
tential equivalents than parallel texts. But parallel texts provide more specific
context-bound equivalents. Consequently, they narrow down translators’
choices significantly, making their work easier and at the same time more accu-
rate.
4.3. Phraseological Units
As has already been mentioned, in order to translate legal texts appropriately

and correctly it is not enough to know the terminology. Knowledge of the phra-
seological units, fixed or frozen expressions is just as important. Most of the
Table 2
Dictionary and parallel text equivalents
Parallel text Dictionary Parallel text
Polish term Dictionary equivalent
equivalent equivalent equivalent
sąd rejonowy district court, local court – – Kerületi Bíró-
ság
sąd okręgowy district court – járásbíróság Megyei Bíróság
wydział cywilny – – – –
przedmiot sporu subject of litigation Civil Division – –
wartość przedmiotu – – – –
sporu
pozew o zapłatę – petition for – marasztalási ke-
(payment) reset
na rzecz to sb, for sb on behalf of javára részére
odsetki ustawowe legal interest, statutory interest pre- – késedelmi ka-
interest scribed by law mat
wnieść pozew file an action, file a suit, file/submit a pe- beidéz keresetet előter-
bring a case, bring a suit tition jeszt
(to court)
dzień zapłaty day of payment, – kifizetés napja
payment day
koszty procesu costs of the proceedings (court) fees – perköltség
normy przepisane – as prescribed by – –
law
przeprowadzić roz- conduct a trial, hear a hear a case – pert megtartani
prawę case, hold a hearing
wezwać na rozprawę – summon – tárgyalásra be-
idézni
uzasadnione żądanie – claims are justi- – –
fied, sound
claims
rygor natychmiastowej order of immediate en- – – azonnali végre-
wykonalności forceability hajtás
w kwocie – in the sum of – összegben
zastępstwo procesowe representation in pro- attorney’s fees – perbeli képvise-
(jako koszty) ceedings at law let

available dictionaries do not contain such expressions, or only a very limited

number. There is no Hungarian–Polish legal dictionary, and there are few stud-
ies devoted to legal language in Polish–Hungarian translation. Therefore, the
task of preparing a terminological dictionary is just as important as the prepara-
tion of a phraseological dictionary including legal word combinations which
may be found in legal texts. Moreover, such expressions should on numerous
occasions be shown in a wider context, as this is a prerequisite for identifying
their exact meaning.
Table 2 contains dictionary and parallel text equivalents in English and
Hungarian for Polish phraseological units.
To sum up, it is much easier to find equivalent phraseological units in par-
allel texts than in dictionaries. The only exceptions are so-called system-bound
terms and expressions that do not have dynamic equivalents in a given target
language. In the case of such terms, translators must resort to the methods of
providing equivalents for non-equivalent terminology (cf. Kierzkowska
2002:117–123, Matulewska and Nowak 2006:65–71). But one should remem-
ber that these methods apply to terminology only. They cannot be applied to
phraseological units which in English are usually called collocations or word
combinations, or to grammatical structures which cannot be calqued (Kierz-
kowska 2005:95).
4.4. Frozen Sentences and Formulaic Sentences
The relief sought is the part of the petition which starts with the words Wnoszę o
(I petition/pray for/ask for), which is followed by a list of things asked for. Ac-
tually, this part of a Polish petition is a sort of a formulaic sentence.
Example 1. Polish petitions:

Wnoszę:
• o zasądzenie od pozwanego Antoniego Krupskiego na rzecz powoda
Jerzego Kraski kwoty 8 000, 00 zł (osiem tysięcy złotych), z
ustawowymi odsetkami od dnia 20 lutego 1998 roku do dnia zapłaty,
• o zasądzenie od pozwanego na rzecz powoda kosztów procesu według
norm przepisanych;
• o nadaniu wyrokowi rygoru natychmiastowej wykonalności.
Nadto wnoszę o:
• o przeprowadzenie rozprawy także pod nieobecność pozwanego,
• o wezwanie na rozprawę i przesłuchanie jako świadków:
Bogumiła Ślęzaka, zam. w Gołonicy, ul. Gnieźnieńska 15/20,
Waldemara Woźniaka, zam. w Kcyni, ul. Żnińska 14b.

Example 2. English petitions:

The petitioner claims:
(a) a divorce;
(b) under the Divorce Act; (…)

Wherefore, the petitioner asks the Court to order payment...
TO THIS HONOURABLE COURT: I hereby petition for a decree of di-
vorce from the Respondent spouse on the grounds and circumstances fol-
lowing: (…)
WHEREFORE, Petitioner respectfully prays for the following: …
Example 6. Hungarian petitions:
A fentiekre tekintettel keresetet terjesztek elő az E. Sz. Autószerelő Kft.
(1222 Budapest, Kossuth u. 120.) alperes ellen, és kérem a T. Bíróságot,
hogy szíveskedjen kötelezni az alperest arra, hogy fizessen meg részemre
850 000 Ft-ot, valamint ezen összeg után 2000. május 8. napjától a kifize-
tés napjáig évi 20% késedelmi kamatot.
Kérem továbbá a T. Bíróságot, hogy szíveskedjen kötelezni az alperest a
jelen perben felmerült költségeim megterítésére.
KÉREM A T. BÍRÓSÁGOT, HOGY AZ ÜGYBEN SZÍVESKEDJEN
TÁRGYALÁST KITŰZNI, ÉS ARRA A PERES FELEKET MEGIDÉZ-
NI.
Kérem a T. Bíróságot, hogy a pert esetleges távollétem esetén is szívesked-
jen megtartani.
Possible versions of translation into English:

Version No. 1
I hereby petition/pray for the following:
1) awarding the Petitioner eight thousand Polish zlotys (PLZ 8 000.00)
with statutory interest accruing from February 29th, 1998 to the day of
payment,
2) awarding the Petitioner costs of proceedings according to the prescribed
rates, and
3) making the judgment immediately enforceable.
Further, I petition for:

4) hearing the case also in the event of the Defendant’s absence,

5) summoning and hearing the following witnesses:
Bogumił Ślęzak, residing in Gołonica, ul. Gnieźnieńska 15/20,
Waldemar Woźniak, residing in Kcynia, ul. Żnińska 14b.
Version No. 2
I petition/pray/ask to:
6) have the sum of eight thousand Polish zlotys (PLZ 8 000.00) with statu-
tory interest accruing from February 29th, 1998 to the day of payment
awarded from the Defendant to the Petitioner,
7) have the costs of proceedings awarded from the Defendant to the Peti-
tioner according to the prescribed rates, and
8) have the judgment made immediately enforceable.
Further, I petition to:
9) have the case also heard in the event of the Defendant’s absence,
10) have the following witnesses summoned and heard:
Version No. 3
I petition/pray that:
11) the Petitioner be awarded the sum of eight thousand Polish zlotys
(PLZ 8 000.00) with statutory interest accruing from February 29th,
1998 to the day of payment,
12) the Petitioner be awarded costs of proceedings according to the pre-
scribed rates, and
13) the judgment be made immediately enforceable.
Further, I petition that:
14) the case be also heard in the event of the Defendant’s absence,
15) the following witnesses be summoned and heard:
Possible versions of translation into Hungarian:

A felperes nevében eljárva, akinek részemre szóló meghatalmazását az ak-
tákhoz mellékelem, indítványozom meghagyási eljárással az alperestől a
felperes javára a következő összegek megítélését:
1) [...] zloty főadósság címén a törvényes kamatokkal együtt, amely össze-
get a következő követelések terhelik:

2) [...] zloty (az 2002. 11. 24. napi számláról a [...] DM ellenértéke a vád-
emelés napján a Narodowy Bank Polski [Lengyel Nemzeti Bank] DM-
középárfolyama alapján) az 1999. november 26-ai számla címén
Nr.222/22/EE, 1999. december 9-étől 2000. november 26-ig 8%-os ka-
mattal, valamint 21% a vádemelés napjától a kiegyenlítés napjáig a to-
vábbi törvényi kamatlábváltozások százalékainak figyelembe vételével.
3) [...] zloty a határátlépési dokumentáció költségeinek megtérítése címén,
4) [...] zloty áruszállítási költség címén a törvényi kamatokkal 1999. de-
cember 1-étől a kiegyenlítés napjáig.
5) [...] zloty manipulációs költség címén,
6) [...] zloty a SAD dokumentum kitöltésének szolgáltatása címén.
7) Perköltségek, ezen belül a jogi képviselet költségei 8.000 zloty összeg-
ben (azaz nyolcezer zloty)
In sum, the comparison of parallel texts does not only provide translators
with dynamic equivalents for formulaic sentences, but also provides informa-
tion on grammatical structures used most often to express certain meanings.
5. CONCLUDING REMARKS
Having compared Polish, Hungarian and English petitions, we have noticed that
the majority of elements which are present in those documents are similar.
However, the information order differs depending on the language.
English petitions are much shorter than the Polish ones and the language in
which they are formulated is much more formal. On the other hand, Polish peti-
tions are much more emphatic and imperative than the English ones (cf. Jopek-
Bosiacka 2008:225). Furthermore, there are many archaisms in them. As a rule,
Polish, Hungarian and English petitions include the following elements: parties
to the proceedings, the court addressed, title of the petition and the relief sought
with grounds. Grounds of the Hungarian petitions are placed at the beginning of
the document before the relief sought. What is more, Polish petitions are more
formalized and succinct than Hungarian ones (cf. Table 2) and, at the same
time, less formal than the English ones. What is characteristic of the Hungarian
petitions is the fact that the utterances are in the form of compound sentences.
At the beginning of the Hungarian petition there is also a courtesy expression
which is not present in English and Polish documents of that sort. Additionally,
at the end of the Hungarian petition one may find a remark on the legal basis for
the court jurisdiction which may also occur in Polish petitions (article 187,
paragraph 1, subparagraph 2 of the Polish Code of Civil Procedure). What usu-
ally constitutes the greatest when translating petitions is usually the part called

grounds. It may be formulated in colloquial language or in a specialized, but

non-legal register, or even in the claimant’s idiolect.
Many non-experienced translators think that a good legal dictionary is
enough to be able to do the job. They do not realize that even the best dictionary
does not contain all the terms they are going to encounter in the course of trans-
lation. And even lexicographers can make mistakes. Consulting a dictionary and
finding some kind of an equivalent does not mean that translators find what they
are looking for. Furthermore, they do not understand the necessity of comparing
the meanings of words given in dictionaries with the meanings attached to the
words in the source language text. Translators are obliged to compare the mean-
ings of words in a given context, as the meanings are context-dependent. It
should be stressed again that in order to choose one term out of the several op-
tions suggested in a dictionary, it is necessary to check them in other sources.
Parallel texts are a perfect tool to verify such equivalents from dictionaries.
Therefore, the usefulness of selections of parallel documents in both source and
target languages cannot be denied. Such selections are extremely important, de-
spite the fact that they are not always enough to provide translators with all the
tools needed to translate legal texts successfully.
Such selections of documents are especially useful in the case of language
pairs having different legal systems and different historical backgrounds. Most
of the available monolingual selections of texts formulated in legal language
contain the most popular legal documents and such texts are an invaluable
source of information on style, register and collocations, terminology as well as
grammatical structures.
It must be stressed here that parallel texts, together with dictionaries can
improve significantly the quality of translations. Legal dictionaries rarely list
phraseological units and prepositional collocations, and in such cases parallel
documents are an invaluable source of terminological and phraseological infor-
mation.
Notes
1
“the aim of any translational action, and the mode in which it is to be realized, are negoti-
ated with the client who commissions the action. A precise specification of aim and mode is es-
sential for the translator. – This is of course analogously true of translation proper: skopos and
mode of realization must be adequately defined if the text-translator is to fulfill his task success-
fully” (Vermeer 2001:221).
2
“the skopos theory thus in no way claims that a translated text should ipso facto conform
to the target culture behaviour or expectations, that a translation must always “adapt” to the target
culture. This is just one possibility: the theory equally well accommodates the opposite type of
translation, deliberately marked, with the intention of expressing source-culture features by target-
culture means. Everything between these two extremes is likewise possible, including hybrid
cases. To know what the point of a translation is, to be conscious of the action – that is the goal of

the skopos theory. The theory campaigns against the belief that there is no aim (in any sense
whatever), that translation is a purposeless activity” (Vermeer 2001:231).
3
“source language orientation. This means that the translator does not make her translation
look like an original target language document, but instead orientates her translation towards the
source language community by imitating the way in which the source language document is de-
signed and by seeking to transfer semantically or near-literally the linguistic choices initially
made by the source text producer. This means that there is a much higher degree of resemblance
with the source text.” (Faber, Hjort-Pedersen and Klinge 1996/97:21).
4
“target language orientation. This means that the translator tries to make her text look as
much as possible like an original target language document. Consequently, she will orient her
translation towards the target language community by imitating the way in which parallel docu-
ments are designed in the target language community and by borrowing linguistic material from
such parallel texts in her translation purposes. In this case, then, there is little resemblance with
the source text” (Faber, Hjort-Pedersen and Klinge 1996/97:21).
References
de Beaugrande, R., Dresler, W. 1990. Wstęp do lingwistyki tekstu. Warszawa: PWN.
Broniewicz, W. 1996. Postępowanie cywilne w zarysie. Warszawa: Wydawnictwa Prawnicze,
PWN.
Delisle, J., Lee-Jahnke, H., Cormier, M. C. (eds) 1999. Translation Terminology. Amster-
dam/Philadephia: John Benjamins Publishing Company.
Dolecki, H. 2005. Postępowanie cywilne. Zarys wykładu. Warszawa: Wyd. LexisNexis.
Ereciński, T., Gudowski, J. 2003. Komentarz do Kodeksu postępowania cywilnego. Warszawa:
Wyd. LexisNexis.
Faber, D., Hjort-Pedersen, M., Klinge A. 1996/1997. Introduction to English Legal Language.
BA Sproglinien.
Kengyel, M. 2000. Magyar Polgári Eljárásjog. Budapest: Osiris Kiadó.
Kierzkowska, D. 2002. Tłumaczenie prawnicze. Warszawa: TEPIS.
Kierzkowska, D. (ed.) 2005. Kodeks Tłumacza Przysięgłego z komentarzem. Warszawa: Wy-
dawnictwo TEPIS Polskiego Towarzystwa Tłumaczy Ekonomicznych, Prawniczych i
Sądowych.
Jodłowski J., Resich Z., Lapierre J., Misiuk-Jodłowska, T. 2002. Postępowanie cywilne. Warsza-
wa: Wyd. LexisNexis.
Jopek-Bosiacka, A. 2008. Przekład prawny i sądowy. Warszawa: PWN.
Joseph, J. E. 1995. Indeterminancy, Translation and the Law. In: Marshall, M. (ed.) Translation
and the Law. Amsterdam/Philadelphia: John Benjamins Publishing Company. 13–36.
Matulewska, A., Nowak, P. 2006. Polskie sprawozdanie finansowe w tłumaczeniu na francuski.
In: Lingua Legis No. 14. 65–71.
Mądrzak, H. (ed.) 1999. Postępowanie cywilne. Warszawa: Wyd. C.H. Beck.
Myrczek, E. 2006. Słownik terminologii prawniczej. Warszawa: C.H. Beck.
Neubert, A. 1996. Textlinguistics of Translation: The Textual Approach to Translation. In:
Gaddis Rose, M. (ed.) Translation Horizons Beyond the Boundaries of Translation Spec-
trum. Translation Perspectives IX. Binghamton: Center for Research in Translation. 87–
105.
Piasecki, K. (ed.) 1996. Kodeks postępowania cywilnego. Komentarz. Warszawa: Wyd. C.H.
Beck.
Pietrzkowski, H. 2005. Zarys metodyki pracy sędziego w sprawach cywilnych. Warszawa: Wyd.
LexisNexis.

Siedlecki, W., Świeboda, Z. 2003. Postępowanie cywilne. Warszawa: Wyd. LexisNexis.

Smith, S. A. 1995. Culture Clash: Anglo-American Case Law and German Civil Law in Transla-
tion. In: Morris, M. (ed.) Translation and the Law. Amsterdam/Philadelphia: John Benja-
mins Publishing Company. 179–197.
www.workerscompensation.com/maine/forms/index.htm
Sources
Bakó, Sz., Bihary, Á., Kovács, K., Klauszmann, K., Ocsiné Naszádos, K. 2000. A polgári per-
rendtartás iratmintatára. Budapest: KJK Kiadó.
Black, H. C. 1979. Black’s Law Dictionary. Fifth Edition. United States of America: St Paul
Minn. West Publishing Co.
Garner, B. A. (ed.) Black’s Law Dictionary. 2004. Eighth Edition. United States of America:
Thomson West.
Csorba, T. (ed.) 1985. Wielki słownik polsko – węgiersk. Budapest: Akadémiai Kiadó.
Łozińska-Małkiewicz, E., Małkiewicz, J. 1997. Polsko–angielski słownik terminologii
prawniczej. Toruń: Ewa.

DOI: 10.1556/Acr.11.2010.1.5
LEGAL TRANSPLANT AND CULTURAL

TRANSFER: THE LEGAL TRANSLATION
IN HONG KONG
LING WANG
The Chinese University of Hong Kong, Shatin, Hong Kong
Phone: +852 26097725
E-mail: ling@arts.cuhk.edu.hk
Abstract: This paper examines the relationship between legal transplant and legal
translation and their roles in cultural transfer. It classifies legal transplants into two kinds:
legal imposition at the socio-political level and legal translation at the socio-linguistic level.
Legal translation is usually the major conduit of legal transplant in the case of legal reform
in the receiving country. Since transplantation involves the transfer of the conceptual think-
ing of the imported law, legal transplant often brings about a transfer of legal culture. Legal
translation as a form of legal transplant always involves the transfer of the legal culture of
the translated law at the socio-linguistic level. Legal translation in the context of Hong
Kong serves as a case of foreignization, requiring both linguistic and conceptual adjustment
of the translating language to accommodate the imported culture.
Keywords: law and translation, cultural transfer, legal transplant, foreignization
1. LEGAL TRANSPLANT AND LEGAL CULTURE
Transposition of law from one society to another is generally known as legal

transplant. While this is an extensively researched area in comparative law in
recent years, legal transplant is as old as the law itself. Earlier legal transplants
such as the transposition of Roman laws to Europe offer a well-known example
(Watson 1974). Since transplantation involves the transfer of the conceptual
thinking of the imported law, legal transplant often brings about a transfer of
legal culture. We will examine the relationship between legal transplant and le-
gal translation, the causes of legal transplant, and its role in the transfer of legal
culture. In so doing we hope to arrive at a better understanding of the theoretical

84 LING WANG
background surrounding the transfer of legal culture in legal translation in gen-

eral and, in particular, the legal translation in Hong Kong.
Watson, the founding scholar in modern legal transplant theory, described
it as “the moving of a rule or system of law from one country to another, or
from one people to another” (1974:21). Watson identified legal transplant with
legal borrowing and argued that the phenomena of legal transplant had “been
common since the earliest recorded history” (1974:21). According to Watson
(2001a), the object of legal transplant is rules. By “rules” he meant not just
statutory rules but also institutions, legal concepts and structures.1 Watson
(2001b:215) held that legal borrowing can take place between societies with
very different political, social, economic and religious conditions and that usu-
ally the borrowing is from the more developed and complex system.2 In analyz-
ing Watson’s concept of legal transplant, Cotterrell (2001:71) held that com-
parative legal history is the primary tool of legal research and borrowing is usu-
ally the major element in legal change.3
Legrand disagreed with Watson’s definition of legal transplant, however,
contending that Watson reduced it to the movement of “rules – which are usu-
ally not defined, but which are conventionally taken to mean legislated texts and,
though less peremptorily, judicial decisions” (2001:55–56). He argued that legal
transplant was in essence impossible since “what can be displaced from one ju-
risdiction to another is, literally, a meaningless form of words” (2001:63). Cot-
terrell agreed with Legrand’s criticism, remarking: “… an emphasis on legal
culture may highlight the difficulty or even impossibility of transplants, since a
legal culture is not easily replaced by a different one and legal rules are under-
stood in relation to legal cultures” (2001:78).
Cotterrell (1997) rejected the concept of legal culture and proposed to re-
place it with the term “legal ideology”. He was thus naturally opposed to the
concept of legal transplant, since this hinged largely on a proper definition of
legal culture. Watson, however, refuted Legrand’s view:
(…) I believe I have shown that massive successful borrowing is common-
place in law (…) Legal borrowing I would equate with the notion of legal
transplant. I find it difficult to imagine that anyone would deny that legal
borrowing is of enormous importance in legal development. Likewise I
find it hard to imagine that anyone would believe that the borrowed rule
would operate in exactly the way it did in its other home (…) I have con-
tinually over more than a quarter of a century insisted that what are bor-
rowed and can be borrowed are legal rules, principles, institutions and even
structure. (Legrand 2001:23–24)
In characterizing the relationship between legal tradition and legal culture

for the development of his legal transplant theory, Watson remarked: “The an-

THE LEGAL TRANSLATION IN HONG KONG 85
swers for understanding the nature of law and its place in society can only be
found in the legal tradition and legal culture” (1991:4).
We shall see that Cotterrell’s dismissal of the concepts of legal culture and
legal transplant is not convincing. Just as legal rules should be understood as an
element of legal culture, the concept of legal culture should be understood as an
indispensable component in legal transplant. Though Watson may well not have
defined legal culture, he did consider it as the basis for understanding the nature
of law and legal transplant. Since we define legal culture in this study as the
conceptual thinking of legal professionals of which legal rules are an integral
part, it is fair to say that legal rules are also an inseparable part of legal culture,
and thus of legal transplant. Though legal transplants may not always be viable,
we cannot simply dismiss them as impossible. History and a fair part of com-
parative law studies show that legal transplants have indeed taken place.
2. LEGAL TRANSPLANT: LEGAL IMPOSITION

AND LEGAL TRANSLATION
Legal transplant takes place for many reasons, such as authority, prestige, po-
litical and economic incentives, and may take different forms in different coun-
tries. In an attempt to explain the phenomenon, Sacco remarked:
There are two fundamental causes of imitation (i.e. legal transplantation):
imposition and prestige. Every culture that has faith in itself tends to spread
its own institutions. Anyone with the power to do so tends to impose his
own upon others. Receptions due to pure force, however, are reversible and
end when the force is removed. (Sacco 1991:398)
Likewise, Berkowitz observed that “some legal transplants were imposed dur-
ing occupation; others were part of a voluntary reform process initiated by the
law receiving country” (2001:8). A fairly wholesale transplantation of legal
systems is possible during an occupation even without any translation of the
imported law into the indigenous language. However, legal translation is usu-
ally the major conduit of legal transplant in the case of legal reform in the re-
ceiving country. Therefore, we classify legal transplants into two kinds in this
study: legal imposition at the socio-political level and legal translation at the
socio-linguistic level.
Comparative legal scholars have carried out extensive studies on the impo-
sition of law since the importation of foreign legal systems is widespread and
poses important theoretical problems. In search of a definition of legal imposi-
tion, Lloyd-Bostock distinguished between “externally imposed law and law
that accords with internalized norms” (1979:10). She remarked:

86 LING WANG
(...) externally imposed law would include cases ranging from particular
instances of law within an established legal system to the importation of an
entire legal system form another culture. It is debatable whether a defini-
tion of imposed law should introduce further distinctions between types of
cases, but there can be no doubt that explanation of compliance will need
to take account of the wider context in which law has been imposed.
(Lloyd-Bostock 1979:10)
Lloyd-Bostock opined that looking into the compliance with imposed law
would be an effective way of understanding the social consequences of legal
imposition. In seeking to define the term legal imposition, Okoth-Ogendo ob-
served that “the use of that phrase might imply concern merely with the norma-
tive and institutional legacies of colonialism” (1979:147). However, his own
view was that legal imposition encompassed “any situation where fundamental
change is contemplated in society through the medium of laws or legal institu-
tions whose content is clearly contrary to the perceived and accepted normative
order of those whose behaviour it seeks to regulate or change” (1979:147).
From this perspective, legal imposition resulting from colonialism always gives
rise to socio-political change in the society that receives the law. Okoth-Ogendo
went on to make an in-depth study of the imposition of English property law in
Kenya, pointing out that “legal imposition is a rampant practice in Africa” and
that the “imposition of law can be seen as an expression of dependency relations
between the Third World (the periphery) and industrialized nations (the metro-
politan centres)” (1979:148).
In a similar vein, Kidder pointed out that “the prototype of imposed law as
it seems most generally to be understood is the colonial situation, where legal
systems are imposed from dominant cultures and forced on indigenous popula-
tions” (1979:289). A case in point here is the imposition of common law in
British colonies in South East Asia. Accompanied by nineteenth-century colo-
nialism, the imposed law radically reshaped and pluralized the law of much of
Africa, Asia, and the Pacific. The research of Harding (2001) provides a thor-
ough description of legal transplant in South East Asia where the imposed law
survived. Following the lead of Watson and other likeminded scholars, he re-
marked: “(…) law in South East Asia has evolved out of legal transplantation,
which has, on the whole, been successful, if judged by the criterion of whether
the law has stuck or come unstuck. In South East Asia the idea that the history
of a system of law is largely a history of borrowing of legal materials from other
legal systems, as maintained by Watson, Pound and others, is proved remarka-
bly accurate” (2001:213).
The wholesale transplant of the common law system in Southeast Asia
also includes the case of Hong Kong since English law was imposed on Hong
Kong after 1843. In Hong Kong, the legal transplant met with a rather benign

reception, and as Epstein (1989) noted, there was little interaction between
Hong Kong’s legal system and the laws of the Chinese Mainland after coloniza-
tion. He remarked:
For a century and a half, British colonial rule has insulated Hong Kong’s
legal system from law and legal change on the Chinese mainland. Although
early provision was made for the application of Qing dynasty law in Hong
Kong, in practice, the Qing codes had little if any impact in Hong Kong
after 1841 and even the role of customary law has been restricted to family
matters and land tenure in the New Territories (quoted in Wacks, 1989:38).
Wesley-Smith (1994) held a different view with regard to the influence of

Chinese customary law, however. In the process of legal transplant, colonial of-
ficials typically endeavoured to eliminate customs they considered repugnant,
such as polygamy, payback killings, suttee and many other kinds of practices
they considered uncivilized. Yet customary laws continued to have some effect
both in Hong Kong and many other countries.4 Wesley-Smith noted that “Chi-
nese law and custom’, despite its decline as a source of law − much of it was
abolished, prospectively, in 1971 − still plays an important role in modern Hong
Kong” (1994:205). In the process of legal imposition, conflicts often emerge
between the indigenous and the imposed law, although, as already noted, the
imposition of the common law on Hong Kong was a fairly well-received legal
transplant, with the imposed law meeting little resistance when it began to
regulate the behaviour of the indigenous inhabitants. However, the legal culture,
i.e. conceptual thinking about the common law, could reside only in the minds
of legal professionals before the common law was translated into Chinese. The
transfer of the legal culture related to the imposed law became fully affective
until the conceptual thinking of the imposed law is translated into the indige-
nous language and made accessible to local laypeople.
3. LEGAL TRANSLATION AND THE TRANSFER OF LEGAL

CULTURE IN HONG KONG
Legal translation as a form of legal transplant always involves the transfer of the
legal culture of the translated law at the socio-linguistic level. It takes place
when a country or region borrows the legal system of another, usually accom-
panied with massive translation of the imported law. Through legal translation
the concepts of the foreign law are introduced to the indigenous people. Com-
pared with legal imposition, legal translation as a form of legal transplant is a
more fruitful way of transplanting legal systems and transferring foreign legal
culture since it imports the underlying legal concepts into the indigenous lan-

88 LING WANG
guage. As Zhang (2003:9) pointed out, legal transplant by translation is the

most common phenomenon in the course of legal development in many coun-
tries. Its history can be traced back to the Old Roman Period. After the medieval
period, many western European countries such as France and Germany trans-
planted Roman law by way of translation. In modern times, many Asian, Afri-
can and American countries have transplanted the laws of western countries
(2003:9). Japan’s legal development also illustrates how the improvement and
modernization of one state’s law may occur through translation. During the
Meiji period, there was massive translation of continental European laws into
Japanese, and their reception took place in a completely non-European cultural,
juridical and religious context.
When Hong Kong became a British colony in 1842, the British brought
along a whole lot of “culture-specific” things, tangible and intangible, of which
the common law was one. The legal transplant, as in the case of Hong Kong,
could pre-empt many of the questions that triggered long debate among scholars.
First, the transplant was not from one jurisdiction to another – it was carried out
within the same common law jurisdiction as Hong Kong became a common law
jurisdiction the moment the British flag was hoisted (or legally, even earlier).
Second, for nearly a century and a half, the law was in the same language as its
home state, namely English. Third, the law was administered and practised by
professionals from its home state, or from other common law jurisdictions, or
from the local community who spoke and were trained in the same language of
the law. In a word, except for some adaptations in areas such as marriage and
succession, the common law was transplanted to Hong Kong en bloc by way of
legal imposition. The translation of the common law into Chinese was therefore
carried out with the purpose of transferring the culture of common law into the
Chinese language.
Successful legal transplant by translation always involves the transfer of
legal culture, i.e. the conceptual thinking of the legal professionals, into the tar-
get language. In the common law, legal concepts are lexicalized or expressed by
legal terms. The translator has to identify the concept and the referent that the
word in the source language represents. In some cases, the legal concepts that
are expressed by the legal terms do not exist in Chinese. There are no words in
Chinese to express some of the most elementary notions of the common law.
The terms “the common law” and “equity” are only two of the examples. There
is no system of “the common law” and “equity” in the Chinese legal system
(either in the PRC or in Taiwan). In addition, many types of institutions specific
to the common law have no direct counterparts in China, e.g. “Magistrate”,
“Lands Tribunal”, and many others. In other cases, partial-equivalent terms also
pose difficulties for the legal translator since one legal term can have both a
specific legal meaning and an ordinary meaning at the same time, e.g. the term

“consideration”. An equivalent for the ordinary meaning, which is shared in

Chinese, can be found, but the specific legal meaning does not exist in Chinese.
Could such a Chinese equivalent, if selected as the translation, convey the same
legal meaning in the common law? For example, transferring the expression
used for seemingly similar institutions, e.g. “high court”, risks blurring the dif-
ferences between these institutions. The common law term “high court” could
be translated into Chinese as gaodeng fayuan (高等法院). However, this very
term as used in the PRC refers to a different legal institution operating under a
socialist legal system. Therefore, the Chinese equivalent gaodeng fayuan (高等
法院) for the common law term “high court” certainly does not mean the same
as the Chinese term gaodeng fayuan (高等法院) as it is already used in the PRC.
Gaodeng fayuan (高等法院) as the translation for the common law term can
only be properly construed with reference to the common law system.
Therefore, translation as cultural transfer usually requires that a choice is
first made between two basic translation strategies, namely domestication and
foreignization, through which the cultural concepts of source language may ei-
ther remain untransferred or be transferred. Foreignizing translation is best ex-
emplified in legal translation since it purports to transfer the cultural concepts of
the source legal text. Legal culture, i.e. the conceptual thinking shared by legal
professionals, is an essential yet inseparable component of any legal system.
Legal translation as foreignization should be understood as translation as a
process of importing, or even transplanting, the culture of the source language
into the culture of the target language. The process necessitates the assimilation
of the legal concepts of foreign laws, as is the case with legal transplants by
translation in China. China has a long history of legal transplants dating back to
the Late Qing dynasty when China transplanted the German system of civil law.
Next, Japan’s legal experience exerted great linguistic and practical influence
on China’s reception of civil law before 1949. Moreover, China transplanted the
Soviet Union’s legal ideas after 1949. The history of legal transplants by trans-
lation in China can usefully elucidate the role that legal translation has played.
Thus in foreignizing translation, the target language accommodates the alien
concepts and adapts to the source culture. The legal translation in Hong Kong
no doubt involves the transfer of the common law culture into Chinese, which
requires both linguistic and conceptual adjustment of the translating language to
accommodate the imported culture.
Notes
1
Watson, A. “Legal transplant and European Private Law”. Ius Commune Lectures on
European Private Law, 2 (electronic version). Dutch Institute of Comparative Law.
(http://www.ejcl.org/44/art44-2.html, accessed on March 15, 2008)

90 LING WANG
2
Watson (2001b:215) also pointed out that law in writing was an obvious source for bor-
rowing: the reception of Roman law (and of canon law) in Western Europe, and the success of the
Sachsenspiegel in medieval Germany, of the French code civil in Europe and Latin America are
all powerful examples.
3
Cotterrell (2001:71–72) summarized Watson’s views thus: firstly, transplantation of legal
rules between legal systems was a principal explanation for the growth of law; secondly, social
need was not the decisive force in legal development; thirdly, legal changes were largely con-
trolled by the internal legal professional elites; fourthly, legal rules survived over long periods
despite significant variation in the social context on which they operate; fifthly, the development
of some important bodies of law was largely the result of legal history.
4
Harding (2001:210) gave an example from Singapore where the famous case known as the
Six Widows Case tried by the colonial court raised a crucial question of what kind of law was to
be applied in cases involving local custom: the common law or the customary law? The court fi-
nally decided the case according to the Chinese polygamous marriage custom.
References
Berkowitz, D., Pistor, K. & Richard, J. F. 2001. Economic Development, Legality and the Trans-
plant Effect. http://www.pitt.edu/~dmberk/bpreerfinal.pdf.
Cotterrell, R. 1997. The Concept of Legal Culture. In: Nelken, D. (ed.) Comparing Legal Cultures.
Aldershot: Dartmouth.
Cotterrell, R. 2001. Is There a Logic of Legal Transplants? In: Nelken, D. & Feest, J. (eds) Adapt-
ing Legal Cultures. Oxford: Hart Publishing. 71–72
Epstein, E. J. 1989. China and Hong Kong: Law, Ideology, and the Future Interaction of the Legal
Systems. In: Wacks, R. (ed.) The Future of the Law in Hong Kong. Hong Kong: Oxford
University Press. 37–76.
Epstein, E. J. 1998. Codification of civil law in the People’s Republic of China: form and sub-
stance in the reception of concepts and elements of western private law. The University of
British Columbia Law Review Vol. 32. 153–198.
Harding, A. 2001. Comparative Law and Legal Transplantation in South East Asia. In: Nelken, D.
& Feest, J. (eds) Adapting Legal Cultures. Oxford: Hart Publishing. 199–222.
Kidder, R. 1979. Toward an Integrated theory of Imposed Law. In: Burman, S. B. & Harrell-Bond,
B. E. (eds) The Imposition of Law. New York: Academic Press. 289–306.
Langer, M. 2004. From Legal Transplants to Legal Translations: The Globalization of Plea Bar-
gaining and the Americanization Thesis in Criminal Procedure. Harvard International Law
Journal. Vol. 1. Winter. 1–64.
Legrand, P. 2001. What ”Legal Transplants"? In: Nelken, D. & Feest, J. (eds) Adapting Legal
Cultures. Oxford: Hart Publishing. 55–69.
Lloyd-Bostock, S. M. 1979. Explaining Compliance with Imposed Law. In: Burman, S. B. &.
Harrell-Bond, B. E. (eds) The Imposition of Law. New York: Academic Press. 9–26.
Okoth-Ogendo, H. 1979. The Imposition of Property Law in Kenya. In: Burman, S.B. &
Harrell-Bond. B. E. (eds) The Imposition of Law. New York: Academic Press. 147–166.
Sacco, R. 1991. Legal Formants: A Dynamic Approach to Comparative Law. American Journal
of Comparative Law Vol. 39. No. 1. 1–34.
Watson, A. 1974. Legal Transplants: An Approach to Comparative Law. Edinburgh: Scottish
Academic Press.
Watson, A. 2001a. Legal transplants and European Private Law. Electronic Journal of Compara-
tive Law (December 2000) Vol. 44. Website: http://www.ejcl.org/ejcl/ 44/44-2.html.

Watson, A. 2001b. The Evolution of Western Private Law. Baltimore: The Johns Hopkins Uni-
versity Press.
Wesley-Smith, P. 1993. The Common Law of England in the Special Administrative Region. In:
Wack, R. (ed.) Hong Kong, China and 1997: Essays in Legal Theory. Hong Kong: Hong
Kong University Press. 5–40.
Wesley-Smith, P. 1994. The Sources of Hong Kong Law. Hong Kong: Hong Kong University
Press.
Zhang, Z. Q. 張中秋. 2003. Bijiao Shiye zhongde Falu Wenhua 比較視野中的法律文化. Bei-
jing: Falu Chubanshe 北京市: 法律出版社.

DOI: 10.1556/Acr.11.2010.1.6
PERCEIVED BENEFITS OF TRANSLATOR

CERTIFICATION TO STAKEHOLDERS IN THE
TRANSLATION PROFESSION:
A SURVEY OF VENDOR MANAGERS1
ANDY LUNG JAN CHAN
Division of Language Studies, City University of Hong Kong
Phone: +852 2788 9118
E-mail: andyljchan@gmail.com
Abstract: Despite a “social turn” in the field of Translation Studies, there have been
few systematic and descriptive studies of the translation profession. In this paper, the theo-
retical framework of information economics is utilized to examine the perceived benefits of
translator certification to stakeholders in the translation profession. The vendor managers
(i.e. the persons who are responsible for hiring translators in translation companies) were
surveyed using an online questionnaire. The translation companies surveyed are a fairly di-
versified sample of the population of translation companies in the world representing small,
medium-sized and large players. The respondents see translator certification as something
that can enhance the overall image of the translation profession. In their view, certification
does bring some benefits to translators who hold it (e.g. increase in number of job offers,
higher self-esteem and respect from co-workers). However, increased monetary benefits
brought by translator certification might be minimal. Translation companies generally wel-
come a system of translator certification, as it provides a relatively reliable signal of appli-
cants’ linguistic ability, and this has made the recruitment process easier and more time-
efficient.
Keywords: information economics, information asymmetry, signalling, translation

profession, translator certification, professionalization
1
This article is adapted from part of my PhD dissertation at the Universitat Rovira i Virgili,
Tarragona, Spain. I would like to thank my supervisors Anthony Pym and Jiri Stejskal for their
help and guidance and two anonymous referees for their constructive comments that substantially
improved the paper.

94 ANDY LUNG JAN CHAN
1. INTRODUCTION
Translation Studies has been quite frequently described as an “interdiscipline”

(e.g. Toury 1980, 1995; Snell-Hornby 1988). Within the field of Translation
Studies, there were conceptual and methodological changes in the late 1970s
and 1980s. After this “cultural turn” in Translation Studies, translation began to
be seen as a cultural process and a number of disciplines have been used to en-
rich the relatively young discipline. As Weissbrod (1998) points out, it is pre-
cisely the incorporation of new perspectives that will help systemic approaches
to Translation Studies to survive and strive.
In its early history, translation research focused on texts, particularly liter-
ary texts (Lambert 1998). It was not until recent years that more attention was
paid to the social contexts in which translation activities take place. Some
scholars state that a “social turn” (Wolf 2006) is emerging in the field of Trans-
lation Studies. Despite this social turn, the study of the translation profession is
still relatively uncharted territory. As compared with other professions or even
semi-professions, the study of translation as a profession is still very limited
(Hermans and Lambert 1998; Gouadec 2007; Dam and Zethsen 2008). In the
words of Chesterman (1998), “Translation Studies (…) [is] not concerned
enough with the real problems at the messy grassroots of life in a big translation
company”. In this paper, the theoretical framework of information economics is
utilized to examine the perceived benefits of translator certification to stake-
holders in the translation profession.
In Section 2 of this paper, a theoretical framework of asymmetric informa-
tion and signaling is offered. It is also explained how this framework can be ap-
plied to the translation profession and translator certification. Section 3 de-
scribes the overall research procedures of the current research project and Sec-
tion 4 discusses the findings.
2. THEORETICAL FRAMEWORK
Information economics is a branch of economics that studies how information
affects economic decisions. This theoretical framework is believed to have great
explanatory power and the 2001 Nobel Prize in Economics was awarded to
three researchers (George Akerlof, Michael Spence and Joseph Stiglitz) for their
contributions to information economics. A key concept from information eco-
nomics for our purpose is asymmetric information. In a state of information
asymmetry, one party in a transaction has more or better information than the
other on which to base decisions. This creates an imbalance in power, which
can adversely affect the transaction.

PERCEIVED BENEFITS OF TRANSLATOR CERTIFICATION 95
George Akerlof, in his seminal paper “The Market for Lemons: Quality
Uncertainty and the Market Mechanism” (1970), uses the example of second-
hand cars to illustrate the problem of asymmetric information. Bad cars are
called “lemons” while good cars are called “plums”. Because it is almost im-
possible for a potential buyer to tell exactly the difference between a good used
car and a bad one by just looking at them, sellers may sell both at the same price
(which is below the “real” worth of the “plums”).
Likewise, in the market of translation service, it is usually difficult for ser-
vice buyers or clients to assess the skills of a translator before any business rela-
tionships have been established. This is particularly true if we take into consid-
eration the fact that the turnover rate for freelance translators is high, as there
are many service providers entering and leaving the translation market. Most of
the time, service buyers or clients can assess the quality of translation only after
they receive the translation or when the translation is used in actual communica-
tion acts for various purposes. Therefore, translation service is an experience
good which is distinguished by the fact that “its quality, and hence its value to
consumers, cannot be precisely determined by buyers at the time of purchase”
(Moav and Neeman 2004:1).
Because of the asymmetric information problem previously illustrated,
translation service buyers cannot effectively distinguish between a “good”
translator and a “bad” one when they need to recruit new translators for their
work. Therefore, they tend to pay a price that is below the “fair” level. It is
worth pointing out however that we are assuming a fixed number of translators
in the market and it needs to be emphasized that the terms “good”, “bad” and
“fair” are for illustration purposes only and based on operational assumptions.
Let us assume good translators are worth $10 for a certain number of
words and bad translators $5. As clients do not have complete information
about the quality of translators they first work with, they are only willing to pay
according to a certain probability distribution. Assuming a probability of 50%
for both types of translators, i.e. half are good and half are bad, clients are only
willing to pay $10(1/2)+$5(1/2) = $7.5. (This distribution is actually a little con-
servative. We often hear complaints about poor translations, but good transla-
tions are seldom praised.) Therefore, the probability that clients will expect a
translator to be bad might be higher than 50%, and the price paid may thus be
even lower, for instance, $10(1/4)+$5(3/4) = $6.25.
One possible result is that most good translators will leave the market and
poorer translators will stay. This phenomenon is called “adverse selection”
(Stiglitz and Walsh 2002). In the illustration above, good translators should be
paid $10 but they receive only $7.5 because of the asymmetric information
problem. We can also observe many highly skilled translators set up their own
agencies and then farm jobs out to freelancers. In this case, the former devote
more time and effort to project management and marketing than to translation

work. Besides, some translators might acquire further academic qualifications

and become translator trainers, as teaching usually gives them a more stable in-
come and probably a higher status.
The problem of asymmetric information in the translation market is detri-
mental to the de facto professionalization of translators. Bell (2000) distin-
guishes between a pseudo-profession (one that resembles or imitates another
profession), para-profession (one that bears a subsidiary and supporting rela-
tionship to a “true” profession), and proto-profession (one that is at an early and
primitive stage of professional development), and points out that translators
should “reject out of hand the ‘pseudo-’ label, resist the servile implications of
the ‘para-’ label and focus on the ‘proto-’ to develop more sophisticated organi-
zation which will grow, in time, to a fully fledged profession” (Bell 2000:147).
One possible way to tackle this problem is to use an effective signal. Ac-
cording to Michael Spence (1973, 1974) who first proposed the theory, signal-
ing is the idea that one party (the agent) conveys some meaningful information
about itself to another party (the principal). In the translation market, translator
certification is often suggested as a signal to differentiate good translators from
bad ones if devised and implemented in a valid and reliable manner. In an
online symposium in January 2000 organized by the Intercultural Group of the
Universitat Rovira i Virgili, Tarragona, Spain titled “Innovation in Translator
and Interpreter Training”, most discussants agreed on the importance of accredi-
tation and certification (Englund Dimitrova 2000). Most of them believed that it
would lead to the professionalization of translators, improving pay and working
conditions, as the development of well-established profession prototypes like
doctors, lawyers, and accountants has shown.
In fact, many countries, especially those in Asia, have developed or are in
the process of developing their own translator certification systems. Even more
mature certification systems offered by well-established professional translator
associations have introduced a number of new initiatives such as mandatory
continuing education requirements to further develop the existing system.
3. RESEARCH PROCEDURES
The population of this translation industry certification survey is the vendor
managers, i.e. the persons who are responsible for hiring translators in transla-
tion companies. In July and August 2006, about 375 e-mail messages were sent
to vendor managers listed on ProZ and TranslatorsCafé, the two largest online
portals for translators, to ask for their assistance in completing the questionnaire
on my3q.com. This is a popular and stable platform for online questionnaires
and respondents should be able to access this site and complete the survey
easily.

Stratified sampling is used in this case to “[ensure] that appropriate num-

bers of elements are drawn from homogeneous subsets of that population”
(Babbie 2007:206). In other words, translation agencies of different sizes are in-
cluded in this survey. For ProZ, translation companies are classified into: fewer
than 5 in-house employees, 6–10 on-site staff, 11–20 on-site staff, 21–40 on-site
staff, and more than 40 on-site staff. For TranslatorsCafé, translation agencies
are classified into small, medium-sized and large, based on a number of criteria
like the number of offices the agency has in various countries, the number of
working languages of the agency, how actively the agency seeks translators, etc.
It is worth pointing out that many links to the translation agencies on ProZ and
TranslatorsCafé are broken. This, to a certain extent, re-affirms the free entry
and exit nature of the translation market.
In addition, e-mails were sent to 487 translation agencies around the world,
using the American Translators Association List of Translation Agencies and
Translators’ Handbook, from June to August 2006. With the assistance of Jiri
Stejskal, the survey was also sent to the translation company division members
of the American Translators Association in June 2006.
The questionnaire contains 34 questions. The first part asks the respondents
to provide information about themselves and their company. The second part is
to elicit responses from the vendors regarding their perception of the effective-
ness of translator certification. Research subjects were asked whether they agree
with some statements related to translator certification on a five-point Likert
scale (1 = strongly disagree, 2 = somewhat disagree, 3 = neither agree or dis-
agree, 4 = somewhat agree, 5 = strongly agree). The questionnaire is reproduced
in the Appendix.
4. FINDINGS AND DISCUSSION

This section reports on the questionnaires sent to persons responsible for hiring
translators in translation companies around the world through the channels de-
scribed above. It asks about views regarding translator certification and its rela-
tionships with translation companies, translators and the translation profession
as a whole. The first sub-section provides a description of the background of re-
spondents (including their job title, the number of years in the current position
and the number of years hiring translators) as well as their companies. The sub-
section that follows looks at some of the hiring characteristics of translation
companies in this survey. There is one question asking which country the com-
pany mainly operates in. Other questions are about the situations of the compa-
nies in hiring in-house and freelance translators. Lastly and most importantly,
research findings regarding the perceived benefits (or the perception of a lack of
benefits) of translator certification to various stakeholders in the translation pro-
fession are presented.

In the questionnaire, the respondents who are responsible for hiring transla-
tors were asked to provide their job titles. The results are shown in Table 1.
Table 1
Job titles of the respondents
Job Title Number (Percentage)

Project Manager 25 (35.71)
CEO/Managing Director 11 (15.71)
Owner/President 8 (11.43)
Operations Director/Marketing Director 7 (10.00)
Translation Manager 7 (10.00)
Senior Translator/Translator/Editor 4 (5.71)
Human Resources Manager 1 (1.43)
Others 6 (8.57)
Not Answered 1 (1.43)
TOTAL 70 (100)
There are a number of interesting observations to make. Perhaps not sur-

prisingly, the most popular title for persons making the translator hiring deci-
sion is Project Manager (25/70). This category includes other similar titles like
Account Manager, Vendor Manager, Project Coordinator or simply Manager. It
was somewhat unexpected that the title Vendor Manager is used only three
times, as vendor management has been a much discussed area in the global
translation and localization industry (DePalma and Beninatto 2008). It is gener-
ally believed that the job duties of a vendor manager are to find, manage and
monitor language service providers and improve relations with them.
It is also worth pointing out that 29/70 respondents are Chief Executive Of-
ficer, Managing Director, President or are the company owners. One may sus-
pect that there are quite a few “mom-and-pop” businesses in this pool and their
owners like to give themselves fancy titles. But, in fact, among these, there are
large companies with 2,500 active freelance translators as well as small compa-
nies with two in-house translators and six freelancers.
In the sample, the mean number of years the respondents have been in their
current position is 6.90, and they have been responsible for hiring translators for
a mean number of 7.94 years. We can interpret this as meaning that the respon-
dents in this sample are fairly experienced in recruiting translators and they
were responsible for hiring before they took up their current position.
In Question 4, the respondents were asked in which country their company
mainly operates. The results are shown in Table 2.

Table 2
Countries or regions the companies mainly operate in
Countries/regions the company mainly operates in Number (Percentage)

USA 15 (21.43)
Hong Kong and China 10 (14.29)
Spain 7 (10.00)
Israel 5 (7.14)
United Kingdom 5 (7.14)
Baltic states 4 (5.71)
Latin America 4 (5.71)
Worldwide 3 (4.29)
Italy 2 (2.86)
Australia and New Zealand 2 (2.86)
Others 10 (14.29)
Not Answered 2 (2.86)
Not Applicable 1 (1.43)
TOTAL 70 (100)
From the table we can see that this group of translation companies is a
fairly diversified sample of the population of translation companies in the
world. First and foremost, the United States accounts for more than one-fifth of
the total sample. North America is arguably the largest segment in the interna-
tional translation industry, and according to the estimate of the Common Sense
Advisory (2005), North America accounted for 42% of the global translation
market in revenue in 2004. To account for this, on one hand, as of 2008, the
United States accepts more legal immigrants as permanent residents than any
other country in the world. In that year alone, the number of persons obtaining
legal permanent resident status totaled 1.1 million (US Yearbook of Immigra-
tion Statistics 2008). On the other hand, the United States lays claim to the larg-
est number of global corporate giants. In the Global 500 Companies survey car-
ried out by the Fortune magazine, 140 of the largest companies in the world
were American in 2008 (Fortune 2009).
Secondly, Hong Kong and China have 10/70 responses. Business activities
in China are booming and the People’s Republic of China is increasingly be-
coming a regional, or even international, software localization centre. Accord-
ing to one estimate (Hong Kong Trade Development Council 2003), the annual
turnover of translation and localization business in China is about €1 billion.
The European market is also quite well represented in this sample (18/70),
with the inclusion of Spain (7/70), the United Kingdom (5/70), Baltic states

(4/70) and Italy (2/70). Latin America and Australasia are also represented in
this sample with 4/70 and 2/70 responses, respectively. Of course, some of these
may be regional branches of international companies.
There are quite a number of translation companies in the category “Oth-
ers”. The countries and regions include South Africa, Malaysia, Thailand and
Singapore, which have shown a varied representation of the international trans-
lation industry. One point that needs emphasizing is that some companies state
that they operate on the Internet, so it might be difficult to categorize which
country they mainly operate in. This shows the increasing influence of the
Internet on the translation business and the emergence of the global village.
However, Canada is also one of the major translation markets, but it is not rep-
resented in this sample.
Question 5 asked the respondents whether their company hires in-house
translators. Thirty-four (34/69) of the translation companies in our sample hire
in-house translators and thirty-five (35/69) said that their companies do not hire
in-house translators. One respondent did not answer this question. The mean
number of staff translators for those who do hire in-house translators is only
6.03. The standard deviation is 8.15, meaning that the difference in the number
of in-house translators hired is quite large. Indeed, most translation companies
hire just one or two staff translators, and only 5/34 of them hire more than 10
in-house translators. The two companies that hire the highest number of in-
house translators mainly operate in the UK and Spain. The UK company has 40
staff translators and, somewhat surprisingly, no freelancer. The Spanish com-
pany has 29 staff translators and 30 active freelancers.
Question 7 asked the respondents whether their company hires freelance
translators. Only four (4/70) of the translation companies stated that they have
not hired any freelancers. After close scrutiny, it was found that only two are
valid responses, because the other two said that they hire neither in-house trans-
lators nor freelance translators. Between the two valid responses, one is the
British company mentioned in the previous paragraph and the other is probably
a “mom-and-pop” translation house in China which has only three translators.
In Questions 8 and 9, respondents were asked the number of freelance
translators they have recruited and the percentage of active freelancers. On av-
erage, the translation companies that hire freelancers have 680 translators in
their pool. The standard deviation is astonishingly high, about 1,224 translators.
The top three employers have recruited 6,000, 5,000 and 4,000 translators re-
spectively, and they are all American companies. In addition, there are eight
medium-sized companies that hire 2,000 freelancers or more. Companies that
hire 100 freelancers or fewer account for the rest (59/70) and they can be re-
garded as small companies. Respondents were asked whether the translators in
their freelance pool are active. The percentage of active translators ranges from

10% (meaning only one-tenth of their freelancers are active) to 100% (meaning
all their freelancers are active) and the mean is 59.13%.
From the questionnaire results, we can see that hiring freelancers has al-
ready become a rule rather than an exception in the translation industry. Virtu-
ally all companies hire freelance translators. However, in-house translators are
not a thing of the past, and talk about the “extinction” of staff translators may be
an exaggeration. There might be two reasons for this. First, in-house translators
are still needed for editing, revision and quality control, as well as offering co-
ordination and administrative support. Second, the presence of staff translators
can act as a signal of quality to the clients. Clients sometimes ask for the num-
ber of in-house translators versus freelancers, in order to evaluate the quality
and reliability of translation service providers.
Moreover, if we use the number of freelancers as a proxy for the size of
translation companies, it can be said this sample consists of large, medium-sized
and small companies. The three US companies with more than 4,000 freelancers
can be considered large. In fact, in recent years, as a result of merger and acqui-
sition activities in the translation industry, a few big market players have been
increasing their market share in the global translation industry. Of course, there
are also many small translation companies in this sample. They can be repre-
sented by those hiring 100 or fewer freelancers, which account for slightly less
than half of the total sample. In between, there are also a few medium-sized
companies, and they may be represented by those having 2,000 to 4,000 free-
lancers in their database.
Another point worth mentioning is that a number of respondents failed to
provide figures regarding the number of in-house translators and freelance
translators. The respondents may consider the information sensitive and have
some reservations about revealing company data.
Questions 10 and 11 asked the respondents about the degree of difficulty in
recruiting in-house and freelance translators.
Table 3
Degree of difficulty in hiring in-house and freelance translators
Strongly Somewhat Somewhat Strongly

Question Mean SD Neither (3)
Disagree (1) Disagree (2) Agree (4) Agree (5)
10. The company

finds it easy to
hire suitable in-
house translators. 2.67 1.24 12 (21.05%) 14 (24.56%) 18 (31.58%) 7 (12.28%) 6 (10.53%)
11. The company
finds it easy to
hire suitable free-
lance translators. 3.30 1.42 8 (12.12%) 17 (25.76%) 6 (9.09%) 17 (25.76%) 18 (27.27%)

In general, the respondents thought that it is more difficult to recruit in-

house translators than freelancers. As shown in Table 4, 35/66 of the respon-
dents believed that it is relatively easy for them to hire freelance translators, but
only 25/66 of the respondents strongly or somewhat agreed that hiring in-house
translators is a relatively straightforward task. Also, the view is more clear-cut
for freelance translators, as far fewer respondents chose “Neither” as their an-
swer.
Table 4
Whether the newly hired translators increasingly have translator certification

12. Newly hired

translators increas-
ingly have certifi-
cation (e.g. MCIL,
ATA, NAATI,
CATTI) 3.01 1.37 12 (17.65%) 14 (20.59%) 15 (22.06%) 15 (22.06%) 12 (17.65%)
These survey results make sense, as hiring in-house translators is a more

long-term and committed investment for the companies. Direct labor costs may
include salaries, medical benefits and travel allowances. Moreover, as there are
peak and off-peak seasons in the translation market, hiring freelance translators
may provide more flexibility to the translation companies. And as some transla-
tion companies provide services in dozens of languages, they cannot have in-
house translators for every single language. More importantly, as revealed, most
translation companies hire only one or two in-house translators. Therefore, staff
translators may need to be responsible for ensuring the quality of the translation
services provided by different freelance workers. These few in-house translators
may need to be, in a sense, masters of all trades. No wonder recruiting in-house
translators is more difficult in the eyes of our respondents.
Questions 13 and 14 ask the respondents the minimum years of experience
they require and prefer their translators to have, respectively. The companies’
minimum requirement is 2.71 years (standard deviation 1.79 years) and their
preference is 3.61 years (standard deviation 1.79 years). It is worth pointing out
that 6/65 respondents do not require their applicants to have any experience but
almost all respondents prefer their applicants to have some years of experience.
The respondents’ preference for job applicants with experience can be explained
as reflecting that if translators come with some on-the-job training, there will be
less need for the translation companies to provide training opportunities. Thus,
training costs can be minimized for the employers.

Question 15 asks the respondents whether their newly hired translators in-
creasingly have translator certification.
As discussed above, more and more countries have developed or are in the
process of developing their own translator certification system. However, as il-
lustrated in Table 5, it is not certain that the newly hired translators increasingly
have translator certification, as the responses of agreement (26/68) are almost
equal to the ones of disagreement (27/68), while 15/68 of them answered “Nei-
ther”.
It might be interesting to see how large translation companies (defined as
those that have 2,000 or more freelance translators in their pool) and small ones
(defined as those that have 50 or fewer freelance translators in their pool) an-
swered this question. Large companies tend to “strongly agree” that the newly
hired translators increasingly have translation certification (arithmetic mean and
standard deviation are 3.82 and 1.17, respectively) while small companies
“somewhat disagree” (arithmetic mean and standard deviation are 2.47 and 0.96
respectively) with the statement. This may be interpreted as suggesting that
large companies tend to employ more translators with certification while small
ones do not. However, it is also possible that the translators who have earned
their certification believe that they are more competitive and apply for a posi-
tion of freelancer in large companies instead of small ones.
Table 5
Identification of job applicant’s linguistic and subject knowledge through translator certification

15. Translator
certification
makes it easier to
identify appli-
cant’s linguistic
knowledge. 3.46 1.35 9 (13.24%) 10 (14.71%) 6 (8.82%) 27 (39.71%) 16 (23.53%)
16. Translator
certification
makes it easier to
identify appli-
cant’s subject
knowledge. 3.03 1.39 14 (20.59%) 11 (16.18%) 13 (19.12%) 19 (27.94%) 11 (16.18%)
Translation service companies are important stakeholders in the translation

market. If the translator certification system is to be successful, support and
even input are needed from them. Therefore, how translation service companies
perceive the translator certification system is an important area to explore.
Questions 15 to 20 ask the respondents whether they think that translator certi-

fication can make the identification of a job applicant’s linguistic knowledge

and subject knowledge easier, and whether they believe that translator certifica-
tion can facilitate the recruitment process and reduce the staff attrition rate.
From Table 5, which shows the responses to Questions 15 and 16, it is
quite clear that most respondents believe that translator certification makes it
easier to identify the applicant’s linguistic knowledge. In answer to Question
15, 43/68 strongly or somewhat agree that they can have a better understanding
of the applicant’s linguistic knowledge through translator certification. The
arithmetic mean and standard deviation are 3.46 and 1.35, respectively.
However, the respondents were not so certain when they were asked
whether translator certification can help them identify applicants’ subject
knowledge. The arithmetic mean is 3.03, indicating that most respondents have
answered “Neither” and/or the answers “Strongly Disagree/Somewhat Dis-
agree” offset those of “Strongly Agree/Somewhat Disagree”. As shown in Table
6, this is indeed the case, as 13/68 of the respondents chose “Neither”. Those
who disagree (30/68) with the statement “Translator certification makes it easier
to identify applicant’s subject knowledge” somewhat exceed those who agree
with it (25/68).
Table 6
Perceived benefits to the recruitment process and staff retention through translator certification

17. Translator
certification
makes the
recruitment
process easier. 3.46 1.31 9 (13.24%) 10 (14.71%) 6 (8.82%) 27 (39.71%) 16 (23.53%)
18. Translator
certification
makes the
recruitment
process
cheaper. 2.35 1.21 14 (20.59%) 11 (16.18%) 13 (19.21%) 19 (27.94%) 11 (16.18%)
19. Translator
certification
makes the
recruitment
more time-
efficient. 3.21 1.40 12 (17.91%) 9 (13.43%) 13 (19.40%) 19 (28.35%) 14 (20.90%)
20. Translation
certification
reduces the
likelihood of
staff turnover. 2.58 1.13 15 (23.08%) 14 (21.54%) 20 (30.77%) 15 (23.08%) 1 (1.54%)

From Table 6, we can see that about 43/68 of the respondents either
strongly agree or somewhat agree that “Translator certification makes the re-
cruitment process easier”, and 33/67 either strongly agree or somewhat agree
that “Translator certification makes the recruitment more time efficient”. How-
ever, when it comes to cost considerations, the respondents are not as certain.
The arithmetic mean for the statement “Translator certification makes the re-
cruitment process cheaper” is only 2.35, meaning that, on average, the respon-
dents do not believe that translator certification can reduce the expenditure of
the recruitment process.
Although the recruitment of more freelancers in the language service in-
dustry may increase the flexibility of personnel needs, it may not significantly
reduce recruitment costs. One possible reason is that the costs of recruitment are
not simply the hiring costs, such as the costs of advertising on various printed
media and the Internet, but also include the indirect costs of a bad hire (Bellizzi
and Hasty 2000). The costs of a bad hire are difficult, if not impossible, to
measure. Increasing quality assurance costs and even reduction in returned
business may be included. Therefore, as more translators have translator certifi-
cation, it might be easier and more time-efficient for language service compa-
nies to locate potential vendors.
In answer to Question 20, “Translation certification reduces the likelihood
of staff turnover”, only one respondent strongly agrees. The arithmetic mean is
2.58. In general, staff turnover is considered undesirable for translation compa-
nies, as it involves various costs. The costs of staff turnover can be significant
and may include administrative costs associated with the leaver’s separation
from the company, administrative costs associated with creating a vacancy,
costs of covering the vacancy until it is filled, advertising and agency costs for
the vacancy, costs involved in the selection process (including but not restricted
to the administration and evaluation of translation tests) and training costs.
According to Table 7, the respondents agreed that a person with translator
certification gets more job offers. In answer to Question 21, 38/66 of the re-
spondents either strongly agree or somewhat agree with the statement. The
arithmetic mean is 3.52. When asked whether a person with translator certifica-
tion is more likely to be hired at the respondents’ firm, the respondents are not
as certain. The mean is only 3.14, and 12/65 of the respondents strongly dis-
agree with the statement. These two results may be interpreted as follows: al-
though translator certification may be a well-appreciated addition to the appli-
cant’s résumé, when a company makes recruitment decisions, other factors
might be considered as well. A very interesting observation is that even fewer
respondents believe that a person with translator certification would receive
higher pay. The arithmetic mean is 3.08. Although 28/66 of the respondents ei-
ther strongly agree or somewhat agree with the positive relationship between
translator certification and pay, 11/66 have the extreme view that translator cer-

tification does not lead to higher pay. To some extent, this shows there is keen
competition in the language service industry and, unlike other professions, al-
most anyone with a university degree and some translation experience can enter
the translation service market.
Table 7
Perceived monetary benefits of translator certification to translators

21. A person with

translation
certification gets
more job offers. 3.52 1.21 5 (7.58%) 9(13.64%) 14 (21.21%) 23 (34.85%) 15 (22.73%)
22. A person with
translator
certification is
more likely to be
hired at my firm. 3.14 1.39 12 (18.46%) 10 (15.38%) 12 (18.46%) 19 (29.23%) 12 (18.46%)
23. A person with
translator
certification
receives higher
pay. 3.08 1.27 11 (16.67%) 9 (13.64%) 18 (27.27%) 20 (30.30%) 8 (12.12%)
Table 8
Perceived benefits of translator certification to translators as regards
the indication of language proficiency and subject knowledge

24. A person with

translator
certification has
better language
proficiency. 2.88 1.26 12 (17.65%) 15 (22.06%) 17 (25.00%) 17 (25.00%) 7 (10.29%)
25. A person with
translator
certification has
better subject
knowledge. 2.75 1.25 13 (19.12%) 17 (25%) 19 (27.94%) 12 (17.65%) 7 (10.29%)
Questions 24 and 25 ask the respondents whether a person with translator

certification has better language proficiency and subject knowledge, respec-
tively. According to Table 8, in general, the respondents do not agree with these

two statements. Twenty-seven (27/68) of the respondents either strongly dis-

agree or somewhat disagree with the statement on language proficiency and
thirty (30/68) with the statement on subject knowledge. This difference is in line
with Questions 15 and 16, which ask respondents whether they agree with the
statements “Translator certification makes it easier to identify applicant’s lin-
guistic knowledge” and “Translator certification makes it easier to identify ap-
plicant’s subject knowledge”.
Table 9
Perceived non-monetary benefits of translator certification to translators
Strongly
Somewhat Somewhat Strongly
Question Mean SD Disagree Neither (3)
Disagree (2) Agree (4) Agree (5)
(1)
29. A person with

translator
certification earns
the respect of co-
workers. 3.23 1.13 7 (12.28%) 7 (12.28%) 16 (28.07%) 22 (38.60%) 5 (8.77%)
30. A person with
translator
certification is more
satisfied with his or
her job. 3.13 0.93 5 (8.77%) 6 (10.53%) 27 (47.37%) 17 (29.82%) 2 (3.51%)
32. A person with
translator
certification has a
higher sense of self-
esteem. 3.62 1.03 4 (7.14%) 2 (3.57%) 17 (30.36%) 24 (42.86%) 9 (16.07%)
Questions 29, 30 and 32 look at the perceived non-monetary benefits of

translator certification to translators. These include peer respect, job satisfaction
and self-esteem. As shown in Table 9, the respondents believe that translator
certification is likely to increase a translator’s peer respect (mean = 3.23) and
self-esteem (mean = 3.62). It should be noted that about 33/56 of the respon-
dents either strongly agree or somewhat agree that translator certification leads
to higher self-esteem. However, far fewer respondents believe that translator
certification can lead to more job satisfaction (mean = 3.13) and most of them
said “Neither”. This makes sense, as job satisfaction may be due to a number of
reasons and translator certification may not be a relevant factor.
Questions 27, 28, 31 and 33 ask the respondents about the perceived bene-
fits of translator certification with regard to translators’ long-term career devel-
opment. According to Table 10, quite a large proportion of respondents either
strongly agree or somewhat agree with Questions 27 and 28, that translators
with certification are more committed to their career (33/55) and less likely to

leave the translation profession (33/56). The means for these two questions are
3.50 and 3.32, respectively. Having translator certification can be interpreted as
having the intention to develop a relatively long-term career in translation.
However, whether that career will be successful might be a different story. That
is why respondents give a much lower rating for Question 31, which states that
“A person with translator certification is more likely to have a successful ca-
reer”.
Table 10
Perceived benefits of translator certification with regards to long-term career development
Somewhat
Strongly Somewhat Strongly
Question Mean SD Disagree Neither (3)
Disagree (1) Agree (4) Agree (5)
(2)
27. A person with
translator certifica-
tion is more able to
deal with change oc-
curring in the indus-
try. 2.84 1.21 14 (24.14%) 6 (10.34%) 18 (31.03%) 16 (27.59%) 4 (6.90%)
28. A person with
tion is more likely to
have a successful ca-
reer. 3.27 1.21 8 (13.79%) 5 (8.62%) 19 (32.76%) 18 (31.03%) 8 (13.79%)
31. A person with
tion is more commit-
ted to his or her ca-
reer. 3.50 1.10 5 (9.09%) 2 (3.64%) 15 (27.27%) 24 (43.64%) 9 (16.36%)
33. A person with
tion is less likely to
leave the translation
profession. 3.32 1.15 6 (10.71%) 3 (5.36%) 14 (25.00%) 25 (44.64%) 8 (14.29%)
One striking observation is that the respondents do not agree that people
with translator certification would be more able to deal with change occurring
in the industry. For Question 33, the mean is at a low of 2.84. Twenty (20/58)
either strongly disagree or somewhat disagree, and eighteen (18/58) answer
“Neither”. This is understandable, as translator certification is usually granted
after a one-off pencil-and-paper translation test. For example, the two well-
established professional translator associations, the American Translators Asso-
ciation [ATA] and the Chartered Institute of Linguists [IoL], United Kingdom,
require candidates to take written tests in an examination hall. However, it is
worth pointing out that for some professional translators associations such as
the Japan Translation Association (2008), candidates are now able to take their
certification test online.

In the real world, professional translators need to be able to use translation

memories and to work with machine-translated texts. What is more, as transla-
tion projects require collaboration and co-operation among translators, revisers
and desktop publishing professionals and have become increasingly team-based,
the inauthentic setting of conventional translation tests might not be an adequate
response to the changes in the translation marketplace. This may explain to
some extent why translator associations have paid more attention to continuing
professional development (CPD) and some even make CPD a mandatory re-
quirement if their members desire to reach a higher status in the membership hi-
erarchy. The designation of Chartered Linguist of the Chartered Institute of
Linguists is a vivid example (Chartered Institute of Linguists 2007). Of course,
as in many other professions, CPD is still a relatively new phenomenon and
there is still much work for the translation profession to make these efforts cater
better to market developments.
Table 11
Perceived benefits of translator certification to the translation profession
34. A person with
translator certification
enhances the image
of the profession. 4.06 0.98 3 (5.17%) 0 (0.00%) 9 (15.52%) 26 (44.83%) 20 (34.48%)
The last question, Question 34, asks the respondents to evaluate the bene-
fits of translator certification to the translation profession. It has the highest
arithmetic mean of all the questions, which is 4.06, and the standard deviation is
relatively small (0.98). A higher percentage of respondents believe that a person
with translator certification enhances the image of the profession: 46/58 of re-
spondents either strongly agree or somewhat agree with the relationship be-
tween translator certification and the image of the translation profession.
5. CONCLUSION AND LIMITATIONS

In sum, we can say that in this sample of small, medium-sized and large transla-
tion companies from around the world, the respondents reported that they main-
tain a pool of freelancers but still hire a few in-house translators. They see trans-
lator certification as something that can enhance the overall image of the trans-
lation profession. In their view, certification does bring some benefits to transla-
tors who hold it (e.g. increase in the number of job offers, higher self-esteem
and respect from co-workers). However, increased monetary benefits brought

by translator certification might be minimal. Translation companies generally

welcome a system of translator certification, as it provides a relatively reliable
signal of applicants’ linguistic ability, and this has made the recruitment process
easier and more time-efficient.
Last but not least, it is also important to point out the limitations of this
online quantitative study. Obviously, there are many advantages of collecting
survey data via the Internet. Apart from cost- and time-efficiency, one of the
most relevant advantages to this research is that the Internet can possibly gener-
ate a global sample and provide access to respondents across geographic
boundaries. However, online surveys have the following limitations and accom-
panying constraints. First, the response rate is low. Among the 375 e-mail mes-
sages sent to vendor managers, only 70 questionnaires were returned represent-
ing a response rate of 18.7%. According to Shaughnessy, Zechmeister and
Zechmeister (2006), the response rates of online surveys were sometimes 90%
before 2000 but have been dropping fast to between 2% and 30% in recent
years. If we compare the response rate of this survey with these figures, it is not
particularly low.
Perhaps a more serious problem is the validity of self-reported data in
online surveys. As noted by Hewson et al. (2003), when questionnaires are ad-
ministered via the Internet, rather than in person, the researcher is less able to
judge the extent to which responses are sincere and genuine. In the research
findings regarding the hiring characteristics of translation companies (Question
8), there was a large translation company which claimed that it has 2,500 active
freelance translators. According to Dr. Jiri Stejskal, President of American
Translators Association, only companies such as Lionbridge (probably the larg-
est multinational which hired 4,600 full-time staff in 2007) language service
provider can possibly have this number of active freelancers. Therefore, it is
quite likely that a few companies may exaggerate their number of freelancers to
make it look like they are large.
In short, the statistical significance of these findings should not be over-
stated. Caution needs to be taken when we make claims that involve generaliza-
bility and representativeness. That having said, well-conducted surveys of a
limited scale still have a contribution to make. As pointed out by Punch
(2003:39), “we can learn something useful from almost any sample as long as
the research is carefully and thoroughly carried out”.
References
Babbie, E. 2007. The Practice of Social Research. Belmont, CA: Thomson Wadsworth.
Bellizzi, J. A. & Hasty, R. W. 2000. The Effects of Hiring Decisions on the Level of Discipline
Used in Response to Poor Performance. Management Science Vol. 38. No. 3. 154–159.
Bell, R. 2000. Pseudo-, Para- or Proto-: What Kind of a Professional is the Translator or Inter-
preter?”. The Linguist Vol. 39. No. 5. 147–150.

Chartered Institute of Linguists, United Kingdom. 2007. Chartered Linguist. Available at:
http://www.iol.org.uk/Charter/cls.asp, last checked on: July 1. 2009.
Chesterman, A. 1998. Description, Explanation, Prediction: A Response to Gideon Toury and
Theo Hermans. In: Schäffner, C. (ed.) Translation and Norms. Clevedon: Multilingual Mat-
ters. 91–98.
Common Sense Advisory. 2007. Ranking of Top 20 Translation Companies for 2006. Available
at: http://www.commonsenseadvisory.com/members/res_cgi.php/070502_Q_Top_20.php,
last checked on: July 1. 2009.
Dam, H. V. & Zethsen, K. K. 2008. Translator Status: A Study of Danish Company Translators.
The Translator Vol. 14. No. 1. 71–96.
DePalma, D. A. & Beninatto, R. S. 2008. Localization Vendor Management. Lowell, MA: Com-
mon Sense Advisory.
Englund Dimitrova, B. E. 2003. Summary of Discussion on “Accreditation”. In: Pym, A., Fallada,
C., Biau, J. R. & Orenstein, J. (eds) Innovation and e-Learning in Translator Training. Tar-
ragona, Spain: Intercultural Studies Group, Universitat Rovira i Virgili. 34–35.
Fortune Magazine. 2009. Global 500. Available at: http://money.cnn.com/magazines/ for-
tune/global500/ 2009/countries/US.html, last checked on: December 15. 2009.
Gouadec, D. 2007. Translation as a Profession. Amsterdam and Philadelphia, MA: John Benja-
mins Publishing Company.
Hermans, J. & Lambert, J. 1998. From Translation Markets to Language Management: The Im-
plications of Translation Services. Target Vol. 10. No. 1. 113–132.
Hong Kong Trade Development Council. 2003. Translation Generates a Multi-Billion Dollar
Business. International Market News. 3 April, 2003. Available at: http://www.hktdc.com/
imn/03040301/markettrends072.htm, last checked on: July 1. 2009.
Japan Translation Association. 2008. Available at: http://www.jta-net.or.jp, last checked on: July
1. 2009.
Lambert, J. 1998. Literary Translation. In: Baker, M. (ed.) Routledge Encyclopedia of Translation
Studies. London and New York: Routledge. 130–133.
Moav, O. & Neeman, Z. 2004. Inspection in Markets for Experience Goods. Discussion Paper Se-
ries dp349, Jerusalem: Center for Rationality and Interactive Decision Theory, Hebrew
University. Available at: http://ratio.huji.ac.il/dp/dp349.pdf, last checked on: July 1. 2009.
Punch, K. F. 2003. Survey Research: The Basics. London: SAGE Publications.
Shaughnessy, J. J., Zechmeister, E. B. & Zechmeister, J. S. 2006. Research Methods in Psychol-
ogy. Cambridge, MA: Harvard University Press.
Snell-Hornby, M. 1988. Translation Studies: An Integrated Approach. Amsterdam and Philadel-
phia, PA: John Benjamins Publishing Company.
Spence, A. M. 1973. Job Market Signaling. Quarterly Journal of Economics Vol. 87. No. 3. 355–
374.
Spence, A. M. 1974. Market Signaling: Informational Transfer in Hiring and Related Screening
Processes. Cambridge, MA: Harvard University Press.
Stiglitz, J. E. & Walsh, C. E. 2002. Economics. New York: W. W. Norton & Company.
Toury, G. 1980. In Search of a Theory of Translation. Tel Aviv: Porter Institute for Poetics and
Semiotics.
Toury, G. 1995. Descriptive Translation Studies and Beyond. Amsterdam and Philadelphia, PA:
John Benjamins Publishing Company.
US Yearbook of Immigration Statistics. 2008. Available at: http://www.dhs.gov/files/statistics/
publications/LPR08.shtm, last checked on: July 1. 2009.
Weissbrod, R. 1998. Translation Research in the Framework of the Tel Aviv School of Poetics
and Semiotics. Meta Vol. 43. No. 1. 35–45.
Wolf, M. (ed.) 2006. Übersetzen – Translating – Traduire: Towards a “Social Turn”. Berlin &
Vienna: LIT Verlag.

Appendix
Translation industry certification survey

1. What is your job title? __________________
2. How long have you been in your current position? _____________ years
3. In total, how many years have you been hiring translators? __________ years
4. In which country does your company mainly operate? __________
5. Does your company hire in-house translators? Yes/No
6. If yes, how many in-house translators does your company currently employ? ____
7. Does your company hire freelance translators? Yes/No
8. If yes, how many translators are there in your freelance pool? ____
9. What percentage of them are active freelancers? _____%
To what extent do you agree or disagree with the following statements? Circle one.

Neither
Disagree Disagree Agree Agree
In general, our company finds it

easy to hire suitable in-house
translators. 1 2 3 4 5
In general, our company finds it

easy to hire suitable freelance
translators. Increasingly, newly
hired translators have certifica-
tion (e.g. MCIL, ATA, NAATI,
CATTI) 1 2 3 4 5
How many months or years of

job experience in translation is
your company’s minimum
requirement? _______________ months ________ years
How many months or years of

job experience does your com-
pany prefer for newly hired
translators? ________________ months ________ years

With specific reference to the organization you currently work for, how much do you agree or
disagree with these statements about the benefits of translator certification?

Translator certification… Neither
Disagree Disagree Agree Agree
Makes it easier to identify applicant’s
linguistic knowledge 1 2 3 4 5
Makes it easier to identify applicant’s
subject knowledge 1 2 3 4 5
Makes the recruitment process easier 1 2 3 4 5
Makes the recruitment process
cheaper 1 2 3 4 5
Makes the recruitment more time
efficient 1 2 3 4 5
Reduces the likelihood of staff
turnover 1 2 3 4 5
Thinking specifically about translators with certification, e.g. MCIL, ATA, NAATI, CATTI
(compared to those without translator certification), how much do you agree or disagree with
these statements? Circle one.
A person with translator Strongly Somewhat Somewhat Strongly

Neither
certification Disagree Disagree Agree Agree
Gets more job offers 1 2 3 4 5
Is more likely to be hired at my firm 1 2 3 4 5
Receives higher pay 1 2 3 4 5
Has better language proficiency 1 2 3 4 5
Has better subject knowledge 1 2 3 4 5
Receives more customer complaints 1 2 3 4 5
Is less able to deal with change
occurring in the industry 1 2 3 4 5
Is more likely to have a successful
career 1 2 3 4 5
Earns the respect of co-workers 1 2 3 4 5
Is satisfied with his/her job 1 2 3 4 5
Is more committed to the career 1 2 3 4 5
Has a higher sense of self-esteem 1 2 3 4 5
Is more likely to leave the translation
profession 1 2 3 4 5
Positively enhances the image of the
profession 1 2 3 4 5
Thank you for your cooperation!

DOI: 10.1556/Acr.11.2010.1.7
ACROSS-Interview
Speculation vs Empirical Research

Sonja Tirkkonen-Condit interviewed by Pál Heltai
The past three decades or so have seen intensive development in Translation

Studies, with many different theories and methods appearing (and proliferating)
in a discipline that used to be rather speculative. One important, or perhaps the
most important development has been the move from speculation to empirical
research. In this, Sonja Tirkonnen-Condit has played an eminent role, doing and
inspiring first-rate empirical research, providing evidence for (or disconfirming)
hypotheses in translation theory. People seem to know a lot about Sonja’s work
and the work of the researchers she has fostered at Savonlinna, but rather less is
known about her professional career, which made Savonlinna a centre of mod-
ern translation studies. To fill this gap, Pál Heltai interviewed her via e-mail.
Heltai: While there are many references to your work in scholarly journals,
there is very little information on your personal background on the internet. You
spent a long time at Savonlinna. Does that mean that you come from that part of
Finland? Could you tell us something about your early education, motivations,
interests, language learning experience?
Tirkkonen-Condit: I was born in Varkaus, where I also went to school.

Varkaus is an industrial city only 100 kilometres from Savonlinna. After finish-
ing school I was accepted by the University of Helsinki to study English, but I
decided, instead, to go to the Helsinki School of Economics. The reason for this
decision was that instead of focusing on just one language I could take up four!
That’s how I ended up in a school which was said to have “a managerial spirit”
hovering above it. My study programme was a three-year B.A. course focusing
on business language. I had English, Swedish, German and Spanish as main
subjects.
Heltai: But you were not interested in management? Your main interest contin-
ued to be languages?
Tirkkonen-Condit: I did not choose the School of Economics because of its

managerial spirit but because it offered a wide range of languages. I soon real-
ized that it was a childish choice.

116 ACROSS-INTERVIEW
Heltai: Your first and second degrees were in economics. How and why did
you switch to text linguistics, English philology and translation studies?
Tirkkonen-Condit: The proportion of economics in my first degree was rather

small, and the level of economic studies was shallow, and it dawned on me
rather early that the career prospects for a person with such a combination of
subjects were secretarial rather than managerial.
Heltai: So you simply wanted to get a higher degree, and once you had a degree
in economics, you continued in economics?
Tirkkonen-Condit: That’s right. My M.A degree required advanced studies in

economics, business administration, accounting and law. My major subject was
English, however, and I wrote my M.A. thesis on English verbs denoting con-
veyance of information. In English, there are more than 20 verbs that can func-
tion as translations of the Finnish verb ilmoittaa (inform, tell, advise, etc.) and
my aim was to account for the circumstances that determine their use. This
topic has relevance to translation, and in fact a career as a translator seemed at-
tractive to me at the time.
Heltai: That sounds like a topic in contrastive linguistics. Was translation still a
branch of contrastive linguistics at the time? Did you ever regard translation as
a matter of regular (linguistic) transfer operations?
Tirkkonen-Condit: Translation was part of all philological and linguistic cur-

ricula at the time. It had no academic or institutional status but was seen as a
practical exercise in the teaching of foreign languages, and of course, as a
method used by contrastive linguistics in its search for linguistic equivalence.
Heltai: Was your economics background an advantage to you in your later ca-
reer?
Tirkkonen-Condit: Yes, it was. After finishing my first degree at the age of 22

I worked as a secretary or secretary-translator for ten years. As I did not want to
be a secretary for the rest of my life I carried on my studies for the M.A.
Heltai: In economics or English?
Tirkkonen-Condit: In both, while working at the same time. I also qualified as

a teacher of English in a commercial college.
Heltai: Was it possible then to get a teaching degree in a college of economics?

ACROSS-INTERVIEW 117
Tirkkonen-Condit: I did my didactic studies and teaching practice in the Fin-

nish Businessmen’s Commercial College in 1970–1971. But there were other
interesting job opportunities as well. The first of these was a translator’s post in
the Department of International Affairs at the Ministry of Education. I had
worked at the Ministry for less than half a year, when the language institutes of
Turku and Savonlinna advertised new vacancies for lecturers in English.
Heltai: What does ‘language institute’ mean?
Tirkkonen-Condit: The Language Institutes were established in Finland in the

late 1960s and early 1970s according to the model from West Germany. They
were colleges financed by the Ministry of Education and offered three-year di-
ploma courses in translation. The diploma was considered to be equivalent of a
B.A. Finland had four such institutes: Turku, Tampere, Savonlinna and Kou-
vola. All of these were joined to universities in 1981. Turku to the University of
Turku, Tampere to the University of Tampere, Savonlinna to the University of
Joensuu, and Kouvola to the University of Helsinki. Their degree system was
integrated into the university system so that their first degree became a 5-year
M.A. degree. They also got their first professors in 1981.
Heltai: So you applied to Savonlinna?
Tirkkonen-Condit: I applied to both. I never got an answer from Turku, and

even now I do not know if my application got lost in the mail. But Savonlinna
appointed me, and so I left my job in the Ministry, and moved from Helsinki to
Savonlinna to start a very challenging and tough career at the Language Insti-
tute. Nine years later we joined the university system and became the
Savonlinna School of Translation Studies.
Heltai: Does this mean that Savonlinna became part of a university without any
other departments?
Tirkkonen-Condit: Savonlinna School of Translation Studies (former

Savonlinna Language Institute) became part of the Faculty of Humanities of the
University of Joensuu in 1981. In 2007 the University made a decision to move
Translation Studies to Joensuu, and already in the autumn of 2007 all new stu-
dents started their studies in Joensuu. The final move took place in the autumn
of 2009. So at the time of writing our Translation Studies are in Joensuu as part
of their Foreign Languages and Translation Studies Section in the Faculty of
Humanities. As from 1st January 2010 the Universities of Joensuu and Kuopio
will be joined together into the University of Eastern Finland, with the main

campus in Kuopio. Then Translation Studies will be a part of the Faculty of Phi-
losophy. Its geographical base will remain in Joensuu.
Heltai: So your teaching career began at Savonlinna?
Tirkkonen-Condit: Yes. I worked here as a lecturer of English until 1979,

teaching students of English. My teaching load included theory and practice of
translation and a variety of other subjects, such as consecutive interpreting and
business studies. It became clear to me very early that I knew virtually nothing
about translation theory or linguistics, and that I could not do my job properly
without further study.
Heltai: Didn’t your background in practical translation make you doubt the
value of theory?
Tirkkonen-Condit: In the late 1970s the Translation Theory I was aware of

was Nida and Catford, and the first practically and empirically oriented piece of
work that I laid hands on was Juliane House’s 1977 book on translation quality
assessment. One cannot say that theory had much to do with practice ... So I
studied Applied Linguistics at the University of Essex, first for one term in
1975 and then a whole academic year in 1977–78. I had married my colleague
Dr. Stephen Condit, who came to Savonlinna in 1974. Our daughter Outi was
born in April 1977, and our little family sailed to England in September. My
husband was finalising his PhD while minding our daughter.
Heltai: Who were the people at Essex University who influenced you most?
Tirkkonen-Condit: I was very impressed by Yorick Wilks, who was at Essex

at that time. He taught us Transformational Grammar, and I am sure he made it
more interesting than the true believers would have done.
Heltai: Wasn’t he a true believer? Did you become one? Do you think TG has
relevance for TS?
Tirkkonen-Condit: He is an Artificial Intelligence man, but the way he taught

TG was great fun. He was a good teacher. I enjoyed reading Chomsky’s Syntac-
tic Structures and Aspects, but I never thought that TG had much to do with
natural language translation. (In fact, neither does Chomsky.) Another impres-
sive teacher was Steve McDonough, who taught psychology and psycholinguis-
tics. And I did a project on the teaching of translation for John Roberts. My re-
port on this project was published in System, and this was my first publication. I

did my M.A. in Applied Linguistics with distinction, and would have liked to
carry on for a Ph.D. in Essex, if only I had had the money.
Heltai: But you did not, so you decided to do a PhD in Finland?
Tirkkonen-Condit: Yes, because in Finland you need not pay fees at univer-
sity. So very soon after coming back to my job at Savonlinna I applied to Jy-
väskylä for an assistant’s job, which made it possible for me to do research for a
licentiate’s degree. My official supervisor was Kari Sajavaara, and other col-
leagues with whom I could share my research problems were Liisa Lautamatti
and Raija Markkanen.
Heltai: Sajavaara has a good name in contrastice linguistics. Didn’t he want

you to go into Contrastive Linguistics?
Tirkkonen-Condit: No. But he told me that translation theory will be at a loss

until Contrastive Linguistics has solved the problem of linguistic equivalence.
Heltai: Do you think that time has proved him right? Isn’t translation theory at
a loss?
Tirkkonen-Condit: Whether translation theory is at a loss or not does not

hinge on issues of linguistic equivalence. Research on linguistics as well as ma-
chine translation has shown that natural language translation cannot even in
principle be fully computerized. In other words, it is not possible to create MC
software that generates hight-quality translations of natural language texts.
Translation is a pragmatic as well as linguistic operation, and however exhaus-
tive our contrastive descriptions might be, they cannot accommodate the prag-
matic information that translation inevitably involves.
Heltai: You were a professor of English before you were appointed Professor
of Theory of Language and Translation. What subjects did you teach in the De-
partment of English? Did you teach English syntax or literature? Or cultural
studies? How did they relate to your research interests?
Tirkkonen-Condit: I taught textlinguistics, pragmatics, theory of translation

and translation studies. I also taught some TG at some point, not because I be-
lieved in its relevance for translation but because I believed that students of
translation should be aware of the main approaches in linguistic theory. I also
supervised M.A. theses and conducted thesis seminars. I also got my first doc-
toral students while working as a professor of English.

Heltai: You have done work on quite a few topics: your first major work was in
text linguistics and translation (Argumentative Text Structure and Translation),
which, I think, made a major contribution to the field of text linguistics itself.
Why did you not continue this line in your further research?
Tirkkonen-Condit: My research topics seem to have developed in a natural

course. The topic for my licentiate’s thesis, translation quality assessment, de-
rived from my work as a teacher of translation. The evaluation of translation
quality was problematic, and empirical work in this domain was virtually non-
existent. The need to describe the structure of entire texts also goes back to
quality assessment. The evaluator needs an overall picture about the whole text
to be able to decide what weight should be given to a particular error or short-
coming in a translated passage, or the other way round, to a particularly suc-
cessful solution in another passage. I started by developing a model for the de-
scription of argumentative text structure, and ended up studying argumentation
as a cultural phenomenon. This is how I became familiar with the international
community of argumentation scholars such as Frans van Eemeren, Rob
Grootendorst and Sally Jackson.
Heltai: Do you think that translation quality assessment can be made objective
in the same way as many aspects of communicative ability have in foreign lan-
guage testing?
Tirkkonen-Condit: I suppose translation skills can be tested as objectively as

foreign language skills but translation quality must be assessed in each context
according to the requirements of the translation brief. It is not possible to give
absolute criteria for quality.
Heltai: You seem to have switched your main line of research several times,
from text linguistics to studying the process of translation by think-aloud proto-
cols, to research on translation universals using (and building) corpora, transla-
tionese, the monitor model of translation, subtitling ... Can you tell us how your
tastes developed over the years?
Tirkkonen-Condit: How did I end up studying translation processes? The first

impetus was in a little seminar organized by Juliane House in Hamburg as early
as 1984. There were some scholars who had collected think aloud data by
video-taping students who wrote a translation on the blackboard while thinking
aloud. When my first M.A. students were looking for topics for their theses, I
had the idea of think aloud as one option. There was one student, Riitta
Jääskeläinen, who took the challenge. Riitta did her M.A. thesis, licentiate’s
thesis and Ph.D. thesis on translation processes using think aloud data as mate-

rial. Our subjects in the pilot studies were students, but it soon became neces-
sary to get professional translators as subjects. Expertise in translation became
the next area of interest in my own research as well as Riitta’s.
Heltai: Does this mean that you have often drawn inspiration from teaching?
Tirkkonen-Condit: Sometimes the problems that come up in teaching may

give inspiration for research, but mostly it is research itself that gives inspiration
for further research.
Heltai: TS is concerned with the products and processes of translation. You

seem to have been concerned with both. How do you see the relation between
the two?
Tirkkonen-Condit: I see this relation as an essential issue in the theory of

translation. Translations of natural language texts come about as a result of hu-
man decision-making, and this decision-making needs to be accounted for by
TS. Corpus-based studies have revealed, for instance, that translated language
tends to be linguistically different from “monolingually” produced language.
Such tendencies have been referred to as translation universals. We cannot say
as yet, if the tendencies discovered so far from comparative corpora are indeed
universal or if they are confined to particular language pairs or language groups.
I have offered the monitor model as one explanation for the linguistic differ-
ences between translated and non-translated language. According to this model,
a professional translator has an automaton in his head which generates transla-
tional equivalents at the level of words, phrases and syntactic structures. If the
equivalent which the automaton offers does not fit the context, the consciously
controlled monitor prevents it from surfacing in the target text. But the monitor
has enough leakages so that translated language turns out statistically significant
differences compared to monolingual production. Traces of the monitor at work
are to be seen in the processes as revealed by think aloud or key-logged data of
translation.
Heltai: So according to this theory, how much does contrastive linguistics have
to do with translation?
Tirkkonen-Condit: According to this model it is possible to speculate that a

translator has a contrastive grammar and dictionary in his head!
Heltai: It seems to me that the monitor model is moving in the direction of psy-
cholinguistics. Is there research in psycholinguistics going onto meet psycho-
linguistics-oriented TS halfway?

Tirkkonen-Condit: Translation scholars are using psycholinguistic testing

methods, such as priming and eye tracking, and in that sense translation schol-
ars may consider themselves psycholinguists. One such translation scholar is
my former colleague Jukka Mäkisalo at Joensuu.
Heltai: And is anything happening in psycholinguistics?
Tirkkonen-Condit: In fact the borderline between psycholinguistics and psy-

cholinguistic translation research is fuzzy, and the interests of translation re-
search and psycholinguistics are united. Translation research with psycholin-
guistic methods sheds new light, e.g. on metaphor theory, bilingualism, and
writing research (Mandelblit 1996, Martikainen, forthcoming and Immonen
2007 and forthcoming).
Heltai: Several of your publications have titles containing the word “empiri-
cal”. What is the significance of this title-giving habit?
Tirkkonen-Condit: In the 1970s, when I started my work as a teacher of trans-

lation, most translation studies literature was theoretical and speculative, and of-
ten normative in its approach. Empirical research was scarce, and so it seemed
that there was a time to advertise its need in book titles and seminar pro-
grammes. And most of my own research is empirical in its orientation.
Heltai: Which does not mean that you abhor theory?
Tirkkonen-Condit: No true researcher can say that they abhor theory. There is
no research without theoretical assumptions, hypotheses, insights or hunches,
and an empirical project is also a theoretical statement.
Heltai: Much research in TS tends to have a practical motivation, and TS is

constantly trying to prove that it can help the practice of translation. Did you
have such motivations?
Tirkkonen-Condit: I felt the need for practically oriented TS very acutely

when I was working as a lecturer with a heavy teaching load of translation exer-
cises. By practically oriented I do not mean normatively oriented but empiri-
cally oriented. It seems that the most benefit for professional practice as well as
translation pedagogy comes from research that is based on authentic material
and from projects with ecological validity. Today we know much more of pro-
fessional translators’ work than we did 30 years ago. We know that there are
many different strategies to achieve high quality and that each student of trans-
lation should be encouraged to find his or her own strategy.

Heltai: What role did you play in organising TS at Savonlinna and in Finland?
My impression is that Finnish scholars play an important role in international
translation studies research. Do you agree? What gave rise to this development?
Am I right in suspecting that you yourself have played an important role in this?
Tirkkonen-Condit: In August 1985 I started as a deputy associate professor of

English at Savonlinna, and decided to take all the challenges that the job of-
fered. One of the challenges was to make Savonlinna a place which attracted
translation researchers from all over the world. Many people laughed at such a
childish idea. But at some point it seemed that this dream in fact had come true.
In the eighties and nineties there were no graduate schools in Finland, and doc-
toral students in different universities did not have many opportunities to meet
each other. To fill this gap I organised summer schools and seminars to which
students and researchers of translation studies were invited.
Heltai: You profited from your managerial skills, after all?
Tirkkonen-Condit: Maybe so. Many of those students are now university lec-
turers and professors. One of the big names was Gideon Toury, who visited
Savonlinna on several occasions. He was first invited to Savonlinna in 1988 as a
plenary speaker to our TRANSIF seminar. In this seminar, there were other par-
ticipants who are now well known but were not celebrities at the time: Kitty van
Leuven-Zwart, Christiane Nord, Miriam Schlesinger, Christina Schäffner,
Kinga Klaudy, Wolfgang Lörscher and Candace Séguinot. The other already
famous people were Nils Enkvist, Albrecht Neubert and Ghelly Chernov.
Largely thanks to these contacts the name of Savonlinna became fairly well-
known among translation scholars, even before we had grown our new crop of
researchers who now go to international conferences and publish their research
abroad.
Later on, the national graduate school system in Finland was established in
the mid 1990s. The graduate school LANGNET, for Ph.D. students in language
studies and linguistics, was established only in 1999. One of LANGNET’s pro-
grams is called Translation Studies and Professional Languages.
Heltai: How is TS in Savonlinna/Finland today? How do you see the status of

TS scholars in universities in Finland and elsewhere, and the status of TS as a
science? Do you think it will achieve the academic respectability of formal lin-
guistics?
Tirkkonen-Condit: As translation industry is growing with globalisation, one

would expect that translation technology benefits from this development. But as

long as translation technology operates largely in isolation from Translation

Studies, it does not seem to make much progress.
Heltai: Are you sure? Isn’t translation technology – CAT tools, translation
memories, etc. – making a lot of progress without looking up what is happening
in TS?
Tirkkonen-Condit: Translation Technology is making progress in isolation –

and so is Translation Studies. And still everyone knows that translation of natu-
ral language texts will never be possible without human contribution. Critical
research on CAT tools is still scarce, because many translation scholars abhor
technology. If TT and TS united their forces to argue for translation as a profes-
sion that requires human expertise, we might see a change in translators’ fees
and professional status. But as long as virtually anyone can present themselves
as translators without a sanction, the status quo will continue.) Since the 1970s
and 1980s Translation Studies has grown enormously in terms of university de-
partments, professors’ chairs, conferences and so on. Translation Studies in
Finland is fairly strong at the moment. There are professors and lecturerers spe-
cialising in TS in the universities of Helsinki, Tampere, Turku, Joensuu and
Vaasa. Translation Studies moved from Savonlinna to Joensuu a year ago, be-
cause geographically it was rather isolated in Savonlinna. In the Joensuu cam-
pus they are have close contacts to Foreign Languages, Psychology, Literary
Studies, etc., and they are doing very well in their new home. At the time of
writing they are preparing an international conference to celebrate their new
base in the University of Eastern Finland. However, this growth does not seem
to benefit the translation profession at all.
Heltai: Perhaps because theory is irrelevant for practical translation?
Tirkkonen-Condit: The graduates from translation schools do not agree with

this statement. Translation takes place on a free lance rather than in-house trans-
lation basis, and translators’ fees do not grow as they should with the greater ef-
ficiency brought about by CAT tools and translation technology at large.
Heltai: Is this because they do not maintain contacts with TS?
Tirkkonen-Condit: This is because translators are not properly organized. This

discrepancy has not disappeared, even though Translation Studies has become
more concerned with professional practices. It is still one of the greatest chal-
lenges of Translation Studies to make its message heard among the major deci-
sion-makers in translation industry.

Heltai: But what is the message of TS?
Tirkkonen-Condit: The message that translation is a profession is not pro-

claimed loud enough. As for translators themselves, I do not believe that their
financial or social status will improve until translation is made into a licenced
profession in the same vein as in medicine and jurisprudence. Trade unionism
among translators is still in its infancy.
Heltai: To return to part of my previous question, how do you see the status of
TS scholars in universities in Finland and elsewhere? Have they got the same
academic respectability as TG linguists? Does TS as a science (or field of
scholarship) have the same standing as other long-established humanities?
Tirkkonen-Condit: It was a major advancement for Translation Studies to

have a place in the national graduate school system. LANGNET ensures Ph.D.
students access to high quality courses and seminars not only in Translation
Studies but in other relevant domains of linguistic and humanistic research.
Now that also Language Technology has joined LANGNET we can hope that
the desired contacts will between TT and TS will be enhanced.
Heltai: Dear Sonja, here you sidestep my question about academic respectabil-
ity. Are you being diplomatic?
Tirkkonen-Condit: Respectability depends on the quality of research rather

than on any institutional declarations. We need high-quality research on theo-
retically interesting issues as well as issues relating to the recognition of transla-
tion as a profession.
Heltai: Is there anything else you would like to say, by way of summary, about
yourself and/or TS?
Tirkkonen-Condit: No.
Heltai: Thank you very much for the interview.

DOI: 10.1556/Acr.11.2010.1.8
Conference Report
Semiotics in Translation: Transferring Cultural Heritage

5th International Imatra Symposium on Semiotics and Translation
Imatra, Finland, June 6–7, 2009
Transferring Cultural Heritage: 5th International Imatra Symposium on Semiot-

ics and Translation took place in the framework of the International Summer
School for Semiotic Studies, Pan European Culture Heritage in the Global World,
in Imatra, Finland, June 6–7, 2009. The seminar dedicated to Translation and
Semiotics, Transferring Cultural Heritage: 5th International Imatra Symposium
on Semiotics and Translation, offered common ground for presenting research
carried out recently by and for people from many walks of life. Presentations on
the semiotic approaches to translation and interpreting served as an introduction
to discussions where both practitioners and researchers participated.
Lotman (in Torop 2008:375) holds: “the elementary act of thinking is
translation. (…) The elementary act of translation is dialogue”. So are the many
processes of thinking, intuition, and dialogue involved in translation and inter-
pretation. What are the parallels between semiotics and translation as seen
through the eyes of presenters?
The seminar was opened by Pirjo Kukkonen’s presentation Translation
on My Mind. Alluding to Ray Charles’ famous piece of music Georgia on My
Mind in her title, Kukkonen stresses that language and our thoughts constantly
remind us of or allude to something else. Referring to Susan Bassnett (1980),
who has stated that “everything is translation”, Kukkonen explains: “thinking is
translation, language is translation, we’ translate’ our thoughts, the world, texts,
culture, etc.” (Abstracts 2009:40). Kukkonen ties the translative idea of thinking
to the different kinds of translation presented in Jakobson (1959/1987: 429): in-
tralingual, interlingual, and intersemiotic translation. Kukkonen (Abstracts
2009:41) holds: “In these processes with the product in mind, we interpret the
source text in semiosis, in a never-ending heuristic process, but we have to stop
somewhere, and decide for semantic and cognitive strategies and solutions in
order to produce our translated or interpreted text, the target text, in another
language and its culture as a pragmatic, i.e. functional new text in its new situa-
tion of communication”. In translation the semiosis, intersemiosis and sign sys-
tems of different cultures meet, they are ‘translated’, ‘transferred’, and ‘medi-
ated’ by translators and interpreters (cf. ibid.). Thus, Kukkonen suggests that
semiotics and translation belong together and should be investigated at a deeper

128 CONFERENCE REPORT
level. Kukkonen invites more researchers to discover and share the results of
their research in the field.
Ritva Hartama-Heinonen’s presentation drew on the results of her re-
cently defended doctoral dissertation Abductive Translation Studies: The Art of
Marshalling signs (2008). Hartama-Heinonen’s presentation was called Semi-
otic Approaches to Translation: the Sixth Paradigm of Translation Studies, and
in tune with the heading, the author suggested that a new paradigm – the semi-
otic paradigm should be entered by translation studies. Not only would this
“embrace an enhancement from the viewpoint of translation studies”. It would
also be “a welcome proposal from the sign-theoretical vantage point, or the field
of semiotics as well”, proposes Hartama-Heinonen (Abstracts 2009:32–33).
Irma Sorvali’s analysis on Who is the Receiver of Translation? focuses
on, as the author puts it, “translations as intermediaries and gateways for his-
tory, literature and culture” (Abstracts 2009:71). In a pedagogically and intel-
lectually interesting visual representation of the translation process as a merry-
go-round, Sorvali explains her views on how far on the merry-go-round the
translation process drives, depending on who works on the text. Depending on
whether it is native readers just reading the text for themselves, translators, or
translation reviewers, the circle of the merry-go-round – in Sorvali’s view – be-
comes respectively longer each time. Sorvali also demonstrated how the wheel
analogy can be used for illustrating processes of translating different texts in
history. Sorvali’s presentation helps to illustrate different possibilities for the
visualization and understanding of the translation processes. It also serves as a
useful pedagogical tool for explaining some translation processes to students.
The afternoon session began with Ritva Hartama-Heinonen’s report
Translation and the Philosophy of Hope. As Hartama-Heinonen suggests: “rea-
soning is interpretation is translation is semiosis”. This is what Umberto Eco
has called an “unlimited semiosis”. Hartama-Heinonen puts forward the idea
that to translate is to resist, or to challenge. To translate is also to learn, to
search for truth, to lie. Translation is a matter of hope. Hartama-Heinonen refers
to Peirce, according to whom inquirers are “animated by a cheerful hope” to be
able to find solutions to all their puzzles (Abstracts 2009:32). Hartama-
Heinonen (ibid.) puts it in this way: “Abduction, or retroducive interference,
therefore “depends on our hope, sooner or later, to guess at the conditions under
which a given phenomenon will present itself”. Through this, Peirce’s theory
“evolves into a hopeful theory of truth, or even better, a theory of hope” (Ab-
stracts 2009:33–32). In research, when also viewing translation as a kind of in-
quiry, Hartama-Heinonen (ibid.) sees this Peircean hope as “manifested in
translational and translatorial Firstness: the possibility of creating abductive hy-
potheses as explanatory solutions, or possibilistic translations, the possibility of
there being deductible consequences and finding facts to confirm our conjec-
tural translations”.

CONFERENCE REPORT 129
Kajsa Wickström’s report Translating Dialects analyzed this challenge of

literary translators based on the example of Väinö Linna’s novel Tuntematon
sotilas. The well-known Finnish author has used a number of Finnish dialects to
convey the idiosyncracies of the different characters in the book who all come
from different regions in Finland. The Finnish Code of ethics for translators
stipulates that the translators should strive to preserve the style of the original.
Wickström studies the translator Nils-Börje Storbom’s cultural and linguistic
choices when rendering the lines into a Swedish dialect from the Kokkola re-
gion. The focus is on the translation of the character Hietanen’s dialogical utter-
ances. An expert in the dialect herself, Wickström holds that the translator has
very well succeeded in conveying the original. The same concerns the novel as
a whole. Wickström concludes that despite the advice sometimes given to trans-
lators not to translate dialects, her study helps to illustrate that successful liter-
ary translation from dialect into dialect can be achieved.
Martina Paatela-Nieminen took us into the colorful world of art. Her
analysis of the Subject in Artistic Process illustrated how modern art is chang-
ing our conception of art, culture and possibly also our world-view. Kristeva
has discussed the subject in culture stating that culture is being artified. Paatela-
Nieminen views the way post-modern art is artified. Paatela-Nieminen’s presen-
tation is based on the intertextual/contextual method she has developed in her
doctoral thesis in 2000 for the field of art education and arts.
As an introduction, Paatela-Nieminen explains the differences between the
modern and postmodern conceptions of art. Modern art is usually seen as
autonomous, independent from culture, as a self-enclosed entity. Post-modern
art, however, is seen as artified. Art is artifying culture. Culture changes art and
art changes culture. Post-modern art, as intertextual text, creates itself in and be-
tween social and cultural networks. In modern art, art historians were resorted
to as an elitarian medium for analyzing a work of art. In post-modern art mean-
ing is created in the viewer. The glocal (local + global) dimensions, the allu-
sions to past works of art and their allusions to past ideals play in. Paatela-
Nieminen (Abstracts 2009:60) interprets ‘a text’ as a visual, verbal, or other
communicative form of culture. Her focus is on the intertextuality – “the way
every text absorbs and transforms other texts”. Paatela-Nieminen holds that
“texts are shaped by immanent time and by divergent temporalities of history
and they are in continuous flux. Text is also a verb, or an action of linking, lay-
ering, weaving and producing connections between ideas, arts, cultures and
people” (ibid.). We are crossing the threshold: today people believe they can
study subjectively what they want. Paatela-Nieminen puts the question on how
far can one go in art, bringing inter-alia examples from the works of Sally
Mann exposing her family (cf. also Paulrud 2007). The borders are also fuzzy
concerning what is today “high art”. Since Andy Warhol “high art” may seem
to be kitsch. In this new world, you tube has become the “bad boys’” literature.

For expressing oneself we often no longer need essays but may speak in images
(as is also the case with graduation “theses” at a Finnish art college). In her
presentation, accompanied by illustrations from famous works of art from his-
tory and modern times, Paatela-Nieminen demonstrated the function of intertex-
tuality in post-modern art, but also in different projects of art. Through these,
she suggests, art can be seen as a medium for understanding one’s own culture
and society in both cultural and intercultural terms.
Päivi Ibl ended the first day of conference bringing us to the quickly
changing world and transformations of identity of Mexican-Americans in the
United States. Starting her discussion with the presentation of the translation of
Sandra Cisneros’ novel The House on Mango Street (1984), Ibl presented an
analysis of a corpus that includes the writer’s whole novelistic production. The
focus of Ibl’s research is on how Chicana/o writers convey their bilingual and
bicultural identity and surroundings, on the conscious or subconscious strategies
chosen for that. Special attention was given to intersentential Code switching as
a conscious strategy. Ibl analyzes the way the authors have chosen their code
“based on individual rationality”, on the other hand being “delimited in their
choices by external norms” (cf. Abstracts 2009:33–34); on how the solutions
are customized, negotiated, and optimized. The main focus was on how mark-
edness and using different linguistic registers yield subtle targeted social mes-
sages, through studying dialogues between texts and systems in the framework
of social semiotics.
The first day of the conference ended with the official opening ceremony
of the International Summer School for Semiotic Studies at Karelia Hall in Ima-
tra Cultural Centre and a reception at Imatra Town Hall.
The second day of the Symposium on Semiotics and Translation was intro-
duced by a lecture by Peeter Torop. His analysis of Semiotics of Translation in
the Context of Cultural Theory established the framework and broader back-
ground for the several topics discussed by presenters. Torop offered a well-
systematized review of the development of semiotics and translation. Torop also
analyzed the modern focuses in translation studies on power, ethics and agents
in a semiotic framework. Torop (Abstracts 2009:78–79) points out that the
“communication and a dialogue between addresser and addressee can be seen
on a deeper level as the autocommunication of culture itself”. Autocommunica-
tive processes, in their turn, “increase the coherence of a culture” (ibid.) and
help identity to be created, to develop and persist. At a meta-level, the on-going
descriptions of (our) cultures, societies, countries, and of different processes,
can be seen as autocommunication and also translation. As Torop (ibid.) sug-
gests, ”culture works in many respects as a translation mechanism and that me-
diation in culture involves both communication and autocommunication”
(ibid.). Thus, “translation semiotics is an important instrument in interpreting
communication processes as cultural autocommunication. Culture translates it-

self to itself in order to constitute and keep its identity. The epistemology of
translation semiotics is based on the distinction of sign systems’ hierarchies,
translatability and translation capacity, and the comparison with intertextual,
transmedial and intersemiotic processes in culture” (ibid.).
Kristina Mullamaa viewed the self-descriptions of liaison interpreters in
Estonia from a semiotic perspective. Kukkonen (2007:24) has pointed to the es-
sential capacity of translation and interpreting to act as semiosis. Mullamaa ana-
lyzed the self-descriptions and the evolving role model of practising interpreters
through this prism. The self-descriptions manifest an interesting document for
studying a specific group. Torop (2008:392) points out: “self-description is a
process of auto-communication, and its result can be a self-modelling that fixes
the dominants, the princples of unification, and the generative language of self-
description”. Lotman has seen self-modelling as “a powerful means for the
‘end-regulation’ of a culture, attributing to it a systematic unity and largely de-
fining its quality as a reservoir of information” (Lotman 1970:420 in Torop
2008:392). Mullamaa views how the participants describe their cooperation
with clients to create specific intricate and interrelated webs of associations; at
the same time taking conscious action as an agent. The results of the study show
that the context in an interpreting situation may extend much further than has
been suggested traditionally. Another side of the semiosis is formed of the se-
miotic tapestry of concepts and a wealth of experience from the socio-political
context. Semiosis is also formed inside each translation act, interpreting situa-
tion, and the formation of a profession specialising in specific translation. The
societies and cultural experiences that have shaped interpeters and translators
continue to shape the formation of semiosis, part of which is also our suggestion
of how to possibly interpret it.
Anja Danska’s report “the Altar-rail is the Signified as a borderline in the
church-hall” analysed “the hierarchical systems in the visible and yet uncon-
scious structures of different values in the religious and architectural cultures”
(Abstracts 2009:26–27). Danska studies the ritual of communion, and how we
translate visual architectural language into spoken language. The framework of
the study is cultural semiotics and existential semiotics. With a rich illustrative
support, Danska demonstrates that our physical environment is in itself a system
of information – a sign. Danska follows the formation of such signs in Christi-
anity, the evangelic-Lutheran church, through the example of the changes in
church architecture in Finland. Danska explains how the Reformation brought
about the individual interpretation of the Bible. The altar-table becomes a sign
of collaboration. According to Luther the holy communion is a sacrament of
love. Bread and wine shared to everybody become signs of the new spirit of col-
laboration. However, even in the 20th and 21st centuries the visual sign sug-
gests that there is a boundary line. The altar-rail that was intended to demon-
strate unity and collaboration between the individual and God has instead be-

come a dividing line, and “the fixture which is supposed to give support to peo-
ple when they knee for prayer is signified as a fence around the altar-table”
(ibid.). Danska is interested in the effects of spaciality on our sense and feel-
ings. In the 1990s, she carried out interviews to find out how the Finnish evan-
gelic-lutheran priests interpret the altar. The answers were not unanimous. The
main religious interpretations were: 1) the altar is the place of God; 2) the altar
is the place and sign of Jesus Christ; 3) the altar represents the communion ta-
ble; and 4) the altar is the sign of total life of the community, congregation and
society. Danska’s research illustrates in an inspiring way how the changing so-
cial values and customs affect (religious) rituals, and put new demands on vis-
ual interior and architecture.
Pirjo Kukkonen’s Interpreting the Singing I in “Kanteletar” constitutes a
thorough analysis of the different approaches to translation and interpreting of
lyrics and poetry in the Finnish collection Kanteletar. The collection, published
in 1840, is based on folk songs and poems. This rich collection of songs is con-
sidered to be a sister collection of the Finnish national epic Kalevala. Kukkonen
focuses on the strategies of cultural translation into Swedish from 1830–1989.
She analyzes the specific capacities of the Finnish language and culture that
have been transferred through the alliteration, parallelisms, and particles in the
lyrics. What solutions have been used to convey their deeper meaning in the
translation? The more specific subject of the study is the ‘I’, the emotions of
man and the strive to make the invisible aspects visible, the importance in how
you express yourself and the ‘I’ and the feelings. There is a number of signs and
sign systems involved. The specifics of a Fenno-Ugric language pose a chal-
lenge for translating lyrics, especially the archaic ones, into Germanic lan-
guages. Kukkonen points out that the Kalevala metre itself, the alliterative, as-
trophic trochaic tetrameter, which forms successive rises and falls in various
combinations, poses intellectual and identity challenges for translators as (based
on Kristeva and Jakobson) phonemes are not only phonemes – they have mean-
ings. With illustrative examples from the corpus Kukkonen shows how for ex-
ample parallelisms, alliterations and particles have been conveyed through
structural markers:
Finnish: Me tulipa – Swedish: de blev ju

meneko – de gick ju
toitko – tog du
Kukkonen’s positions is that “the Skopos of poetic translation is to shape the il-
lusion, the effect, the experience, the life and soul, mind and spirit, the emotion,
the poetic idea of the source text in the target text”. And the answer to the ques-
tion posed by Kukkonen (Abstracts 2009:39): “Is the Finnish singing I different
in Swedish, or English interpretations?”, seems to be in the affirmative.

The seminar session was ended by Peeter Torop’s presentation Semiotics

of Translation from the Pedagogical Point of View. Torop analysed the trend to
regard processes in culture as translational. He also discussed the “semiotic
turn” in approaches to translation. Torop presented an overview of the new con-
ceptions of disciplinary history and the modern developments in intersemiotic
translation. For example, as we know, today filmic versions may deteremine
what works will be regarded by young people as cultural masterpieces. “Read it
before Hollywood does it!”, can be a relevant warning as the filmic versions
(Torop illustrated this using the example of the Hollywood version of Troy) of-
ten erase the deeper dimensions and nuances from both the plot and feelings,
and leave the audience just the “love and fight” story. Torop puts the question
what effects such simplifications have on future generations’ perception of cul-
ture. The multimedial solutions and conceptions compete, but also co-create and
support our memories, associations and concepts of the work of art as a whole.
When speaking about semiotics of culture in the context of the curriculum,
and as a system of knowledge, Torop inter alia describes the principles for
teaching semiotics at the university of Tartu. As was also confirmed in profes-
sor Deeley’s plenary lecture at the end of the seminar day, in many universities
in the world, semiotics is taught only at the MA or even PhD levels. It is be-
lieved that it is only after a certain length of studies in one’s “own” discipline
that the level of maturity necessary for delving into the semiotic aspects is
achieved. In Tartu, however, semiotics is taught as a discipline in its own right
from the very scratch, i.e. from the level of BA studies. In the autumn of 2009
an international MA course in semiotics was launched at Tartu Univesrsity.
The plenary lectures by Solomon Marcus and John Deeley winded up the
conference session for the translation section. The seminar day ended with Pro-
fessor Deeley’s vision about the future of semiotics: in the future semiotics
might become the discipline that will be the common platform for continuing
research beyond the level of one’s more narrow discipline; in the same way as
Latin once used to be a common platform in universities across the globe for
even starting research and studies.
References
Abstracts. International Summer School for Semiotic Studies. Pan European Heritage in the
Global World. Imatra, Finland. June 5–9, 2009.
Bassnett-McGuire, S. 1980. Translation Studies. London & New York: Methuen.
Hartama-Heinonen, R. 2008. Abductive Translation Studies: the Art of Marshalling Signs. Acta
Semiotica Fennica XXVIII. Imatra: International Semiotics Institute & Helsinki: The Semi-
otics Society of Finland.
Kristeva, J. 1969/1980. Word, Dialogue, and Novel. In: Roudiez, L. (ed.) Desire in Language.
A Semiotic Approach to Literature and Art. New York: Columbia University Press. 64–91.

Jakobson 1959/1987. On Linguistic Aspects of Translation. In: Krystyna Pomorska & Stephen
Rudy (eds) Language in Literarature. Cambridge, Massachusetts & London, England: The
Belknap Press of Harward University Press. 428–435.
Lotman, J. 1970. Problema “obutseniya kulture” kak tipologitseskaya kharakteristika. [The prob-
lem of “learning to understand culture” as a typological characteristic]. In: Semiosfera. St.
Petersburg: Isskustvo. SPB. 417–425.
Paulrud, A. 2007. Döden heter Amerika. http://aftonbladet.se/kultur/article482439.ab. Accessed in
August 2009.
Torop, P. 2008. Translation as Communication and Auto-communication. In: Torop, P., Lotman,
M, & Kull, K. (eds) Sign System Studies. Vol. 36. No. 2. Tartu: Tartu University Press.
375–397.
Kristina Mullamaa
Tartu, Estonia
E-mail: kristina.mullamaa@ut.ee

DOI: 10.1556/Acr.11.2010.1.9
Book Review
Harriet S. Hill: The Bible at Cultural Crossroads

From Translation to Communication
Manchester: St. Jerome Publishing, 2006, 280p. ISBN 1-900650-74-4 (pbk)
Translation Studies is an interdisciplinary field. There is nothing new in this

statement, but its truth is undeniable, as demonstrated by Harriet S. Hill’s The
Bible at Cultural Crossroads, which draws on the results of several fields of
scientific inquiry. When reading this book, one has the impression that under-
standing the anthropology of the Adioukrou, familiarity with first century Jew-
ish beliefs or Relevance Theory (RT), as well as the use of statistical methods
are not requirements set by an established research paradigm, but follow from
the author’s desire to provide a firm basis for her work – to remain “down to
earth” in questions of the “unseen world”.
Hill has the right, the qualifications and the experience to do provide such
an analysis of the problems of Bible Translation. There are few people in the
world who have completed a similar journey. This book represents her life ex-
perience spanning nearly 30 years.
Hill obtained a B.A. in theology in 1984. Since 1980, she has worked in
several linguistic and anthropological projects in Africa, two of them concerned
with Bible translation: the Anyi Bible Translation Project in Côte d’Ivoire be-
tween 1980–1983, and the Bible Translation for the Adioukrou in Côte d’Ivoire
between 1984–1998. This book is based on the latter project. She lived together
with the Adioukrou, a tribe of around 100,000 members, speaking the language
of Adioukrou in the North-East area of the Ivory Coast. Parallel to this, she was
also coordinator of anthropology on the Ivory Coast and taught in Nairobi be-
tween 1994 and 1996 at the Pan African Christian College.
Working and living with native Africans, Hill understood that it is not
enough to provide a translation of the Bible for the Adioukrou or the Anyi (an-
other Côte d’Ivoire tribe) based on the code model, decoding the source text
and encoding it in the target language. The author explains that the cultural con-
text of the target audience must also be considered, and this is something that
the code model does not do, while the relevance theoretic model can, and
does. Setting up parallels between the first century Jewish cultural context and
that of the Adioukrou, she finds similarities in concepts and terms and argues
that the meaning of already existing concepts in the target culture can be
enlarged. According to her survey, this is a more fruitful method than introduc-
ing new terms that are not bound in any way to the cultural concepts of the tar-

136 BOOK REVIEW
get community. Hill’s method of conceptual adjustment is the backbone of

this work. Using three different kinds of Bible texts (previously translated unad-
justed texts; texts with the adjustments incorporated in the text; and texts with
adjustments in footnotes) she tested the understanding of the Bible with the
Adioukrou and found significant differences. Her work is also a didactic guide
for translators, helping them to involve in the translation process communities
in need of a translation product.
To understand the scientific background behind Hill, it is worth having a
look at the SIL (Summer Institute of Linguistics)1. This is an organisation that
helps communities all around the world to preserve their own ethnic culture,
and if they wish, to participate in Bible translation projects to the point where
they can read and understand the Bible in their mother tongue. Almost 1,000
people with academic degrees work in different language programmes and
translation projects to achieve this goal. Hill is one of them, as is her mentor R.
Daniel Shaw2, who is also engaged in Anthropology and Bible translation at the
Fuller Graduate School of Intercultural Studies. Hill’s other consultant was
Ernst-August Gutt3, who implicated Relevance Theory in Bible Translation.
Gutt is also a member of the SIL. The “family tree” can ultimately be traced
back to one of the founding fathers of modern Bible translation and one of the
greatest theorists in Translation Studies, Eugine Nida4, who joined the SIL in
1936 and revolutionized Bible translation with his theory of dynamic equiva-
lence. Hill walks in their footsteps and comes to new conclusions.
In the first three chapters Hill gives an overview of relevance theory. She
cites examples from her own experience among the Adioukrou to illustrate
Sperber and Wilson’s theory.
In Chapter 1 (Relevant Communication) Hill explains what relevant com-
munication means: first of all, getting the audience’s attention and then keeping
it up with relevant products. In connection with the communication of the
Scripture she suggests that a Bible translator should find the point of interest
of a given community. The Book of Genesis proved to be interesting for the
Australian Ninggubuyu tribe because they had always been interested in how
the world was created, while the Kuwawa in Liberia listened to the stories of the
Bible only after they had the chance to tell their own stories. In this way, infor-
mative intent is one thing, and communicative intent, i.e. how to get the audi-
ence’s attention is another. Both are keys to successful communication. To get
the Adioukrou’s attention, Hill learned their language, since it is only through
dialogue with the audience that one can determine what kind of translation
product will attract their attention. The cognitive effects the communicator
wishes to convey may result in three different reactions from the listener: 1. it
can contradict their previously held assumptions, 2. it can modify their strength,
and 3. it can derive new implications from already existing assumptions. Hill
gives examples from the New Testament to demonstrate the above. For exam-

BOOK REVIEW 137
ple, when Jesus called the tax collector to follow him, the pious Jews were upset
– something that most readers today would not understand, because such a reac-
tion would contradict their previous assumptions. On the other hand, when Je-
sus says “Rejoice and be glad, because great is your reward in heaven”, he
strengthens the assumptions of those readers who believe in the afterlife. There
are also many examples for the third type, showing how Jesus enriched the al-
ready existing assumptions of his listeners about the Kingdom of God. Hill pro-
vides an interesting analysis of how Jesus’ utterances might have affected 1st
century Jews. Quoting Sperber and Wilson, she points out that the cognitive
processing costs for positive cognitive effects must be low in order to achieve
relevant communication. If this is beyond the audience’s ability to understand,
the communication will be irrelevant. There are several factors that increase
costs: linguistic complexity, the accessibility of the communication (depending,
e.g. on the ability to understand a given language, or having access to technical
devices that ensure communication), the availability of background (contextual)
information and the amount of new information. Hill concludes with some ex-
amples of how relevance can be increased through increasing expectations of
cognitive effects (e.g., “when Bible translation is carried out in the context of a
church mission ministry where people experience God’s presence in their life”
(p. 8.)), or if the Bible translator is a trusted person of the community, or
through decreasing the expectations of processing effort (e.g. by building con-
textual knowledge or by using familiar media).
In Chapter 2 (Context and Comprehension) Hill gives a quick overview of
the development of communication models, starting from Shannon and
Weaver’s code model to the models based on new insights into language and
communication, such as pragmatics, speech-act theory, semiotics, Grice’s max-
ims and Relevance Theory. According to Hill, RT is extremely helpful in Bible
translation because it “provides conceptual tools for understanding how context
is selected and built, and contextual differences between first-century Jewish
and contemporary non-Jewish audiences of Scripture are large. The same text
has different meanings in different contexts. This is the proof, says Hill, that
context plays an important role in communication. Then she goes on to explain
explicatures, which is the ad hoc, temporary meaning that the hearer constructs
in the context, in which the utterance occurs and implicatures, which are 1. the
implicated premises, the assumptions the audience brings to the communication
and 2. the implicated conclusions, the implications that the audience can infer
from the communication. In this way we get to the problem of context in the
Scripture. The authors of the Gospel themselves explained the context when
they knew that the audience had a lack of information. Luke, in Acts: 23, pro-
vides an explanation about the Sadducees and Pharisees, otherwise his audience
would not have understood why Paul’s statement “I hope in the resurrection of
the dead” had caused such an uproar between the two parties. But often there

138 BOOK REVIEW
are no explanations in the Scripture, or readers may have assumptions different

from those underlying the Scripture, and in such cases communication cannot
succeed. According to Hill, the context might be unintended, listeners can hold
their assumptions too strongly or too weakly about the context, and it can hap-
pen that people are unable to find a context through which to process the text.
Thus, in many cases, failures in communication are related to context.
A basic claim of RT is that, to achieve the right Cognitive Effect, the right
context must be selected. Hill introduces a mutual cognitive environment matrix
that is divided into four quadrants showing when communicators or listeners
get the wrong or right context. The four quadrants are: 1. the intended context,
when the listener thinks it is shared and it is actually shared, 2. the unrecognized
context, when the listener does not think it is shared, but it is actually shared; 3.
the unintended context, when the listener thinks it is shared, but actually it is not
shared; 4. the missing context, when the listener does not think it is shared, and
it is actually not shared. This matrix proves to be a useful tool in the author’s
analysis of mismatches in communication. Hill argues that the picture becomes
even more complicated if we consider how many different communications take
place at the same time in Bible translation: God is the original and ultimate
communicator who speaks to all people of all time through the Scripture; the
human author is a communicator, who lived in a specific place and time and
communicated to people in a specific situation; within the text, people commu-
nicate with one another, and contemporary translators are also communicators,
actively involved in bringing the original message to a new audience in the be-
lief that it will improve their cognitive environment. The audience may share, or
think they share, a different mutual cognitive environment with each of these
authors. (p. 30)
Selecting the right context also depends on speakers’ and listeners’ abilities
and preferences. Following Sperber and Wilson, the author distinguishes be-
tween the nai’ve and cautious optimism strategy that a listener may employ in
inferring meaning. A naïve audience tries to infer meaning directly from the
Scripture, in their own cognitive environment; a cautious optimist audience, on
the other hand, is conscious about the fact that the first audience’s and their own
cognitive environments differ. Therefore, the more they learn about the first au-
dience, the better their inferences will be.
Chapter 3 (Context Building) is probably theoretically the strongest part of
the book. We could summarize its main points as follows. Translators must un-
derstand how source text context and target text context function. Although this
is a complicated task, it is manageable, since every human being is capable of
meta-representation that is, making thoughts about their own and other people’s
mental representations. After detecting the differences in cognitive environ-
ments, translators not only need to adjust context to make it accessible for the
target audience, but they need to understand the assumptions of the target audi-

BOOK REVIEW 139
ence as well. The author uses here the mutual cognitive environment matrix,
giving examples about the divergences in assumptions and about their possible
adjustment.
Table 5. The Goal of Contextual Adjustment (p. 38)
Hearer Thinks It Is Shared Hearer Does Not Think It Is

Shared
Actually Shared 1. Intended assumptions 2. Unrecognized assumptions
Not Actually Shared 3. Unintended assumptions 4. Missing assumptions
In the first case, the intended assumption coincides with the listener’s as-
sumptions (due to commonality of human experience, e.g. we all eat, become
ill, die etc.). The second possibility is when there is a common area between the
first and second audience’s cultural contexts, but it is hidden. For example,
‘phylacteries’ are mentioned in Matthew 23:5. These religious items (leather
boxes fixed on the arm and head) are worn by religious Jews during prayer,
containing words from the Torah. The Adioukrou have a similar concepts: in
some amulets they have powerful words written on paper. At this point the au-
thor goes on to suggest that sometimes it may be preferable to borrow expres-
sions from the vernacular language, extending their meaning based on the
source text, since these expressions may refer to concepts that are already part
of the target group’s cognitive environment. This means that they only need to
be adjusted, whereas introducing a new expression (e.g. by borrowing from the
source language) involves the risk that it will remain a separate entity, unrelated
to other concepts and outside the already existing cognitive environment of the
target audience, and will not help to bring Christianity nearer to the target audi-
ence.
The third case is when there are some inferred meanings that are reached
by already existing assumptions, but these are not the intended ones. When in
the Gospel some people lower down their friends through the roof to the feet of
Jesus, the target audience, e.g. the Adioukrou might not know that in ancient
Palestine roofs were flat, and people spent time on them. They might think of
their own huts and cannot imagine the situation. With additional information
this type of contextual mismatches can be fixed. Translators often hesitate –
says Hill – to use local terms because they want to be accurate. Instead they
should be relevant.
The fourth case is the most demanding one. New concepts need to be in-
troduced very carefully. The author calls this bridge building and defines three
criteria for this type of communication: relevance, truth and acceptability. Rele-
vance refers to the amount of information the translator communicates to the
readers and the costs necessary to invest in processing meaning. “Only the in-

140 BOOK REVIEW
formation that is necessary to derive the explicatures and implicatures of the

original is relevant” (p. 46). The criterion of truth refers to the use of context ad-
justment materials, which must be accurate. Finally, the criterion of acceptabil-
ity means that the Bibles adjusted in this way should be accepted by all the
churches.
Whether the context adjustment materials are correct, i.e. the audience un-
derstands what was meant should be checked by testing the audience, says Hill.
On the other hand, we must also be conscious of the fact that context building
takes time, it is a process and it is dynamic in nature since the source and target
audiences’ cognitive environments change continuously.
Chapter 4 (The Design and Effects of Relevant Products), similarly to
chapter 1, deals again with relevance for the audience, but in this case not from
the point of view of attracting the attention of a given group, but from the point
of view of the relevance of the whole translation product for the given audience.
Processing effort varies from audience to audience. It depends on their expecta-
tions of relevance and their background knowledge. At this point the author
tries to give a definition of Bible translation and says that “a text is considered a
translation because the communicator intends it to achieve relevance as a retell-
ing of the original, not because of its degree of resemblance to the original”.
Emphasizing the degree of resemblance to the original – points out the author –
is code model thinking. So in her view “preparing a product that requires more
processing energy than the audience has available will put the product outside
their processing budget”. Her suggestion is that context must be provided using
various means: teaching, preaching, literature and media and other context ad-
justment products.
Hill undertook field research in 2001 to find out how contextual adjustment
materials should be presented to make them the most effective. She chose 4 dif-
ferent passages from the Gospel which she prepared in 3 different ways: Mode
1 was a straight translated text without any adjustment, Mode 2 was the same
text with adjustment as footnotes, and Mode 3 was an amplified text, in other
words, adjustment material was inserted into the text. She had 271 respondents
among the Adioukrou, and each respondent listened to or read one of the texts.
After checking comprehension through questions, she found that Mode 2 was
the most relevant, in which respondents had a comprehension score of 62.9%,
Mode 3 had a comprehension score of 58.7% and Mode 1 was the worst with
28.2%. Hill came to the conclusion that the “contextual adjustment materials in-
creased relevance not by reducing processing costs, but by increasing the bene-
fits”. She further showed that the respondent’s role in the church gave them
some advantage in Mode 1, whereas those with higher education had better
comprehension scores in Modes 2 and 3 because they knew how to learn.
Church affiliation also influenced comprehension in Mode 1: Methodists per-
formed better because they had had more exposure to biblical background in-

BOOK REVIEW 141
formation than the members of other churches. All in all, the most important
factor affecting comprehension was the use of contextual adjustment materials.
Furthermore, Hill describes that after using the adjustment materials, respon-
dents were able to come to conclusions and could answer questions for which
no adjusted materials existed. The author calls this the “spreading effect of
comprehension”. Analysing the role of the media, Hill found that 47% of those
who could not read had worse results in comprehension than those who could.
In fact, the second most important factor influencing comprehension was the
ability to read Adioukrou.
Chapter 5 (Ways of Supplying Context: Out-of-Text and In-Text Solutions)
groups contextual adjustment materials first into out-of-text and in-text solu-
tions, then both are divided into simplified and full text variants. From the out-
of-text adjustment materials the author deals with those completing full texts,
just like her Mode 2 versions. The types of out-of-text solutions are as follows:
introductions (giving socio-cultural, historical and geographical information),
section titles (indicating the topic of a passage and providing readers with men-
tal maps), using various formats, illustrations, glossaries, maps and charts, and
notes. I put ‘notes’ at the end of this list because the author regards “notes as the
best means to provide information that is needed for the dynamics of pas-
sages”(p. 74), (e.g. in John 13 Jesus gives the bread to Judas and the text says
that Satan entered Judas. In the original giving food to the guest was a sign of
Jewish hospitality, whereas Adioukrou believe that the food was enchanted. In
such cases the footnote addresses the dynamics of the scene and explains the
difference in perception). (Oral footnotes, marked by the use of a different
voice, were too difficult to make; the written version was more successful.) Un-
dertaking an analysis in educational level, Hill found that Mode 2 readers were
not better educated than Mode 3 readers, their level of education did not affect
their ability to use the notes. Notes in the adjusted Mode 2 texts were of 3 basic
types: 1. contextual notes (explaining concepts, customs beliefs and references
to the Old Testament); 2. textual notes (with the level geared to audience and
the purpose of translation), and finally 3. life application notes (how the audi-
ence can apply the text in their lives).
In-text-solutions are woven into the text, bridging the audience’s knowl-
edge of the Bible and their biblical background information. These products are
based on the Bible, but without verse numbers. In this version communicators
may add contextual information, they may omit details of the story that are not
relevant to the receptors, they may re-order events into actual chronological or-
der, or they combine several accounts of the same event. These texts are faithful
in relevant respects.
The author illustrates the reordering of different parts of the story with a
short text adapted by herself. There are no verses in the Gospel texts adapted in
this way, and the author is aware that some kinds of information cannot be in-

142 BOOK REVIEW
corporated into the text using this technique, since the adjustments must be
blended naturally into the story.
A comparison of the comprehension scores of Mode 2 and Mode 3 showed
that when respondents were listening to the text, and not reading it Mode 3
yielded better results.
In Chapter 6 (Cultural Research) the author argues that making cultural re-
search both in the source and target contexts are at least as important as analys-
ing the source and target languages. All too often, translators fail to carry out
cultural research, although we know that context contributes as much to com-
munication as the text itself. Hill formulates this as follows: “Translators that
focus on linguistic research may well strain at a gnat while swallowing a
camel.” Since cultural research can be interpreted in different ways, Hill sug-
gests that the purpose of such research should be clearly defined. Her own pur-
pose was “to measure the effect of the terms used in Adioukrou Scripture for
God, angels, demons, and the devil on the Adioukrous’ perceptions of the un-
seen world”.
She further argues that translators need to develop research questions. Her
own were the following: 1. What were the first-century Jewish perceptions of
the unseen world? 2. What were the Adioukrou perceptions of the unseen world
before the arrival of Christianity in the area? 3. What terms were used for God,
the devil, angels and demons in the Adioukrou Scriptures? 4. How do the Adi-
oukrou understand the terms used for God, the devil, angels and demons today?
To find answers to her questions, she developed a work chart divided it
into three columns: a. the first audience’s assumptions; b. Bible text and c. Adi-
oukrou assumptions. In the first column she entered terms like meals, death,
slaves etc., and in the third column she wrote open-ended questions to find out
about the Adioukrous’ existing assumptions and about the assumptions the Bib-
lical texts evoked in them. This research helped her to identify contextual mis-
matches. In further parts of this chapter Hill gives some useful advice on re-
search methodology in this field.
In chapter 7 (First-Century Jewish Perceptions of the Unseen World) Hill
gives an insight into 1st century Jewish beliefs. This represents the kind of re-
search that can help translators to understand the context of the source text. The
author finds parallels between 1st century Jewish beliefs of the unseen world and
that of the Adioukrous’ world today. Hill quotes 1st century texts to show that
Israel’s life, because of the loss of political autonomy, exile and Roman oppres-
sion, was difficult and tensions grew between the rich and the poor. The focus
of religious activity was very practical. Hill echoes Ehrman: “In a world that is
helpless against the elements, the gods play a major role”. The perceptions of
God, spirits, and Satan had changed during the centuries. In the time of the Old
Testament, the presence of angels was very limited, almost insignificant. In the
time of the New Testament, the opposite was true: demons and angels were

BOOK REVIEW 143
very common. Supernaturalism, magic, exorcism and healing were much more
in the centre of attention than the Torah. From a certain point of view, Jesus was
only one of many magicians and exorcists. When reading about 1st century be-
liefs, one finds more similarities with the religiousness of primitive nations than
with the world-view of today’s western civilization.
The essence of chapter 8 (The Effect of the Key Term Choices on Adi-
oukrou Theology) could be summarized as the contrast between two choices:
whether to introduce new terms or fill the old ones with new meanings. Hill ar-
gues that the terms used in a translation evoke contextual assumptions, but we
must make sure that they are the intended ones. Some authors warn translators
against using words taken from the vernacular to translate the name of God
since that would lead to misunderstanding: the word’s original meaning would
alter the Biblical message. This way of thinking can be related to the early mis-
sions in the 1900s and to the first Adioukrou translations of the New Testament:
Mark in 1930, John in 1957 and Matthew in 1976. New, formerly unknown ex-
pressions were used in these early translations, with the result that the Adi-
oukrou could not connect them with their existing world view.
Following John Beekman, Hill subscribes to the opposite view, saying that
the vernacular expressions should be kept and their meaning will change with
time. “Rather than polluting Scripture, Scripture has purified the traditional
categories and redeemed them. In this way, translated Scripture has converted
not only individuals but also cultural categories and worldviews” (p. 147). This
was the view that Hill adopted when she started translating the Bible into Adi-
oukrou in the 1980s. Using the Adioukrou term Nyam for the Christian God
transformed the original meaning of the word Nyam, but it had a strong position
in the network of concepts in the Adioukrou way of thinking.
This chapter also gives an interesting analysis of the concept of witchcraft.
In this case the term agn, or witch was initially very different from the biblical
category of devil, but its meaning can be modified: “some elements of the tradi-
tional perception are being minimized, others maximized” (Table 26. on p.
151). According to surveys among the Adioukrou the term agn on the mutual
cognitive environment matrix has changed position from not actually shared to
actually shared.
Finally, the author gives a definition of the word translator, which reflects
changes in the perception of this profession. Her own credo as a translator is
this:
Translators are often described as traitors, because they cannot communi-
cate the full meaning of the text in the receptor language. Rather than trai-
tors, perhaps they are frontline soldiers who bring God’s word where it has
not yet gone and see the transformation it brings, both to individuals and
worldviews. It is a prime means of discipling the nations.

144 BOOK REVIEW
In Chapter 9 (Identifying and Adjusting Contextual Mismatches in John 13:1-

30) the author provides an example of “Identifying the contextual mismatches
between the first and secondary audience’s mutual cognitive environments” (p.
155). The author explores the contextual assumptions the text evoked for 1st
century Jews, then she compares this with the assumptions the text evoked in
the Adioukrou. She proposes contextual adjustments at several points in the text
(e.g. concerning what Passover meant for the Jews, what eating together and
foot washing meant in the 1st century Jewish context, what were the messianic
expectations of these people and of the disciples). Then the author measures the
impact of these materials both in Modes 2 and 3. At the end of the book we find
three more texts from the New Testament with contextual adjustments and their
measured effect on the Adioukrou audience. (Appendixes A–C.)
The traditional view of the translator is that he/she works on her own.
Chapter 10 (The Communication Context – Working with the Church Commu-
nity) opposes this view and gives an account of Hill’s experience: translating
cannot be solitary work, since without joint efforts made together with the
community concerned the translation will not be accepted, and it will not
achieve its goal. Regarding the translation, Hill emphasises: “natural translation
can only be drafted by a mother-tongue speaker; an appropriate programme can
only be designed by members of the community” (p. 179). Furthermore, she ar-
gues that Bible translation is not a strategy on its own but “Bible translation is
part of a missiological strategy since its ultimate goal is to change people’s life
as a result of understanding Scripture” (p. 181). So translation is one of the mis-
siological activities of the community like teaching, preaching and fellowship.
A translator, in addition to having the necessary competence of a translator in
technical aspects (knowledge of linguistics, biblical languages, translation prin-
ciples and exegesis), must be involved in the programme of the church, other-
wise the translation will not succeed.
The last subchapter On Change Processes gives an overview about how a
change, i.e. the introduction of a mother-tongue Scripture for the first time, is
adopted by a community.
As a conclusion to The Bible at Cultural Crossroads by Harriet S. Hill, we
can say that this book is a significant step in proving the practical usefulness of
RT. The author does not want to lead a discussion between traditional TS and
RT, she simply demonstrates on an empirical basis that context is as important
as the text itself. Perhaps this is the reason why the translation-technical part
comes short, e.g. the book does not give information on the source language
from which the translators worked. All the Scripture examples are in English.
The author answered my query on this point by email, saying that the source
language of the translation was Hebrew and Greek, but the Bible translating
team also consulted translations into other languages to see how linguistic diffi-
culties were solved in those translations.

BOOK REVIEW 145
Hill’s work cannot be disconnected from the genre of the text. According
to Katharina Reiss, Biblical texts are appellative in their nature; they motivate
people to change their life5. This intention is clear also from the author’s
“credo” quoted above. In this sense we see a fruitful combination of faith and
scientific approach.
The present reviewer believes that there is at least one thing that is, and
should be, common to both researchers and religious people: humility. Hill ap-
proaches her research task with humility, giving clear descriptions and careful
analyses of contexts, assumptions and cognitive environment. She exhibits the
same humility towards the Scripture: she does not go along with the demands of
the feminist approach for a gender fair translation of the Bible, based on the
claim that times have changed, and men and women are now equal and have the
same human Rights, so the creator is also a she and this must appear in transla-
tion as well. Instead, she undertakes time-consuming surveys to understand the
cognitive environment of the target audience and adjusts the text with footnotes
(Mode 2) or extra information (Mode 3) in order to transmit the message. To
transmit the message by hard work, rebuilding the importance of mother-tongue
translations, keeping and not negating familiar terms of the target audience is
Harriet S. Hill’s humility as a believer.
Notes
1
www.sil.org/sil/
2
www.sil.org/sil/roster/shaw_dan.htm
3
www.sil.org/SIL/roster/gutt_ernst-august.htm
4
www.nidainstitute.org/vsItemDisplay.dsp&objectID=0920A817-28AA-4D6F-
9B9F70012FE3A462&method = display
5
Reiß, K. 1983. Texttyp und Übersetzungsmethode. Heidelberg: Julius Gross Verlag
János Nagy
Pécs, Hungary
E-mail: glagonya@hotmail.com

DOI: 10.1556/Acr.11.2010.1.10
Books Received
The editors of ACROSS welcome review copies of publications and

also appreciate candidates for reviewing incoming copies.
Published in 2009
Rodica Dimitriu, Miriam Shlesinger (eds): Translators and their Readers. In Homage to
Eugene A. Nida. Brussels: Lés Éditions du Hazard. 2009. 363 pp. ISBN 2-930154-23-3
Inger M. Mees, Fabio Alves, Susanne Göpferich (eds): Methodology, Technology and
Innovation. A Tribute to Arnt Lykke Jakobsen. Copenhagen Studies in Language
Vol. 38. Copenhagen: Samfundslitteratur. 2009. 411 pp, ISBN 978-87-593-1476-0
Susanne Göpferich, A. Lykke Jakobsen, Inger M. Mees (eds): Behind the Mind.
Copenhagen Studies in Language Vol. 37. Copenhagen: Samfundslitteratur. 2009.
264 pp, ISBN 978-87-593-1462-0
Published in 2008
Petra Szatmári, Dóra Takácsa (Hrsg.): “…mit den beiden Lungenflügeln atmen” Zu Eh-
ren von János Kohn. Munchen: Lincom, 2008. 291 pp. ISBN 978-3-89586-148-2
Erika Kegyes (Hrsg.) unter Mitarbeit von Ágnes Huszár: Genderbilder aus Ungarn
(Ergebnisse der ungarischen Genderforschung). Hamburg: Verlag Dr. Kovač,
2008. 298 pp. ISBN 978-3-8300-3775-2
John Kearns (ed.): Translator and Interpreter Training. Issues, Methods, Debates. Con-
tinuum Studies in Translation. London: Continuum Publishing Group, 2008. 223
pp. ISBN 978-08264-9806-9 (reviewed in Vol. 10. No. 1. by Dénes Neumayer)
Kaisa Koskinen: Translating Institutions. UK & Kinderhook, NY, USA: St. Jerome
Publishing, 2008. 188 pp. ISBN 1-905763-08-5 (pb)
Federico Zanettin (ed.): Comics in Translation. UK & Kinderhook, NY, USA: St.
Jerome Publishing, 2008. 336 pp. ISBN 1-905763-07-7 (pb)
Radegundis Stolze: Übersetzungstheorien. Eine Einführung. 5. Auflage. Narr Studien-
bücher. Tübingen: Gunther Narr Verlag, 2008. 285 pp. ISBN 978-3-8233-6431-3
Susanne Göpferich: Translationsprozessforschung. Stand − Methoden − Perspektiven.
Translationwissenschaft. Band 4. Tübingen: Gunther Narr Verlag, 2008. 313 pp.
ISBN 978-3-8233-6439-9
148 BOOKS RECEIVED
Published in 2007
Ida Klitgard: Fictions of Hybridity. Translating Style in James Joyce’s Ulysses. Univer-
sity Press of Southern Denmark, 2007. 282 pp. ISBN 978-87-7674-193-8
Riitta Jaskelainen, Tiina Puurtinen & Hilkka Stotesbury (eds): Text, Processes, and
Corpora: Research Inspired by Sonja Tirkkonen-Condit. Publications of the
Savonlinna School of Translation Studies 5. Joensuu: Savonlinna School of Trans-
lation Studies. ISBN 952-458-840-4
Theo Hermans: The Conference of the Tongues. UK & Kinderhook, NY, USA: St.
Jerome Publishing, 2007. 190 pp. ISBN 1-905763-05-0 (pb)
Maria Tymoczko: Enlarging Translation, Empowering Translators. UK & Kinderhook,
NY, USA: St. Jerome Publishing, 2007. 360 pp. ISBN 1-900650-66-5 (pb)
Willy Vandeweghe, Sonia Vandepitte & Marc Van de Velde (eds): The Study of Lan-
guage and Translation. Belgian Journal of Linguistics 21. Amsterdam: John Ben-
jamins Publishing Company, 2007, v, 200 pp. ISBN 978 90 272 2681 5 (pb) (re-
viewed in Vol. 10. No. 1. by Pál Heltai)
Jan Walsh Hokenson & Marcella Munson: The Bilingual Text: History and Theory of
Literary Self-Translation. Manchester, UK & Kinderhook, NY, USA: St. Jerome
Publishing, 2007. 246 pp. ISBN 978-1-900650-93-9 (reviewed in Vol. 10. No. 1.
by Katalin Juhász)
Helen T. Frank: Cultural Encounters in Translated Children’s Literature. Images of
Australia in French Translation. Manchester, UK & Kinderhook NY, USA: St.
Jerome Publishing, 2007. 262 pp. ISBN 978-1-905763-03-0
Jorge Díaz Cintas & Aline Remael: Audiovisual Translation: Subtitling. Manchester: St.
Jerome Publishing, 2007. 272 pp. ISBN 978-1-900650-95-3 (reviewed in Vol. 9.
No. 2. by Károly Polcz)
Judy Szöllősy: Hunglish into English. The Elements of Translation from Hungarian into
English. Budapest: Corvina Books Ltd., 2007. ISBN 978 963 13 5556 7 (reviewed
in Vol. 9. No. 2. by Anette Márta)
Rudolf Pölzer: Kein Land des Übersetzens? Studie zum österreichischen Über-
setzungsmarkt. Wien: Lit Verlag, 209 pp. ISBN 978-3-7000-0721-0 (Österreich)
Books to be mailed to:
ACROSS Editorial Office

Department of Translation and Interpreting
Faculty of Humanities, Eötvös Loránd University
Múzeum krt. 4, Building “F”, H-1088, Budapest, Hungary
Phone: +36 1 411 6500/5894, Fax: +36 1 485 5217
E-mail: kklaudy@ludens.elte.hu

Journal Across Languages and Cultures, A Multidisciplinary Journal

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal Across Languages and Cultures, A Multidisciplinary Journal

Uploaded by

Copyright:

Available Formats

Across Languages and Cultures 11 (1), pp.

ABANDONING THE NOTION OF

Abstract: One of the best-known hypotheses of translation studies, the Explicitation

Keywords: explicitation, implicitation, translation universals, translation theory

1585-1923/$ 20.00 © 2010 Akadémiai Kiadó, Budapest

The present article is a plea to abandon the dogma of translation-inherent

Implicitness is the non-verbalization of information that the addressee might

Across Languages and Cultures 11 (1) (2010)

inference fails – which obviously happens occasionally in conversation. For the

Explicitness is the verbalization of information that the addressee might be

To put it somewhat informally, explicitness means saying something that the

It should be pointed out that the definitions of implicitness, explicitness

Across Languages and Cultures 11 (1) (2010)

Two important points deserve to be mentioned in connection with Klaudy’s

Across Languages and Cultures 11 (1) (2010)

stance of this type of explicitation is supposed to look like. In this connection it

4.1. The Birth of the Explicitation Hypothesis

Blum-Kulka proposed her famous Explicitation Hypothesis in a seminal and

Blum-Kulka tells us that the process of interpretation that invariably occurs in

Across Languages and Cultures 11 (1) (2010)

ferences between “linguistic” systems) and optional explicitations3 (due to dif-

4.2. Problems with the Explicitation Hypothesis

There are three serious problems to be noted in connection with Blum-Kulka’s

Problem 1. Why do I criticize that it is not clear where the “argument” is

Across Languages and Cultures 11 (1) (2010)

Problem 2. Blum-Kulka’s Explicitation Hypothesis clashes head-on with

Across Languages and Cultures 11 (1) (2010)

For example, the Explicitation through Simplification Hypothesis sketched

Problem 3. In a later passage of her paper, Blum-Kulka paraphrases the

Across Languages and Cultures 11 (1) (2010)

5.1. Subconscious Processes of Syntactic Explicitation?

1. Olohan and Baker conducted their research using a preliminary, work-

Across Languages and Cultures 11 (1) (2010)

a higher incidence of the optional that in translated English would provide

Across Languages and Cultures 11 (1) (2010)

z Source language interference.7 As has been mentioned above, the

z Conservatism. It has been suggested that translators tend to use more

Across Languages and Cultures 11 (1) (2010)

Both of these approaches, which draw on proposed translation universals other

5.2. Explicitation in English–Norwegian and Norwegian–English

Across Languages and Cultures 11 (1) (2010)

bility.” If translation-inherent explicitation were indeed the result of a transla-

Across Languages and Cultures 11 (1) (2010)

Øverås’ inclusion of (3) as evidence for the Explicitation Hypothesis is particu-

Across Languages and Cultures 11 (1) (2010)

Explicitation 347 248

Across Languages and Cultures 11 (1) (2010)

strategy inherent in the process of language mediation” (Blum-Kulka 1986:21)

6. ABANDONING THE EXPLICITATION HYPOTHESIS

Across Languages and Cultures 11 (1) (2010)

This of course entails the danger of counting pseudo-explicitations such

The Asymmetry Hypothesis (modified version):

The Asymmetry Hypothesis claims that translators display a tendency to

Across Languages and Cultures 11 (1) (2010)

6.1. Motivating the Asymmetry Hypothesis

It is well known that human communication is driven by two competing forces,

Across Languages and Cultures 11 (1) (2010)

face-to-face communication, albeit in a spatially and temporally “dilated” man-

Across Languages and Cultures 11 (1) (2010)

Property 2: The communicative situation underlying translation is typi-

Across Languages and Cultures 11 (1) (2010)

6.2. What Explicitational Asymmetry Looks Like in Practice

We have now provided a motivation for postulating the Asymmetry Hypothesis:

Across Languages and Cultures 11 (1) (2010)