Cognitive Linguistics

and Translation

Advances in Some Theoretical

Models and Applications

Edited by
Ana Rojo and Iraide Ibarretxe-Antunano

DE GRUYTER
MOUTON
Sandra L. Halverson
Implications of Cognitive Linguistics for
Translation Studies*

1 Introduction
As outlined in the introduction to this volume, Translation Studies (TS) has
reached the point at which a number of traditional questions and issues can
and should be either rethought or jettisoned altogether. The situation is this:
the inability of structuralist or generativist linguistic theories to adequately
address some of the basic questions of translation has led many scholars to
reject linguistic approaches altogether. At the same time, translation scholarship
within the descriptive-empirical paradigm has been accused of being unable to
generate theoretical innovation (Hermans 1999: 160; Pym 2010: 86). And finally,
at the same time as new methods and types of data are being introduced by
those who do advocate a linguistic approach, accusations of naive empiricism
or insufficient reflexivity threaten the position of those scholars who have not
been able to fully articulate a post-positivist, empirical epistemology (see Arrojo
1998; Hermans 1999: 159-160; Pym 2010: 85).
The publication of a volume such as the present one presents an opportunity
to take stock of the current situation, to highlight key areas of persistent concern
for linguistic approaches to translation, and to showcase the potential of cogni­
tive linguistic approaches in addressing translational issues. In this particular
contribution, the emphasis is on the second and third of these objectives, and
in discussing the implications of Cognitive Linguistics for Translation Studies,
the focus w ill be on three specific areas: theory development, methodology,
and epistemology.
The perspective taken in the following discussion is that of a scholar of
Translation Studies, and this vantage point w ithin TS has driven the selection
° f questions and issues. Accordingly, the focus in the second section is on
those translational issues that are of urgency with regard to future theorizing
a°d empirical study. Please note that the selection is by no means exhaustive;

This paper was written during a sabbatical stay at Purdue University, West Lafayette, IN,
during the 2009-10 academic year. Thanks to Purdue and the Interdepartmental Program in
linguistics for their generous hospitality during that year. Thanks are also due to the Meltzer
Foundation for financial support.
34 Sandra L Halverson

a number of other issues could have also been selected. Furthermore, the
breadth of Cognitive Linguistics w ill not be adequately represented here: the
main thrust of what follows w ill be taken from the theory of Cognitive Grammar,
as articulated by Langacker (primarily [1987] 1991, 2008) and from the accounts
of Cognitive Linguistics given in Croft and Cruse (2004), Lakoff (1987), Johnson
(1987, 2007), and Lakoff and Johnson ([1980] 1995,1999). This is primarily due to
limitations of scope in the current paper, and is remedied by the breadth of
material in the volume as a whole.
In the remainder of the introduction, a brief sketch of a cognitive linguistic
view of translation is given, and some of the requisite elements from theories of
bilingualism are also singled out. In the second section, a small set of transla­
tional issues w ill be considered in light of cognitive linguistic theory. In the
third section, the growing use of multiple methods and combinations of data
in one or a set of studies is discussed as an innovation in research methodology
in Translation Studies. In the fourth section, two epistemological issues are ad­
dressed: (i) the need for “ embodied realism” in Translation Studies and (ii) epis­
temology and definitions of the object of study. The final section presents a few
concluding remarks, including perspectives on evolving conceptualizations of
“ translation” itself.

1.1 A cognitive linguistic view of translation: Translation as

usage event and dynamic meaning construal

The creation of a translation,1 in whatever medium, is recognized by translation

scholars as an instance of discourse; that is, as a communicative event situated
in historical, cultural, and personal circumstances and impacted by the particu­
lars of those very real circumstances. Different theoretical approaches put more
or less emphasis on the various sources of contingency: history, culture, or per­
sonal politics, status, or position (or lack thereof). To oversimplify matters quite
a bit, some sociological and cultural frameworks prioritize cultural relation­
ships, ideology and hegemonic structures (e.g. Calzada Pérez 2003; Inghilleri
2003, 2005; Simeoni 1998; Wolf 1997, 2002; Wolf and Fukari 2007), while other
postmodernist theories tend to emphasize translator status and agency (e.g.
Pym 1998; Tymoczko 2006; Venuti 1995). To the extent that linguistic theories

1 In this paper, the term translation” is used to cover to what Jakobson referred to as “ in­
terlingual translation” , including interpreting, as well as “ intersemiotlc translation” , involving
other, non-linguistic semiotic systems (1959).
 Implications of Cognitive Linguistics for Translation Studies ----- 35

have emphasized language-internal or local contextual elements they have been

limited in their ability to incorporate the broader context of history and culture,
ideology and individual agency.2 In addition, both literary theory-based and
linguistics-based descriptive approaches have been criticized as being overly
concerned with “ texts and systems” , rather than people and their actions (Pym
2010: 84). A common denominator running through much of this criticism is a
call for increased focus on the role of the human translator and of translator
agency. From a cognitive perspective, an emphasis on the translating individual
must also involve a framework to deal with the specifics of linguistic cognition.
Indeed, in order to fully understand the complex interplay of the many sources
of impact in the act of translation, it is imperative to look at the human basis
in translational cognition, where all of these forces are brought together and
where the causal struggle is actually played out (see also Chesterman 2000).
How, then, could a cognitive linguistic theory of translation grapple with the
translational act? How might it be conceived? Starting with the consensual
agreement that translation is a discursive act, one might consider Langacker’s
outline of discourse, quoted here at some length:

A discourse comprises a series of usage events: instances of language use in all their com­
plexity and specificity. A usage event has no particular size; depending on our analytical
purpose, we can segment a discourse into words, clauses, sentences, intonation groups,
conversational turns, and so on. An event is bipolar, consisting in both conceptualization
and means of expression. On the expressive side, it includes the full phonetic detail of an
utterance, as well as any other kinds o f signals, such as gestures and body language (con­
ceivably even pheromones). Conceptually, a usage event includes the expression s fu ll con­
textual understanding - not only what is said explicitly, but also what is inferred, as well
as everything evoked as the basis for its apprehension. (2008: 457-458, author’s emphasis)

The view of language in discourse is given in the continuation:

Conventional linguistic units are just one resource exploited in usage events. In speaking
and understanding, we draw on our full range of knowledge, mental abilities, and interper­
sonal skills. Also essential is our apprehension of the context, one facet of which is the
ongoing discourse itself. The various factors contributing to usage events should not be
thought of as separate and discrete. In particular the specific contributions of language
cannot be segregated or precisely delimited. The linguistic meaning of a word, for example,
is not a distinct and self-contained entity, divorced from other knowledge and cognitive
abilities - instead it recruits and exploits them. (2008: 458, my emphasis)

2 importantly, not all linguistically oriented approaches to translation fall to this criticism.
Notable exceptions include the discourse-based work of Hatim and Mason (1990) and partic-
UlarlV Hatim and Mason (1997).
36 Sandra L Halverson

In this view, discourse, the series of usage events, draws on a broad range of
human meaning-making capacities and in the process makes use of the entire
embodied knowledge of the speaker/hearer, only some of which is knowledge
of conventional linguistic units (hence the emphasis above). From this perspec­
tive, discursive acts are meaningful in ways that far exceed traditional views of
linguistic meaning and that allow for the incorporation of personal, ideological,
cultural, contextual considerations in a new way. Discursive acts are still situ­
ated in time and space, but the parameters for the discourse are created by
the knowledge (conscious and subconscious) of the human interlocutors, not
solely by systems, linguistic or otherwise, texts, or structural relations between
other cultural entities.
A compatible cognitive linguistic account which is particularly amenable to
conceptualizing the translational act is that presented in Croft and Cruse (2004:
97) under the heading “ the dynamic construal of meaning” . The authors argue
for a view in which

[ . . .) words do not really have meanings, nor do sentences have meanings: meanings are
something that we construe, using the properties of linguistic elements as partial clues,
alongside non-linguistic knowledge, information available from context, knowledge and
conjectures regarding the state of mind of hearers and so on. (2004: 98)

This is very similar to the position outlined in Langacker, as cited above. In this
view also, it is important to note the description of linguistic elements as “ partial
cues” in meaning making, and also the significance of knowledge of various
kinds. Croft and Cruse describe the dynamic construal of meaning in terms of
four key concepts: (i) contextualized interpretation, (ii) purport, (iii) constraints,
and (iv) construal operations. From this perspective, an act of meaning making
represents a contextualized interpretation (Croft and Cruse 2004: 98). Such inter­
pretation involves purport or a “ body of conceptual content” associated with
words (Croft and Cruse 2004: 98). “ Purport” is to be conceived of as the raw
material of meaning, which is then transformed in specific instances of meaning
making. As the authors describe it, “ purport is to interpretation as egg is to
omelette, or flour to bread: it is of a different ontological category. Purport is
an ingredient of meaning, not a constituent” (2004: 101). In addition, the mean­
ing making/interpretation process is subject to a number of constraints, includ­
ing human cognitive capacities, the nature of reality, convention, and context
(linguistic, physical, social, and knowledge-related) (2004: 101-103), thus
grounding the act in the human cognizer in a discursive situation as s/he con­
ceives of it. The final element of this view of dynamic meaning is the notion
of construal, which consists of a series of cognitive processes which turn the
raw material of “ purport” into “ fully contextualized meanings” (2004: 103),
 Implications of Cognitive Linguistics for Translation Studies ----- 37

operating under the constraints listed above. Grammar reflects the operation and
conventionalization of construal processes. The crucial differences between this
view and other linguistic theories lie in the integrated nature of linguistic and
other kinds of knowledge and the integration of social, historical, and contex­
tual influences through their basis in the knowledge and cognitive processes
of the human communicator. This view of meaning creation is not deterministic
or mechanical: it is human in every regard.
If linguistic communication works through dynamic meaning construal,
then how do these key concepts operate in a translational scenario? Is there any­
thing that distinguishes a translational process from a non-translational one? If
so, what could it be? Though little technical integration of theory has been
accomplished, it is possible, at this stage, to tentatively adopt a view of transla­
tion that builds on the basic premises of the approaches to meaning making out­
lined above, while retaining the insights gained through several decades of
research in TS. Thus, the starting point is this: given the theory of language
sketched above, what does this imply for translation? First, it is necessary to
Posit an anterior3 text, spoken or written, linguistic, signed, or expressed in
some other symbolic form. (Incidentally, this is also true of other linguistic
acts such as paraphrase and/or the writing of a synopsis/abstract/precis [. . .])
The anterior text is expressed (in Langacker’s terms [2008: 457-458]) in a differ­
ent form than the ensuing translation. The creative translation process involves,
nhnimally, (i) an emerging and dynamic contextualized interpretation of the
anterior text (including knowledge of the relevant aspects of its discursive, his­
torical, cultural context), (ii) constraints in the translational situation itself, and
(iii) in the discursive situation for which the emerging translation is destined (in­
cluding rich knowledge of the sort mentioned for the anterior text), (iv) a con­
ceptualization of the translational act itself, both generally and in the specific
Present, and (v) the dynamic construction of a new text. Underlying and sup­
porting all of this is the encyclopedic knowledge base of the specific translator.
In other words, the translational act is comprised by the activation and selective
use of several particular kinds of knowledge, including linguistic knowledge, all
of which contribute to and constrain the process.
As cited above, constraints in contextualized interpretation include human
cognitive capacities, the nature of reality, convention, and context (linguistic,

^ The term “ anterior text” is preferred to the more traditional "source text or the older
original". Both of the latter suggest some sort of logical primacy or inherent status that is not
consistent with the view to be presented here. By referring to an “ anterior” text, the chrono-
,0gicat aspect is highlighted.
38 Sandra L. Halverson

physical, social and knowledge-related)” (Croft and Cruse 2004:101-103). Thus,

in addition to knowledge-related contextual constraints such as those men­
tioned above, there are also contextual constraints of a more social nature.
Indeed, much work w ithin functionalist TS in the 1980s and 1990s aimed at il­
lustrating the importance of social factors (for surveys, see Munday 2008; Nord
1997; Pym 2010; Schaffner 2009). Examples of such social constraints include
the specifics of a translation brief or an assumed brief, i.e. specific instructions
or guidelines provided by the translation commissioner or initiator, the status or
position of the translator w ithin an organization or in the translational situation
itself (e.g. in a courtroom, in a police station, in a war zone, etc.), and the use of
technology, among other things. Importantly, however, these constraints are
filtered through the cognitive process of the translator, and as such are also
knowledge-related. This cognitive perspective on context is not one that is unique
to the cognitive linguistic theorists drawn on here, nor indeed to the field of
linguistics itself. Several of the relevant perspectives with regard to translation
are introduced in Baker (2006), who argues for a more dynamic perspective
on the process of contextualization, as opposed to static delineation of contex­
tual elements and their relations. Baker discusses both social and cognitive
theories of context, and emphasizes that the two represent alternative and com­
patible perspectives (2006: 324). A related, though separate and larger issue, is
the demarcation of mind and world in cognitive processes (Clark 1997, 2008). In
short, it is argued that, in many ways, it is not feasible to draw clear and definite
boundaries between a cognitive operation and the situation in which it is em­
bedded.4 At the very least, this suggests that cognitive and socially-oriented
theories should both bring analyses to the table, and that there are currently
interesting ways in which they may inform one another.

1.2 Cognitive linguistic theory and bi- or multilingualism

In its earliest days, the theoretical framework of Cognitive Linguistics primarily

modeled linguistic communication among monolingual speakers of the same
language. In more recent years, a growing body of work in Contrastive Cogni­
tive Linguistics has investigated cross-linguistic differences w ithin a cognitive
framework (e.g. Valenzuela and Rojo 2003) and work w ithin second language

A Within Translation Studies work on "situated cognition” has been done by Risku (2002).
These perspectives are also integrated into the work being done by the PACTE group (2003,
2005).
 Implications of Cognitive Linguistics for Translation Studies ----- 39

acquisition is increasingly adopting cognitive linguistic perspectives. None­

theless, there has not as yet been a systematic attempt to model the particular
type of bilingual communication that is translation.
Even monolingual communication is complicated enough: as Langacker
points out, the usage event experienced by the speaker is not precisely identical
to that experienced by the hearer, as the knowledge base of each participant is
personal and unique, allowing for variations in what is inferred and evoked
(2008: 458). As he points out, however, “ substantial overlap is usually enough
for successful communication” (458). The conceptualizations of two speakers of
the same speech community are, however, grounded in somewhat shared envir­
onments, and successful linguistic interaction over time provides a basis for
achieving sufficient “ overlap” . It is obvious, however, that in intercultural com­
munication, whether monolingual, bilingual, or multilingual, the assumption of
shared lived environments does not hold: the knowledge bases that may be ac­
tivated and evoked, as well as the conventional paths of inference, may differ
widely. Thus it seems that a cognitively oriented theory of translation must in
some way incorporate a theory or model of knowledge, linguistic and otherwise,
■n bi- and/or m ultilingual people.
In a recent article (Halverson 2010a), it was argued that a model of bilingual
representation such as that put forward in Jarvis and Pavlenko (2008) and
Pavlenko (2009) is a viable source for this purpose, as it shares both cognitive
assumptions and some conceptual elements with the cognitive linguistic
theories mentioned above, for example, a means of capturing prototype effects
and patterns of connectivity, such as polysemy, and the knowledge-based, cul­
turally situated view of meaning. The model includes distinctions between con-
ceptual, semantic, and phonological representational levels, which are required
for the study of translational data (see Section 2.2). A detailed discussion of
these models is beyond the scope of the current paper, as much work is needed
'n order to bring the insights of bilingualism research and Cognitive Linguistics
together in specific and testable ways. Work in that area is in progress, as
witnessed by work cited above (see also Robinson and Ellis 2008 and the
2009 special issue of the Annual Review o f Cognitive Linguistics for comparable
developments in second language acquisition). For the purposes at hand, it is
vital to single out the cognitive aspects of bilingualism that are of most impor­
tance to further work on translational cognition. At this stage, at the very least,
cognitive theories should not violate or come into conflict with current knowl-
edge of bilingual representation and processing. In the following three central
areas of research are touched on.
The first area of investigation concerns the modeling of conceptual repre-
Sentations in a bilingual, more specifically, the relationship between the
AO Sandra L. Halverson

semantic/conceptual level5 and the level at which linguistic forms in a b ilin ­

gual’s two languages are stored. One of two dominant models of bilingual concep­
tual representation, the so-called Revised Hierarchical Model (Kroll and Stewart
199A) posits links between a conceptual tier that is common to a bilingual’s two
languages, and a lexical tier, where the formal elements of the two languages
are stored separately. So-called “ conceptual links” run between the conceptual
tier and the lexical tier, and so-called “ lexical links” run between the lexical
items in the two languages. According to the model, the links within this system
are not all of equal strengths: there is asymmetry. This is outlined as follows:

The revised hierarchical model makes two critical assumptions about the strength of con­
nections between words and concepts in bilingual memory. The first is that LI words are
assumed to be more strongly connected to concepts than are L2 words. The second is
that L2 words are assumed to be more strongly connected to their corresponding translation
equivalents in LI than the reverse. (Kroll and Tokowicz 2005: 546)

The developmental aspect is integral to this model, as increasing L2 proficiency

is claimed to strengthen the links between L2 words and the conceptual level.
According to the model, over time, proficient bilinguals w ill also be able to pro­
cess L2 words conceptually, through stronger links to that level, though asym­
metry w ill remain for “ all but the most balanced bilinguals” (Kroll and
Tokowicz 2005: 5A6). The authors proceed to present experimental evidence in
which the assumptions of the model were tested with regard to the lexicon.
The results provided support for the model, but also suggested that results
are not uniform across all word types, with differences for high and low fre­
quency words, for example (Kroll and Tokowicz 2005: 5A8). A similar concern
for differences across word types was one of the motivations for Pavlenko’s artic­
ulation of an alternative version of the model, dubbed the “ Modified Hierarchi­
cal Model” (2009: 146-151). Notably, the latter model allows words in different
languages to share conceptual representations to differing degrees.6
This model is in some ways an oversimplification of some of the relevant
characteristics of bilingual conceptual representation (Pavlenko 2009: 142-146),

5 The use of the terms “ semantic" and “ conceptual” representations is inconsistent in the
bilingualism literature (see Francis 2005). The terms are sometimes used interchangeably, and
sometimes one or the other is used. In this paper, the two will be used to refer to two separate
levels, in line with Jarvis and Pavlenko (2008).
6 This partial sharing is also a feature of the other main model of bilingual representation, the
“ Distributed Feature Model" (de Groot 1992.1993). Though it does cater for this distinction, the
model suffers from other shortcomings (see Jarvis and Pavlenko 2008: 225-226), which has
repercussions for its viability for the study of translation (Halverson 2010a).
 Implications of Cognitive Linguistics for Translation Studies ----- 41

and a cognitive theory of translation must ultimately address these issues in a

detailed and specific way. However, the model outlined above serves as an
illustration of some of the relatively uncontroversial aspects of bilingualism
that cognitive translation scholars must take on board. Even a surface level
understanding of the types of models mentioned here, and the evidence on
which they are based, makes it clear that translational outcomes are affected
by at least the following factors: (i) the degree of language dominance in the
translator and how this is related to (ii) the direction of translation (into or
out of the dominant language), (iii) the developmental trajectory (order of
acquisition of languages, stage of proficiency), and (iv) the specific linguistic
item(s) or task(s) in question. On this account, even bilinguals sharing the
same two languages may translate differently, depending on the degree of
proficiency in a given language, the individual’s own developmental trajec­
tory, the direction of the translation, and the item(s) and task specifications
involved in translation. A ll of these factors are compounded in the case
of m ultilingual speakers, a group which probably constitutes a relatively siz­
able percentage of professional translators (cf. Jarvis and Pavlenko 2008:
228-229).
A second issue of current interest in the bilingualism literature is the joint
activation of a bilingual’s two languages in language production. There is
quite compelling evidence to support the claim that both of a bilingual’s lan­
guages are activated in language production (Kroll et al. 2008: 427; Hartsuiker,
Costa, and Finkbeiner 2008: 413), and that this is true also for quite proficient
bilinguals (Kroll et al. 2008: 427). Current work focuses on identifying and inves­
tigating the detailed control mechanisms that ultimately support selection of the
appropriate target language. For a cognitive theory of translation, the issues
Mentioned here raise numerous questions with regard to the conceptualization
° f linguistic knowledge and the processes by which this knowledge is used.
Some of these issues w ill be considered in the following section.
Finally, w ithin translation process theories, there has been some discus-
sion of whether the process of interpreting the anterior text and the process
° f creating the new translation are either serial processes, parallel processes
0r a hybrid of the two (see Ruiz et al. 2008 for review). With regard to the
hybrid model, it has been suggested that varying degrees o f experience or
Gaining would determine the type of processing favored by any individual
translator (Ruiz et al. 2008: 491 and sources cited there). In their study, Ruiz
et al. (2008) provided experimental evidence in support for a parallel processing
rnodel: in other words, ST comprehension and TT activation processes were
found to run in parallel. Further studies w ill undoubtedly shed more light on
this issue.
42 Sandra L. Halverson

2 Implications of Cognitive Linguistics

for Translation Theory
In this section, a small selection of key translational issues is in focus. In each
subsection, relevant background and the current state of knowledge in TS are
outlined first; then the new perspectives provided by Cognitive Linguistics are
sketched out. The aim is to single out those elements of the cognitive linguistic
and bilingualism frameworks that might provide the necessary grounding for
more specific theorizing and empirical investigation.

2.1 Two sides of the same coin: Equivalence and shifts

2.1.1 Equivalence in Translation Studies

In its emergent and early stages in the 1960s up until the 1980s, contemporary
Translation Studies was primarily identified with the study of relationships
between source and target structures. The concept of translation itself was de­
fined with reference to a concept of “ equivalence” , and translations were iden­
tified as texts that met the requisite equivalence criteria (Halverson 1997; Koller
1995, Munday 2008: 38ff; Pym 2010: 6ff). In many ways, early work represented
a variety of ways of coping with what Jakobson referred to as “ equivalence in
difference” (1959), or the problem of how different languages constrain the
ways in which their speakers may express “ the same thing” . Work w ithin this
paradigm focused both on the identification of the invariant (the tertium compar-
ationis) on the basis of which equivalence was to be established, and on the
elaboration of the various linguistic levels or units that might be affected. Cat-
ford (1965) was careful to distinguish between formal correspondence (between
languages) and translational equivalence (between texts or textual elements).
He proceeded to elaborate a detailed system for the description of translational
relationships that included reference to linguistic levels and structural cate­
gories, e.g. phrase, clause, sentence, text and “ shifts” between them, emphasiz­
ing variability of equivalence types within a text. Nida (1964) recognized the
need for a means of capturing pragmatic elements in the relationship taxonomy
and distinguished between 'formal equivalence” , or equivalence of form and
message, and dynamic equivalence” , which catered for equivalence of response
in source and target contexts. Koller provided an even more detailed taxonomy
of equivalence types, including denotative, connotative, text-normative, prag­
matic and formal equivalence (1989). Each type identifies the invariant on the
basis of which an equivalence relationship between source and target is
 Implications of Cognitive Linguistics for Translation Studies — 43

established: here the invariants are extralinguistic content, lexical connotations,

text type, communicative effect, and linguistic form, including wordplay and
typical characteristics of individual authors, respectively (1989). Evolving
theories of language brought with them new approaches to defining the equiv­
alence relationship, as discourse and pragmatic concerns entered into the anal­
yses, while the focus remained on the identification of appropriate invariance
types and units. The entire equivalence paradigm gradually fell into disfavor,
due to the lingering uncertainty w ith regard to the appropriate “ unit of transla­
tion” and the inevitable proliferation of equivalence types, as well as the con­
cept’s perceived inability to address the real concerns of TS (Snell-Hornby
1988). The various equivalence typologies also proved unable to fully capture
the reality of translational data (Koller 1995).
In a review of the history of the equivalence concept in TS, Pym (2007,
2010) distinguishes between what he refers to as “ natural” and “ directional
concepts of equivalence. In Pym’s view, the work sketched above is concerned
with “ natural equivalence” , or relationships between textual/linguistic elements
that are determined by a notion of naturalness of expression that may be
achieved in spite of structural or pragmatic differences between the two involved
languages. According to Pym, natural equivalence “ should not be affected by
directionality: it should be the same whether translated from language A into
•anguage B or the other way around” (2010: 7). “ Directional equivalence” , on
the other hand, assumes some sort of directional asymmetry. Pym introduces
the term “ to refer to all those cases where an equivalent is located on one
side more than the other, at least to the extent that they forget to tell us
about movements that could go either way” (Pym 2010: 28). The asymmetry of
the relationship is often illustrated in binary pairs of translation types where
each of the members in a pair captures the tilt of the equivalence balance
towards either the source or target culture/language/discourse complex, for
Sample foreignizing versus domesticating, semantic versus communicative, doc­
amentary versus instrumental, resistant versus fluent, etc. (Pym 2010: 30-33).
While these types of relationship are different from those focused on natural-
aess” , they share the overall emphasis on articulating the nature of the
source-target relationship (cf. Pym 2010: 28).
In his review, Pym states that Toury (1995), Gutt (2000) and Pym (1992) rep
resent a genuinely different approach to the idea of equivalence through a com-
mon focus on what he refers to as “ equivalence beliefs” (2010: 37-38). These
authors all posit means of investigating source-target relationships on the
basis of historically contingent practices or beliefs, and they all involve some
SOrt °fp o s i hoc identification process, rather than the a priori definitions posited
Previously. For Toury, all translations are by definition equivalent to their
44 Sandra L Halverson

sources, and the scholar’s task is to describe and explain how equivalence is
realized. Gutt, from a relevance theoretic perspective, argues that translations
are those texts which present themselves as interpretations of anterior texts,
and are consequently believed to be so. Pym emphasizes the historical contin­
gency of equivalence-making practices and beliefs. The shift from a priori defi­
nition of objective relationships to post hoc belief practices is the only viable
approach to “ equivalence” in TS, according to Pym (2010: 40-41).
To sum up: the history of the equivalence concept in TS is the story of evo­
lution from the search for a set of relationships between objectively identifiable,
“ natural” linguistic structures and/or contextual features towards dynamic,
changing, historically situated conceptualizations of source and target and the
balance between them. Over time, “ equivalence” has been reconfigured from
“ essential characteristic of translations, defined a priori” , to “ quality of all trans­
lations, to be characterized post hoc” . This evolution, in fact, amounts to an epis­
temological recategorization. The concept is no longer locked into an essentialist
epistemology; it is, in its latest version, a subjective belief concerning a text or
texts, relativized in time and place (Pym 2010: 37; see also Halverson 1997).

2.1.2 Equivalence from a cognitive perspective: From correspondences

to knowledge constraints

W ithin a dynamic meaning construal approach to translation such as that out­

lined in Section 1.1, there is little room for distinct and uniquely isolable entities
such as the traditional, systemic source-target equivalents, of whatever size or
kind. To state the obvious, the first, and in some sense the only real, place
where the source and target entities involved in a specific translation task actu­
ally come together is in the mind of the translator (though analysts may attempt
to reconstruct this meeting later on). And while there are, clearly, relevant inter­
lingual links involved, these are by no means the same, either in terms of how
they are cognitively represented or cognitively processed, as the kinds of binary
equivalence relationships that have been posited in TS. Indeed, as mentioned
above, psycholinguistic studies of bilingualism demonstrate that, from a cogni­
tive perspective, a bilingual’s two languages are not completely separate (de
Groot 1992, 1993; Dong, Gui, and Macwhinney 2005; Kroll et al. 2008: note 1;
Pavlenko 2009). Instead, the picture emerging from studies of bilingual repre­
sentation suggests that there are complex patterns of linkage involving repre­
sentational elements at conceptual, semantic and formal levels, and that the
conceptual level is to some degree shared by a multilingual speaker’s languages
(Jarvis and Pavlenko 2008). Pavlenko (2009) also addresses the question of
 Implications of Cognitive Linguistics for Translation Studies ----- 45

language/culture-specific knowledge, or knowledge that is linked to linguistic

items in only one of a multilingual speaker’s language, and incorporates such
cases into a model that caters for varying degrees of conceptual overlap, ranging
from zero to completely shared. The point to be made here is that from a cognitive
perspective, the interaction of the languages that a bi- or multilingual speaker
knows is much more complex than any binary correspondence (equivalence)
view could ever capture.
In the cognitive view, a linguistic act is conceived of as dynamic, using con­
ceptual raw material and construal operations, under pertinent constraints.
Adopting this view does not preclude some associative linking of elements of
one language w ith elements of another w ithin the speaker’s cognitive system;
but as argued in the bilingualism literature cited above, these links are complex,
involving several representational levels. As a consequence, the task of a cogni­
tive translation scholar is to look for ways in which specific linguistic items may
be represented and activated in the language of a bilingual and how various re­
presentational characteristics might impact translational outcomes. In transla­
tion, an antecedent, linguistically activated, conceptualization in the mind of
3 translator is an integral part of the dynamic creation of another text. Linguistic
correspondences, of the type previously posited as equivalence relationships,
instead must be situated w ithin a cognitive model of linguistic knowledge,
which means that their status must be reassessed. Complex, multilevel, interlin­
gual links must be conceived of as one of the many types of knowledge that are
3ctivated and utilized in the translational act (Langacker 2008: 458; Croft and
Cruse 2004: 98), and as such these links function as a constraining factor in
the process. They do not constitute it.
In short, it would seem that a cognitive theory of translation does not need a
concept of source-target equivalence for future theorizing. However, as men-
tioned in Section 1.1, in addition to linguistic knowledge, the translational act
also involves a translator’s conceptualization of the translational act itself,
and this conceptualization, in turn, w ill serve as a contextual constraint on
the translational act.
Studies of various “ translation” concepts (Chesterman 2006 and references
cited there; Halverson 1999a, 2008; Tymoczko 2007 and studies cited there,
Taker 2002) provide relevant examples of different conceptualizations, even
though these studies were not carried out w ith this in mind. Another relevant
example is Toury’s “ initial norm” , described as the translator s choice of basic
orientation towards either target or source (towards either being a text [. . .]
0r ‘constituting a representation [. . .]” ) (1995: 56). As pointed out in Halverson
(1999a: 22), this notion actually captures the translator’s concept of translation.
The conceptualization that a translator maintains of the very activity in which
46 Sandra L. Halverson

s/he is engaged could, then, serve as the theoretical site for what Pym referred to
as “ equivalence beliefs” . In other words, if the idea of equivalence is to have any
role at all, then it may be w ithin the concept of “ translation” that a translator
has, and as such it functions as a knowledge-based constraint, on a par with
a number of other such constraints. In this view, the causal force of equivalence
beliefs lies in the mind of the translator. But such belief systems must also be
investigated in terms of the cultural, historical, and social contexts in which
the translator is embedded and in which these beliefs are shared (see
Halverson 2008).

2.1.3 Shifts in translation theory

Another concept used in studying and accounting for source-target relationships

is the notion of “ shifts” (Molina Martinez and Hurtado Albir 2002; Chesterman
2005; Halverson 2007; Marco 2007). As these authors point out, “ shift” is one
of a variety of terms that are meant to capture either translational procedures,
conscious or subconscious, routine or automatized, problem-oriented or not,
or relationships between translation products and their sources. Terms often
vary also in their scope: whether they cover global or local procedures or struc­
tures. Terms such as “ procedure” , “ technique” , “ strategy” , “ method” , “ shift” ,
etc. are used differently by different authors, and different terms are also used
to refer to the same concept. In several of the analyses mentioned above, the
authors argue for conceptual clarification and terminological consensus, though
they disagree on the requisite distinctions. Chesterman, for example, proposes a
four-way conceptual framework (2005: 26-27), distinguishing between “ method” ,
“ strategy” , “ technique” , and “ shift” , while Marco argues that all that is needed
is a two-way conceptual and terminological distinction between translational
procedures and post hoc relationship categories (2007: 262-263).
Cross-cutting the question of whether the source-target relationship is
procedural or related to end-products is the underlying issue of change. There
is a fundamental tension inherent in the notion of “ shift” as used to refer
to “ [. . .] changes which occur or may occur in the process of translating” (Bak-
ker, Koster, and Van Leuven-Zwart 2009: 269). As the authors point out, the
notion of “ shift” or “ change” implies that there is an invariant, and shifts or
changes are identified and characterized relative to that invariant. One well-
known representative of this approach is Popovic, who defines shifts as “ All
that appears as new with respect to the original, or fails to appear where it
may have been expected [.. (1970: 79). Thus procedural approaches may sug­
gest operations that result in changes, while product-based categories classify
 Implications of Cognitive Linguistics for Translation Studies ----- 47

change types. The most well-known frameworks of the former type are Vinay
and Darbelnet (1995) and Klaudy (1996), which posit such processes as, modula­
tion, adaptation, transposition, etc., operating on linguistic structures of various
sizes and type. The most well-known of the latter type are arguably Catford
(1965) and Van Leuven-Zwart (1989, 1990). Catford identified shifts as “ depar­
tures from formal correspondence” (1965: 73), and his framework allows for
the various types of formal differences, often involving either category shifts
(e.g. syntax to lexis or vice versa) or rank shifts (e.g. from clause to phrase,
from phrase to lexeme, etc.). Van Leuven-Zwart’s framework involved the deri­
vation of a common semantic denominator linking ST and TT segments, and
shifts were then classified in terms of extent of change from the invariant and
the level and type of linguistic element affected.
Chesterman illustrates the problem of operationalizing concepts of shift or
invariance by demonstrating how shift analysis often ends up with overlapping
or indeterminate categories, which is related to the problem of identifying the
“ unit of translation” (2005: 24-25). This problem was also implicated in the in ­
creasing number of equivalence types, mentioned in Section 2.1.1. Also relevant
to the issue of invariance is the fact, identified by both Chesterman (2005) and
Marco (2007), that a number of the types of strategies/techniques/operations/
Procedures may end up in either similarity or difference. On this basis, Chester-
oran is able to elaborate a set of parallels between types of equivalence and
types of shifts (2005: 27). Chesterman advocates the term “ solution type” ,
Munched by Zabalbeascoa (2000: 122) to cover post hoc relationships of both
similarity and difference.
To sum up: the term “ shift” has traditionally been applied to translational
°Peraticns or actions and to categories of source-target relationships identified
after the fact, though at present some scholars are calling for a clearer distinc­
tion between the two. Similar and related terms, such as “ method” , “ strategy ,
“ operation” , and “ technique” , have also been used with reference to this con-
cePtual territory, and the term “ shift” has traditionally been used to refer to
cases where there is some change from an identified invariant of some kind,
though this is problematic in cases where a given operational translational pro­
cedure or strategy may result in either similarity or difference. There is no obvious
0r agreed solution to the question of how an invariant may be established.

* ’*•4 Shifts as construal operations

mentioned in the discussion of equivalence, it is not immediately obvious

what role invariance or elaboration of source-target relationships of similarity
48 Sandra L. Halverson

or dissimilarity should have in a cognitive linguistic theory of translation. Con­

sequently, the notion of “ shift” , in the sense of post hoc product categories
established on the basis of an identified invariant, is problematic for the
same reasons as “ equivalence” is. As suggested above, in a dynamic meaning
construal account, interlingual relationships are much more complex than
theories of equivalence or shifts have previously allowed. Moreover, linguistic
cues represent only part of the raw material used in meaning making.
If the notion of shifts is considered in its procedural sense, starting with the
dynamic contextualized interpretation of the anterior text and without an
assumption of an invariant core, it becomes easier to situate w ithin a cognitive
theory of translation. In an earlier discussion (Halverson 2007), it was argued
that many of the proposed procedural categories of shift/procedure/strategy/
etc., especially those in Vinay and Darbelnet (1995) and Klaudy (1996), may
be conceived of as “ construal operations” , as described in cognitive theories
(Croft and Cruse 2004: 40-73; see also Langacker 1987). Recall that construal op­
erations transform the raw material of conceptual purport into contextualized
meanings that are conventionally linked to specific forms. At this level, transla­
tional acts are no different from monolingual communicative acts: conceptual
content is transformed and linked to linguistic form. Indeed, as pointed out
earlier, in one sense translation is construal (Halverson 2007).
In a cognitive linguistic translation theory, construal operations would be
implicated in three ways: (i) in the translator’s contextualized interpreting of
the anterior text, (ii) in the translator’s knowledge of the inventories of conven­
tionalized construals in the languages s/he knows, and (iii) in his/her online,
creative generation of a translation, as mentioned above. If translation is seen
as dynamic meaning construal, then (i) and (iii) are actually part of the same
ongoing process, though they may represent two alternating areas of attention.7
With regard to (ii), however, it is important to avoid the risk of returning to static
source-target comparisons. There are, naturally, inventories of conventionalized
linguistic structures, and translators are in some ways constrained by them. If
these inventories of linguistic elements are reassessed within a cognitive frame­
work, they are reduced to the level of one of a number of constraints in the trans­
lator’s creative process, operating alongside other knowledge-based constraints.
Viewing construal operations as integral to, as actually constituting the transla­
tion process, allows us to maintain many of the insights of previous work on
translational procedures such as Vinay and Darbelnet’s methodology (1995) or

7 This issue is the subject of ongoing research into serial and parallel processing, as men­
tioned in Section 1.2.
 Implications of Cognitive Linguistics for Translation Studies ------ 49

Klaudy’s translational operations (1996), while at the same time emphasizing the
creative, non-deterministic nature of the process. Translational alternatives are
seen as creatively generated, while choice is constrained by cognitive factors,
convention, and context.
A view similar to the above is articulated in detail in Lewandowska-
Tomaszczyk (2010). We differ in some of the particulars, but in the main, the
argument for a construal-oriented approach is the same. A full comparison of
the two perspectives is beyond the scope of the present discussion.

2.2 “Translation universals”: From description to

explanation

2.2.1 Describing patterns

One of the areas in which linguistic approaches to translation have had a recent
resurgence is corpus-based Translation Studies. Starting with the development
of a number of parallel corpora in the early 1990s, for example the English-
Norwegian Parallel Corpus and the English-Swedish Parallel Corpus, ultimately
joined under the Oslo Multilingual Corpus (http://hf.uio.no/ilos/OM C/), and
the first so-called “ comparable corpus” , the Translational English Corpus in
Manchester in the mid-1990s (http://www.monabaker.com/tsresources/Trans
lationalEnglishCorpus.htm), a number of translation scholars have used corpus
data and corpus-based methods (see Laviosa 2002; Olohan 2004; Kenny 2009 for
surveys). Many, though not all, of the studies in this paradigm involved inves­
tigation of so-called “ translation universals” ,8 defined by Baker as features
which typically occur in translated text rather than original utterances and
which are not the result of interference from specific linguistic systems
(!993: 243). The list o f candidate universals suggested by Baker included:

~ simplification (the idea that translators subconsciously simplify the lan­

guage or message or both)
explicitation (the tendency to spell things out in translation, including, in its
simplest form, the practice of adding background information)

8 Baker’s use of the term "universal” has been criticized as conceptually unclear (Pym 2008)
and as inappropriate for the task it was selected to do (Malmkjaer 2008). For these and other
reasons, in the following the term “ features of translation” (or features of translated text) used
by ®l°han (2004) will be adopted here also.
50 Sandra L Halverson

- normalization or conservatism (the tendency to conform to patterns and

practices which are typical of the target language, even to the point of exag­
gerating them)
- levelling out (the tendency of translated text to gravitate around the centre of
any continuum rather than move towards the fringes) (1996: 176-177)

Patterns such as these may be investigated through primarily two kinds of inves­
tigations: comparing translations with their sources, or comparing translations
with non-translated texts in the same language (Chesterman 2004). It was pri­
marily the latter kind of comparison that was advocated by Baker, through
the introduction of the comparable corpus methodology (Baker 1993, 1995).
While Baker’s program served as an impetus to a number of studies (e.g. the col­
lections edited by Mauranen and Kujamaki 2004; Anderman and Rogers 2008),
the conceptual framework came under pressure from several directions. Mal-
mkjaer (2008) argues that patterns such as those mentioned above are better
conceived of as norms, rather than universal (2008: 57). Pym (2008) argues
that Baker’s universals are not conceptually distinct and that they ultimately
reduce to the two “ laws of translation” originally put forward by Toury (1995),
i.e. “ the law of increasing standardization” and the “ law of interference” . Hal­
verson (2003) argues that the paradigm lacks an adequate theoretical founda­
tion from which to hypothesize about potential explanations. In later work,
Halverson (2010b) picks up on an element of the earlier paper, and argues
that many of the proposed universal patterns may, in fact, not be unique to
translation, and that they ought to be conceived of as natural effects of bilingual
language production.
In later work, Sonja Tirkkonen-Condit put forward another candidate for the
status of translation universal in her work on the “ unique-items hypothesis”
(Tirkkonen-Condit 2004, 2005). This hypothesis states that linguistic items
which “ lack straightforward linguistic counterparts” (2004: 177) in the source
language w ill be underrepresented in translated language, as opposed to non-
translated language. The hypothesis is supported in both studies and by Eskola
(2004) and Kujamaki (2004), and is advocated as a candidate for universal status
also by Malmkjaer (2008).

2.2.2 Features of translation: Towards a possible explanation

The types of patterns suggested by Baker in the 1993 article, while arguably
lacking in conceptual clarity, seem to capture some recurring and recognizable
features of some translational and learner data (some of which is reviewed in
 Implications of Cognitive Linguistics for Translation Studies ----- 51

Halverson [2003]). However, unless the paradigm is able to put forward an ade­
quate explanatory model, conceptual clarity w ill also remain elusive. In Hal­
verson (2003), it was argued that an appropriate explanatory framework may
be found in the theory o f cognitive grammar. More specifically, it was proposed
that the features of normalization/conservatism (Englund-Dimitrova 1997;
Kenny 2001), conventionalization and exaggeration o f target language features
(Baker 1993) may be the effects of a what was referred to in that paper as
“ gravitational p u ll” , or the effect of the cognitive salience of high-level sche­
mas and/or prototypes w ithin schematic networks. The hypothesis is that
linguistic forms linked to highly salient cognitive structures would be overre­
presented in translated, as opposed to non-translated text (Halverson 2003:
218f).
In deriving the gravitational pull hypothesis, several studies were cited
that provided evidence for effects of category structure on language production
m learner language (e.g. Hasselgren 1993; Ijaz 1986; Kellerman 1978, 1979).
These studies showed that learner language showed effects of both LI category
structure and what Ijaz referred to as “ the semantic equivalence hypothesis”
(1986: 433). In her study, Ijaz found that “ ESL learners [ . . . ] approximate native
speakers more closely in the meaning they ascribe to typical or central in ­
stances o f semantic categories than in the meaning ascribed to noncentral
ones” (433). Effects of category structure (here prototypicality) on cross-
Imguistic influence is also discussed in Jarvis and Pavlenko (2008: 186-188),
and evidence is given for an effect of prototypicality on L2 word choice (Jarvis
1998).
The gravitational pull hypothesis, as put forward in Halverson (2003), also
aftempted to account for the data emerging from studies of the unique items
hypothesis by suggesting that schematic networks without prototypes or highly
salient schemas, or w ith “ weak” or “ distant” links w ithin bilingual networks,
a^ay tend towards underrepresentation of TL structures. In a subsequent project,
attempts to operationalize “ network linkage” revealed that the model presented
ltl Halverson (2003) did not have the distinctions needed to capture the differ-
ence between effects of network salience and effects of differences in network
configuration (type or strength of the links). This was hinted at in Section 1.2
(see Halverson 2009, 2010a). The problem may be resolved through the integra-
hon of a newer model of bilingual conceptual representation, with two distinct
levels to capture these two difference sources of potential effect, as suggested in
A ction 1.2.
The ongoing project to test the gravitational pull hypothesis (outlined in
Halverson 2009, 2010b, in progress b) has emphasized the need for further
VVOrk on the integration of the Cognitive Grammar and bilingualism models. If
52 Sandra L. Halverson

the hypothesis is supported through this and other studies, then this would be a
demonstration of how cognitive theory can contribute to the explanation of the
linguistic characteristics of translated text. Success in this endeavor rests, not
least, on a number of important methodological issues, some of which w ill be
addressed in Section 3.

2.3 Summing up: The implications of Cognitive Linguistics

for Translation Theory

This section has been organized around a selection of well-known issues

which have been central to the study o f translation for the past sixty years:
equivalence, shifts, and more recently, translation universals. To reiterate:
this does not exhaust the list of relevant or interesting issues. These have
been selected due to their historical and current status in the field and the
ensuing necessity for any linguistic theory to take a stance w ith regard to them.
Translation scholarship has moved past the stage at which the key objective
was to elaborate systemic frameworks for the identification and classification of
source-target relationships on the basis of a stable invariant core, though
there is s till some unease about how to tackle the issue of equivalence or sim­
ilarity. Pym (2010) has suggested that the answer is to consider the role of
equivalence as a belief: this is entirely feasible, and indeed necessary if we
adopt the cognitive approach outlined above. Every translator is constrained
by his/her conceptualization of what “ translation” is: in many cultures, an
equivalence belief is part o f this. Importantly, however, this may not be u ni­
versally true. Different cultures, at different times, may have conventionalized
different belief systems, as suggested by, among others Tymoczko (2006) and
Chesterman (2006).
The bilingualism literature provides at least one framework that might
inform cognitive linguistic theory in extending it to the investigation of bilingual
cognition. Having considered some of the findings from this work, it becomes
imperative to ask whether Translation Studies actually needs, or is at all served
by the kind of binary relational framework that underlies the concepts of equiv­
alence, similarity, and shifts (in its post hoc product sense). The sources of this
mindset are relatively obvious: the TRANSFER metaphor that structures our
understanding of the “ translation” concept (Halverson 1999b; Tymoczko
2007), the highly visible and tangible nature of both ST and TT, as well as the
analytical separability of the two involved language systems. But, if translation
is a kind of dynamic meaning construal, (where linguistic and non-linguistic
knowledge, only partially separable and jointly activated, is constrained by
 Implications of Cognitive Linguistics for Translation Studies ------ 53

the real world), cognition itself, and a variety of contextual factors, themselves
the subject of online construction, then this process is a much richer, a much
more creative and organic one than a structural comparison, binary relation
view could ever capture. The very basis for comparison, the tertium comparatio-
nis, vanishes in this view, or at the very least, is buried deep in the cognitive
links, the activation patterns established between words, constructions or
other kinds of knowledge in the mind of the translator. It is not needed to
explain translation.910
The concept of translation shift, if reframed in one of its two current mani­
festations as a set of procedural operations, can be incorporated within a cogni­
tive theory through its conceptual overlap with construal operations, as argued
in Halverson (2007).'° The notion of translation “ universal” or translational pat­
terns or features requires further clarification, but this clarification must be rel­
ative to a theory w ith some explanatory power. It has been argued (Halverson
2003) that the theory of cognitive grammar has explanatory potential here,
and that, if combined with the insights of current work on cross-linguistic influ­
ence such as that by Jarvis and Pavlenko (2008), could prove even more fruitful.
Thus both procedural “ shifts” and “ universals/features” remain central to a
cognitive theory of translation.
From the perspective of the cognizing human translator, rather than of, say,
’niplicated languages or cultures, the need is for a more organic translation
theory with a more unitary focus on one engaged and situated mind. From the
translator's perspective, languages, cultures, situations and conventions, indeed
that s/he knows, is embodied in the personal mind. In this view of translation,
there is no invariant, no movement, no transfer, no carrying over. There is one
creative process, drawing on and utilizing the cues provided by the anterior
text and the rich cognitive resources of the translator, subject to a set of

9 Tabakowska (1993) also adopted a cognitive linguistic approach to translation, and in her
view, (he “ image” or conceptualization, could serve as the tertium comparationis. This was a
liv a b le contribution to thinking in TS, as it highlighted the role of cognition in translation. In
adopting a view that attempts to bridge cognitive linguistic and bilingualism research, I argue
that we do not need an invariant, or basis for source-target comparison, and that the continuing
Search for it hinders further theoretical innovation.
10 In that article, the process/product distinction was not adequately made, and the analyses
actually may suggest a product-type interpretation. The operations are to be considered pro­
cedural and dynamic, while further work is needed to elaborate a means of relating these
°Perations to a set of terms to refer to their results. Actually, this problem is itself a matter of
c°nstrual, in that a change of perspective from the path to its endpoint does not change the
underlying process, just the view taken of it (see also Fawcett 1997: 50-51).
54 Sandra L. Halverson

constraints. While all of the constraining elements listed above, both knowledge-
related and contextual, have been theorized individually, they remain to be inte­
grated into a testable holistic theory with a cognitive base that provides for their
working in a unique, real translational act. A cognitive translation theory could
incorporate a number of the findings of previous functionalist work, though
w ith a realignment of conceptual status for some of the key notions in order to
integrate them into a human-driven knowledge-based causal nexus.
Some might view the above as an exercise in merely moving translation
theory from outside to inside the head of the translator, and not really saying
anything new in the process. Indeed, the equivalence paradigm has been de­
clared dead several times, so yet another kick in that direction w ill probably
not do too much damage. But this is not another kick. In reframing equivalence,
in line w ith Pym, as a belief that some cultures and the translators within them
may share, it is given the potential to exercise causal force in the only way in
which it can. Similarly, shifts and universals/features have been reframed in
order to begin further work at conceptual clarification and operationalization
for empirical study. Both of these issues are of current and ongoing interest,
and new theoretical perspectives are definitely called for.

3 Implications of Cognitive Linguistics for

Translation Studies methodology
For the last fifty-odd years, empirical work in Translation Studies has often been
categorized as either “ process- or product-based” , according to whether the spe­
cific object of study was the process itself or its textual product (cf. Holmes 1988;
Toury 1995). While process-oriented studies have always been oriented towards
psycholinguistic methods and models from cognitive psychology, such as think-
aloud protocols (see Jaaskelainen 2009 for a review), product-oriented studies
have relied heavily on case study methods (Susam-Sarajeva 2009: 37). Quantita­
tive methodologies also became more frequent with the advent of corpus-based
studies (cf. Kenny 2009; Laviosa 2002; Olohan 2004).
One of the most promising innovations in research methods in TS was the
introduction of keystroke logging, most often in the form of TRANSLOG technol­
ogies (Jakobsen 2006). I his technology has allowed for collection of data related
to the online production of translations, including all keystroking (backspace,
editing, etc), as well as indications of the duration and (textual) location of
all pauses. Newer versions of the software also allow for the integration of an
eye-tracker and/or a screen monitor, such that data on several process variables
(e-g- gaze fixation, gaze direction) may be collated and linked to the final
 Implications of Cognitive Linguistics for Translation Studies ----- 55

translated product. Keystroke logging thus represents the first and only current
data collection tool that allows for the integration of process- and product-
related data on one and the same production sequence.
At present, the full potential of combined methods research in TS is only
beginning to be explored. The most innovative research groups with an empirical
focus are those working on process-oriented research. These w ill be very briefly
mentioned in Section 3.1. The methodologies adopted by these groups remain
predominantly experimental, with the possible exception of the PACTE group,
which also integrates more observational data. In other areas of Translation Stu­
dies, empirical work remains focused on either case studies or corpus-based
methods. W ithin Cognitive Linguistics, a recent development among corpus lin ­
guists is emphasis on combining corpus-based and experimental methods. This
development, and its repercussions for TS, w ill be briefly discussed in Section 3.2.

3.1 Experimental methods in TS: Recent developments

There are currently three major research groups working on translation process
Projects: The Centre for Research and Innovation on Translation and Translation
Technology (CRITT) in Copenhagen, the Eye-to-lT project led from Bulgaria, and
the PACTE group in Spain. A ll of these groups have adopted a combination of
Psycholinguistic methods, many involving combinations of data types. Exam­
ples include studies on textual segmentation in translation (e.g. Dragsted
2°04; Jakobsen 2003), ST-TT attention (Jensen 2009), and cognitive effort and
metaphor (Sjorup 2009). Among other things, Dragsted found that, with rela­
tively easy texts, novices and professionals tended to segment texts differently,
with professionals demonstrating larger units and novices smaller. These differ­
ences were somewhat neutralized given a more demanding task, however (2004:
358)- Jakobsen (2003) found that concurrent think-aloud had an effect on seg­
mentation, w ith concurrent think-aloud resulting in smaller segments. Jensen
(2009), in ongoing work using eye-tracking," has found both similarities and
differences between professionals and novices in the distribution of attention
between ST and TT and in the duration of gaze with regard to ST and TT seg­
ments. Also in ongoing work, Sjprup (2009) studies the gaze time involved in
tbe translation of metaphor, using this factor as an indicator of cognitive load.
Much of the Copenhagen work is published in successive issues of Copenhagen
Studies in Language.

3 See also Rojo and Valenzuela, this volume, for another study with eye-tracking technology.
56 Sandra L. Halverson

The Eye-to-IT project, which includes several of the Copenhagen research­

ers, is a primarily experimental research program, and combines data collected
through eye-tracking, EEG, and keystroke logging (http://cogs.nbu.bg/eye-to-it/
?home). Its primary objective is to develop technologies that enable human-
computer interaction during the translation process. Numerous psycholinguistic
variables have also been the subject of investigation under the auspices of this
project, for example, segmentation and the coordination of reading and writing
(Dragsted and Hansen 2008; Jakobsen and Jensen 2008), the effects of time
pressure and text complexity on cognitive processing (Sharmin et al. 2008), cor­
respondences in the bilingual lexicon (Stamenov, Gerganov, and Popivanov
2009), and the effect of cognate status on cognitive load (Lachaud 2009) to
mention a few.
The PACTE group adopts a cognitive theoretical framework, and is engaged
in the study of translation processes and the development of a model of trans­
lator competence (Munoz Martin 2006; PACTE group 2003, 2005, 2010, 2011).
This group also incorporates a range of data types and methodologies in its
work, including textual material, translation protocols, direct observation,
questionnaires, and retrospective interviews (PACTE group 2005: 611).
The three groups mentioned here have been selected as indicative of meth­
odological development and innovation in TS. All of these groups work within
so-called process-oriented TS; that is, the main object of investigation is the
translation process. Consequently, the selection of data and analytical method
is driven by process-derived variables, such as reaction time, pause length,
gaze fixation, EEG patterns, etc. However, as seen even in this superficial sketch,
all of the groups include textual (product) data types in their portfolios, enabling
combinations of textual (product) variables and process variables. The clearest
implementations of this strategy are found in the Eye-to-IT work: textual vari­
ables such as syntactic segmentation, idioms, metaphor or text complexity are
linked to process variables such as cognitive load, indicated by such measures
as pause length, reaction time, or gaze fixation. The full potential of this wealth
of data remains to be exploited through the integration of a theoretical model
that would allow both product and process variables to be brought together
w ithin one and the same paradigm.

3.2 Corpus-based Cognitive Linguistics and innovation in TS

In a recent state-of-the-art review, Gilquin and Gries (2009) review the back­
ground and current state of affairs as regards data and methods in linguistics.
 Implications of Cognitive Linguistics for Translation Studies ----- 57

They single out the rapidly emerging use of combinations of corpus-based and
experimental methods as a promising development, stating:

Because the advantages and disadvantages of corpora and experiments are largely comple­
mentary, using the two methodologies in conjunction with each other often makes it pos­
sible to (i) solve problems that would be encountered if one employed one type o f data only
and (ii) approach phenomena from a m ultiplicity of perspectives. (2009: 9)

Within Cognitive Linguistics, the need for both types of data is particularly
acute, given the offline nature of corpus data and the psychological/cognitive
nature of the theoretical constructs put forward to explain them. As stated by
Tummers, Heylen, and Geeraerts, the problem is this: “ Given the offline nature
of corpus data, they are not suited to support claims about the cognitive or neu­
rological mechanisms underlying language use” (2005: 233). This issue has been
the subject of considerable debate w ithin the Cognitive Linguistics community
(Sandra and Rice 1995; Croft 1998; Sandra 1998). More recently, Tummers, Hey­
len, and Geeraerts (2005) and Heylen, Tummers, and Geeraerts (2008) have also
addressed the issue. At present, it would seem that while there is shared recog­
nition of the need for several types of data, there is no general consensus on
how different types of data should be brought together. Heylen, Tummers,
and Geeraerts (2008) summarize the situation as follows:

Relating to the second property of corpus data that we discussed, viz. offline usage, there
are still questions about which aspects o f the language system can be fruitfully studied
through corpus analysis. More generally, there is a need to integrate quantitative corpus
analysis with other empirical methodologies into a coherent methodological framework
for research into the cognitive aspects of language. [. . -1 A division of labour could be
to first identify relevant variables through quantitative corpus analysis and then investigate
the online properties o f variables in a more targeted way through psycholinguistic experi­
ments. On the other hand, corpus analysis has the advantage that it relies on spontaneous
naturally occurring language use, whereas psycholinguistic experiments are often forced
to use constructed examples. Validating experimental results against corpus data can be
a way to avoid experimental artefacts. (2008: 121)

While either of these tacks are feasible, i.e. corpus to experiment or vice versa,
the questions of how the data shall be integrated or what to do with diverging
results remains. As pointed out by Gilquin and Gries (2009: 17), studies that do
Use combined methods have ended up with both converging and diverging
results. As they put it,

Possible explanations have been offered to account for the differences between corpus and
experimental data, and suggestions have been made to bring them closer to each other, but
'• is still true that the relation between the two types of data remains unclear and that
identity cannot be taken for granted. (2009:121)
58 Sandra L. Halverson

The authors continue to point out that efforts toward integration should ideally
proceed through recourse to linguistic theory.
Several recent papers exemplify the trend towards combining psycholin­
guists experimentation and corpus analysis (e.g. Divjak and Gries 2008; Ellis
and Simpson-Vlach 2009; Mollin 2009; W ulff 2009). Only the second of these
studies implemented processing variables of the sort used in translation process
analysis, and all of them speak to issues of semantic representation. The studies
mentioned here cover, respectively, the organization of the mental lexicon, pro­
cessing of formulaic language, the relationship between corpus collocation and
word association, and linguistic factors used in idiomaticity judgments.
The process-oriented TS work sketched above could represent the kind of
methodological step combining experiment with corpus suggested by Heylen,
Tummers, and Geeraerts (2008). One study which initiates efforts in this direc­
tion is Alves et al. (2010), in which keystroke logging and eye-tracking data
are brought together with corpus annotation and alignment data in an investi­
gation of translational units. This study is one of the first of its kind. Further
work w ill be required to isolate the appropriate links between online and offline
data, as process variables such as pause duration, location and duration of gaze,
etc. are to be operationalized relative to a broad, (cognitive) linguistic theory
that would also allow for corpus analysis.
As sketched in Section 2.3, corpus-based methods have represented an
important addition to empirical Translation Studies since the early 1990s. It is
fair to say that corpus-based approaches to the study of translation have
grown alongside corpus methodologies w ithin linguistics at large. At present,
however, the implementation of combined methods in the corpust-to-experiment
direction has been limited. This is probably due to the lack of pressure from cog­
nitive theories: most corpus-based studies of translation have drawn on other
discourse analytical theories. Thus, there are few examples of the opposite
movement, in the corpus to experiment direction. One example is the ongoing
work mentioned in Section 2.3 and outlined in Halverson (2009). In this work,
the gravitational pull hypothesis is first investigated through a series of corpus
studies involving the light verb get in the English-Norwegian Parallel Corpus
(ENPC), the British National Corpus (BNC), and the Translational English
Corpus (TEC). Thf' tests comprise a series of statistical analyses, including tra­
ditional bivariate analyses to test for overrepresentation o f a prototypical
sense in translated (as opposed to non-translated text). Additional m ultivari­
ate analysis is also introduced to look for additional factors affecting the dis­
tribution of senses across corpora. Elicitation data is then gathered using the
methodology outlined in Cuyckens, Sandra, and Rice (1997), involving a word
 Implications of Cognitive Linguistics for Translation Studies ----- 59

prompt and sentence generation task. At a later stage, also keystroke data
w ill be collected.
The introduction of cognitive linguistic theory has clear and immediate con­
sequences for Translation Studies. The most obvious consequence is the need to
combine online and offline methods, a requirement that stems from the nature
of the theory and the data, as mentioned above. This is the price that must be
Paid for the explanatory power that cognitive linguistic theory has to offer
with regard to the kinds of translational patterns currently under investigation
in Translation Studies.
The second consequence of Cognitive Linguistics for TS methodology is less
obvious, though potentially more radical: that is the breaching of the long­
standing barrier between process and product perspectives. W ithin Cognitive
Linguistics itself, there is still a clear recognition of the difference between
describing semantic or conceptual representation and studying language pro­
cessing. Langacker himself is quite clear in situating Cognitive Linguistics
within the former category (2008: 31). However, he also makes it clear that
the theory of cognitive grammar is commensurate with certain aspects of proces­
sing (2008: 31). In the bilingualism literature, there is an increasing interest in
attempting to bring together the disparate perspectives of representation and
Processing (Hartsuiker, Costa, and Finkbeiner 2008: 413).
Within TS itself, it would seem that product and process are currently being
brought together by technological advances and the ambitious research pro­
-a m s that utilize them. Even so, the potential offered by keystroke logs to
Serve as a bridge between psycholinguistic investigation and corpus-based stu­
dies has yet to be fully explored. Cognitive linguistic theory may provide the
type of grounding needed to integrate these perspectives: the onus is on TS
scholars to take an active part in furthering this development.

^ Implications of Cognitive Linguistics

for TS epistemology
u is common in Translation Studies today to identify two apparently conflicting
and supposedly incommensurable epistemological frameworks. These have
been referred to as the empirical and postmodern approaches or as empirical
descriptive and postmodern cultural studies approaches (Delabastita 2003; Ches-
terman and Arrojo 2000). The attempt to identify “ shared ground” between these
iw ° broad philosophical frameworks was the subject of an extended debate con-
Ucted in Target from 2000 to 2002, in which a range of scholars expressed their
lews’ though perhaps less shared ground was identified than was initially hoped.
60 Sandra L. Halverson

As was pointed out in the debate (cf. Simeoni 2000; Malmkjaer 2000; Sela-
Sheffy 2000; Halverson 2000a), the attempt to distinguish between the two
frameworks with reference to an essentialist or non-essentialist foundation is
an inappropriate tack. As Simeoni puts it:

A good many approaches to cultural diffusion and translation are at the same time empir­
ical, descriptive and very much aware of the dynamics of culture and language. Meaning is
certainly not taken by all descriptivists to be stable, neither substantially nor even for­
mally. Nor does every researcher who considers him/herself a postmodern culturalist reject
empirical work. (2000: 337)

Tymoczko makes the same point when she describes the “ postpositive” stage of
contemporary TS (2007). This point is also made by Delabastita, who identifies a
common platform between the two approaches in “ awareness of the impossibil­
ity of constructing knowledge of cultural reality that is observer-independent
and value-free” (2003: 23). In his paper, Delabastita also points out that the
two approaches, while sharing this common epistemological platform, make d if­
ferent methodological assumptions on the basis of it. The result is opposing
positions with regard to two fundamental questions: (i) Should we search for
hypothetical conceptual models of reality that are as observer-independent as
possible? (ii) Should we search for hypothetical conceptual models of reality
that are as value-free as possible? In Delabastita’s account, empirical ap­
proaches comprise a position referred to as “ epistemological utopianism” , in
which the response to the two questions is yes and no, respectively. Postmodern
approaches, on the other hand, answer no and yes, respectively, and represent a
program of “ ethical utopianism” (2003: 23). Thus these two basic orientations
end up with diametrically opposed views regarding the objectives and actual
conduct of empirical research.
Delabastita’s account is particularly helpful in illum inating how, given d if­
ferent methodological preferences and emphases, quite diverging views and
practices may develop even given a shared starting point. The contribution of
Cognitive Linguistics to 1 ranslation Studies lies in its potential to further enrich
this common platform through more explicit use of the philosophy of “ embodied
realism” (Lakoff and Johnson 1999; Johnson and Lakoff 2002).

4.1 TS and embodied realism

The philosophy of embodied realism is described by Johnson and Lakoff as

follows:
 Implications of Cognitive Linguistics for Translation Studies ------ 61

Embodied realism, as we understand it, is the view that the locus of experience, meaning,
and thought is the ongoing series o f embodied organism-environment interactions that
constitute our understanding o f the world. According to such a view, there is no ultimate
separation of mind and body, and we are always in touch w ith our world through our
embodied acts and experiences. (2002: 249)

This view is presented as the alternative to traditional epistemological oppositions.

According to Harder,

By claiming that mental function is grounded in the body and imaginatively structured, in
a way that reflects specifically human experience, Cognitive Linguistics has defined a new
position in the philosophical landscape. This position contrasts, on the one hand, with
a belief in absolute objective foundations o f knowledge and, on the other hand, with a
deconstructionist rejection of any kind o f foundation whatsoever. (2007: 1253)

As also pointed out by Harder (2007: 1261), a philosophically related position is

Putnam’s “ internal realism” (1992). This position too represents an alternative to
both objectivist and deconstructionist/relativist programs, and grounds this
middle position in a perspective that is “ internal” to the human interacting
with the world around him/her.
A cognitive linguistic theory of translation must build on an epistemology of
this type. Indeed, the consequences of adopting such a position are amply clear
*n the theoretical apparatus outlined in Section 1.1. The perspective is always
that of the situated, cognizing translator, and the conceptual framework is
built on current understanding of general and linguistic cognition. On the
°fher hand, the implications of this philosophy for Translation Studies epistemol­
ogy have not been sufficiently explored or articulated. For that reason, the gap, or
difference of direction, that separates descriptive translation scholars and post­
modern translation scholars keeps re-emerging as a false dichotomy between
°bjectivists and relativists. In outlining the anti-descriptivist position, Pym
summarizes as follows:

This argument basically sees the descriptive paradigm as an exercise in positivism.

The paradigm would require belief in a neutral, transparent, objective knowledge about
translation, and that progress w ill come by accumulating that knowledge. A great deal
of armor is bu ilt around that belief. However, the armor cracks at several of the points
we have seen: in the problem o f defining translations, in the problem of how to use de­
scriptions of norms, in the possibility that the various levels of description are themselves
translations o f a kind [ . . . ] and in the general emphasis on the role of context [ . . . ] At all
these points, some attention is required to the role of the observer, the person doing the
describing. The descriptive paradigm has not really been able to rise to that challenge.
(2010: 85)
62 Sandra L. Halverson

In a similar vein, Tymoczko describes the current situation as follows:

Failure to understand how all the branches of Translation Studies relate to postpositivist
epistemology and how they represent complementary aspects o f postwar investigations
o f language and text has contributed to the tendency of some translation scholars to posi­
tion themselves rigidly w ithin one domain or another and to see their approaches as
antithetical to those o f other branches of the discipline. (2007: 52)

Both of the above are fair descriptions of the current state of affairs in Transla­
tion Studies. While there is ample empirical work being done, some scholars
view this work with suspicion, as a clear alternative to a relativist epistemology
has not been fully worked out or adequately articulated. The response to this
criticism, the alternative to defeated positivism (objectivism) and unrestrained
relativism lies in the middle ground of embodied realism. From this standpoint,
there is no one objective truth, no reality that is unambiguously given. There is
a world “ out there” , but our access to it and understanding of it is always
mediated by, and subject to, the workings of our own interpretive processes.
This makes it necessary for us to understand how these processes work, as
well as how they ultimately are shared. In this view, scientific knowledge is fun­
damentally human knowledge, the realism is “ internal” to, or “ embodied” in us.
An epistemology of embodied or internal realism has been advocated in
Translation Studies earlier in Halverson (1998, 2002). However, at present
exploratory work in this area has been pursued furthest for the purposes of TS
in the debate concerning the definition of “ translation” . For that reason, this
debate w ill be briefly outlined in the next section in order to give an indication
of the current state of affairs.

4.2 TS and approaches to defining “translation”

The most extensive discussion of the “ definitional impulse” in Translation

Studies is found in Tymozcko’s 2007 volume. Tymoczko presents a broad inves­
tigation of the history and current cultural diversity of a concept of “ transla­
tion” , and uses this discussion both as an epistemological diagnostic and as
support in making her case for a broader field of enquiry and a more empowered
position for translators themselves. Before returning to the particulars of her
approach, it is helpful to reiterate the two most important definitional traditions
that preceded it w ithin TS.
As mentioned in Section 2.1.1, some of the earliest scholars in contemporary
Translation Studies approached the definition issue through adopting equiva­
lence as the defining characteristic of translations. In other words, “ translations”
 Implications of Cognitive Linguistics for Translation Studies — 63

were defined as texts that were “ equivalent to” a source text in another lan­
guage, and the task of the scholar was to elaborate equivalence frameworks
in order to circumscribe the translation category. As also mentioned above,
this pursuit was subsequently abandoned along with its underlying objectivist
philosophy, and a relativistic approach was adopted. This radical turn in the
study of translation was captured in Toury’s well-known notion of “ assumed
translation” , where “ translations” were identified as texts that were presented
or recognized as translations in a given culture at a given time (Toury 1995).
This notion has been hotly debated (see Halverson 2008 for a discussion), and
there is still some disagreement regarding its utility. The point to be made here
is that Toury’s proposal ushers in the potential for a wholly relativistic approach,
in which culturally or historically situated definitions remain isolated and incom­
parable. The same consequence stems from the deconstructionist approach of
Arrojo (1998). The pursuant challenge remains: how should a discipline do with
isolated, culturally and historically relative descriptions?
At present, there are two proposals that aim to provide a framework by
which to grapple with contingent conceptualizations. These are the prototype
approach (Snell-Hornby 1988; Halverson 1999a, 1999b, 2002) and Tymoczko’s
(2007) “ cluster concept” approach. Both of these build on insights drawn
from cognitive science, though they take slightly different forms. Moreover,
both approaches aim to provide a framework by which historically contingent,
culturally relative conceptualizations may be fruitfully brought together. In
fact, it is not entirely certain what the key differences in the two approaches
are,12 given that some authors use the terms “ prototype” and cluster concept
synonymously (Armstrong, Gleitman, and Gleitman 1983). Be that as it may,
both authors argue for the investigation of the conceptualizations that different
cultures have for a concept of “ translation” , and Tymoczko presents a fascinat­
ing collection of work of this kind (2007: 68-77 and sources cited there). Halver-
s°n (2008) argues along the same lines, that the words used by different cultures
0r times must be the relevant starting point for conceptual analyses across
cultures, which all lie w ithin the remit of the discipline, and which jointly
constitute the terrain of investigation. The conceptual analyses, using the

^ In her book, Tymoczko argues against the prototype view (2007: 90 100). The Full
resP°nse to these arguments is beyond the scope of the present paper, and some of the
resPonse is also implied in Halverson (2008), in the discussion of cross-cultural comparison.
arguments are harder to understand, such as the claim that a prototype approach implies
requiring the exclusion of empirical data from the past or from situations that are not Euro-
er|tric (2007: 97). For further details see Halverson (in progress a).
64 Sandra L. Halverson

framework of Cognitive Linguistics, may illuminate how different members of

different cultures draw on a common cognitive apparatus, within the constraints
of cultures and contexts, to create and develop different models of communica­
tion, interlingual and otherwise. Thus, the method used to elucidate relation­
ships between foundational concepts and the philosophical framework on
which that task is based, emerge from Cognitive Linguistics and embodied real­
ism. For example, the models that have been put forward for the English trans­
late (Halverson 1999b) and the Norwegian oversettelse (Halverson 2000b) made
use of image-schematic structure and prototype effects, and are based on empir­
ical investigations o f several types of data, more specifically etymological
evidence, elicitation data and corpus data.
The epistemological goals of Tymozcko and Halverson seem to coincide in a
call for an informed and human-based epistemology that would provide ade­
quate grounding for empirical work while avoiding radical relativism. The two
differ most, perhaps, in their predominant allegiances to epistemological uto­
pianism (Halverson) and ethical utopianism (Tymoczko). As a result, the former
advocates more empirical investigation using cognitive linguistic tools (Halver­
son 2008) while the latter calls for increased reflexivity and self-awareness
(Tymoczko 2007). At the same time, neither scholar has adequately addressed
the complex relationship between individual conceptual knowledge and shared
cultural representations, even though both are aiming at cross-linguistic sharing
of scientific knowledge w ithin a research community. A full discussion of these
weaknesses in the work cited here is beyond the scope of the present paper.
A detailed treatment is currently in progress (Halverson in progress a).

5 Concluding remarks
The primary objective of this paper has been to demonstrate specific ways in
which Cognitive Linguistics has implications for Translation Studies in terms
of theory development, methodology, and epistemology. In the domain of
theory, it was suggested that traditional views of equivalence should be jetti­
soned, but that equivalence beliefs may be incorporated through the constrain­
ing force of a tianslators own conceptualization of what s/he is doing. The
notion of translation shift may be reframed as construal and the notion of trans­
lation universals, or features of translational language, may be more clearly con­
ceptualized and empirically investigated by adopting a cognitive linguistic
translation theory. In the discussion of methodology, it was argued that a cog'
nitive linguistic translation theory w ill require increased used of combined
methods, with the concomitant focus on the methodological questions that
 Implications of Cognitive Linguistics for Translation Studies ----- 65

arise from this. In this area, a Cognitive Linguistic Translation Studies could
make valuable contributions to ongoing debates w ithin Cognitive Linguistics
itself. With regard to epistemological issues, it was argued that a clearer articu­
lation of embodied realism w ithin Translation Studies would be a fruitful contri­
bution to a rather stale and artificial anti-descriptivist stance w ithin TS and at
the same time constitute a constructive alternative to deconstructive relativism.
As was pointed out in Section 2, a cognitive linguistic theory of translation
posits a view of translation that is much more organic than previous accounts,
in that it takes the cognizing translator as the locus of the situated event.
The individual translator is taken as the source of the translated text, and the
theory must then account for the ways in which the text emerges as a contextua-
•ized interpretation of an anterior text and a re-expression of it using alternative
forms. The situated translator is a permeable cognizer, however, and his/her
integration and interaction with the surrounding world are also part of the pic­
ture. The specific translational usage event draws on a number of different knowl­
edge types and is subject to a variety of constraints, as discussed in Section 1. This
organic and unitary view is at odds with many traditional conceptualizations of
translation based on long-standing metaphors of movement, transfer and source-
target relationships. It is a different conceptualization in that the prominence
of source and target diminish as the prominence of the translator grows.
In many ways, the path forward in developing a cognitive linguistic frame­
work for TS w ill require, and thus open for contributions to, ongoing efforts in
breaching a number of interdisciplinary divides. First of all, it is clear that con­
siderable effort must go into developing a cognitive linguistic theory that is
capable of accounting for patterns of language demonstrated by language lear­
ners and highly proficient speakers of two or more languages. As mentioned in
Section l, efforts in this direction are underway, but much work needs to be
done. At present the aim must be for theory developments to at least not violate
fundamental knowledge w ithin other fields. Second, as pointed out in Section 3,
developments in bilingualism and in TS seem to point towards a more unified
aPProach to representational and processing aspects of bilingualism, areas
which up until now have remained relatively separate. This could be paralleled
W efforts to combine and integrate so-called product- and process-based ap­
proaches in TS. Work on identifying the relevant theoretical concepts, on oper-
at>onalizing these for empirical research, and on extensive testing w ill require a
concerted effort across disciplines in the years to come. Finally, it may also be
u^gued that w ithin Cognitive Linguistics and Literary Studies, the traditional
wide between the study of language and the study of literature is currently
e'n^ cr°ssed by several scholars working on “ cognitive poetics’ (for a review
See Freeman 2007; see also Turner 1991,1996; Lakoff and Turner 1989; Stockwell
66 Sandra L. Halverson

2002; Tsur 2003). This particular area has not been dealt with earlier in this
paper, but there is clearly room for rapprochement within TS also, as linguistic
and literary approaches often find themselves on opposite sides of the epistemo­
logical divide described in Section 4. If cognitive translation scholars are w illing
and able to turn their analytical lights to texts and discursive practices of all
kinds, then this too would facilitate intradisciplinary discourse.
In closing, it is important to point out that perhaps the most striking conse­
quence of adopting a cognitive perspective is that it places human cognition and
human agency at the center of the causal picture. This is another area in which
the present author shares the concerns of, among others, Tymoczko (2007). This
may be empowering, as Tymoczko claims, but it w ill also be better science (in
the view of a humble epistemological utopian). At the same time, it is expedient
to point out that not all studies must start here. As mentioned previously, also
social factors are of the utmost importance, and there are numerous ways of
going about studying such things, as witnessed by the recent “ sociological
turn” in TS.13 In other words, this is not a call to reduce Translation Studies to
the study of translational cognition, but a call to augment Translation Studies
to include cognition in a broader and more comprehensive way. A cognitive lin ­
guistic approach to translation has the potential to enrich the discipline by
introducing a philosophically grounded framework to enable theoretical and
methodological innovation.

References
Alves, Fabio, Adriana Pagano, Stella Neumann, Erich Steiner and Silvia Hansen-Schirra 2010
Translation units and grammatical shifts: Towards an integration of product- and process-
based translation research. In: Gregory Shreve and Erik Angelone (eds.). Translation and
cognition, 109-141. Amsterdam: )ohn Benjamins.
Anderman, Gunilla and Margaret Rogers 2008 Incorporating corpora: The linguist and the
translator. Clevedon, UK: Multilingual Matters.
Armstrong, Sharon Lee, Lila R. Gleitman and Henry Gleitman 1983 What some concepts might
not be. Cognition 13: 263-308.
Arrojo, Rosemary 1998 The revision of the traditional gap between theory and practice and the
empowerment of translation in postmodern times. The Translator 4: 25-48.
Baker, Mona 1993 Corpus linguistics and translation studies: Implications and applications. In:
Mona Baker, Gill Francis and Elena Tognini-Bonelli (eds.), Text and technology: In honour
of John Sinclair, 233-250. Amsterdam: |ohn Benjamins.
Baker, Mona 1995 Corpora in translation studies: An overview and suggestions for future
research. Target 7.2: 223-243.

13 See Chesterman (2000) and Brownlie (2003) on multiple causation in translation.

 Implications of Cognitive Linguistics for Translation Studies ----- 67

Baker, Mona 1996 Corpus-based translation studies: The challenges that lie ahead. In: Harold
Somers (ed.), Terminology, LSP and translation: Studies in language engineering, 175-186.
Amsterdam: )ohn Benjamins.
Baker, Mona 2006 Contextualization in translator- and interpreter-mediated events. Journal of
Pragmatics 38.3: 321-337.
Bakker, Matthijs, Cees Koster and Kitty M. Van Leuven-Zwart 2009 Shifts. In: Mona Baker and
Gabriela Saldanha (eds.), Routledge encyclopedia of translation studies, 269-274. 2nd
edition. London: Routledge.
Brownlie, Siobhan 2003 Investigating explanations of translational phenomena: A case for
multiple causality. Target 15.1:111-152.
Calzada Pérez, María 2003 Apropos of ideology. Translation studies on ideology - ideologies in
translation studies. Manchester: St. Jerome.
Catford, |. C 1965 A linguistic theory of translation. London: Oxford University Press.
Chesterman, Andrew 2000 A causal model for translation studies. In: Maeve Olahan (ed.),
Intercultural faultlines: Research models in translation studies. Textual and cognitive
aspects, 15-27. Manchester: St. Jerome.
Chesterman, Andrew 2004 Beyond the particular. In: Anna Mauranen and Pekka Kujamdki
(eds.), Translation universals: Do they exist? 33-49. Amsterdam: John Benjamins.
Chesterman, Andrew 2005 Problems with strategies. In: Krisztina Károly and Agota Fóris (eds.),
New trends in translation studies: In honour of Kinga Klaudy, 17-28. Budapest: Akadémiai
Kiadó.
Chesterman, Andrew 2006 Interpreting the meaning of translation. In: Mickael Suominen, Antti
Arppe, Anu Airola, Orvokki HeinSmdki, Matti Miestamo, Urho Madtd, Jussi Niemi, Kari K.
Pitkdnen and Kaius Sinnemaki (eds.), A man of measure: Festschrift in honour o f Fred
Karlsson on his 60th birthday. Special supplement, SKY Journal of Linguistics 19: 3-11.
Chesterman, Andrew and Rosemary Arrojo 2000 Shared ground in translation studies. Target
12.1: 151-160.
dark, Andy 1997 Being there: Putting brain, body and world together again. Cambridge:
Cambridge University Press.
Clark, Andy 2008 Supersizing the mind: Embodiment, action, and cognitive extension. Oxford:
Oxford University Press.
Croft, William 1998 Linguistic evidence and mental representations. Cognitive Linguistics 9.2:
151-173.
Croft, William and D. Alan Cruse 2004 Cognitive linguistics. Cambridge: Cambridge University
Press.
Cuyckens, Hubert, Dominiek Sandra and Sally Rice 1997 Towards an empirical lexical
semantics. In: Birgit Śmieja and Meike Tasch (eds.). Human contact through language and
linguistics, 35-54. Frankfurt am Main: Peter Lang.
e Groot, Anette M. B. 1992 Bilingual lexical representation: A closer look at conceptual
representations. In: Ram Frost and Leonard Katz (eds.), Orthography, phonology,
morphology, and meaning, 389-412. Amsterdam: North Holland.
e Groot, Anette M. B. 1993 Word-type effects in bilingual processing tasks: Support for a
mixed representational system. In: Robert Schreuder and Bert Weltens (eds.). The bilingual
lexicon, 27-51. Amsterdam: John Benjamins.
e>abastita, Dirk 2003 Translation studies for the 21st century: Trends and perspectives.
Génes/s. Revista científica do ISAI 3: 7-24.
68 Sandra L. Halverson

Divjak, Dagmar and Stefan T. Gries 2008 Clusters in the mind? Converging evidence from near
synonymy in Russian. The Mental Lexicon 3.2: 188-213.
Dong, Yanping, Shichun Gui and Brian Macwhinney 2005 Shared and separate meanings in the
bilingual mental lexicon. Bilingualism: Language and Cognition 8.3: 221-238.
Dragsted, Barbara 2004 Segmentation in translation and translation memory systems.
Copenhagen Working Papers in LSP 4. Copenhagen: CBS/Handelshojskoen i Kobenhavn.
Dragsted, Barbara and Gitte Hansen 2008 Comprehension and production in translation: A pilot
study on segmentation and the coordination of reading and writing processes.
Copenhagen Studies in Language 36: 9-29.
Ellis, Nick and Rita Simpson-Vlach 2009 Formulaic language in native speakers: triangulating
psycholinguistics, corpus linguistics and education. Corpus Linguistics and Linguistic
Theory 5.1: 61-78.
Englund-Dimitrova, Birgitta 1997 Translation of dialect in fictional prose - Vilhelm Moberg in
Russian and English as a case in point. In: J. Falk (ed.), Norm, variation and change in
language: Proceedings of the centenary meeting of the Nyfilologiska sdlskapet, Nedre
Manilla, 22-23 March 1996, 49-65. Stockholm: Almqvist and Wiksell.
Eskola, Sari 2004 Untypical frequencies in translated language: A corpus-based study on a
literary corpus of translated and non-translated Finnish. In: Anna Mauranen and Pekka
Kujamaki (eds.), Translation universal: Do they exist? 83-100. Amsterdam: John Benjamins.
Fawcett, Peter 1997 Translation and language: Linguistic theories explained. Manchester:
St. Jerome.
Francis, Wendy 2005 Bilingual semantic and conceptual representation. In: Judith Kroll and
Annette M. B. de Groot (eds.). Handbook of bilingualism: Psycholinguistic approaches,
251-267. Oxford: Oxford University Press.
Freeman, Margaret 2007 Cognitive linguistic approaches to literary studies: State of the art in
cognitive poetics. In: Dirk Geeraerts and Hubert Cuyckens (eds.), The Oxford handbook of
cognitive linguistics, 1175-1202. Oxford: Oxford University Press.
Gilquin, Gaetenelle and Stefan T. Gries 2009 Corpora and experimental methods: A state-of-the-
art review. Corpus Linguistics and Linguistic Theory 5.1: 1-26.
Gutt, Ernst-August 2000 Translation and relevance: Cognition and context. 2nd edition.
Manchester: St. Jerome.
Halverson, Sandra L. 1997 The concept of equivalence in translation studies: Much ado about
something. Target 9.2: 207-233.
Halverson, Sandra L. 1998 Concepts and categories in translation studies. Unpublished PhD
dissertation, University of Bergen, Norway.
Halverson, Sandra L 1999a Conceptual work and the “ translation" concept. Target 11.1: 1-31.
Halverson, Sandra L. 1999b Image schemas, metaphoric processes and the “ translate” concept.
Metaphor and Symbol 14.3: 199-219.
Halverson, Sandra L. 2000a The fault line in our common ground. Target 12.2: 356-362.
Halverson, Sandra L. 2000b Prototype effects in the “ translation” category. In: Andrew
Chesterman, Natividad Gallardo San Salvador and Yves Gambier (eds.), Translation in
context, 3-16. Amsterdam: John Benjamins.
Halverson, Sandra L. 2002 Cognitive models, prototype effects and “ translation": The role of
cognition in translation (meta)theory. Across Languages and Cultures 3.1: 21-43.
Halverson, Sandra L. 2003 The cognitive basis of translation universals. Target 15.2: 197-241.
Halverson, Sandra L. 2007 A cognitive linguistic account of translation shifts. Belgian Journal of
Linguistics 21:105-119.
 Implications of Cognitive Linguistics for Translation Studies ----- 69

Halverson, Sandra L. 2008 Translations as institutional facts. An ontology for “ assumed

translation” . In: Anthony Pym, Miriam Shlesinger and Daniel Simeoni (eds.), Beyond
descriptive translation studies: Investigations in homage to Gideon Toury, 343-361.
Amsterdam: John Benjamins.
Halverson, Sandra L. 2009 Elements of doctoral training: The dynamics of the research process.
The Interpreter and Translator Trainer 3.1: 79-106.
Halverson, Sandra L. 2010a Cognitive translation studies: Developments in theory and method.
In: Gregory Shreve and Erik Angelone (eds.), Translation and cognition, 349-369.
Amsterdam: John Benjamins.
Halverson, Sandra L. 2010b Translation universals or cross-linguistic influence: Conceptual and
methodological issues. Paper presented at the 6th EST Conference, Leuven, Belgium,
September 23-25, 2010.
Halverson, Sandra L. in progress a Prototypes, stereotypes, clusters, norms, and universals:
Conceptual frameworks in translation studies.
Halverson, Sandra L. in progress b Testing for gravitational pull: The case of get.
Harder, Peter 2007 Cognitive linguistics and philosophy. In: Dirk Geeraerts and Hubert
Cuyckens (eds.), The Oxford handbook of cognitive linguistics, 1241-1265. Oxford: Oxford
University Press.
Hartsuiker, Robert )., Albert Costa and Matthew Finkbeiner 2008 Bilingualism: Functional and
neural perspectives. Acta Psychologica 128: 413-415.
Hasselgren, Angela 1993 Right words, wrong words and different words: An investigation into
the lexical coping of Norwegian advanced learners of English. Unpublished master’s
thesis. Department of English, University of Bergen.
Hatim, Basil and Ian Mason 1990 Discourse and the translator. London: Routledge.
Hatim, Basil and Ian Mason 1997 The translator as communicator. London: Routledge.
Hermans, Theo 1999 Translation in systems: Descriptive and system-oriented approaches
explained. Manchester: St. Jerome.
Heylen, Kris, Jose Tummers and Dirk Geeraerts 2008 Methodological issues in corpus-based
cognitive linguistics. In: Gitte Kristiansen and René Dirven (eds.). Cognitive
sociolinguistics: Language variation, cultural models, social systems, 91-128. Berlin:
Mouton de Gruyter.
H°lmes, James S. 1988 Translated! Papers on literary translation and translation studies.
Amsterdam: Rodopi.
Haz, Helene 1986 Linguistic and cognitive determinants of lexical acquisition in a second
language. Language Learning 36.4: 401-451.
lnShilleri, Moira 2003 Habitus, field and discourse: Interpreting as a socially situated activity.
Target 15.2: 243-268.
nghilleri, Moira 2005 Mediating zones of uncertainty: Translator agency, the interpreting
habitus and political asylum adjudication. The Translator 11.1: 69-85.
àaskeiainen, Riitta 2009 Think-aloud protocols. In: Mona Baker and Gabriela Saldanha (eds.),
Routledge encyclopedia o f translation studies, 290-293. 2nd edition. London: Routledge,
lakobsen, Amt Lykke 2003 Effects of think-aloud on translation speed, revision and
segmentation. In: Fabio Alves (ed.), Triangulating translation, 69-95. Amsterdam: John
Benjamins.
akobsen, Arnt Lykke 2006 Research methods in translation - TRANSLOG. In: Kirk P. H. Sullivan
and Eva Lindgren (eds.), Computer keystroke logging: Methods and applications, 95-105.
Oxford: Elsevier.
70 Sandra L. Halverson

Jakobsen, Arnt Lykke and Kristian T. H. lensen 2008 Eye movement behaviour across four different
types of reading tasks. In: Susanne Gdpferich (ed.), Looking at eyes: Eye tracking studies of
reading and translation processing, 103-124. (Copenhagen Studies in Language 36.)
Copenhagen: Samfundslitteratur.
Jakobson, Roman 1959 On linguistic aspects of translation. In: R. A. Brower (ed.), On
translation, 232-239. Cambridge, MA: Harvard University Press.
)arvis, Scott 1998 Conceptual transfer in the interlingual lexicon. Bloomington: Indiana
University Linguistics Association Publications.
Jarvis, Scott and Aneta Pavlenko 2008 Cross-linguistic influence in language and cognition.
New York: Routledge.
Jensen, Kristian T.H. 2009 Distribution of attention between source text and target text during
translation. Paper presented at “ Source language influence” , PhD seminar, University of
Bergen, Norway, November 11-14, 2009.
Johnson, Mark 1987 The body in the mind. Chicago: University of Chicago Press.
Johnson, Mark 2007 The meaning of the body. Chicago: University of Chicago Press.
Johnson, Mark and George, Lakoff 2002 Why cognitive linguistics requires embodied realism.
Cognitive Linguistics 13.3: 245-263.
Kellerman, Eric 1978 Giving learners a break: Native language intuitions as a source of
predictions about transferability. Working Papers on Bilingualism 15: 60-92.
Kellerman, Eric 1979 The problem with difficulty. Interlanguage Studies Bulletin 4: 27-48.
Kenny, Dorothy 2001 Lexis and creativity in translation. Manchester: St. Jerome.
Kenny, Dorothy 2009. Corpora. In: Mona Baker and Gabriela Saldanha (eds.), Routledge
encyclopedia of translation studies, 59-62. 2nd edition. London: Routledge.
Klaudy, Kinga 1996 Concretization and generalization of meaning in translation. In: Marcel Thelen
and Barbara Lewandowska-Tomaszczyk (eds.). Translation and meaning. Part 3. Proceedings
of the 2nd International Maastricht-todz Duo Colloquium on “ Translation and Meaning",
Maastricht, the Netherlands, 19-22 April 1995,141-163. Frankfurt am Main: Peter Lang.
Koller, Werner 1989 Equivalence in translation theory. Translated by Andrew Chesterman. In:
Andrew Chesterman (ed.), Readings in translation theory, 99-104. Finland: Oy Finn Lectura.
Koller, Werner 1995 The concept of equivalence and the object of translation studies. Target 7.2:
191-222.
Kroll, Judith F., Susan C. Bobb, Maya Misra and Taomei Guo 2008 Language selection in
bilingual speech: Evidence for inhibitory processes. Acta Psychologica 128: 416-430.
Kroll, Judith F. and Erika Stewart 1994 Category interference in translation and picture naming:
Evidence for asymmetric connections between bilingual memory representations. Journal
of Memory and Language 33: 149-174.
Kroll. Judith F. and Natasha Tokowicz 2005 Models of bilingual representation and processing:
Looking back and to the future. In: Judith Kroll and Annette M. B. de Groot (eds.).
Handbook of bilingualism: Psycholinguistic approaches, 531- 553. Oxford: Oxford
University Presa.
Kujamaki, Pekka 2004 What happens to “ unique items” in learners' translations? "Theories"
and “ concepts” as a challenge for novices' views on “ good translation” . In: Anna
Mauranen and Pekka Kujamaki (eds.), Translation universal: Do they exist? 187-204.
Amsterdam: John Benjamins.
Lachaud, Christian M. 2009 EEG, EYE and KEY: Three simultaneous streams of data for
investigating the cognitive mechanisms of translation. Paper presented at the IATIS
Conference, Monash University, Melbourne, Australia, July 8-10, 2009.
 Implications of Cognitive linguistics for Translation Studies ----- 71

Lakoff, George 1987 Women, fire and dangerous things. Chicago: University of Chicago Press.
Lakoff. George and Mark ]ohnson [1980] 1995 Metaphors we live by. Chicago: University of
Chicago Press.
Lakoff, George and Mark Johnson 1999 Philosophy in the flesh. New York: Basic Books.
Lakoff, George and Mark Turner 1989 More than cool reason. Chicago: University of Chicago
Press.
Langacker, Ronald [1987] 1991 Foundations of cognitive grammar. Vols. 1-2. Stanford, CA:
Stanford University Press.
Langacker, Ronald 2008 Cognitive grammar: A basic introduction. Oxford: Oxford University Press.
Laviosa, Sara 2002 Corpus-based translation studies: Theory, findings, applications.
Amsterdam: Rodopi.
Lewandowska-Tomaszczyk, Barbara 2010 Reconceptualization and the emergence of discourse
meaning as a theory of translation. In: Barbara Lewandowska-Tomaszczyk and Marcel
Thelen (eds.), Meaning in translation, 105-147. Frankfurt am Main: Peter Lang.
Malmkjaer, Kirsten 2000 Relative stability and stable relativity. Target 12.2: 341-344.
Malmkjaer, Kirsten 2008 Norms and nature in translation studies. In: Gunilla Anderman and
Margaret Rogers (eds.), Incorporating corpora: The linguist and the translator, 49-59.
Clevedon, UK: Multilingual Matters.
Marco, Josep 2007 The terminology of translation: Epistemological, conceptual and intercultural
problems and their social consequences. Target 19.2: 255-269.
Mauranen, Anna and Pekka Kujamdki (eds.) 2004 Translation universals: Do they exist?
Amsterdam: John Benjamins.
Molina Martínez, Lucia and Amparo Hurtado Albir 2002 Translation techniques revisited: A
dynamic and functionalist approach. Meta 47.4: 498-512.
Mollin, Sandra 2009 Combining corpus linguistic and psychological data on word co­
occurrences: Corpus collocates versus word associations. Corpus Linguistics and
Linguistic Theory 5.2: 175-200.
Monday, Jeremy 2008 Introducing translation studies: Theories and applications. 2nd edition.
London: Routledge.
Muñoz Martín, Ricardo 2006 Expertise and environment in translation. Paper presented at the
Second IATIS Conference, “ Intervention in Translation, Interpreting and Intercultural
Encounters” , University of the Western Cape, South Africa, July 11-14, 2006.
Nlda- Eugene 1964 Towards a science of translating. Leiden: Brill.
Nord, Christiane 1997 Translating as a purposeful activity: Functionalist theories explained.
Manchester: St. jerome.
^°han, Maeve 2004 Introducing corpora in translation studies. London: Routledge.
PAcTE group 2003 Building a translation competence model. In: Fabio Alves (ed.). Triangulating
translation, 43-66. Amsterdam: John Benjamins.
A ctE group 2005 Investigating translation competence: Conceptual and methodological
issues. Meta 50.2: 609-619.
aETE group 2010 Results of the validation of the PACTE translation competence model.
Translation project and dynamic translation index. In: Sharon 0 Brien (ed.), Cognitive
explorations of translation, 30-56. London: Continuum.
ACTE Sroup 2011 Results of the validation of the PACTE translation competence model:
translation problems and translation competence. In: Cecilia Alvstad, Adelina Hild and
Elisabet Tiselius (eds.), Methods and strategies of process research, 317-343. Amsterdam:
l°hn Benjamins.
72 Sandra L. Halverson

Paker, Salitha 2002 Translation as terceme and nazire-. Culture-bound concepts and their
implications for a conceptual framework for research on ottoman translation history. In:
Theo Hermans (ed.), Crosscultural transgressions: Research models in translation studies
II. Historical and ideological issues, 120-143. Manchester: St. Jerome.
Pavlenko, Aneta 2009 Conceptual representation in the bilingual lexicon and second language
vocabulary learning. In: Aneta Pavlenko (ed.), The bilingual mental lexicon:
Interdisciplinary approaches, 125-160. Bristol, UK: Multilingual Matters.
Popovic, Anton 1970 The concept of “ shift of expression” in translation analysis, in: James S.
Holmes (ed.). The nature of translation: Essays on the theory and practice of literary
translation, 78-87. The Hague: Mouton.
Putnam, Hilary 1992 Renewing philosophy. Cambridge, MA: Harvard University Press.
Pym, Anthony 1992 Translation and text transfer: An essay on the principles of intercultural
communication. Frankfurt am Main: Peter Lang.
Pym, Anthony 1998 Method in translation history. Manchester: St. Jerome.
Pym, Anthony 2007 Natural and directional equivalence in theories of translation. Target 19.2:
271-294.
Pym, Anthony 2008 On Toury’s laws of how translators translate. In: Anthony Pym, Miriam
Shlesinger and Daniel Simeon! (eds.), Beyond descriptive translation studies:
Investigations in homage to Gideon Toury, 311-327. Amsterdam: John Benjamins.
Pym, Anthony 2010 Exploring translation theories. London: Routledge.
Risku, Hanne 2002 Situatedness in translation studies. Cognitive Systems Research 3.3:523-533.
Robinson, Peter and Nick C. Ellis 2008 Handbook of cognitive linguistics and second language
acquisition. New York: Routledge.
Ruiz, C., N. Paredes, P. Macizo and M. T. Bajo 2008 Activation of lexical and syntactic target
language properties in translation. Acta Psychologica 128: 490-500.
Sandra, Dominiek 1998 What linguists can and can’t tell you about the human mind: A reply to
Croft. Cognitive Linguistics 9.4: 361-378.
Sandra, Dominiek and Sally Rice 1995 Network analyses of prepositional meaning: Mirroring
whose mind - the linguist’s or the language user’s? Cognitive Linguistics 6.1: 89-130.
Schaffner, Christine 2009 Functionalist approaches. In: Mona Baker and Gabriela Saldanha (eds.),
Routledge encyclopedia of translation studies, 115-121. 2nd edition. London: Routledge.
Sela-Sheffy, Rakafet 2000 The suspended potential of culture research in TS. Target 12.2:
345-355.
Sharmin, Selina, Oleg Spakov, Kari-Juoko Rdiha and Arnt Lykke Jakobsen 2008 Effects of time
pressure and text complexity on translators’ fixations. In: Stephen N. Spencer (ed.),
Proceedings of Eye Tracking Research and Applications Symposium (ETRA08),
26-28 March, Savannah, Georgia, 123-126. New York: ACM.
Simeoni, Daniel 1998 The pivotal status of the translator’s habitus. Target 10.1: 1-39.
Simeoni, Daniel 2000 When in doubt, contextualize . . . Target 12.2: 337-341.
Sjprup, Annette C. 2009 Cognitive effort in metaphor translation: an eye-tracking study. Paper
presented at Source linguistic influence” , PhD seminar, University of Bergen, Norway,
November 11-14, 2009.
Snell-Hornby, Mary 1988 Translation studies: An integrated approach. Amsterdam: John
Benjamins.
Stamenov, Maxim, A. Gerganov and I. D. Popivanov 2009 Shortcuts to meaning: Finding out
meaning correspondences in the bilingual lexicon during a word translation task. Paper
presented at Eye-to-IT Conference, Copenhagen, Denmark, April 28-29, 2009.
 Implications of Cognitive Linguistics for Translation Studies — 73

Stockwell, Peter 2002 Cognitive poetics: An introduction. London: Routledge.

Susam-Sarajeva, $ebnem 2009 The case study research method in translation studies. The
Interpreter and Translator Trainer 3.1: 37-56.
Tabakowska, Elżbieta 1993 Cognitive linguistics and the poetics of translation. Tübingen,
Germany: Gunter Narr.
Tirkkonen-Condit, Sonja 2004 Unique items - over- or underrepresented in translated
language? In: Anna Mauranen and Pekka Kujamaki (eds.). Translation universal: Do they
exist? 177-186. Amsterdam: lohn Benjamins.
Tirkkonen-Condit, Sonja 2005 Do unique items make themselves scarce in translated Finnish?
In: Krisztina Käroly and Agota Fóris (eds.), New trends in translation studies: In honour of
Kinga Klaudy, 177-189. Budapest: Akademiai Kiadó.
Toury, Gideon 1995 Descriptive translation studies and beyond. Amsterdam: John Benjamins.
Tsur, Reuven 2003 On the shore of nothingness: A study in cognitive poetics. Exeter, UK: Imprint
Academic.
Tummers, Jose, Kris Heylen and Dirk Geeraerts 2005 Usage-based approaches in cognitive
linguistics: A technical state of the art. Corpus Linguistics and Linguistic Theory 1.2:225-261.
Turner, Mark 1991 Reading minds: The study of English in the age of cognitive science.
Princeton, N|: Princeton University Press.
Turner, Mark 1996 The literary mind. New York: Oxford University Press.
Tymoczko, Maria 2007 Enlarging translation, empowering translators. Manchester: St. Jerome.
Valenzuela, Javier and Ana Rojo (eds.) 2003 Contrastive cognitive linguistics. Special Issue,
International Journal of English Studies 3.2.
Van Leuven-Zwart, Kitty M. 1989 Translation and original: Similarities and dissimilarities, I.
Target 1.2: 151-181.
Van Leuven-Zwart, Kitty M. 1990 Translation and original: Similarities and dissimilarities, II.
Target 2 .1: 69-95.
Vanuti, Lawrence 1995 The translator's invisibility. New York: Routledge.
Vinay. lean-Paul and Jean Darbelnet 1995 Comparative stylistics of French and English: A
methodology for translation. Translated and edited by Juan C. Sager and M.-J. Hamel.
Amsterdam: John Benjamins.
Wulf, Michaela 1997 Translation as a process of power: Aspects of cultural anthropology in
translation. In: Mary Snell-Hornby, Zuzana Jettmarova and Klaus Kaindl (eds.), Translation
os intercultural communication: Selected papers from the EST Congress, Prague 1995,
123-133. Amsterdam: John Benjamins.
Wolf, Michaela 2002 Culture as translation - and beyond: Ethnographic models of
representation in translation studies. In: Theo Hermans (ed.), Crosscultural
transgressions: Research models in translation studies II. Historical and ideological
issues, 180-192. Manchester: St. Jerome.
Wolf, Michaela and Alexandra Fukari (eds.) 2007 Constructing a sociology of translation.
Amsterdam: John Benjamins.
ulff, Stefanie 2009 Converging evidence from corpus and experimental data to capture
idiomaticity. Corpus Linguistics and Linguistic Theory 5.1:131-159.
a albeascoa, Patrick 2000 From techniques to types of solutions. In: Allison Beeby, Doris Ensinger
and Marisa Presas (eds.), Investigating translation, 117-127. Amsterdam: John Benjamins.
^¿bieta Tabakowska
(Cognitive) grammar in translation:
^°Nn as meaning

1 Preliminaries
In c°ntemporary scholarship the difficult relation between translation theory
applied Translation Studies (TS) on one hand and linguistics on the other
ad not been running very smoothly; in the 20th century, fascination with gen-
erat've models o f language as infallible paths leading to equivalence (Nida 1964)
ended in disillusion. Some of those who were (rightly) defending the autonomy
0f the discipline opted (wrongly) for a total ban on linguistic intrusions. As late
^008 Routledge announced “Translation Studies, a New Journal for 2008” ,
a'n'ed at exploring

Promising lines of work within the discipline of Translation Studies while placing a special
rntphasis on existing connections with neighbouring disciplines. In addition to scholars
w''hin Translation Studies, [the editors) invite those as yet unfamiliar with or wary of
Translation Studies to enter the discussion. Such scholars w ill include people working
*n Mtenry theory, sociology, ethnography, philosophy, semiotics, history and historiography,
Zoology, gender studies, postcolonialism, and related /¡eMs.( www.inform aworld.com /
translationstudies, emphasis added)

nSuistics was simply not there - not even among “ related fields” [. . .]
obv.^°r*Una*ely> 1° many people working w ithin the field of TS it seems rather
^ Vlous that translation theory cannot exist without a theory of language and
^ atTunar - indeed so very obvious that TS are often considered to be synony-
atJ Us Applied Linguistics, where the “ applied” part is supposed to provide
PerfVerS ^ 0W linguistic material is to be handled when translation is being
Pitied, and the “ linguistics” part is expected to tell the translators what the
la er'al *s actually like. In recent years, the advance of the cognitive theory of
ihg Ua^e an<^ emergence o f cognitive models of grammar have been herald-
the 9 cbange in the overall attitude of TS scholars towards linguistics. Out of the
dj0s^ es Presently available, that behind Cognitive Linguistics (CL) seems the
■bar ^r° m' s' n§: *1 incorporates precisely those aspects of language and gram-
Prac i a^ ^ave always frustrated language-oriented theorists of translation and
kctj S*n^ iranslators. Indeterminacy of grammar (Langacker 2009), inherent sub-
lty meaning, the existence of motivated relationships between meaning
 230 Elżbieta Tabakowska

and form, ubiquity o f metaphor (Lakoff and Johnson 1980; Lakoff 1987) and con­
ceptual integration (Fauconnier [1985] 1994; Fauconnier and Turner 2002) - al1
these postulates strike a sympathetic note in the hearts of translation scholar
as does the cognitivist statement that culture is embodied in grammar, which
marks the “ cultural turn” in linguistics, running parallel to the “ cultural turn
in TS as heralded in Mary Snell-Hornby’s seminal paper (1990).
Methodological consequences are pertinent: if TS is to accept the theoretical
framework o f CL, it also has to admit - in agreement with CL tenets - that thefe
is no qualitative difference between “ common” and “ literary” language, an
assumption that may be considered as the strongest pier supporting the bridge
to be built over the gap which has traditionally - and unfortunately - separate
linguistics from literature. According to the cognitivist stance, literature, and
poetry in particular, is simply an extreme manifestation of language use. Litef
ary (poetic) texts contain more instances of what is traditionally defined as I*1'
erary (poetic) devices than ordinary everyday usage, with writers and P°ets
pushing linguistic expressions up along the scale of linguistic creativity. Hence’
ultimately, the methodological turn in TS makes “ literary translation” qualitatively
comparable to “ translation as such” .
I here are at least three arguments that may be used to defend the significar,t
role of linguistics in TS. At the first stage o f the translation process, systematic
reference to language, and to grammar in particular, helps to understand the
original text or discourse by focusing the reader’s/translator’s attention on f°*'
mal minutiae that ultimately decide about the overall meaning of the origin3 •
In other words, focusing on grammar underlies the necessary phase of ‘ ci°s
reading” .1 At the second stage - that o f translation proper - focus upon gra^
mar in its traditional guise of contrastive linguistics helps to find TL counter-
parts o f ST units. Finally, in the final phase o f translation assessment and
criticism, it makes it possible to judge the results of the translator’s effort® in
a less-than-usually impressionistic way.
All these arguments find support in the cognitivist theory of language and
mguistic structure, that is, in Cognitive Grammar (CG). Underlying the cogn>tlVt
theory of grammar, as developed w ithin the framework of CL, there are s ° f
basic questions which seem relevant for TS. The first one, most significant f°
ot c isciplines, is the question concerning the nature of meaning, which c°$
mtivist theories treat as tantamount to conceptualization. Since c o n c e p t1'
zation is necessarily the product of an individual cognizant mind, niean'0
erently subjective. Consequently, the semantic value of an expresS'°

etter known to language teachers under the French name explication du texte.
 (Cognitive) grammar in translation: Form as meaning 231

involves not only the properties of what the expression refers to, but also the
c°nceptualizer’s individual way of looking at things. And since conceptualiza-
tl°n results from a particular selection of a point of view and of the resulting per­
fective, it never covers all aspects of things perceived; hence meaning is also
'nherently metonymic in character.
The second basic question to be asked concerns the nature of language,
which - from the cognitivist point of view - is largely motivated by perceptual
^ ilitie s of the human mind. It is assumed that Gestalt principles of perception
are reflected in linguistic structure. For instance, perception involves scanning
of the visual field, and different types of scanning (syncretic scanning for things
lis tin g in space and sequential scanning for processes which develop in time)
Underlie the basic categorial opposition between nouns and verbs, found in
m°st natural languages. The principles of sim ilarity and proximity, whereby
thlngs that are seen as similar and close together are perceived as belonging
to§ether, account for collectivity as the grammatical property of lexemes such
as the English nouns archipelago or alphabet. The principle of area, which states
,hat the smaller of two overlapping or adjacent elements being perceived is seen
as a figure against the background formed by the larger one underlies entire lin-
8uistic structure. For example, it is reflected in the choice of a lexeme for the role
the sentential subject: we tend to say the tree near the house rather than the
h°Use near the tree. Finally, since meaning is inherently metonymic, linguistic
fressions of concepts are indeterminate: they only provide guidelines that
ead recipients of linguistic messages towards an interpretation. This claim tal-
es with modern translation theories, which claim that the object of the transla-
. 0r* Process is not the original text (or discourse), but the translator s own
nterPretation thereof.
( Although the “ founding father” of CG openly rejects “ uninformed commen-
*0rs claim that all meaning is “ based on space or visual perception (Lan-
8acker 2008: 55), there are well justified (theoretically, but also empirically)
j! °P°sals that some kind of visualization does indeed underlie conceptualiza-
^°0s. Thus perception and conception are two interconnected domains of sen-
r'al and mental activities; taken together, they constitute a larger domain of
ePfi'°n, which covers all cognitive processes, both conscious and subconscious
er° r explanation of the notion of -ception, see Talmy 1996: 245-248). In ref-
r*ce translation, it is not unreasonable to assume that a visualization might
Q 0 ar’Se in the translator’s mind to function as the notorious tertium compar-
Vj ° n's involved in the translation process (cf. e.g. Kussmaul 2005). Appeal to
in ^a''Zafi° n seems particularly well grounded in the analysis presented further
this Paper (in Section 3), since - like many texts the illustrative text
ahal
Vzed is predom inantly descriptive.
232 Elżbieta Tabakowska

2 Imagery - conventional and unconventional

2.1 Dimensions

Another fundamental question that underlies cognitivist models of langu3^

concerns the nature of grammar and grammatical structure. As was said in Sec'
tion 1 above, the basic principle that underlies CG states that the human mind
has the ability to construe the scene, that is, the situation that is conceived , 1,1
alternate ways. This ability is called imagery.
In CG the ability to create - and to choose between - alternate constru^ 5
does not pertain exclusively to lexical semantics. Grammar also contributes t°
the meaning of expressions, and its semantic contribution resides in what lS
defined as conventional imagery: an inventory of alternate scene construals sa*1^
tioned by social conventions, which are traditionally described as grammat'ca
rules. In this sense grammar is symbolic, since it constitutes a part o f the over3
semantic value of expressions: the meaning of an expression is not only its c011
ceptual content, but also the way in which the expression is construed, that >•'
its grammatical construction. Obviously, such a claim undermines the tra
tional dichotomy between form and meaning and validates the cognidv's
description of grammatical units as cognitive routines. In CG the status of the ^
guistic unit is granted to any conventional pairing of meaning with a for111
has been established well enough to be “ evoked as an integrated whole’ •
initiated, it is carried out without much conscious reflection - very much 1'
following a routine path (Langacker 1987: 100). Thus - unlike in other gramr* 1^
ical theories - a unit can be a lexeme, a bound morpheme, a complex word (
instance the plural form of a countable noun), an idiomatic phrase or, at the se®
tence level, a grammatical construction with a schematic meaning, e.g. Suffi
verb-object (realized as, for instance, the sentence The bird ate the ongle^01^ ^
Close relation between such units of language and the “ units of translation
postulated by contemporary translation theories seems fairly obvious. I he ea ^
postulate by Vinay and Darbelnet ([1958] 1995) that the unit of translation sh°u ^
be defined as the smallest segment of the text which has to be translated 3
whole has been criticized for its prescriptivism and for being “ based on ¡dta
translations (Kenny 2008: 304). However, the definition found in a more reCe
presentation reads as if it were taken verbatim from a textbook of CG: “ A tranS
tion unit is a segment of a text that a translator treats as a single cognitive did* ,g
the purposes of establishing an equivalence. The translation unit may be a s' ^ j
word, a phrase, one or more sentences, or even a larger unit” (Banjar Yousef11- ^
Similarly, considered to be one of the key terms in translation theory, the uu1
 (Cognitive) grammar in translation: Form as meaning — 233

translation is defined in a glossary of computer-aided translation terms as “ the

smallest entity in a text that carries a discrete meaning” which “ varies all the
time, ranging from individual words and sentences right up to entire para­
graphs” (TRANSLATUM, online). Establishing the status of cognitive routines,
m the source and the target languages, could certainly be significant for the
developrnent of translation theory, practice and pedagogy. The issue deserves
m°re detailed discussion which, however, cannot be pursued w ithin the lim its
this essay.
first defined in Langacker’s programmatic article (1988), the notion of imag­
ery has since been somewhat modified. In recent works Langacker and his fol-
lovvers tend to replace mental imagery with the term construal, which includes a
''’•der array of construal operations, some of which are less easily discussed
u>ider the rubric of the imagistic dimensions of linguistic structure (cf. Lan-
Sacker 2008; Croft and Cruse 2004; Taylor 2002). However, the main assump-
h°ns and definitions remain unchanged: “ Cognitive Linguistics incline to [. . .]
'maSistic accounts” (Langacker 2008: 32). Mental images are “ representa-
d°ns of specific, embodied experiences” (Croft and Cruse 2004: 44).
I he full inventory of aspects of conventional imagery and construal phe-
a°mena has been amply discussed in the literature (for a recent account, see
p a c k e r 2008: 55-89). The framework for the case study below is provided
y those grammatical aspects of construal that are intuitively most obviously
.^S istic in nature, on the assumption that it is grammatical imagistic symbol-
‘Sm that is most relevant for TS. The dimensions, or in Langacker’s terminology,
°CQI adjustments (Langacker 1987: 116-138), are

*evel of specificity
figure-ground alignment
Perspective, involving point of view and subjectivity.

these aspects are convincingly explained in terms of the visual metaphor that
p a c k e r uses to refer to what he defines as viewing arrangement (2008: 73-78),
t f tde relationship between the conceptualizer (the speaker and the hearer, me-
P °rized as “ viewers") and the object of conceptualization (metaphorized as the
atiri^ ^ default viewing arrangement - naturally presupposed by the speaker
Sg tfle hearer - is the situation in which the interlocutors are located in the
He*16 ^ ace and “ view” things at the same time, like an audience in the theatre.
the0»6.1*16 extension o f the basic viewing metaphor: whatever is located w ithin
» visual field” of the conceptualizers, is said to be located onstage. The
atrSUal” aspects o f construal are thus seen as derivatives of the viewing
tangement.
234 Elżbieta Tabakowska

2.2 Construal: Specificity

In Langacker’s model of CG, the term specificity refers to the level of accuracy
and precision that the conceptualizer chooses in order to characterize a g*vel1
situation. Thus they can say, for instance, that a bird came down the ^ a,k
(see the analysis of Dickinson’s poem in Section 3 below), but they can als°
choose a more detailed account: a small magpie ran down the narrow g o ^e,]
path. In reference to the overall viewing metaphor, it can be stated that
scale running from low to high levels o f specificity (i.e. from most schema*11-
to most detailed construals) is connected with the situation (spatial and/or te111
poral) of the viewer and the viewed. This parameter, metaphorically referred
as “ distance” , finds its direct counterparts in linguistic structure. For in sta n t
we tend to say the bird unrolled his wings rather than the bird unrolled his fea^
ers, as the concept of “ feathers” requires a reduction of the visual field to
direct scope of observation, i.e. the wings (of a bird). Conventionally, this meat3
(and linguistic) operation, metaphorically referred to as zooming, requires adhe
ence to what things are in the w orld” : feathers are a part of wings, and win§s a(
a part of a bird. As exemplified below, Dickinson’s play with the level of spec'
city breaks the convention: the readers have to supply the missing link then1^
selves, which makes their task more complex and at the same time justifieS *
classification of the expression as belonging to “ the language of poetry” - ,
In translation, shifts in levels of specificity appear as effects of transfer
strategies used in dealing w ith cases of non-equivalence, notably those resid**a
from interlingual differences in the hierarchical structure of concepts fe-S- ^
notorious problems with culture-specific taxonomies). Such local stra,e^
involve translation by a subordinate term (magpie vs. bird), cultural substiW*'0
overt explanation or paraphrase, etc. (for a discussion, see, e.g. Baker 1992)-

2.3 Construal: Figure-ground alignment: Trajector

and landmark

As was said above (Section 1, the principle of area), the figure-ground d i s t i « ^

comes from Gestalt psychology. Introduced into CL by Leonard Talmy*'' ‘
opposition is used to account for asymmetries o f linguistic structures,
speakers tocus upon certain elements of conceived scenes by foregf°un j
them at the cost of other elements, which are backgrounded. Elements seleC

2 For an overall survey, see Talmy (2000).

 (Cognitive) grammar in translation: Form as meaning 235

for attention and consequently given more prominence are figures; the back-
bounded ones serve as the ground. Thus in the sentence A bird came down
tf,e walk “ a bird” is a figure, and “ the w alk” is the ground. Like in visual percep-
tior>, figures tend to be smaller, more mobile, structurally simpler, more salient
and more recently called upon than grounds (Talmy 2000: 315-316; cf. also
Section l above).
In Langacker’s CG the figure-ground opposition is inherent in the concepts
Qf R e c to r and landmark, which w ill be used in the analysis in Section 3 below.
In a conceptualization, the trajector is the most prominent element - the object
of Primary focus. The entity of secondary prominence (or secondary focus) goes
by the name of landmark (Langacker 2008: 70). Two expressions might have the
Same semantic content, and still differ in meaning because of the difference in
the trajector-landmark alignment. The semantic role of figure is found to coin-
Clde with that of the sentential subject. For instance, in The bird glanced with
rQPid eyes, also from Dickinson’s poem in Section 3, it is “ the bird” that is the
bgUre (and, also, the grammatical subject) and “ rapid eyes” are the landmark
(the indirect object). In the almost-synonymous The bird’s rapid eyes glanced
relationship is reversed. In traditional terms, the first sentence is about the
while the second one “ is about” the bird’s eyes (i.e. the sentences show
*be reversal of topic and comment selection).
ia translation, shifts in the trajectory-landmark structure can be imposed
Pon the translator by systematic differences between source and target gram-
tT*ars>but they can also result from lack of sufficient attention to semantic nuances
n (be part of the translator.

^ Construal: Perspective

P conceptualization, perspective is closely connected with the viewing arrange-

s 6nt: expressions, and especially spatial descriptions, depend crucially on the
isPeaker’s position, defined as their relative point of view. The phenomenon itself
v ° f c°urse well known: semantic oppositions exemplified by such pairs of ad-
,herbs as behind vs. in front of, or under and above, have been amply discussed in
Qle ''terature. But CG employs the term in a wider sense, to cover, for instance,
s0 ° ^ernP°ral and epistemic aspects. It is not only the so-called shifters (per
a9l or demonstrative pronouns) that establish viewpoints, but also grammat-
trg tenses>which establish temporal perspective relative to the speaker. Mental
Psfer, in some ear,ier theories of grammar discussed under the rubric of
^P athy (Runo and Kaburaki 1977)> accounts for shifts of perspective, which
y even occur w ithin a single expression. For instance, while the expression
236 Elżbieta Tabakowska

A bird came down the walk establishes the viewpoint as that of the speaker,
whom the bird is approaching, an alternative construal A bird approached
the poet would bring in the point of view of an external observer, while yet
another alternative, e.g. A bird inspired the poet would shape the conceptualizer’s
perspective through mental transfer.
An important parameter modifying perspective is the opposition between
subjectivity and objectivity of construal. The two concepts were introduced and
developed in Langacker’s model of CG (Langacker 1999: 297-315). The opposi­
tion that they express refers to the way and the extent to which the speaker is
present in the conceptualization. In other words, the speaker can perform - to
varying extent - the role of either the sujet parlant, that is, the subject of the con­
ceptualization, or - while retaining their role of the speaker - become also the
object of what is being said. To come back to the viewing metaphor, in such a
case the speaker is present onstage. The phenomenon can be observed in
numerous manifestations of linguistic structure, and the scale ranges from
high levels of objectivity of an expression, as for instance in Mommy will give
you a sweet (said by mother to child) to high levels of subjectivity, as in I will
give you a sweet, uttered in the same pragmatic context, with the pronoun
I bringing the speaker onstage.3
Like other dimensions of scene construal, perspective introduces subtle,
but often very significant, aspects of meaning, irrespective of the actual seman­
tic content of utterances. And like the other dimensions, it tends to be prob­
lematic in translation, because of either systematic differences between
languages (e.g. obligatory or optional presence of personal pronouns) or cultural
differences (e.g. social conventions concerning the presence of the speaker
onstage).
A ll the dimensions o f construal find their way into grammatical systems
through conventionalization. With particular ways of seeing the world becom­
ing conventionalized due to repeated use, the speakers of the language begin
to see things through the “ optical glasses” o f convention. This statement, one
of the fundamental assumptions underlying the cognitive model of grammar
tallies w ith the principle o f linguistic relativism. Banned as contradictory t0
algorithmic conceptions of grammar, the principle of relativism, known as
the weaker version o f the Sapir-Whorf hypothesis, finds its place again in con
temporary linguistics. We look at the world w ith the eyes of a generalize
(conventional) observer. But poets are different. Extending or breaking ^ e

3 In Dickinson s poem there is no example to illustrate objective construals, which tallieS

the interpretation proposed in Section 3.
 (Cognitive) grammar in translation: Form as meaning ----- 237

conventions is the crux of poetry, as well as the stumbling block for poetry
translators.

3 Case study
Although TS scholars who arrived from the “ linguistic camp” postulate creating
closer ties between theoretical assumptions and their practical applications,
works devoted to the actual application of linguistics to translation (process
and product) are few and far in between (Halverson 2007; O’Brien 2011; McElha-
non 2005; Tabakowska 1993a, 1993b, 1993c, 1996, 1997, 2000, 2004; Wójcik-
Leese 2000). The difficulty is to a large extent purely technical. Notwithstanding
their claims to the contrary, grammatical descriptions made w ithin the CL frame­
work rarely go beyond the level of sentence as the largest unit of consideration:
longer stretches of text and/or discourse used as illustration would meet theoret­
ical requirements of the model, but at the same time become unmanageable and
make high demands on the reader’s time and patience. The difficulty increases
when focus on translation calls for more languages - especially if one of them
does not happen to be a world language - e.g. Polish, whose knowledge is still
classified in job application forms as an “ exotic s k ill” .
Hence the decision taken by the present author to choose for her case study
an analysis of a short poem written in English (328 by Emily Dickinson) and its
Polish translation by Stanisław Barańczak (Dickinson 1990). The analysis is an
attempt to substantiate two claims: firstly, that much of the meaning of a
poem resides in its grammar, and secondly, that the poetry of imagery is built
of the prose of grammar. I w ill investigate selected aspects of the grammatical
structure of the original and compare them with those found in the translation,
hoping to reveal (some of the) reasons why the translation - although in the
opinion of Polish readership it “ reads w ell” - prompts an interpretation different
from that imposed upon the reader by the original poem.

3*1 The original poem

Emily Dickinson: 328

1 A Bird came down the Walk -
He did not know 1 saw -
He bit an Angleworm in halves
And ate the fellow, raw,
238 Elżbieta Tabakowska

5 And then he drank a dew

From a convenient Grass -
And then hopped sidewise to the Wall
To let a Beetle pass -

9 He glanced with rapid eyes

That hurried all around
They looked like frightened beads, I thought
He stirred his Velvet Head

13 Like one in danger, Cautious,

I offered him a Crumb
And he unrolled his feathers
And rowed him softer home -

17 Than Oars divide the ocean,

Too silver for a seam -
Or Butterflies, off banks of Noon
Leap, plashless as they swim.

3.2 The Polish translation

1 Ptak przyskakał po Ścieżce -

Nieświadom, że go widzę -
Wpierw rozpolowił dziobem
I zjadł na surowo Dżdżownicę -

5 Potem popił ją Rosą

Z dogodnego źdźbła Trawy -
Pod Mur uskoczył - spłoszony -
Jakimś Żukiem niemrawym -

9 Bystre oczy strzelały

Dookoła spojrzeniami -
Czarne Paciorki Trwogi -
Jego Głowy Aksamit

13 Drgał, targany przez Groźby

Rzuciłam ostrożną Ręką
Okruch - rozpostarł pióra
I odpłynął tak miękko,
 (Cognitive) grammar in translation: Form as meaning
239

17 Jak Wiosła krojące Ocean -

Zbyt srebrny, by szwy pozostały -
Lub Motyl - co spada bez plusku
Z Urwisk Dnia w bezdenne Upały.

3.3 The original poem and the translation: Alternate construals

3.3.1 Specificity

The original o f Dickinson’s 328 begins with

(1) a. A Bird came down the Walk -

which is rendered in the translation as

(1) b. Ptak przyskakal po Ścieżce -

Bird przy.jumped.iMPKRF along path
‘A/the bird approached jumping repeatedly along the path’

ln (lb) the verbal prefix przy■ - derived from the corresponding spatial pre­
position przy, and corresponding to the English prepositions at or (near)by -
contributes to the meaning of the entire expression by bringing in the sense
equivalent to that brought into the phrasal verb in (la) by the element down:
fhe direction of motion towards the observer. However, while the English
c«me down does not specify the particular kind of this motion, its Polish coun-
terPart does: the bird moved towards the observer (direction), jumping along the
path (manner). Thus in the Polish translation the scene is “ painted” in more
detail. Increased specificity can thus imply a more careful observation, made
p°ssible by a shorter distance between the subject and the object/* The image
ls enhanced, in the Polish translation, by the explicit reference to the bird’s
keak (apparently seen in detail); compare:

^ a. He bit an Angleworm in halves

b. Wpierw rozpolowil dziobem Dżdżownicę
At.first halved beak.iNSTR earthworm
‘At first he halved an earthworm with his beak’

4
^0r a detailed CG analysis of przy, see Przybylska (2002: 491 515).
240 Elżbieta Tabakowska

The translator’s choice o f a higher level o f specificity has yet another conse­
quence as far as the interpretation of the poem is concerned. The translation
of (la) might have been

(1) c. Ptak przyskoczył po Ścieżce

Bird przy.jumped.PERF along path
‘The bird approached by making a single jump along the path’

Przyskakał and przyskoczył are both perfective (in the sense of the term “ perfec­
tive” commonly used in Slavic studies for aspectual oppositions, although not in
the sense in which it occurs in reference to English verbs analysed in CG, where
the opposition “ perfective/imperfective” overlaps the distinction between simple
and continuous tenses; cf. Langacker 1991: 351). However, these two Polish verb
forms display a subtle semantic difference. Przyskakał in (lb) is rendered perfec­
tive by adding the perfectivizing prefix przy- to an imperfective iterative verb
skakał, while in przyskoczył in (lc) the stem itself (skoczył) is non-iterative and
perfective, and hence its lexical meaning is that of a single and completed
action.5 While both verbs focus on the aspect of completion,6 the iterative
verb in (lb) expresses a repeated action of jumping, while the non-iterative
verb in (lc) relates to a single jump. By choosing (lb) over (lc) the translator
“ prolonged" the process, thus implying a longer lasting observation of the ap­
proaching bird on the part of the sujet parlant. The original construal does
not carry analogous implications. But the choice seems to be a welcome decision
on the part of the translator, as the iterative verb reflects the process of sequential
scanning, with the act of observation extending over a period of time.
As was stated above, in visual perception the level of specificity varies rel­
ative to the distance between the observer and the object of their observation-
In her readers’ and critics’ opinion (cf. e.g. Sewall 1963; Freeman 1997), Emily
Dickinson excelled at describing minute details. However, noticing minute
details implies being positioned close to the object of one’s observation-
Such is indeed the construal in lines 5-12 of 328. For instance, lines 5 and
6 read:

(3) a. And then He drank a Dew/From a convenient Grass -

5 In traditional terminology used by Polish linguists, the former would be an "aspectual”

the latter a creative” formation (cf. e.g. Janowska and Pastuchowa 2005: 150).
6 Cf. Vendler’s "accomplishment" in Vendler (1967).
 (Cognitive) grammar in translation: Form as meaning ----- 241

which is rendered in Polish as

(3) b. Potem popił ją 7 Rosą z

Afterwards washed.down her dew.iNSTR from
dogodnego źdźbła Trawy
convenient blade grass.GEN
‘And then he washed it down with dew from a convenient blade of grass’

In terms of prescriptive textbook rules, (3a) is ungrammatical: dew and grass

designate8 substances and as such are uncountable; their linguistic labels
must not be preceded with the indefinite article. But - in perfect agreement
with cognitivist principles of alternate construal (see Section 2 above) - the
Poet breaks the convention (twice) in order to render the two nouns countable,
so that she can then make reference to individual tiny portions of each of the
substances which they designate. English offers a conventional possibility of
doing this by what is in CG defined as external bounding: a portion of substance
selected for reference can be made countable by adding a notional (and linguis­
tic) contour that provides a boundary. Thus we can (and conventionally do) talk
about drops o f dew and blades o f grass. But Dickinson does not make use of this
conventional option. Her choice of construal may be interpreted as the intention
to a linguistic counterpart of what in film shooting is called zooming, that is,
reduce the distance with a consequent narrowing o f a visual field and focusing
uPon a little fragment of the scene.9 The Polish translation fails to preserve this
dement of the meaning - partly due to the absence in Polish of morphological
barkers of countability. In the first of the two cases in (3a) the countable a dew
changes in the translation into the unbounded substance rosa-NOM ‘dew’. In
tbe second case, the translator compensates for the lack of the indefinite article
by adding explicit external bounding, which results in a strongly conventiona-
ll2ed set phrase źdźbło trawy - ‘a blade of grass’. The result is an image incon-
s'stent in terms of specificity and (partly) devoid of the non-conventional
Pagery, which in the original justifies Dickinson’s famous penchant for “ minute
details” .

e noun rosa ‘dew’ is feminine.

n terms or, in CG terms, profile,
linguistic counterpart of zooming as a technique in film shooting; cf. Section 2.
242 Elżbieta Tabakowska

3.3.2 Trajector-landmark alignment

The first four stanzas of 328 have a clear structure, enhanced both by lexical
choices and by grammatical structuring. The sujet parlant notices a bird coming
down the path and she begins to watch him. She sees him catch and eat a worm
and then zooms in to watch him drink some dew off the grass close to him. She
sees him look round anxiously, and zooms in again, to focus upon his frightened
eyes. The fourth stanza brings a scene of a larger scope, with the observer offer­
ing the bird a crumb, and the bird flying away. However, an analysis of the
Polish translation reveals a somewhat different construal.
In the original poem, the bird is the trajector - also in the grammatical
sense, as he is the grammatical subject of all structures in lines 1-15: A b ird
came down (line 1); he did not know (line 2), he bit the Angleworm (line 3), he
drank a dew (line 5), /he/ hopped to the Wall (line 7), he glanced (line 9), he
stirred his Velvet Head (line 12), he unrolled his feathers (line 15), he rowed
him (line 16). In the translation the arrangement is sometimes changed:

(4) a. He stirre d his Velvet Head

is rendered as:

(4) b. Jego Glowy Aksamit/ Drgal targany przez

His head.GEN velvet/ trembled jerked by
Grozby
threats
‘The Velvet of his Head trembled, jerked by threats’

and

(5) a. He glanced with rapid eyes

as

(5) b. Bystre oczy strzelaly

Quick eyes shot
‘His quick eyes shot rapid glances’

The three metonymies (velvet for ‘the bird’s head’ ; the bird’s head for ‘the bird’ >n
[4b| and eyes o f the bird for ‘the bird’ in [5b]) alter the trajector/landmark alig11
ment of the original image: in these two cases, (4b) and (5b), the function of ha
jectors is ascribed not to the bird itself as the overall object of observation, but 1°
 (Cognitive) grammar in translation: Form as meaning ----- 243

what CG refers to as active zones, that is, those parts of the objects which are
made cognitively activated (cf. e.g. Langacker 2008: 331-334), or focused upon.
In the Polish version, the observer immediately focuses upon the active zones:
the velvet (of the bird’s head), the quick eyes (of the bird). In the original, the
observation begins with the bird as a “ Gestalt” entity, and - like earlier in the
poem - the eyes and the head are (gradually) focused upon through zooming in.
Similarly, where in lines 13 and 14 the original text has:

(6) a. Cautious,/1 offered him a Crumb

in the translation we read:

(6) b. Rzuciłam ostrożną Ręką/ Okruch

(I) threw cautious.iNSTR hand.iNSTR/ crumb.ACC
‘I threw a crumb with a cautious hand’

Once again, the Polish phrase in (6b), retranslated into English as ‘a cautious
hand’, makes reference to an active zone - the part of the observer’s body
which directly performs the action of offering a crumb to the bird. Yet the orig­
inal has the “ holistic” personal pronoun /. Lines 13-16 depict a direct confron­
tation between the bird and the bird-watcher, with both participants onstage.
This widening o f the visual field is significant because o f what follows: the
’ast stanza brings further widening of perspective, which now embraces the
°cean and the “ banks of Noon” .
In the original poem the bird and the woman are conceptually (and syntac-
tically) linked together by the adverb cautiously, which refers both “ forwards”
and “ backwards” . Both participants onstage are cautious - unlike in the Polish
Version, where the property is metonymically attributed to the observer alone,
^he interaction between the two participants of the encounter becomes in the
P°lish translation a rather conventional scene of a woman watching a bird
and then offering it a crumb, cautious not to scare it away.

3-3.3 Perspective

Was already said, the poem reveals a clear spatial orientation: the poet begins
Watch a bird, which she notices coming her way along the path. The orienta-
t'0'1 *s signalled by lexical semantics (came down, hopped sidewise to the Wall,
.0lVed him [.. .¡home), but also, as was seen in Section 3.3.2 above, by construals
ltTlplying shifts in spatial distance. The temporal orientation is marked just as
2 44 Elżbieta Tabakowska

clearly - again, by the time adverb then (line 5) and the repeated use of the
sequential conjunction and (lines 4, 5, 7, 15, 16).101 Perspective is shaped by
the consistent use, in the original poem, of simple past tense, rendered in
English by perfective verbs (iterative or non-iterative). However, there is one
significant departure, in line 2:

(7) a. He d id not know l saw -

rendered as

(7) b. Nieświadom, że go widzę

Unaware that him.Acc see.PRES
‘Unaware that I see him ’

The past participle nieświadom erases time characteristics, and the finite verb
widzę is used in the present tense. Unlike English, Polish does not require adher­
ence to the principle of consecutio temporum. The choice of present rather than
the past

(7) c. nieświadom, że go w idziałam

Unaware that h im .A c c s a w .P A S T

‘Unaware that 1 saw/had seen h im ’

brings in the “ here and now” of the actual time of the observation, instead o f the
time o f the poet’s reporting the event. In terms of Fauconnier’s theory of mental
spaces," the verb in the present tense becomes a space builder, an expression
that establishes a new mental space. Evoking the mental space of the “ here
and now” of immediate observation makes the description more vivid, but, on
the other hand, breaks the consistency of description of the English original.
Throughout the poem there are other grammatical signals shaping perspeC'
tive, notably indefinite and definite articles. Thus the poem opens with

(8) a. A Bird came down the Walk -

b. Ptak przyskakał po Ścieżce
Bird przy.jumped.PKRF along path
‘A/the bird approached jumping repeatedly along the path’

10 For the iconic sequential interpretation of and in natural language, see, e.g. Enkvist (199^'
11 For a detailed description, see Fauconnier ([1985] 1994).
 (Cognitive) grammar in translation: Form as meaning ___ 245

Later on, in line 7 of the poem,

(9) a. It hopped sidewise to the Wall.

b. Pod Mur uskoczyl -
Under wall u . ju m p e d . P E R F

‘He dodged towards the w all’

the lexical semantics of the phrasal verb came down in (8a) and o f the adverb
sidewise in (9a) establish the viewing arrangement relative to the positioning
of the viewer (the poet) and the viewed (the bird). The situatedness o f the
observer relative to the things observed shapes her epistemic perspective,
be. the knowledge that he shares - or does not share - with the reader. In
languages like English, epistemic perspective is most readily expressed
by the opposition between definite and indefinite articles. Thus, a bird pro­
files an entity new to both the observer-speaker and to the reader; the Walk
ar>d the Wall, although non-identifiable for the latter, are most probably
well known to the former. The reader - especially if he knows Dickinson’s
biography - would interpret the walk and the wall as belonging to Emily’s
own garden.
In the Polish translation the bird is just this, the bird, identifiable (or, as
bangacker would have it, mentally accessible, cf. e.g. Langacker 2008: 284);
definite reference is made through the syntactic position of the noun, cf.

b. Ptak przyskakal po Ścieżce -

Bird przy.jumped.iMPHRF along path
‘A/the bird approached jumping repeatedly along the
path’
c. Po Ścieżce przyskakal Ptak
Along path przy.jumped.iMPERF Bird
‘A bird approached jumping repeatedly along the path’

lh o construal, most probably imposed by prosody requirements, changes the

" lterPretation of the poem quite significantly: although “ unaccessible” to the
eader, the bird is known to the poet - from another occasion, or from earlier
. ta§es of the observation described. Epistemic perspective of the poem changes
lts value.
In the original poem the articles shape the ground relative to the speaker
position her within the viewing arrangement only in an indirect way; in
I 's 'vny the construal is subjective (in the sense defined in Section 2.4). Ana-
^Ztd against the objectivity scale, the poem brings the suject parlant directly
246 Elżbieta Tabakowska

onto the stage by direct pronominal reference: / saw (2), I thought (11), I offered
(14). In the translation, out of the three instances - each marking the “ non­
zoom” holistic confrontation of the bird and the observer - two (lines 2 and
14) are weakened by the grammatical convention that requires in Polish that
person is morphologically marked on the verb, w ith the personal pronoun
being verbalized only in non-neutral contexts (for semantic contrast). In the
third case - the I thought in line 11 - the whole phrase disappears. This appar­
ently tiny detail causes a significant shift in meaning: I thought is a typical
space builder, and as such it signals the passage from pre- to con-ception,
or from direct observation to conceptualization (making a metaphorical
comparison).
In the original the bird’s rapid eyes look to the poet “ like frightened beads",
giving rise to a creative simile. In the translation

(10) Czarne Paciorki Trwogi

Black beads fear.GEN
‘Black beads of fear’

are a “ ready” metaphor rather than an ad hoc mapping prompted by immediate

observation. It might be argued, however, that the fact that the Polish reader
must himself recreate the metaphor makes mental operations more complex,
whereby the description becomes “ more poetical” . The question whether this
is an advantage or a disadvantage should be answered by a translation critic
rather than a linguist.
Yet another device that may be used for establishing the point of view and
perspective is word order, conforming to what CL defines as the principle of
experiential iconicity, whereby the order in which constitutive elements of an
expression are combined reflects the order of perception (see, e.g. Tabakowska
2005). An example is offered in lines 3-4 of the poem:

(11) a. He b it an Angleworm in halves/And ate the fellow, raw

b. Wpierw rozpolowił dziobem i zjadl na
At.first (he)halved beak.iNSTR and ate.PBRP on
surowo Dżdżownicę
raw earthworm.Acc
‘At first he halved with his beak and ate up raw an earthworm’

1 he thing that the observer sees the bird bite in halves is first identified as a11
“ Angleworm", and only then she watches the bird eat the earthworm up. I *1e
image consists of two “ sub-images” , and the structure is reinforced by the tW°
 (Cognitive) grammar in translation: Form as meaning ----- 247

parts being structurally parallel: Subject-Verb-Object-Indirect Object. Splitting

the worm in halves is actually perceived; realizing that something that was
alive a second earlier must have been eaten raw, is a reflection, a mental activ­
ity. The interpretation is actually enhanced by the grammar: the adjective raw
comes at the end of the verse, between commas, as an afterthought rather
than a direct effect of visual perception. In the Polish translation the image is
somewhat different. The time adverb wpierw (line 3, ‘at first’), absent in the orig­
inal, is contrasted with potem (a lexical equivalent o f then, line 5). In conse­
quence, the actions of biting and eating are less distinctly separated in time.
In addition, the Polish expression zjadl na surowo (‘[he] ate raw’) is almost a
unit - a cognitive routine, a set phrase, processed by the reader automatically
with the semantic content being taken for granted.

4 Conclusions
With its orientation towards the text/discourse, the cognitive model of grammar
seems to substantiate the claim that TS have been making all along: verbal
expression is an interpretation rather than a reflection of things, and - since lan­
guage is indeterminate by definition - translation becomes an interpretation of
an interpretation. In this sense, the interpretation o f the original version of Dick­
inson’s 328, as proposed above, differs from that offered by her Polish translator.
11 does not mean that the translation analyzed above is a bad translation; it just
^eans that the Polish reader is offered a poem somewhat different (though not
Necessarily worse!) than that at the disposal of the English speaking public.
It might be argued that in both versions of the poem instructions provided
by the grammar prompt an interpretation whereby the poet sees a bird ap­
proaching, gets interested in its activities, some of which she watches closely,
Narrowing the visual field to selected salient details. Then there comes a holis-
dc confrontation of the bird and the woman, which finally leads the poet to a
reflection of a more general nature. But it might also be argued - which I hope I
have managed to do - that painting techniques used to depict this content make
*be two images differ in detail, and that the differences result from construal
Phenomena, mostly inherent in grammatical conventions.
I hope that the analysis justifies a couple of more general conclusions.
Ustly, a linguistic theory of translation is tantamount to a theory of language
Wrhich is selected as its underlying set of assumptions; its shape crucially de-
Pends on this selection. Secondly, out of those presently available, the theory
CG seems best suited for the purposes of TS, because it incorporates precisely
° Se aspects of language that have always frustrated language-oriented
 248 Elżbieta Tabakowska

theorists, i.e. the subjectivity of meaning, the motivated relations between

meaning and form and (some form) of language relativism. Using the CG tools
makes it possible to substantiate the claim that various grammatical devices
“ conspire” to build up the overall meaning of expressions. And thirdly, focusing
on grammar in the context of translation results in accepting a syllogism,
whereby since grammar is choice and style is choice, grammar is style. This
claim, however, goes beyond the problems of translation, and as such it also
extends the scope of this essay.

References

Baker, Mona 1992 In other words. A coursebook on translation. London: Routledge.

Banjar Yousef, Shanja n.d. Unit of translation. Available at: http://www.slideshare.net/dr.
shadiabanjar/translation-unit-by-dr-shadia-yousef-banjar (accessed August 29, 2011).
Croft, William and Alan D. Cruse 2004. Cognitive Linguistics. Cambridge: Cambridge University
Press.
Dickinson, Emily 1990 100 wierszy [100 poems]. Translated by Stanisław Barańczak. Kraków,
Poland: Arka.
Enkvist, Nils Erik 1990 Discourse comprehension, text strategies and style. Journal of the
Australasian Universities Modern Language Association 73: 166-180.
Fauconnier, Gilles [1985] 1994 Mental spaces. Cambridge: Cambridge University Press.
Fauconmer, Gilles and Mark Turner 2002 The way we think: Conceptual blending and the mind's
hidden complexities. New York: Basic Books.
Freeman, Margaret H. 1997 Grounded spaces; deictic -self anaphors in the poetry of Emily
Dickinson. Language and Literature 6: 7-28.
Halverson, Sandra 2007 A cognitive linguistic approach to translation shifts. Belgian Journal of
Linguistics 21.1: 105-121.
Janowska, Aleksandra and Magdalena Pastuchowa 2005 Slowotwórstwo czasowników
staropolskich. Kraków, Poland: Universitas.
Kenny, Dorothy 2008 Unit of translation. In: Mona Baker and Gabriela Saldanha (eds.),
Routledge encyclopedia of translation studies, 304-306. London: Routledge.
Kuno, Susumu and Etsuko Kaburaki 1977 Empathy and syntax. Linguistic Inquiry 8: 627-672-
Kussmaul, Paul 2005 Translation through visualisation. Meta 50.2: 378-391.
Lakoff, George 1987 Women, fire and dangerous things. Chicago: University of Chicago Pres*'
Lakoff, George and Mark Johnson 1980 Metaphors we live by. Chicago: University of Chicago
Press.
Langacker, Ronald W. 1987 Foundations of cognitive grammar. Vol. 1, Theoretical prerequisite*■
Stanford, CA: Stanford University Press.
Langacker, Ronald W. 1988 An overview of cognitive grammar. In: Brygida Rudzka-Ostyn
Topics in cognitive linguistics, 3-48. Amsterdam: John Benjamins.
a igacker, Ronald W. 1291 Concept, image and symbol: The cognitive basis of grammar. Betl'"'
Mouton de Gruyter.
Langacker, Ronald W. 1999 Grammar and conceptualization. Berlin: Mouton de Gruyter.
angac er, Ronald W. 2008 Cognitive grammar: A basic introduction. Oxford: Oxford Univer5'ty
 (Cognitive) grammar in translation: Form as meaning ----- 249

Langacker, Ronald W. 2009 Metonymic grammar. In: Klaus-Uwe Panther, Linda Thornburg and
Antonio Barcelona (eds.), Metonymy and metaphor in grammar, 45-71. Amsterdam: John
Benjamins.
McElhanon, Kenneth A. 2005 From word to scenario: The influence of linguistic theories upon
models of translation. Journal of Translation 1.3: 29-67.
Nida, Eugene 1964 Toward a science of translating. Leiden: Brill.
O’Brien, Sharon (ed.) 2011 Cognitive explorations of translation. London: Continuum.
Przybylska, Renata 2002 Polisemia przyimków polskich w świetle semantyki kognitywnej.
Kraków, Poland: Universitas.
Sewall, Richard B. (ed.) 1963 Emily Dickinson: A collection of critical essays. Englewood Cliffs,
NJ: Prentice Hall.
Snell-Hornby, Mary 1990 Linguistic transcoding or cultural transfer? A critique of translation
theory in Germany. In: Susan Basnett and Andre Lefevere (eds.). Translation, history and
culture, 79-86. London: Cassell.
Tabakowska, Elżbieta 1993a Articles in translation: An exercise in cognitive linguistics. In:
Richard A. Geiger and Brygida Rudzka (eds.), Conceptualizations and mental processing in
language, 785-800. Berlin: Mouton de Gruyter.
Tabakowska, Elżbieta 1993b Cognitive linguistics and poetics of translation. Tübingen,
Germany: Gunter Narr.
Tabakowska, Elżbieta 1993c Image as grammar: Some linguistic aspect of translating "The
Asian journal" of Thomas Merton. In: Elżbieta Górska (ed.), Images from the cognitive
scene, 89-100. Kraków, Poland: Universitas.
Tabakowska, Elżbieta 1996 The linguist and the poet: Nouns and verbs in a Polish translation
of Emily Dickinson’s poetry. In: Barbara Lewandowska-Tomaszczyk and Martin Thelen
(eds.), Translation and meaning, 267-282. Maastricht, the Netherlands: Opleiding Tolk-
Vertaler.
Tabakowska, Elżbieta 1997 Conditionals as an instance of figure/ground alignment. In: Angeliki
Athanasiadou and Rene Dirven (eds.). On conditionals again, 273-288. Amsterdam: John
Benjamins.
Tabakowska, Elżbieta 2000 Is (cognitive) linguistics of any use for (literary) translation? In:
Sonja Tirkkonen-Condit and Riitta Jaaskelainen (eds.), Tapping and mapping the processes
of translation and interpreting: Outlooks on empirical research, 85-95. Amsterdam: John
Benjamins.
Tabakowska, Elżbieta 2004 Point of view in the theory of literature and in linguistics: Grammar
or a “ logic of reading". In: Barbara Lewandowska-Tomaszczyk and Alina Kwiatkowska
(eds.), Imagery in language, 693-710. Frankfurt am Main: Peter Lang.
Tabakowska, Elżbieta 2005 Iconicity as a function of point of view. In: Costantino Maeder, Olga
Tischer and William Herlofsky (eds.). Outside-in - inside- out: Iconicity in language and
literature 4, 375-388. Amsterdam: John Benjamins.
almV. Leonard 1996 Fictive motion in language and - ception. In: Paul Bloom, Mary A.
Peterson, Lynn Nadel and Merrill F. Garrett (eds.), Language and space, 211-276.
Cambridge, MA: MIT Press.
almy, Leonard 2000 Toward a cognitive semantics. Vol. 1, Concept structuring systems.
T Cambridge, MA: MIT Press.
ylor, John 2002 Cognitive Grammar. Oxford: Oxford University Press.
end|er, Zeno 1967 Linguistics in philosophy. Ithaca, NY: Cornell University Press.
250 Elżbieta Tabakowska

Vinay, Jean-Paul and lean Darbelnet [19581 1995 Comparative stylistics of French and English: A
methodology for translation. Amsterdam: John Benjamins.
Wojcik-Leese, Elzbieta 2000 Salient ordering of free verse and its translation. Language and
Literature 9/2: 170-181.
http://www.informaworld.com/translationstudies (accessed |une 15, 2009).
http://www.translatum.gr/etexts/translation-theory.htm (accessed August 29, 2011).