RSS feed

Intertextuality in specialised translation: citations as semantic markers in social science1

Tanja Collet, University of Windsor


Specialised texts, social scientific texts in particular, possess by virtue of their exclusive and exclusionary nature embedded features that can enhance or, conversely, impede a reader’s ability to fully access their content, such as domain-specific lexis or terminology, and conventional textual formats. One such feature is examined in this article: manifest intertextuality or citation, i.e. the incorporation of other texts in the construction of a specialised text, a conventional textual format that is a distinguishing characteristic of specialised writing, specifically of academic writing. The article focuses on citations that are mainly engaged in meaning construal, and aims to show how these semantic markers can be beneficial for the specialised translator during the pre-translation phase of text analysis. Disciplinary ‘outsiders,’ such as specialised translators, may find it challenging to fully comprehend the content of specialised texts, especially when written by a disciplinary ‘insider’ for other ‘insiders.’ Semantic markers, however, behave in the text as ‘interactive resources’ that allow a writer to control the flow of information by responding to plausible needs for additional information on the part of an ‘imagined’ readership. A typology of semantic markers in specialised discourse, specifically social scientific discourse, and an analysis of their behaviour in the text are included in the discussion.


Specialised translation, social science translation, specialised discourse, discourse analysis, citation analysis, citations, semantic markers, term meaning, reading comprehension.

1. Introduction

In today’s globalised world, the market for specialised translation far exceeds the demand for ‘general’ and literary translation (Gotti and Šarčević 2006). Indeed, the great majority of professional translators provide specialised translation services. They work with a wide spectrum of specialised text types, from technical manuals and reports to research articles and other academic texts, in a great variety of disciplines.

This article focuses on a form of specialised translation that tends to receive insufficient attention in Translation Studies (Price 2008: 348): the translation of social scientific texts, i.e. of texts, that like their scientific and technical counterparts, are of a pragmatic nature but emanate from ‘soft’ disciplines, such as philosophy, economy, political science, sociology, linguistics, translation studies, etc. Few scholars have studied the challenges particular to this form of specialised translation. Notable exceptions, however, include Wallerstein (1981, 1996), whose foundational writings on the translation of social science texts inform much of the current thinking on the subject, from Price (2008) to the Social Science Translation Project (SSTP) undertaken by the American Council of Learned Societies (Heim and Tymowski 2006; Poncharal 2007).

At least four parameters distinguish specialised text types from their non-specialised (either ‘general’ or literary) counterparts: their purpose, subject matter, target audience, and text structure. These parameters, first formulated by Meyer and Russell (1988), also apply to social scientific texts.

Purpose: Specialised texts typically rely on language to accomplish at least two basic goals: the dissemination of data and ideas and the construction of knowledge. As such, they are central to any research-oriented enterprise, whether undertaken within academe or the private sector.

Target audience: Specialised texts tend to emanate from and target an exclusive subgroup of the general population, an in-group of members who in turn act as readers or writers, called a “discourse community” in the literature (Swales 1990: 21-32). Discourse communities are discipline-specific entities that typically set threshold levels of expertise that their members are required to meet.

Subject matter and text structure: Discourse communities devise, in addition, mechanisms of intercommunication (e.g. text genres, formats and lexis) that are exclusive to their members and that aim at furthering the communities’ communicative goals: mainly providing information about the research-related activities that are at their core.

Social scientific texts present, in addition to these four parameters, a fundamental distinguishing characteristic, summarised as follows by Price (2008: 348): “they traffic in concepts.” Indeed, social scientific texts use concepts to evoke specific views of reality that are ideologically based, i.e., that hold within a given theoretical framework. These concepts, however, are not, to quote Wallerstein (1996: 108) “a small thing.” In fact, “each of these concepts is a whole theory” or a “theory of [the concept’s] history” (Wallerstein 1996: 108), a ‘theory’ which is never completely finished but keeps evolving over time, incorporating changes and amendments made by the original author in subsequent writings or by contemporary or later authors in their works. The ‘word-symbols’ or terms that refer to these concepts can be rather idiosyncratic, i.e. specific to an author and his or her work. As Wallerstein (1996: 108) explains:

In some cases, the author literally invented the word (or phrase). In other cases, the author took a word which was already to be found in dictionaries and used it as he or she saw fit. This usage was in some instances how most other authors of the time would have used it, but sometimes the author gave the word-symbol distinctly new (even unusual) content.

Social scientific concepts are, furthermore, “more or less clearly defined” (Wallerstein 1981: 88). On the one hand, they are “shared references of meaning” (Wallerstein 1981: 88), manipulated by the members of a discourse community, but on the other, they are not necessarily “universally shared” or accepted (Wallerstein 1981: 88) but are quite often the subject of much debate, even conflict, among the members of that same community.  

The ill-defined and dynamic nature of social scientific concepts and their centrality in social science texts have led many scholars to argue that the translators of these texts should be scholars themselves or, if not, be highly familiar with the fields or subfields to which these texts belong (Wallerstein 1981; Rochlitz 2001/2; Heim and Tymowski 2006; Poncharal 2007). It is, indeed, a challenging task to fully retrieve the truly intended meaning of such texts, even for disciplinary insiders. Translators, however, have a much greater responsibility than ordinary readers: “after all, if an individual reader misreads, he suffers the consequences individually, [but when] a translator misreads, he leads innumerable others astray” (Wallerstein 1996: 116).

This article concerns the readability of social science texts for translators, who may not be scholars and consequently may not have in-group status, in spite of their extensive professional knowledge of the fields or discourse communities from which these texts emanate2. Social science texts, by virtue of their exclusive nature, possess embedded features that can enhance or, conversely, impede a reader’s ability to fully access their content (Collet 2011; Collet forthcoming). One such feature, relevant for the interpretation of social scientific concepts, will be examined in this article: manifest intertextuality (Fairclough 1992: 271) or citation, i.e. the incorporation of other texts in the construction of a specialised text, a conventional textual format that is a distinguishing characteristic of specialised writing, specifically of academic writing in the social sciences where the practice is virtually mandatory (Thompson and Tribble 2001; Hyland 2004).

Manifest intertextuality is generally thought of as a means for writers of specialised texts to mark their intellectual indebtedness to other authors (Bavelas 1978), to support their claims and pre-empt opposition (Latour 1987), and to enhance the overall persuasiveness of their arguments (Gilbert 1977). As such, citations play a role in the social construction of knowledge, in the sense that they create a map of the linkages tying the writer’s text to the works of other authors and scholars.

However, as I will show in this article, citations can also be used differently, not just as indicators of intellectual linkages but also as markers of semantic content. In this role, citations become a device for tracing semantic indebtedness by identifying similar or identical usages, or for signalling how the writer’s intended meaning for a particular term deviates from the one, sometimes more canonical, constructed by others in their texts. In other words, in social science texts, citations can act as semantic markers that trace and identify aspects of a concept’s ‘theory of history.’ The example below is a case in point. In it, the author attempts to situate two related concepts, “overt translation” and “covert translation,” by providing an historical context (the reference to the German philosopher Friedrich Schleiermacher, 1768-1834) and by indicating what distinguishes the two concepts from other similar concepts used by other authors in their works.

The distinction between an “Overt Translation” and a “Covert Translation” goes back at least to Friedrich Schleiermacher’s famous distinction between “verfremdende” and “einbürgernde” Übersetzungen, which has had many imitators using different terms. What sets the Overt-Covert distinction apart from other similar distinctions and concepts is the fact that it is integrated into a coherent theory of translation criticism, inside which the origin and function of the two types of translation are consistently described and explained. (House 2001: 245)

In his seminal writings about the challenges of social science translation, Wallerstein (1996: 108) briefly mentioned that authors may seek under certain circumstances to provide their readership with information about a concept’s ‘theory of history,’ without however identifying a specific method, such as the use of citations.

[…] when the author employs a concept, the reader is presumed to know the implicit theory of history […]. Or if the author does not think the reader knows what the concept implies, a sensible author will seek to explain it.

As for the existing body of literature on citations (see section 2), it has in general focused very little on this particular function of citations, though it is sometimes alluded to. Garfield (1977a), for instance, enumerated fifteen reasons for citation, including the identification of publications in which an idea or concept was discussed. More recently, Harwood (2009) identified eleven citation functions, ranging from supporting citations that help authors justify their research topic and methods, to credit citations that allow authors to acknowledge their sources, and position citations that allow authors to explicate their viewpoint in some detail. The latter category, according to Harwood (2009: 505), “can also be used when writers want to specify what they understand by a particular term.”

In the sections that follow, we will first examine the concept of manifest intertextuality or citation and then provide a typology or classification of citations that act as semantic markers in social science texts. The typology is based on a corpus of more than 200 citations extracted from texts published in the translation journal Meta (see section 4.1). This corpus seemed justified since scholarly works in the area of Translation Studies can themselves become the object of a translation. Recent examples include the translation into French of the Guidelines for the Translation of Social Science Texts (Heim and Tymowski 2006), and On Translator Ethics, published in 2012 by John Benjamins and translated from the original French by Heike Walker in collaboration with Anthony Pym, the author of the original, Pour une éthique du traducteur, published in 1997 by Artois Presses Université and Presses de l’Université d’Ottawa. The role of semantic markers in social science translation will be examined following the discussion of the typology described above.

2. Intertextuality in Specialised Texts

Specialised texts, such as social science texts, do not exist in isolation. They are interconnected in networks that link current texts to prior texts and that anticipate, in addition, links to texts that have yet to be written. In fact, it has been noted that texts are shaped by the ways in which they respond to prior texts, on the one hand, and attempt to predict the reactions of subsequent texts, on the other (Fairclough 1992: 270). Consequently, it can be said that social science texts are fundamentally heterogeneous or “intertextual,” i.e. “constituted by elements of other texts” (Fairclough 1992: 270), past, present, and, to some extent, future.

Because of their inherent intertextuality, social science texts possess a certain historical quality or ‘historicity,’ since each text constitutes, as it were, a link in the historical chain of text production. Kristeva (1980) argues, in this regard, that the historicity of texts is in reality complex, entailing not only that texts are links in the chronological chain of text production but also that the past, i.e. ‘history,’ is embedded in the body of texts. Fairclough (1992: 270) takes the latter to mean that texts “absorb and are built out of texts from the past” in addition to their inherent chronological tendency to “respond to, reaccentuate, [and] rework past texts” and attempt to shape future texts. He uses, as mentioned earlier, the term “manifest intertextuality” to refer to those instances of heterogeneity where the presence of “texts from the past” is “‘manifestly’ marked or cued by features on the surface of the text” (Fairclough 1992: 271).

Citation, the most common form of “manifest intertextuality,” is a major topic of research in at least three different disciplines: Information Science, Sociology of Science and Applied Linguistics.

Information scientists became interested in the phenomenon as early as the 1960s. Pioneering studies were carried out by Eugene Garfield and Norman Kaplan. Garfield (1977a and 1977b) researched citations within the context of the production of citation indices, i.e. of databases that collect statistical information by tracing citations from later documents to earlier documents. Kaplan (1965), on the other hand, was more interested in citer motivations, i.e. in explaining citation as a practice. He suggested that citation is in essence an ethical practice used as a reward system by the members of discourse communities who need to acknowledge intellectual indebtedness by giving credit where credit is due.

Kaplan’s theory became the starting point of many theories formulated by sociologists of science, intrigued by a fact, left unmentioned by Kaplan (1965), that citer motivations could also be more complex, in fact that citation could be multifunctional and serve purposes other than recognition and reward. Proponents of the social construction of knowledge, for instance, argued that citation is a collaborative ‘ritual’ whereby citing authors “participate in a discipline’s collective process of creating knowledge” (Rose 1993/1994: 28). Citation, then, can be construed as representing “a dialogue among citing authors on the ‘meaning’ of earlier texts” (Rose 1993/1994: 28). Small (1978) suggested, in this regard, that authors only cite works that “embody ideas” they are discussing and stressed the importance of viewing these citations as “interpretations” of the cited works. These “interpretations” naturally play a role in the social determination of knowledge, because, as Small (1978: 338) observed:

The interesting question is not whether the cited work is ‘correct,’ or whether the citing author has made a ‘correct’ interpretation of it, but rather whether the interpretation given is in accord or at variance with the interpretations others have given it. It is the process of acquiring a standard or conventional interpretation that is crucial for the social determination of scientific ideas.

The social-constructionist viewpoint according to which citation is a collaborative tool that members of discourse communities employ to advance their disciplinary knowledge is nowadays well accepted. However, sociologists of science have added that when one wishes to explain citation, one should not ignore the fact that discourse communities are also competitive entities with hierarchical structures whose members often “struggle for recognition and promotion” (Harwood 2004). This understanding of discourse communities has led to the claim that citation also serves another purpose: enhancing the rhetorical persuasiveness of one’s work to make it more appealing to one’s peers. The view of citation as a tool of persuasion was first proposed by Gilbert (1977).

The need to persuade and the role citations play in that regard in no way contradict the collaborative interpretation of citation but, in fact, tie in with the social-constructionist view of knowledge determination. Indeed, as Gilbert (1977: 115) put it:

A scientist who has obtained results which he believes to be true and important has to persuade the scientific community (or, more precisely, certain parts of that community) to share his opinions of the value of his work. For it is only when some degree of consensus among his colleagues has been achieved that his research findings will become transformed into scientific knowledge.

Authors, then, cite strategically to increase the likelihood of their works being cited in turn, and ultimately incorporated into their discourse community’s general store of knowledge or, to use Wallerstein’s terminology, into the shared “theories of history” of the concepts their research focuses on.

Based on the work carried out in Information Science and the Sociology of Science, applied linguists such as Harwood (2004) claim that citation can be viewed as a “multifunctional pragmatic act,” in the sense that it allows a normative, social-constructionist and rhetorical interpretation. As a “multifunctional pragmatic act,” citation acknowledges the property rights of writers (reward), creates intellectual or conceptual linkages (social construction of knowledge), and increases the overall rhetoricity of texts (persuasion).

In general, however, the focus of applied linguists has been geared more towards the ways in which writers linguistically integrate citations into their texts. John Swales, for instance, who pioneered the study of citation analysis in the mid-1980s in applied linguistics, particularly in the areas of discourse analysis and English for Academic Purposes (or EAP), has argued that citations can be classified based on their surface structure in the sentence. He has introduced the well-known and much used distinction between “integral” and “non-integral” citations (Swales 1990) (see 4.2). Studies conducted in applied linguistics have focused, furthermore, on the rhetorical value of reporting verbs (compare show, establish and suggest, propose), and on the expression of author commitment to the cited information (Tadros 1993; Thomas and Hawes 1994; Hyland 2004) (see also 4.2).

The focus, in the remainder of this article, will be on citations that acknowledge a specific type of intertextual linkage: the historical quality of social science concepts and of their “word-symbols” or terms. These citations, henceforth semantic markers, have all of the hallmarks of citations discussed above. Like all citations, they are “multifunctional pragmatic acts” that, as Kaplan (1965) argued, constitute “social devices” for discourse communities “for coping with problems of property rights and priority claims.” However, unlike citations in general, semantic markers are primarily concerned with term meaning: they are used when writers want to specify what they understand by a term, particularly when that term has already been used or defined elsewhere. Consequently, from the point of view of the text’s content, their contribution is essentially semantic in nature, concerned more with the precision of the language in which the text’s arguments are phrased, than with the identification of a theoretical framework or method, or with the rallying of supporting evidence for the claims made by the text. In fact, these citations acknowledge, among other things, that the concept of intellectual ownership can apply to the meaning and even to the form of a term in specialised language, as opposed to that of a word in everyday language. Semantic markers, then, play an important role in one of the main qualitative features of social science texts: the fact that these texts tend to engage with meanings that are defined by the texts themselves and less so by the physical context of the outside world (Hyland 2009), or that they, as Price (2008: 348) has suggested, “traffic in concepts” and “utilize [these] concepts as [their] central mode of communication” (Wallerstein 1981: 88).

As tools for meaning construction in social science texts, semantic markers should be of interest to the translators that are called upon to translate them. Indeed, translators, whether they are themselves scholars or not, generally have to analyse the texts they have been given before embarking upon their translation, “since this appears to be the only way of ensuring that the texts have been […] correctly understood” (Nord 2005: 1). An understanding of how semantic markers operate in social science discourse may be of help in this respect.

3. Citations as Semantic Markers

Semantic markers, or citations inserted by a writer who wishes to recognise “semantic” indebtedness, bear a resemblance to “knowledge-rich contexts”. The concept of “knowledge-rich context” or “KRC” was developed by Meyer (2001) within an area linked to specialised translation, namely computational terminology and specifically semi-automatic knowledge extraction. Knowledge-rich contexts are text fragments that express “conceptual information for a term” (Meyer 2001), i.e. information that can be used by a language specialist, such as a translator, to acquire domain knowledge for a term. As a consequence, they have received considerable attention in recent years, and are often included in attempts to create terminology resources aimed specifically at assisting translators in the comprehension and production of specialised texts (L’Homme and Marshman 2006; Marshman, Gariépy and Harms 2012; Marshman 2014).

Knowledge-rich contexts are generally composed of a term and a knowledge pattern (KP), i.e. a formula (such as is, means, consists of, etc.) introducing the text items that can be useful for conceptual analysis. Theytypically define (defining KRC) or explain (explanatory KRC) terms or the concepts that they designate.

Defining KRCs resemble the classical, Aristotelian definition: X = Y + differentiating characteristics. Like the latter, they are made up of a definiensthat associates differentiating characteristics to a hypernym or superordinate term (Y) in order to define the definiendum(X). Meyer (2001) provides the following example, in which the definiendum and the definiens are linked by the copula is:

Composting is a biological decay process which converts organic wastes into a crumbly, sweet smelling earth-like substance. (Meyer 2001: 285)

Explanatory KRCs, on the other hand, do not respect the form of the classical Aristotelian definition, but provide information about the X in the right-hand part of the formula without mentioning Y, i.e. the hypernym.

Compost contains nutrients, nitrogen, potassium and phosphorus. (Meyer 2001: 287)

Semantic markers can relay both types of information, defining or explanatory, but the conveyed information is always reported, either directly quoted or summarised by the citing author. The first semantic marker given as an example below conveys defining information structured in the classical Aristotelian fashion: Van Dijk and Kintsch (1983: 51) propose that {X are Y + differentiating characteristics}.

Van Dijk and Kintsch (1983: 51) propose that inferences are cohesive mechanisms pertaining to a separate level of analysis: not textual, but situational. (Pinto 2001: 296)

The second semantic marker below, however, conveys explanatory information, in that it seeks to describe X without linking it to a superordinate term, i.e. the Y of the Aristotelian formula: Saussure (1969) suggested that {X is comprised of…}.

[…] Saussure (1969) suggested that the linguistic sign is comprised of the signifier and the signified […]. (Clifford 2001: 368)

Semantic markers that define or explain are similar to Pearson’s “defining expositives:” a term introduced by Pearson (1998: 105), also in the area of computational terminology, to refer to those instances “where definitions which already exist are being repeated or rephrased for the purposes of clarification or explanation.” Pearson (1998: 118) observed that “defining expositives” may be “explicitly signalled” in the text by surface features that serve to identify the original author, such as would be the case with semantic markers, but that often they are not. The absence of surface features, however, could be indicative of the status of the definition, i.e. that it has been absorbed into the background knowledge of the field and no longer has to be linked to its originator, due to the so-called phenomenon of “obliteration through incorporation” (Garfield 1977b). In fact, many of Pearson’s examples seem to fall into that category, as can be judged from the two defining expositives below taken from her corpus.

Skeletal muscle consists of bundles of muscle fibres held together by connective tissue. (Pearson 1998: 147)
Transpiration is a process whereby plants lose water into the air through the stomata in their leaves. (Pearson 1998: 153)

Semantic markers, however, do not only define or explain term meanings. By establishing linkages to earlier works, they also provide etymological information, identify synonyms, signal alternate meanings, indicate equivalent expressions in other languages, etc. When semantic markers do the latter without defining or explaining, they resemble Pearson’s “partial defining expositives:” text segments “that may simply provide information about synonyms or the correct term to be used in a particular context” (Pearson 1998: 119). But unlike partial defining expositives, all semantic markers are citations: text segments that link the highlighted semantic content to its original source. The semantic marker below, for example, identifies a synonymous relationship between “interpénétration horizontale des langues,” a somewhat idiosyncratic expression, and “alternance de codes,” the more common term in the literature.

[…] les théoriciens se réfèrent presque toujours aux textes comprenant ce que Lewis appelle « interpénétration horizontale » des langues (Lewis 2003 : 411), synonyme inédit de l’expression plus courante « alternance de codes » (Kürtösi 1993 : 107). (Stratford 2008: 459-460)

To summarise, semantic markers are a form of manifest intertextuality in specialised discourse, particularly social scientific discourse, that are used, during the process of text creation, to identify, repeat or rephrase passages from past works that help with the negotiation of term meaning within the confines of the new text. These ‘reported’ passages or text fragments may define or explain term meaning, or may simply relate ‘partial’ semantic information pertaining to the term (etymology, synonymy, hyponymy, etc.). They can, therefore, be considered comparable, but only to an extent, to Meyer’s “knowledge-rich contexts” and Pearson’s “defining expositives.”

4. Classification of Semantic Markers
4.1. Corpus

This study is based on a corpus of more than 200 semantic markers extracted from articles published, in English or in French, in the scholarly journal Meta dedicated to the field of translation studies and published by Les Presses de l’Université de Montréal. The articles appeared over a 10-year period, from 2001 to 2011, in the following issues: 46(2) - 2001, 47(1) – 2002, 49(1) – 2004, 53(3) – 2008, 53(4) -2008, 55(3) – 2010, 56(1) – 2011, 56(2) – 2011, 56(3) – 2011, 56(4) – 2011. The issues and articles were selected at random.

The semantic markers were extracted by hand by a small team of highly dedicated research assistants enrolled in upper-level linguistics and translation courses at the University of Windsor (Ontario, Canada).

The extracted semantic markers were indexed using the following criteria: title of journal (Meta), year of publication, volume, number, and page.

4.2. Classification

The classification uses two sets of criteria. The first set is comprised of textual features that describe the surface structure of the semantic marker. They seek to capture, on the one hand, how the semantic marker is integrated into the body of the citing text and, on the other, how it structurally relates to the text fragment of the original text. The second set of criteria focuses on the content of the semantic marker, that is on the nature of the semantic information it transfers from the earlier text. Taken together, the two sets of criteria seek to illustrate how semantic markers operate in specialised discourse. The surface structures, for instance, identified by the textual features, have various rhetorical values, which the citing writers tend to manipulate to the benefit of their texts.

4.2.1. Set of textual features

The classification applies a set of 5 textual features:

  • a. Integral citation: the name of the cited author is embedded in the structure of the sentence, as a subject, or as an object of a passive sentence, for instance. Integral citations give greater prominence to the cited author rather than to the reported message.

    As Nida (1994: 157) p uts it, culture refers to “the total beliefs and practices of a society.” (Meta 2010 55(3): 570)

  • b. Non-integral citation: the name of the cited author is not incorporated into the structure of the sentence, but put in parentheses or referenced by superscript numbers or any other indirect method. Non-integral citations place greater emphasis on the reported message.

    Cultural equivalence, which means replacing a cultural term in the SL with a TL one (Newmark 1988:83). (Meta 2010 55(3): 573)

  • c. Direct Quotation: The words of the cited author are reproduced as is in the citing text. Direct quotations embed into the text the voice of the cited author, but may be introduced by reporting verbs that express stance or writer commitment to the content of what is being reported.

    Within the field of translation studies, Vermeer (1992: 38) defines culture as “the whole of norms and conventions governing social behaviour and its results.” (Meta 2010 55(3): 546)

  • Direct quotations may be of variable length. Short quotes are generally less than or equal to 10 original words quoted, and are incorporated into the citing sentence (Chang 2008). Long quotes, on the other hand, exceed 10 words, but remain embedded in the citing sentence (Chang 2008). Block quotes, finally, correspond to extensive fragments of original wording, and are often separated from the body of the text by indentation (Chang 2008).

  • d. Summary: The citing writer summarises in their own words the cited author. This represents a strategic rhetorical choice as it allows the writer greater flexibility to emphasise and interpret the comments of the cited author; or, put differently, summary allows the writer to employ or manipulate the cited fragment in such a way that it most effectively supports the argument the citing text is putting forward (Hyland 2004).

    Crucially, therefore, the recipients of any […] text need to make multiple […] links to create coherence, i.e., the general impression of a continuity of sense in a text (Beaugrande and Dressler 1981: 84). (Meta 2011 56(3): 646)

  • e. Reference: The cited author is referenced but the author’s comments are neither quoted nor summarised. Semantic markers often rely on this basic format to convey partial semantic information related, for instance, to a term’s etymology.

    The widespread use today of the term discourse community (Swales 1990) reflects a retreat from the view that the concept of culture is limited to a question of national cultures or stereotypes. (Meta 2010 55(3): 548)

4.2.2. Set of semantic features

The classification also employs a set of 8 semantic features:

  • a. Etymology: the aim of the citation is to identify the originator of a concept and/or its name.
  • b. Naming: the aim of the citation is to name a concept.
  • c. Definition: the aim of the citation is to define or explain the meaning of a term (see section 3).
  • d. Polysemy: the aim of the citation is to identify an alternate meaning for a term.
  • e. Synonymy: the aim of the citation is to identify an alternate name for a concept.
  • f. Hyponymy: the aim of the citation is to identify the hyponyms of a superordinate term, or hypernym.
  • g. Term usage: the aim of the citation is to indicate in which context a term is used.
  • h. Translation equivalent: the aim of the citation is to identify equivalents in other languages.
4.2.3. Types of semantic markers

All semantic markers possess two textual features: one describing the syntactic integration of the semantic marker, particularly of the name of the cited author, into the citing text or sentence (integral or non-integral), and the other signalling the structural overlap between the original text fragment and the citing text (direct quotation, summary or reference). All semantic markers possess in addition at least one semantic feature, though semantic features can also be clustered in groups of two or three. Examples of clusters of semantic features, encountered in our corpus, are, for instance: Etymology / Naming, Etymology / Naming / Definition, and Naming / Hyponymy / Definition.

What follows is a select list of examples of semantic markers encountered in our corpus. The labels assigned to the semantic markers first list the textual and then the semantic features. The comment section, beneath the examples, examines the interplay between the textual and semantic features of the semantic markers.

  • Integral, Reference, Etymology
[…] the initial hypothesis holds that the vast majority of the articles might either mediate the source-text ideology unchanged or fall within the category of minimal mediation. We are referring here to the term coined by Hatim and Mason (1997) to refer to those target texts in which the scale of introduced translational shifts is relatively small. (Meta 2011 56(4): 764)

The highlighted segment (in italics) corresponds to the actual semantic marker. It identifies the term, minimal mediation, and then proceeds to identify the authors who “coined” it. The emphasis is on the creators of the term and that information is foregrounded by using an integral citation that simply references the two originators of minimal mediation.

  • Integral, Reference, Naming/Etymology
The act of deliberately choosing not to observe one or more of the maxims for the purpose of communicating something Grice called flouting, and the product of this act he called implicature. (Meta 2011 56(3): 540)

This semantic marker contains two instances of naming: one for flouting, and the other for implicature. Both naming instances are presented as well-known, as can be judged from the fact that they are linked to the cited author, Grice, by way of an integral non-citation, i.e. a citation that establishes a reference but without identifying a particular work. The emphasis, here, is on the instances of naming and less so on the originator of these instances.

  • Integral, Direct Quotation (Block Quote), Definition
This description of editing goes well with Mossop’s (2001: 166) definition of the term:
Editing: The process of checking a non-translational text for error and making appropriate amendments, with special attention to making the text suitable for its readers and intended use.
(Meta 2008 53(4): 806)

Mossop’s definition of editing is reproduced verbatim as a block quote, a distinct body of text separated from the paragraphs of the main text surrounding it. When a writer attaches great importance to a definition, he may opt for a block quote. This type of direct quotation allows for extensive use of the original wording of the cited author, and attracts visual attention to the cited text fragment through the optics of indentation. Block quotes are, moreover, often introduced by integral citations, as in the semantic marker above. By syntactically incorporating the name of the cited author in the introductory sentence, integral citations contribute to the overall effect of the foregrounding of the cited author’s voice that block quotes aim to accomplish.

  • Integral, Summary, Definition
Meanwhile, Kenny (1998: 515) defines ‘sanitization’ as the suspected adaptation of a source text reality to make it more palatable for target audiences.
(Meta 2001 46(2): 350)

By summarising or rephrasing the cited author’s definition of sanitization,” the writer softens the foregrounding of the cited author accomplished by the integral citation and blends their voice with that of the cited author. It becomes difficult, then, for the reader to disentangle both voices: they become one in the text.

  • Non-integral, Direct Quotation (Long Quote), Definition
Also, rhetorical devices are “the means by which the writer makes known his vision to the reader and persuades him of its validity” (Chatham 1990: 190). (Meta 2004 49(1): 41)

Through non-integral citation, the voice of the writer overlaps with the voice of the cited author in the above semantic marker which proposes a definition for rhetorical devices. The quotation marks, however, clearly signal to the reader that the wording of the definition is that of the cited author. The rhetorical aim, here, is to place greater emphasis on the reported definition and less so on the identity of its creator.

  • Non-integral, Summary, Definition
Needless to say, ideology is taken here in a sense not limited to the political sphere; rather, ideology would seem to be that grillwork of form, convention, and belief which orders our actions (Jameson 1974: 107). (Meta 2008 53(4): 869)

The writer of the above semantic marker is delineating the meaning of ideology tosupport the arguments or claims the citing text will be making. The source of the reworded definition is acknowledged but only the voice of the writer of the citing text shines through.

  • Non-integral, Reference, Synonymy
A translation strategy (Delisle, Lee-Jahnke et al. 1999) – or translation method, as some scholars (Newmark 1988) call it – refers to a coherent plan of action adopted by translators in translating a given text. (Meta 2010 55(3): 571)

Two synonyms, translation strategy and translation method, are identified by two non-integral citations that reference the authors that employ the terms in their respective writings.

  • Integral, Direct Quotation (Long Quote and Block Quote), Polysemy/Definition
While she herself is concerned with précis-writing in a specialised sense (“a written text, of a prescribed length, that accurately summarizes a longer passage”), Russel (1988: 3) also mentions that précis-writing may be used in a broader sense:

[…] a summary of the contents of a document or series of documents, a summary of a series of events, or a summary of the proceedings of a meeting or conference.
(Meta 2008 53(4): 805)

The semantic marker contrasts two definitions of the same term, namely précis-writing. By using two different formats, an embedded long quote between brackets and a block quote, the reader’s attention is naturally drawn to the second (broader) definition provided in the block quote and foregrounded by an integral citation.

  • Non-integral, Summary, Hyponymy/Definition
Unlike soft news, which revolves around human-interest stories, hard news generally refers to those news stories that are timely, factual, important and serious on issues such as politics, economics, business and major crime (Fedler, Bender et al. 2001: 121). (Meta 2011 56(1): 120)

The meanings of two hyponyms, soft news and hard news, are compared in the above semantic marker. The passage is dominated by the citing writer’s voice. The non-integral citation, placed at the very end of the sentence, signals to the reader that the definitions, which seemingly carry the writer’s voice, are in reality rewordings of earlier definitions formulated by another author and multiple co-authors.

  • Integral, Reference, Translation Equivalent/Term Usage
[…] Croft (2003: 187) suggère de réserver le terme de projection domaniale: « domain mapping » à la métaphore […]. (Meta 2008 53(4): 755)

The semantic marker identifies the original English term, domain mapping, employed by the cited author, Croft, and juxtaposes it to the French equivalent, projection domaniale, used by the citing writer. This is, however, not the primary aim of this semantic marker, but a consequence of the creation of a French-language text that relies on English-language source material to explain how the term projection domaniale/domain mapping should be used according to Croft, the cited author identified in the integral citation at the start of the sentence.

  • Non-integral, Summary, Term Usage
Translation procedures differ from translation methods in that the former relate to whole texts while the latter are used for sentences and the smaller units of language (Newmark 1988). (Meta 2010 55(3): 571)

Finally, in this last semantic marker, the writer explains the correct usage of translation procedure and translation method. The writer’s voice dominates the passage, but as the non-integral citation, at the end of the sentence, indicates, its words have been inspired by another author.

5. Semantic Markers in Social Science Translation

The focus in this section will not be on issues of translation (how to translate semantic markers, for instance – though that would be a valid question)3, but on issues of text comprehension, that is on the readability of specialised texts, specifically social science texts, for translators, who may or may not be scholars.

Social science texts, as seen earlier in this article, are a distinctive form of specialised discourse that communicates through concepts which constitute theory or domain specific interpretations of what exists in the outside world. These concepts, however, are highly labile, their content constantly shifting due to continually on-going intellectual inquiry and debate. Much of this intellectual activity is reported on in publications, such as social science texts, where scholars trace, interpret, revise and alter existing concepts or introduce new ones. It follows, then, that every social scientific concept “is [itself] a whole theory” (Wallerstein 1996: 108).

Concepts are not truths; they are interpretations of reality. In practice, the symbol we use to represent the concept is used differently not only by different persons at the same time, but often also by the same person at different times. Concepts have histories. They not only evolve historically but they cannot be understood without reference to the total historical process. (Wallerstein 1996: 115)

The labile quality of social scientific concepts and of the terms that designate them constitutes a major challenge for any reader of social science texts, translators included. Indeed, because of it, social scientific concepts require for their interpretation a familiarity not only with the subject matter of the text that contains them but also with the literature of the field or subfield to which the text is tied and through which the concepts evolve over time. However, even when a reader possesses the required knowledge, there generally remains a fuzzy area between the truly intended meaning of the text and the reader’s ability to appreciate and fully retrieve it (Collet 2009; Collet 2011; Wallerstein 1996).

Writers of social science texts are generally aware of the major challenge to text interpretation posed by the concepts through which they communicate, and of the rather relative nature of meaning retrieval. They tend to make assumptions about the knowledge of their readership, and generally insert into their texts aids that aim to facilitate text interpretation. For instance, as observed by Wallerstein (1996: 110), writers will use “bare” concepts when they assume that their “theories of history” are known by the reader, or, conversely, provide explanations when they assume that the “theories of history” are less known.

Semantic markers can be said to represent such a textual aid. Principally engaged in meaning construal, they embed, as shown in section 4, background information about concepts and their terms directly into the text. By doing so, semantic markers can also act as beacons during the reading process, and help to guide the reader through the text, and particularly through the content of the segments in which they have been inserted. Semantic markers, then, do not only signal the citing author’s knowledge of a concept’s historical evolution (its “theory of history”), but also betray a certain awareness of the needs of potential readers, whether “disciplinary insiders” or not. In this sense, they can be considered an “interactive resource” (Thompson 2001), i.e. a resource that is available to any writer who has to manage the flow of information throughout the text under construction by constantly anticipating the likely reactions and needs of imagined readers. Indeed, as Thompson (2001: 61) puts it: “writers make assumptions about the questions that might plausibly be asked by the reader and construct the text to provide answers.” It appears that one way of providing these answers is by way of semantic markers that expand on the meaning content of key terms employed in the text or on other semantic characteristics of these terms. Consequently, a valuable reading strategy for readers would undoubtedly be to look for instances of semantic markers throughout the text, and to process these instances not just as citations, i.e. as links to the earlier literature, but also as attempts by the writer to reach out to the reader and respond to a probable or expected need for additional semantic or conceptual information.

Within the context of social science translation, it makes sense to rank semantic markers among the “intratextual factors” (Nord 2005) that need to be carefully parsed during the pre-translation analysis of a specialised text. They are “knowledge-rich contexts” that are an invaluable aid to the interpretation, the exegesis, of the source language text, and identify, in addition, sources that can be consulted if additional research about a concept’s history is deemed necessary. In short, they are highly advantageous for the hermeneutical stage in the translation of a social science text, even though, as Poncharal (2007: 103) argues, the resultant text interpretation will always remain subjective to a certain degree:

Afin de traduire un texte de sciences humaines […], il faut véritablement que le traducteur se l’approprie, c’est-à-dire, qu’il en fasse une interprétation personnelle, nécessairement subjective, et qu’il réfléchisse au meilleur moyen de restituer cette lecture dans sa propre langue, la langue cible donc.
Dans le cas des sciences humaines, il nous semble que la traduction ne peut pas être chose qu’une lecture critique – au sens noble du terme – de l’œuvre originale. Il n’y a aucun moyen de produire un texte qui serait une « réplique » à l’identique du texte original dans une autre langue.

For their retrieval during the pre-translation phase, however, translators may have to be alerted to the different ways in which semantic markers can be structurally integrated in a text. Indeed, semantic markers present various textual or structural features, such as integral versus non-integral citation, or direct quotation versus summary, as seen in section 4.

To examine the role semantic markers can play in the hermeneutical stage of a social science translation, consider the following excerpt from our corpus:

One main reason for choosing to observe effects on these four particular abilities is that we believe them to reflect the four cognitive stages of solving a translation problem. (Meta 2008 53(4): 785)

The author of the excerpt refers to a well-known phenomenon in translation, that of the translation problem. Translation problems, however, can be of many different types (lexical, syntactic, semantic, cultural, etc.) and, as a consequence, this seemingly transparent term allows for more than one interpretation. The semantic marker that immediately follows the passage is particularly helpful in this instance. It directs the reader to a source – Lörscher, W. (1991): Translation Performance, Translation Process, and Translation Strategies. A Psycholinguistic Investigation. Tübingen: Gunter Narr – that proposes a rather general definition of the term which it summarises:

Following Lörscher’s conclusions from his empirical study in 1991, we consider a ‘translation problem’ to be present in any text segment that requires conscious attention from the translator, because he/she is not able to transfer it automatically (i.e., without having to stop and think about it). (Meta 2008 53(4): 785)

It is clear that the strategic presence of this semantic marker greatly alleviates the hermeneutical challenge initially posed by the text. Indeed, it facilitates text interpretation in at least two different ways. On the one hand, it signals to the translator how translation problem is to be understood in the text to be translated, and on the other, it references the literature in which the term first received that definition, thus allowing for additional research on the concept’s historical evolution. In short, the semantic marker ensures the ‘correct’ interpretation of translation problem in the segment under consideration, while also identifying the location of additional information that could prove useful at different times in the translation process.

Semantic markers that, unlike the one just analysed, do not define, but provide other types of terminological information are also quite useful at the hermeneutical stage of a social science translation. Indeed, they generally identify aspects of a concept’s historical evolution that concern the often rather idiosyncratic nature of the concept’s word-symbol or term. Semantic markers that focus on the word-symbol’s origin (etymology and naming), on the use of other word-symbols by different authors (synonymy), or on the presence of alternate meanings or interpretations in other works (polysemy) fall into that category. Semantic markers that aim to situate the concept in relationship to others, such as those identifying subordinate or superordinate concepts (hyponymy), are yet another example. All of these semantic markers educate the reader, in this instance the translator, in the potential terminological challenges posed by the lexis of the area to which the text belongs, while at the same time referencing sources of additional information for the translator.

Although they are generally beneficial for translators, semantic markers may on occasion reduce readability by excessively blurring the voices of the writer and the cited author, or by providing information that is either ambiguous or incomplete. For instance, when writers cite – a semantic marker is a citation, – they invite readers’ lookups. Readers may or may not accept that invitation, but if they do, they expect the lookup to bear out their expectations (White 2011), i.e. they expect to be able to localise the needed information without too much effort and they expect that information to be useful.

In the example below, the writer provides etymological information by signalling that the terms domestication and xénisation are the French equivalents of the English terms domestication and foreignisation employed by another author, namely Venuti.

Quand les différences dans les normes sont qualitatives et portent par exemple […] sur des stratégies de « domestication » ou de « xénisation » (« domestication » et « foreignization » dans la terminologie de Venuti), la question est effectivement délicate. (Meta 2001 46(2): 392)

The writer mentions the cited author, Venuti, only by name and omits any reference to a specific work, usually identified by a date or year, thereby signalling that the provided etymological information should be common-knowledge among insiders and that a full reference will not be included in the reference section of the article. This citing format, called “non-citation” in the literature (Thompson and Tribble 2001), is an obvious obstacle for any reader who is not an insider and who wishes to perform a lookup to maximise his understanding of the citing text.

All in all, however, semantic markers can, because of their dialogic and interactive quality, act as beacons that facilitate text interpretation, one of the major challenges of social science translation.

6. Conclusion

Citations have multiple functions in specialised discourse. By inserting the text under construction in the historical chain of text production, they acknowledge intellectual indebtedness, partake in the social construction of knowledge, and enhance the text’s overall persuasiveness. They can furthermore help to delineate the meaning of specific terms within the confines of the new text. As semantic markers, citations acquire a dual dialogic quality: they are oriented towards the writers of the previous texts to which they refer and with whom they engage in a negotiation of term meaning, but they also ‘interact’ with the reader by responding to possible questions or to plausible needs for more precise semantic or conceptual information. In this latter role, they become invaluable for the social science translator, since they tend to positively impact the overall readability of a text that is generally written for a select readership, a group of disciplinary ‘insiders’ that may or may not include the translator.

  • Bavelas, Janet Beavin (1978). “The Social Psychology of Citations.” Canadian Psychological Review 19(2), 158-163.
  • Chang, Yu-Ying (2008). “Citation and Disciplinary Knowledge: A Comparison between Two Fields.” Concentric: Studies in Linguistics 34(2), 123-152.
  • Clifford, Andrew (2001). “Discourse Theory and Performance-Based Assessment: Two Tools for Professional Interpreting.” Meta : Translators’ Journal 46(2), 365-378.
  • Collet, Tanja (2009). “La manière de signifier du terme en discours.” Meta: Translators’ Journal 54(2), 279-294.
  • — (2011). “Semantic Hurdles to the Successful Reading of a Specialised Text.” Maria Luisa Carrio et al. (eds) (2011). Proceedings of the 10th Conference of the European Association of Languages for Specific Purposes. La investigacion y la ensenanza aplicadas a las lenguas de especialidad y a la tecnologia. Valencia: Universitat Politecnica De Valencia, 43-51.
  • — (forthcoming). “Obstacles lexico-sémantiques à la lecture réussie d’un texte de spécialité.” To appear in a special issue of TTR (Traduction, Terminologie, Rédaction) 26(2).
  • Fairclough, Norman (1992). “Intertextuality in Critical Discourse Analysis.” Linguistics and Education 4, 269-293.
  • Garfield, Eugene (1977a). “Can Citation Indexing be Automated?” Eugene Garfield (ed.) (1977). Essays of an Information Scientist. Philadelphia: ISI Press, 84-90.
  • — (1977b). “The ‘Obliteration Phenomenon’ in Science – and the Advantage of Being Obliterated!” Eugene Garfield (ed.) (1977). Essays of an Information Scientist. Philadelphia: ISI Press, 396-398.
  • Gilbert, Nigel (1977). “Referencing as Persuasion.” Social Studies of Science 7(1), 113-122.
  • Gotti, Maurizio and Susan Šarčević (eds) (2006). Insights into Specialised Translation. Bern: Peter Lang.
  • Harwood, Nigel (2004). “Citation Analysis: a Multidisciplinary Perspective on Academic Literacy.” Mike Baynham et al. (eds) (2004). Applied Linguistics at the Interface. London: BAAL and Equinox, 79-89.
  • (2009). “An Interview-based Study of the Functions of Citations in Academic Writing across Two Disciplines.” Journal of Pragmatics 41, 497-518.
  • Heim, Michael Henry and Andrzej W. Tymowski (2006). Guidelines for the Translation of Social Science Texts. American Council of Learned Societies, (consulted 18.06.2015), translated into French by Bruno Poncharal (2006) Recommandations pour la traduction de textes de sciences humaines, (consulted 10.05.2016)
  • House, Juliane (2001). “Translation Quality Assessment: Linguistic Description versus Social Evaluation.” Meta: Translators’ Journal 46(2), 243-257.
  • Hyland, Ken (2004). Disciplinary Discourses. Social Interactions in Academic Writing. Ann Arbor, MI: The University of Michigan Press.
  • — (2009). Academic Discourse. English in a Global Context. London/New York: Continuum International Publishing Group Ltd.
  • Kaplan, Norman (1965). “The Norms of Citation Behaviour: Prolegomena to the Footnote.” American Documentation 16(3), 179-184.
  • Kristeva, Julia (1980). “Word, Dialogue and Novel.” Julia Kristeva (ed.) (1980). Desire in Language. A Semiotic Approach to Literature and Art. New York: Columbia University Press, 64-91.
  • Latour, Bruno (1987). Science in Action. Cambridge, MA: Harvard University Press.
  • L’Homme, Marie-Claude and Elizabeth Marshman (2006). “Extracting Terminological Relationships from Specialized Corpora.” Lynne Bowker (ed.) (2006). Lexicography, Terminology, Translation: Text-Based Studies in Honour of Ingrid Meyer. Ottawa: University of Ottawa Press, 67-80.
  • Lörscher, Wolfgang (1991). Translation Performance, Translation Process, and Translation Strategies. A Psycholinguistic Investigation. Tübingen: Gunter Narr.
  • Marshman, Elizabeth; Julie L. Gariépy and Charissa Harms (2012). “Helping Language Professionals Relate to Terms: Terminological Relations and Termbases.” JoSTrans, The Journal of Specialised Translation 18, 30-56.
  • Marshman, Elizabeth (2014). “Enriching Terminology Resources with Knowledge-Rich Contexts.” Terminology 20(2), 225-249.
  • Meyer, Ingrid and Pamela Russell (1988). “The Role and Nature of Specialised Writing in a Translation-specific Writing Program.” TTR: traduction, terminologie, rédaction 1(2), 114-124.
  • Meyer, Ingrid (2001). “Extracting Knowledge-Rich Contexts for Terminography. A Conceptual and Methodological Framework.” In Didier Bourigault et al. (eds) (2001). Recent Advances in Computational Terminology. Amsterdam/Philadelphia: John Benjamins Publishing Company, 279-302.
  • Nord, Christiane (2005). Text Analysis in Translation. Theory, Methodology, and Didactic Application of a Model for Translation-Oriented Text Analysis.  Amsterdam: Editions Rodopi B.V.
  • Pearson, Jennifer (1998). Terms in Context. Amsterdam/Philadelphia: John Benjamins.
  • Pinto, María (2001). “Quality Factors in Documentary Translation.” Meta : Translators’ Journal 46(2), 288-300.
  • Poncharal, Bruno (2007). “Le Social Science Translation Project et la traduction des sciences humaines.” Hermès 49, 99-106.
  • Price, Joshua M. (2008). “Translating Social Science: Good versus Bad Utopianism.” Target 20(2), 348-364.
  • Pym, Anthony (1997). Pour une éthique du traducteur. Arras: Artois Presses Université and Presses de l’Université d’Ottawa, translated into English by Heike Walker (2012). On Translator Ethics. Principles for mediation between cultures.Amsterdam: John Benjamins.
  • Rochlitz, Rainer (2001/2). “Traduire les sciences humaines.” Raisons politiques 2, 65-77.
  • Rose, Shirley K. (1993/1994). “Citation Rituals in Academic Cultures.” Issues in Writing 6(1), 24-37.
  • Small, Henry G. (1978). “Cited Documents as Concept Symbols.” Social Studies of Science 8, 327-340.
  • Stratford, Madeleine (2008). “Au tour de Babel! Les défis multiples du multilinguisme.” Meta 53(3): 457-470.
  • Swales, John (1990). Genre Analysis: English in Academic and Research Settings. Cambridge, UK: Cambridge University Press.
  • Tadros, Angele (1993). “The Pragmatics of Text Averral and Attribution in Academic Texts.” Michael Hoey (ed.) (1993). Data, Description, Discourse. Papers on the English Language in Honour of John McH Sinclair on His Sixtieth Birthday. London: Harper Collins, 98-114.
  • Thomas, Sarah and Thomas Hawes (1994). “Theme and the Rhetorical Function of Citation in Research Articles.” INTERFACE. Journal of Applied Linguistics 9(1), 15-27.
  • Thompson, Geoff (2001). “Interaction in Academic Writing: Learning to Argue with the Reader.” Applied Linguistics 22(1): 58-78.
  • Thompson, Paul and Chris Tribble (2001). “Looking at Citations: Using Corpora in English for Academic Purposes.” Language Learning & Technology 5(3), 91-105.
  • Wallerstein, Immanuel (1981). “Concepts in the Social Sciences: Problems of Translation.” Marilyn Gaddis Rose (ed.) (1981). Translation Spectrum. Essays in Theory and Practice. Albany (NY): State University of New York Press, 88-98.
  • Wallerstein, Immanuel (1996). “Scholarly Concepts: Translation or Interpretation?” Marilyn Gaddis Rose (ed.) (1996). Translation Horizons Beyond the Boundaries of Translation Spectrum. Translation Perspectives IX. Binghamton (NY): State University of New York at Binghamton, 107-117.
  • White, Howard D. (2011). “Relevance Theory and Citations.” Journal of Pragmatics 43, 3345-3361.

Collet portrait

Tanja Collet holds a Ph.D. in Linguistics from the Université de Montréal (2001) and is an Associate Professor of linguistics at the University of Windsor, where she offers upper-level courses in applied linguistics and translation. Her research is focused primarily on specialised discourse and the lexicon. She adopts a discourse analytic and/or text linguistic approach to study the syntactic and semantic properties of words in text, specifically of terms in specialised discourse. Her work has appeared in Meta, Terminology, Linguistica Antverpiensia, Le langage et l’homme, La banque des mots, etc.


Note 1:
This article is based on a paper read at the seventh biennial conference of the American Translation and Interpreting Studies Association (ATISA), Where Theory and Practice Meet, held at New York University on April 3-5, 2014. A thoroughly reworked version of this paper was also presented at the international conference celebrating the 60th anniversary of the translation journal, Meta, Colloque international 60e anniversaire de META, Les horizons de la traduction : retour vers le futur, held at the Université de Montréal on August 19-21, 2015.
Return to this point in the text

Note 2:
For more information on the selection of translators for social science texts, see Heim and Tymowski (2006).
Return to this point in the text

Note 3:
The Guidelines for the Translation of Social Science Texts (Heim and Tymowski (2006)) contain a few paragraphs on technical issues such as the translation of quotes and footnotes.
Return to this point in the text