The Cognitive and Communicative Functions of Term Variation in Research Articles: A Comparative Study in Psychology and Geology

The Cognitive and Communicative Functions of Term Variation in Research Articles: A Comparative... Abstract This article reports a study about the behavior and functions of term variation in research articles (RAs) in Geology and Psychology. The aim of the study was twofold: first, to investigate the role of intra-textual term variation as a device for the representation and transfer of specialized knowledge; second, to explore whether there are disciplinary differences and whether these differences can be chalked up to different perceptions of this phenomenon among subject field experts. Two methods of data collection were combined: corpus-based analysis of 38 RAs in Spanish; and semi-structured interviews with six experts. Results show that the incidence of term variation is higher in Psychology, although both groups manifested a positive attitude toward variation. Corpus analysis and interviews confirm that term variation is used as a cognitive device, to provide information about the concept’s characteristics and relationships with other concepts; and as a communicative device to avoid repetition, to accommodate to the audience and to generalize. In sum, term variation proves to be an important device for constructing and communicating specialized knowledge for both disciplines. 1. INTRODUCTION Terms are key elements in scientific discourse because they represent specialized knowledge in texts. An adequate and consistent use of terminology is one of the most challenging requirements of specialized discourse because terms guarantee precision and accuracy, both characteristics that specialized communication should search for (Sager 1990; Kocourek 1991; Hoffmann 1998). Terminological variation, or the existence of different terms referring to the same concept, was traditionally considered as an obstacle for the effective transfer of specialized knowledge, and its use was discouraged in terminology and language for specific purposes (LSP) manuals (Wüster 1979; Rondeau 1984; Gutiérrez 2005). However, corpus-based descriptions of terminology in specialized texts have shown that term variation is a pervasive phenomenon in specialized discourse. The reasons for this phenomenon are both functional and cognitive in nature. Functionally, term variation is the result of the accommodation of language to the geographic, social, and functional sub-registers of specialized communication. Cognitively, term variation has been studied in relation to the dynamic and flexible nature of specialized knowledge. Experientialist approaches to cognition have shown that human knowledge is not a reflection of how the world really is, but of how it is experienced by a group of people that are situated in a sociotemporal context. This is supported by a growing body of neuroscientific evidence (Barsalou 2003; Gallese and Lakoff 2005) and is assumed in recent approaches to the study of specialized knowledge and terminology (Temmerman 2000; Fernández-Silva 2011; Tercedor 2011; Faber 2012). However, the idea that term variation is unnecessary and an obstacle to the precise communication of scientific thought is well entrenched in some scientific circles, especially those that are rooted in the objectivist tradition. Objectivism relies on the classificatory capacity of the mind to capture the structure of the real world and views language as a tool for the objective representation of reality (Ortony 1999; Holden and Lynch 2004). In this article, we present a study of term variation in two disciplines of the Natural and Social Sciences: Geology and Psychology. The purpose of this study was to describe the behavior and functions of term variation in research articles (RAs) of these two disciplines and explore whether these differences could be chalked up to different perceptions of this phenomenon among subject field experts. For that purpose, we carried out an analysis of intra-textual term variation in a corpus of 38 RAs published in Chilean journals in Spanish. We analyzed the incidence of term variation and the types of term variants according to the semantic change and the semantic distance with respect to the base term, to describe the type of information provided by term variation and its role in specialized knowledge construction and transfer. The corpus-illustrated analysis was complemented with semi-structured interviews to six subject field experts, with the purpose of obtaining their views about term variation and its function(s) in the discourse of their disciplines. After this introduction, the article is followed by a discussion about variation in terminological studies, and the communicative and cognitive functions of intra-textual term variation. Then follows the methodological section, which reports the corpus, the method of term variant detection, the semantic analysis, and the interviews with experts. The results section presents the findings of the incidence and behavior of term variation in RA and the attitudes and perceived functions of term variation according to experts. In the conclusion section, we highlight the main contributions of this study, its limitations, and suggest future research directions. 2. VARIATION IN TERMINOLOGICAL STUDIES The existence of terminological variation has been a subject of much debate in terminological studies because it falls within the wider topic of the relationship between reality, knowledge, and language. Different and opposing positions toward variation and its role in specialized communication have been proposed during the development of the discipline. The first and most influential theory of Terminology, the General Theory of Terminology (GTT) (Wüster 1979), was built under the objectivist paradigm and inherited the tenets of logical positivism concerning reality, knowledge, and language. For the GTT, reality can be objectively apprehended by the human mind, and specialized knowledge is clear-cut and universal. The goal of terminology is to ensure precision and unambiguous representation of specialized knowledge, by means of standardization of terms and concepts: Terminology work takes the concept as its point of departure, with the objective of establishing clear delimitations among them. Terminology views as independent the realm of concepts and denominations (=terms). (Wüster 1979: 21)1 Language is only viewed in its naming capacity and treated as ‘a necessary evil which needs to be constricted’ (Temmerman 2000: 60). In the search for a universal language and uniformity of communication, terms are treated as context-independent labels for concepts, and variation is considered an obstacle to effective communication: Ideally, the objective of term-concept assignment in a given special language is to ensure that a given term is attributed to only one concept and a given concept is represented by only one term, a condition called monosemy. This condition reduces ambiguity while homonymy and synonymy can lead to ambiguity. (ISO/704 2000: 24) The GTT remained the most influential theory of terminology until the 1990s, and its legacy is unquestionable today. The principles and methods of terminology work are still implemented in the technical terminology standardization work that is carried out by the International Organization for Standardization (ISO). However, the limitations and problems emerged as a result of implementing the GTT to other subject fields or activities where terminology management is involved, such as translation, language planning, or science popularization. For example, Riggs (1993) reflects on the contradictions between the urge for unambiguous communication and the accommodation to audiences with different levels of knowledge in the Social Sciences: Social scientists experience two contradictory requirements when writing their research reports. First, they need precise concepts that can be designated unambiguously. However, they also think of themselves as writing about human beings and their relationships to each other, matters which ought to be explained as much as possible in familiar everyday language. (Riggs 1993: 195) Alternative theories of terminology emerged in the 1990s as a reaction to the GTT, claiming the need to take into account the cognitive, linguistic, and sociocommunicative dimensions of terminology (Sager 1990; Gaudin 1993; Daille et al. 1996; Cabré 1999; Temmerman 2000). These theories share the view that variation is present in the way reality is experienced, structured, and expressed linguistically, and therefore, variation is an inherent characteristic of the term (Cabré 1999). Specialized knowledge is dynamic and flexible; it is embedded in the historical and cultural experience of specialized communities and reflects similarities and differences between social groups and individual speakers (Gaudin 1993) as well as their evolution over time (Temmerman 2000). Furthermore, terms are not only units that represent subject-specific concepts but ‘are a means of expression and communication’ (Cabré 2003: 21) when they are used in a variety of written and spoken specialized discourses. When terms are observed in their natural environment (i.e. texts), they show redundancy, conceptual, and synonymic variation, which ‘shatters the idealized view that there can or should be only one designation for a concept and vice-versa’ (Sager 1990: 58–59). From a textual perspective, it has been shown that textual factors such as level of specialization, text genre, textual function, theme, or rhetorical purpose determine the choice, treatment, and formal and conceptual variability of the term (Messineo 2002; Ciapuscio 2003; Collet 2003; Pecman 2012). In sum, term variation is the result of the dynamics of categorization and knowledge structuring (Fernández-Silva 2011; Freixa and Fernández-Silva 2017), and of the accommodation of language to the geographic, social, and functional sub-registers of specialized communication (Freixa 2002; Picton and Dury 2017). Today, term variation is widely recognized in descriptive approaches to terminological theory and is a widespread topic of research (Daille 2017; Drouin et al. 2017). Furthermore, handling term variants ‘contributes to the improvement of several terminology-oriented applications’ (Daille 2005: 194), such as specialized dictionaries (Collet 2004), terminology acquisition (Daille 2017), or automatic text indexing (Jacquemin 2001). However, the idea that term variation is contrary to the precision of scientific language still prevails in some scientific circles and LSP manuals. For example, Gutiérrez (2005: 22) suggests that ‘to ensure precision one can use always the same term for a given concept, no matter how many times it is reiterated’. Fuentes (2006) insists that bi-univocity is the only valid relationship between concepts and terms: Ideally, there should be a unique terminology system that departs from fundamental ‘axiomatic terms’, from which the rest of terms would be derived, forming a structure similar to that of Mathematics. In this terminology, bi-univocity would be the only valid relation between concepts and terms. Fuentes (2006: 241)2 3. INTRA-TEXTUAL TERM VARIATION Terminological variation within the same text cannot be related to differences in conceptualization, dialectal, or functional factors because it is produced by the same author(s) who write for the same audience(s). That is why it has been deemed a stylistic device to avoid repetition and redundancy in order to produce ‘a text that is stylistically acceptable’ (Freixa 2005: 10). From a textual perspective, term variation performs an important function in text construction because it is one of the main devices of lexical cohesion that contribute to the general texture of the text. If we take Halliday and Hasan’s (1976) typology of cohesion, term variation could be identified with reiteration, ‘a form of lexical cohesion which involves the repetition of a lexical item, at one end of the scale; the use of a general word to refer back to a lexical item, at the other end of the scale; and a number of things in between—the use of a synonym, near-synonym, or superordinate’ (Halliday and Hasan 1976: 278). Term variation is seen as a process of reiteration whereby the writer uses different terminological units to express the same concept (Kerremans 2017). As Collet (2003: 4) states, the base term and its variants are equally able to refer to the same concept within the boundaries of the text ‘in light of this potential for co-referentiality’. From a cognitive–semantic perspective, term variation can be functional in shaping understanding and transferring knowledge because the coexistence of term variants allows focusing on different aspects of the concept in different contexts. Several authors have suggested this idea. Temmerman (2000: 227) states that ‘synonymy and polysemy appear to be functional in the process of progress of understanding’; Fernández-Silva (2016) observed how intra-textual term variation allows the author to present different but complementary views of the same concept within the same text by selecting variants with different semantic information; and Pecman (2014: 10) suggested that: The terminological variation could undoubtedly be explained in relation to cognition and the process of knowledge construction in progress: a speaker, a scientist, would thus resort to terminological variation as a device for transferring his or her experience into language, and consequently into knowledge. 4. METHODOLOGY This study was conducted within the framework of the VARTERM research project, which aims to investigate the role of term variation in academic discourse across four disciplines of Natural and Social Sciences and Humanities.3 For this particular study, a two-step methodology was designed which combined corpus-based analysis of term variation in Geology and Psychology RAs and semi-structured interviews with field experts. 4.1 Corpus-based analysis of term variation in RAs The study is based on a small corpus of 38 RAs written in Spanish (358,425 words). These were published in three Chilean peer-reviewed journals indexed in the open-access digital scientific library Scielo Chile4: Andean Geology (Chilean National Service of Geological Sciences and Mining Engineering), Terapia Psicológica (Chilean Society of Clinical Psychology), and Psykhe (Catholic University of Chile). These are the only journals available in Scielo Chile in both fields that publish articles in Spanish, and all three cover a wide range of topics. Andean Geology publishes research on general topics of broad interest concerning the geology of South and Central America and Antarctica, and particularly of the Andes; Psyche covers all topics of scientific Psychology, and Terapia Psicológica focuses on all topics and subareas of Clinical Psychology. Articles were selected at random from issues published between 2004 and 2014. 4.2 Detection of term variants This study adopted an onomasiological and contextual approach to term variation (Geeraerts et al. 1994; Hamon and Nazarenko 2001). The first lexical occurrence of a given concept was considered the base term, and all subsequent lexical instantiations of the same concept were treated as term variants. Under this approach, terminological units with varying degrees of fixedness were included, and conceptual equivalence among them was considered within the boundaries of the text. Table 1 shows the terminological variants of one concept in our corpus. Table 1: Detection of term variants Status  Term  Context  Base term  vínculo inseguro  [abstract] Estudios recientes han sugerido que el TDAH podría darse en el contexto de un vínculo inseguro. [Recent studies suggest that ADHD can arise in the context of an insecure bond]  Term variant [validated]  relación de apego inseguro  La investigación de las interacciones actuales entre padres e hijos [intro] con TDAH revelan patrones similares a los observados en el contexto de relaciones de apego inseguro. [Research into relationships between parents and ADHD children reveal patterns similar to those observed in insecure attachment relationships]  Term variant [rejected]  patrón vincular inseguro  [intro] Los autores sugieren una asociación entre TDAH y vínculo inseguro. El patrón vincular inseguro en este grupo se caracterizaría por una alta expresividad emocional. [Authors suggest an association between ADHD and insecure bond. The insecure bond pattern is characterized by a high emotional expressiveness]  Term variant [validated]  apego inseguro  [results] La distribución del tipo de apego inseguro en el grupo de niños con TDAH es la siguiente: [The distribution of types of insecure attachment in ADHD children is shown below]  Status  Term  Context  Base term  vínculo inseguro  [abstract] Estudios recientes han sugerido que el TDAH podría darse en el contexto de un vínculo inseguro. [Recent studies suggest that ADHD can arise in the context of an insecure bond]  Term variant [validated]  relación de apego inseguro  La investigación de las interacciones actuales entre padres e hijos [intro] con TDAH revelan patrones similares a los observados en el contexto de relaciones de apego inseguro. [Research into relationships between parents and ADHD children reveal patterns similar to those observed in insecure attachment relationships]  Term variant [rejected]  patrón vincular inseguro  [intro] Los autores sugieren una asociación entre TDAH y vínculo inseguro. El patrón vincular inseguro en este grupo se caracterizaría por una alta expresividad emocional. [Authors suggest an association between ADHD and insecure bond. The insecure bond pattern is characterized by a high emotional expressiveness]  Term variant [validated]  apego inseguro  [results] La distribución del tipo de apego inseguro en el grupo de niños con TDAH es la siguiente: [The distribution of types of insecure attachment in ADHD children is shown below]  Table 1: Detection of term variants Status  Term  Context  Base term  vínculo inseguro  [abstract] Estudios recientes han sugerido que el TDAH podría darse en el contexto de un vínculo inseguro. [Recent studies suggest that ADHD can arise in the context of an insecure bond]  Term variant [validated]  relación de apego inseguro  La investigación de las interacciones actuales entre padres e hijos [intro] con TDAH revelan patrones similares a los observados en el contexto de relaciones de apego inseguro. [Research into relationships between parents and ADHD children reveal patterns similar to those observed in insecure attachment relationships]  Term variant [rejected]  patrón vincular inseguro  [intro] Los autores sugieren una asociación entre TDAH y vínculo inseguro. El patrón vincular inseguro en este grupo se caracterizaría por una alta expresividad emocional. [Authors suggest an association between ADHD and insecure bond. The insecure bond pattern is characterized by a high emotional expressiveness]  Term variant [validated]  apego inseguro  [results] La distribución del tipo de apego inseguro en el grupo de niños con TDAH es la siguiente: [The distribution of types of insecure attachment in ADHD children is shown below]  Status  Term  Context  Base term  vínculo inseguro  [abstract] Estudios recientes han sugerido que el TDAH podría darse en el contexto de un vínculo inseguro. [Recent studies suggest that ADHD can arise in the context of an insecure bond]  Term variant [validated]  relación de apego inseguro  La investigación de las interacciones actuales entre padres e hijos [intro] con TDAH revelan patrones similares a los observados en el contexto de relaciones de apego inseguro. [Research into relationships between parents and ADHD children reveal patterns similar to those observed in insecure attachment relationships]  Term variant [rejected]  patrón vincular inseguro  [intro] Los autores sugieren una asociación entre TDAH y vínculo inseguro. El patrón vincular inseguro en este grupo se caracterizaría por una alta expresividad emocional. [Authors suggest an association between ADHD and insecure bond. The insecure bond pattern is characterized by a high emotional expressiveness]  Term variant [validated]  apego inseguro  [results] La distribución del tipo de apego inseguro en el grupo de niños con TDAH es la siguiente: [The distribution of types of insecure attachment in ADHD children is shown below]  The terminological analysis focused on the central concepts of each article, that is those appearing in the title, abstract, and keywords sections. This assumption is supported by extensive research on RA agreeing that these sections summarize the main content and state the central concepts of the article (Swales 1990; Hartley and Kostoff 2003). A combination of manual and semi-automatic searches with concordance software5 was used to identify term variants in texts (Fernández-Silva 2011). Manual identification involved reading the whole text and highlighting the term variant candidates based on formal and/or semantic similarity and co-referential analysis—identification of reformulation procedures (Ciapuscio 2003; Kerremans 2017). Semi-automatic searching was based on discourse markers of synonymy relation—such as known as, also called (Suárez 2004)—and search based on constituent elements of term variants (Hamon and Nazarenko 2001). For example, the base term uso de drogas [drug use], was split into constituent elements and the occurrences of each element—uso and drogas—were used to search for other variants in the corpus. The search based on the head element returned the variants uso de sustancias [substance use] and uso de tabaco, alcohol y drogas ilícitas [tobacco, alcohol and illicit drug use]; a search based on the modifier returned consumo de drogas [drug consumption]. Then, contexts of occurrence of term variants were automatically retrieved and manually validated by members of the research team (n = 5). Finally, all terms and equivalence relations were validated by field experts (n = 42). Expert validation is crucial for confirming conceptual equivalence between terms, since it requires expert field knowledge to ultimately determine whether two expressions refer to the same concept. Each expert was given a document with all the variants found in a text grouped by concepts (on average each Geology RA had 15.42 concepts per text and 8.36 in Psychology). They were asked to confirm the equivalence relation by examining the term variants embedded in their contexts of occurrence (see Table 1). The experts’ answers were analyzed again by the research team and, in case of disagreement between both opinions, the text was handed out to a second validator—that was the case with three texts in Geology and two in Psychology. Based on this procedure, 452 concepts, 1,766 term variants, and 6,566 occurrences were identified (see Table 2). Table 2: Raw frequency of concepts, term variants and contexts by discipline   Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1,766  Contexts  3,253  3,313  6,566    Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1,766  Contexts  3,253  3,313  6,566  Table 2: Raw frequency of concepts, term variants and contexts by discipline   Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1,766  Contexts  3,253  3,313  6,566    Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1,766  Contexts  3,253  3,313  6,566  4.3 Semantic classification of term variants Term variants were classified according to the type and degree of semantic change from the base term to assess a possible cognitive function of term variation. This involved comparing the conceptual information displayed by terms and variants (Kageura 2002; Fernández-Silva et al. 2011). Simple and complex terms were interpreted as onomasiological structures, that is combinations of concepts within the conceptual structure. The head element indicates the general category the concept belongs to (or hypernym), whereas the modifier—whether simple or complex—states the defining feature(s), that is the feature(s) that differentiate(s) the concept from its co-hyponyms. For example, the term theropod footprint is analyzed as a concept belonging to the general category of footprints—as expressed in the head element—whose defining feature is that it belongs to a theropod dinosaur—as stated in the modifier—. For single-word terms, the conceptual information was either deduced from the morphological structure—for example, hypsometry is the measure (general concept) of altitude (defining feature) and erosion is the act (general concept) of eroding (defining feature). Based on this analysis, eight variation types were identified (Fernández-Silva 2016). Then, variation types were ranked along a semantic distance gradient and grouped into three categories (see Table 3): Table 3: Classification of term variation according to semantic distance Semantic distance  Variation type  Example (base terms; term variants)  Maximum semantic distance  Change of conceptual configuration  paternidad tradicional/modelo hegemónico de paternidad [traditional paternity/hegemonic fatherhood model]  Hypernymic/hyponymic selection  embarazadas primigestas/población [primigravidae/population]; relaciones de amistad/díadas de amistad [friendship relations/dyadic friendship]  Change of category  domo/estructura dómica [dome/dome structure]  Medium semantic distance  Change of defining feature  grupo clínico/grupo TDAH [clinical group/ADHD group]  Change of non-defining feature  huella de terópodo/huella de dinosaurio terópodo [Theropod (dinosaur) footprint]  Reduction  nanofósiles calcáreos/nanofósiles [calcareous nannofossils/nannofossils]  Minimum semantic distance  Synonymy  estudio/investigación [study/research]  Formal variation  sistema estuárico distal/DES [distal estuarine system]  Semantic distance  Variation type  Example (base terms; term variants)  Maximum semantic distance  Change of conceptual configuration  paternidad tradicional/modelo hegemónico de paternidad [traditional paternity/hegemonic fatherhood model]  Hypernymic/hyponymic selection  embarazadas primigestas/población [primigravidae/population]; relaciones de amistad/díadas de amistad [friendship relations/dyadic friendship]  Change of category  domo/estructura dómica [dome/dome structure]  Medium semantic distance  Change of defining feature  grupo clínico/grupo TDAH [clinical group/ADHD group]  Change of non-defining feature  huella de terópodo/huella de dinosaurio terópodo [Theropod (dinosaur) footprint]  Reduction  nanofósiles calcáreos/nanofósiles [calcareous nannofossils/nannofossils]  Minimum semantic distance  Synonymy  estudio/investigación [study/research]  Formal variation  sistema estuárico distal/DES [distal estuarine system]  Table 3: Classification of term variation according to semantic distance Semantic distance  Variation type  Example (base terms; term variants)  Maximum semantic distance  Change of conceptual configuration  paternidad tradicional/modelo hegemónico de paternidad [traditional paternity/hegemonic fatherhood model]  Hypernymic/hyponymic selection  embarazadas primigestas/población [primigravidae/population]; relaciones de amistad/díadas de amistad [friendship relations/dyadic friendship]  Change of category  domo/estructura dómica [dome/dome structure]  Medium semantic distance  Change of defining feature  grupo clínico/grupo TDAH [clinical group/ADHD group]  Change of non-defining feature  huella de terópodo/huella de dinosaurio terópodo [Theropod (dinosaur) footprint]  Reduction  nanofósiles calcáreos/nanofósiles [calcareous nannofossils/nannofossils]  Minimum semantic distance  Synonymy  estudio/investigación [study/research]  Formal variation  sistema estuárico distal/DES [distal estuarine system]  Semantic distance  Variation type  Example (base terms; term variants)  Maximum semantic distance  Change of conceptual configuration  paternidad tradicional/modelo hegemónico de paternidad [traditional paternity/hegemonic fatherhood model]  Hypernymic/hyponymic selection  embarazadas primigestas/población [primigravidae/population]; relaciones de amistad/díadas de amistad [friendship relations/dyadic friendship]  Change of category  domo/estructura dómica [dome/dome structure]  Medium semantic distance  Change of defining feature  grupo clínico/grupo TDAH [clinical group/ADHD group]  Change of non-defining feature  huella de terópodo/huella de dinosaurio terópodo [Theropod (dinosaur) footprint]  Reduction  nanofósiles calcáreos/nanofósiles [calcareous nannofossils/nannofossils]  Minimum semantic distance  Synonymy  estudio/investigación [study/research]  Formal variation  sistema estuárico distal/DES [distal estuarine system]  Minimum semantic distance: Variants that do not involve changes in the conceptual content displayed by terms (i.e. semantically equivalent variants) fall into this category. Formal variation can involve graphical changes (sistema estuárico distal/DES [distal estuarine system]), morphological changes (vínculo inseguro/vinculación insegura [insecure attachment]), and morphosyntactic changes (programa de prevención/programa preventivo [prevention/preventive program]). Synonymy—replacing one lexical item with another with a similar meaning in the language system (Freixa 2002: 90)—entails lexical changes to the whole term (estudio/investigación [study/research]) or constituent elements of multiword terms (vínculo inseguro/apego inseguro [insecure attachment/bond]). Medium semantic distance: Variants involving changes in the modifier—where the concept’s defining feature is expressed—were ranked in the category of medium semantic distance. In reductions, the modifier is suppressed and the resulting term variant stands as a hypernym (nanofósiles calcáreos/nanofósiles [(calcareous) nannofossils]); in the next variation type, with greater semantic distance than the last, a non-defining feature is added to (huella de terópodo/huella de dinosaurio terópodo [Theropod (dinosaur) footprint]) or removed (coeficiente de correlación de Pearson/coeficiente de Pearson [Pearson (correlation) coefficient]) from the original head–modifier structure. In the last category, a different defining feature is selected. For example, the defining feature of grupo clínico [clinical group] is the role the group of individuals plays in the study, whereas the defining feature in grupo TDAH [ADHD group] is their pathology, that is attention-deficit/hyperactivity disorder. Maximum semantic distance: Variants under this category involve changes in the term’s head or in both head and modifier. These variations were considered to be cognitively more complex, since they change the concept’s conceptual category. Within this group, four categories with increasing semantic distance were identified. The first variation type involves a change of category. This change in head selection does not place the concept in a different category but rather emphasizes a given perspective of conceptualization. For example, the variant estructura dómica [dome structure] adds the word structure to the base term domo [dome] to emphasize this perspective of conceptualization. The next variation type selects a hypernymic category (embarazadas primigestas/población [primigravidae/population]) or a hyponymic category (relaciones de amistad/díadas de amistad [friendship relations/dyadic friendship]). Finally, the term variants marked by the highest degree of semantic distance entail a change of conceptual configuration: both the main head and modifier represent taxonomically unrelated categories, such as paternidad tradicional [traditional paternity] and modelo hegemónico de paternidad [hegemonic fatherhood model]. 4.4 Semi-structured interviews with field experts The second method for data collection involved semi-structured interviews with field experts. The aim of this interview was to obtain information about experts’ attitudes toward term variation and to dig deep into the functions of term variation in RAs according to their opinion. A three-part questionnaire was designed for the interview (see Supplementary Appendix 1), which combined direct and indirect methods for measuring language attitudes that are common in language attitudes research (Garrett 2010). The interview was structured as follows: Part 1. Experts were asked to read two excerpts from an RA from the corpus (300–500 words) containing a set of terminological variants for a given concept in bold letters. Then, they had to answer a set of open questions aimed at obtaining their opinion about term variation (e.g. in your opinion the quality of the text would have improved if the author had used fewer terms to refer to the same concept?), or the reasons for term variation (e.g. Why did the author use these variants to refer to the same concept?). They had to answer the questions twice, once for each text excerpt. Part 2. Experts had to read two excerpts of an RA from the corpus authored by them with a set of terminological variants for a given concept in bold letters. As in the previous part, they had to answer a series of questions aimed at eliciting their opinion on their use of term variation (e.g. If you could rewrite the text, would you use less different terms for the same concept and why?) and the reasons for choosing specific variants (e.g. Why did you use ‘early parenthood’ instead of repeating ‘adolescent parenthood’?). Again, they had to repeat the same exercise twice, one for each excerpt. Part 3. Experts were given 12 statements expressing different opinions on term variation and had to put a checkmark against those statements they agreed with (e.g. using different term variants for the same concept helps me to better explain its content). The questionnaire combines direct and indirect elicitation of attitudes. In the third part participants are asked to articulate explicitly what their attitudes are toward term variation; in Parts 1 and 2, on the contrary, they are not aware that they are doing an attitude rating task. The use of both methods allows obtaining information about private, unconscious attitudes (covert attitudes) and those that are conscious and overtly declared by respondents (overt attitudes). This helps reduce the social desirability bias—‘the tendency for people to give answers to questions in ways that they believe to be socially appropriate’ (Garrett 2010: 45)—and the acquiescence bias—the tendency to agree with an item as a way of gaining the researcher’s approval. Both are pervasive phenomena in language attitudes research, whose incidence is more significant in face-to-face interviews (Oppenheim 2000; Gass and Seiter 2003). The interview was administered to six experts (three psychologists and three geologists) who were single or first authors of RA from the corpus, and are affiliated to Academic Institutions in Chile. The experts who answered the questionnaire were different from those who validated the variants, to prevent circularity. The informed consent form was read and signed before the interview began. The interviews lasted between 45 and 75 min. In the first and second parts, respondents were asked the questions orally by the interviewers, who wrote the answers and the various comments made by the respondents on the questionnaires. The third questionnaire was answered directly by the participant. The analysis proceeded by identifying the main views on term variation that emerged from the interviews. For instance, comments such as ‘this term expands the concept of virtual reality’, and ‘to emphasize that debris flow is a dynamic process’ were grouped into a function category called ‘explaining the concept’. 5. INCIDENCE AND BEHAVIOR OF TERM VARIATION IN GEOLOGY AND PSYCHOLOGY RAs: CORPUS FINDINGS To compare the incidence of term variation in Psychology and Geology RA, two indexes were calculated: Terminological variability index: It is obtained by dividing the number of term variants per the number of concepts. It indicates the average of term variants each concept has. The higher the terminological variability index, the greater the variability of the corpus. Type/token ratio: It is calculated by dividing the number of term variants per the number of occurrences of these variants, and expressed in percentages. The more term variants (types) there are in comparison to the number of occurrences (tokens), the greater is the terminological variability in the corpus. As can be seen in Table 4, the terminological variability index is higher in Psychology RA, with 4.84 variants per concept. However, the type/token ratio is higher in Geology, with 30.62 per cent. This means that in the Psychology corpus there are more term variants per concept, but each variant is repeated more times than in Geology. This could be related to the fact that the terminological density in Psychology RA is lower—8 concepts per abstract in average, as opposed to 15 concepts per abstract in Geology. This might explain why specialized concepts in Psychology are reiterated more frequently throughout the text. Table 4: Raw frequency of concepts, term variants and contexts and terminological variability measures in Geology and Psychology RA   Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1766  Contexts  3,253  3,313  6,566  Terminological variability index  3.40  4.84  3.91  Type/token ratio (per cent)  30.62  23.24  26.90    Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1766  Contexts  3,253  3,313  6,566  Terminological variability index  3.40  4.84  3.91  Type/token ratio (per cent)  30.62  23.24  26.90  Table 4: Raw frequency of concepts, term variants and contexts and terminological variability measures in Geology and Psychology RA   Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1766  Contexts  3,253  3,313  6,566  Terminological variability index  3.40  4.84  3.91  Type/token ratio (per cent)  30.62  23.24  26.90    Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1766  Contexts  3,253  3,313  6,566  Terminological variability index  3.40  4.84  3.91  Type/token ratio (per cent)  30.62  23.24  26.90  As a next step, the incidence of semantic variation types was examined in the two corpora. Figure 1 shows the percentage frequency of occurrence of each variation type in texts along with the percentage frequency of occurrence of the base term: Figure 1: View largeDownload slide Distribution of variation types in Psychology and Geology Figure 1: View largeDownload slide Distribution of variation types in Psychology and Geology This analysis shows that the incidence of term variation is higher in Psychology than in Geology: in Psychology RA, the authors used the base term consistently in 36.7 per cent of cases, whereas in Geology repetition of the base term amounts to 47.2 per cent. These differences proved statistically highly significant (P < p-value = 0.00 using Pearson’s chi-square test). Our results seem to indicate that the naming function of terminology, which aims at quickly and efficiently identify the concept by its denomination (Sager 1990), is more frequent in Geology than in Psychology, where the repetition of the base term only occurs one-third of the time. However, in none of the cases it can be concluded that the univocity principle prevails because term variation accounts for more than 50 per cent of terminological occurrences in Geology and more than 60 per cent in Psychology RA. Variants with minimum semantic distance have a small impact in both corpora. They account for 7.22 per cent in Geology and 10.42 per cent in Psychology. The function of these variants is merely stylistic because they do not provide additional information to the reader about the concept. For example, replacing a multiword term (sistema estuárico distal [distal estuarine system] or representaciones sociales [social representations]) by its acronym (DES and RS) is a frequent strategy in academic texts that favors verbal economy and avoids the repetitive use of long, cumbersome multiword terms. Formal variants involve graphical changes but also morphological changes—for example, suffix changes such as in vínculo inseguro/vinculación insegura [insecure attachment])—or morphosyntactic changes, mostly concerned with the alternation between the syntagmatic structure N P N and N A, like in conjunto de sierras/conjunto serrano [system of mountain ranges]. This variation has no cognitive consequences because the way the concept is represented does not change; they can have a stylistic function of avoiding repetition or respond to the author’s expressive needs or carelessness in the consistent use of terminology. Synonyms have little impact in the corpus, both in Geology (3.1 per cent) and Psychology (4.3 per cent). They can involve replacement of words having the same meaning in general Spanish language, such as estudio/investigación [study/research]; in other cases, more frequently in Geology, synonymy is the result of the cohabitation of different term formation processes, such as vernacular derivation or compounding, English borrowing (antepaís fracturado/broken foreland), or classical compounding with Greco-Latin root words (mineral arcilloso/argilomineral [clay mineral]). The potential benefits of using these synonyms do not lie in their cognitive value because they have equivalent meanings; their co-occurrence inside a text seems more a matter of stylistic choice or the cohabitation of different practices of term formation in the discipline. Variation types with medium semantic distance, that is involving lexical changes in the term’s modifier, do have cognitive consequences in the way the concept is presented. The term’s modifier reflects the concept’s defining feature(s), so modification of its constituent elements involves adding, deleting, or changing information about the concept’s content. Reductions are the most frequent type of term variation in Geology texts (15.9 per cent) and the second more frequent in Psychology (14.1 per cent), although the differences between both are not statistically significant (p-value = 0.1185). Reduced variants are, conceptually speaking, hypernyms (conglomerado sintectónico/conglomerados [(syntectonic) conglomerates]. They provide intercategorial information about the concept, that is to say, about its position within the conceptual structure (Temmerman 2000). Besides this cognitive function, reduction is also a textual strategy that ensures verbal economy and facilitates cohesion and coherence, both requirements for textuality (Halliday and Hasan 1976; Collet 2003). Term variants reflecting a different defining feature (relleno de la cuenca/relleno volcanosedimentario [basin filling/vocanosedimentary filling]) or selecting an additional, non-defining feature (huella de terópodo/huella de dinosaurio terópodo [theropod (dinosaur) footprint]) provide information about the intracategorial aspects of the concept—its internal characteristics—which are useful for explaining its content (Temmerman 2000). However, their frequency is low in the corpus in both disciplines (3.3 and 3.2 per cent in Geology and 3.1 and 2.1 per cent in Psychology). Finally, the maximum semantic distance category—which includes variation types involving changes in the head element or in the whole term—is the most frequent in both corpora. It accounts for 23.21 per cent of occurrences in Geology and 33.67 per cent in Psychology. However, it is also where more distributional differences are found between corpora. In Geology, variants selecting a different category in the head element of the term are the most frequent after reductions (11.34 per cent). This variation provides different perspectives of categorization of a given concept, and in our opinion, it could reflect the dynamic nature of categorization and the prototypical structure of concepts (Temmerman 2000; Faber 2012). For example, when the term lago somero [shallow lake] is replaced by the term variant sistema lacustre somero [shallow lake system], it emphasizes the complexity of this concept in the field, covering hydrological, geomorphological, physical, chemical, and biological conditions; when the term variant cuerpo lacustre somero [shallow lake body] is selected, the lake is categorized as a body of water. As can be seen, these three categorizations cross-cut the conceptualization of lake in the field. Hypernymic and hyponymic variants represent 8.7 per cent in Geology and 18.99 per cent in Psychology, where it stands as the most frequent variation type. These variants fulfill a cognitive function at the intercategorial level because they specify the relationships the concept bears with others within the domain’s conceptual structure. For example, when the term afectividad [affection] is replaced by its hypernym interacción emocional [emotional interaction], it informs about the classification of this concept in Psychology. However, it is also often the case that hypernymic variants correspond to generic terms, such as phenomenon as a variant of domestic violence, case for HIV patients, or subjects for university students. In those cases, it seems that the cognitive function of term variation is not so obvious, and one could argue that this variation functions as a textual strategy for establishing co-referential relations within the text (Halliday and Hasan 1976; Collet 2003). Finally, variants with a change in the conceptual configuration, that is variants involving a change in the term’s head and modifier, are not so common in Geology RA (3.17 per cent) but stand in the third position in the Psychology texts, with 9.03 per cent of occurrences. The equivalence of these variants with the base term is limited to the scope of the text. In a previous study (Fernández-Silva 2016), it was observed that this kind of variation fulfills a rhetorical function in Psychology RA. It appears mostly in the method and results sections of RA, and it is used to link the theoretical concepts identified in the introduction with the specific sample investigated in a given study, which is representative of the general concept. For example, in an RA on the relationship between quality of life and health care for patients with HIV/AIDS, the base term pacientes diagnosticados con VIH/Sida [patients diagnosed with HIV/AIDS] is used in the title, introduction, and conclusion sections and is replaced by the term variants sujetos del estudio [subjects of the study], participantes del estudio [participants], or pacientes de nuestro estudio [patients of our study] in the method and results sections. This shows the ‘particularizing function of term variation’ (Fernández-Silva 2016: 73), which is used to ‘making transition from the general field or context of the experiment to the specific experiment’ (Hill et al. 1982, in Swales 1990: 134). In sum, the corpus-illustrated analysis has revealed differences in the use of term variation between disciplines. The incidence of term variation is quantitatively higher in Psychology RA than in Geology, and differences in the distribution of variation types proved statistically significant for all categories with the exception of reductions and variants with changes in non-defining feature(s). Furthermore, the semantic analysis shows that variants with medium and maximum semantic distance prevail in both corpora. The use of variants with semantic changes provide additional information about the concept, either selecting different defining and non-defining feature(s), selecting a different head category, or indicating its position in the domain’s conceptual structure—when the base term is replaced by hypernyms, hyponyms, or reductions. Term variation is used to explain the concept’s content throughout the text, and therefore, it could be claimed that it fulfills a cognitive function. In addition to this, variation is also the result of accommodating technical vocabulary to the textual environment, where it establishes relations with other linguistic elements that may cause transformations in their form and structure—such as reductions—or where the representative function of terms can be overrun by other rhetorical purposes of the textual genre, such as generalizing or particularizing the specific study to the general research field. In sum, these results reveal that term variation is not only a stylistic resource for avoiding repetition but it also fulfills a cognitive and communicative function in both corpora. However, some additional questions remain that the corpus analysis cannot answer. For example, does the higher incidence of term variation in Psychology correlate to a more positive attitude toward this phenomenon in the disciplinary community, as compared to Geology? Are the cognitive and communicative functions of term variation consciously handled by the members of the disciplinary community? The results of the questionnaire will shed light on these questions, helping to better assess the role of term variation in both disciplines. 6. THE PERCEIVED FUNCTIONS OF TERM VARIATION IN SPECIALIZED COMMUNICATION: RESULTS OF THE INTERVIEWS WITH EXPERTS The content of the interviews was analyzed in relation to two aspects: experts’ attitudes toward term variation and the perceived functions of intra-textual term variation according to their opinion. Overt attitudes toward term variation were retrieved from the questionnaire (Part 3), where 10 of the 12 assessments reflected either positive or negative attitudes toward this resource. Covert language attitudes were extracted by analyzing the participants’ reflections, as they were doing Parts 1 and 2. Table 5 shows, for each participant, the number of positive and negative judgments selected in the questionnaire (overt) and formulated during the interview (covert). It is also presented in percentages to compare the proportion of positive and negative statements expressed by each participant. The right column provides the percentages for each discipline. Table 5: Positive and negative attitudes toward term variation by discipline   GEO1 (per cent)  GEO2 (per cent)  GEO3 (per cent)  PSY1 (per cent)  PSY2 (per cent)  PSY3 (per cent)  GEO (per cent)  PSY (per cent)  Positive  Overt  5 (45.4)  6 (50)  4 (57.1)  4 (33.3)  2 (11.1)  5 (22.7)  24 (80.6)  30 (58.8)  Covert  3 (27.3)  4 (33.3)  2 (28.6)  5 (41.7)  4 (22.2)  10 (45.5)  Negative  Overt  1 (9.1)  0 (0)  0 (0)  1 (8.3)  4 (22.2)  2 (9.1)  6 (19.4)  22 (41.1)  Covert  2 (18.2)  2 (16.7)  1 (14.3)  2 (16.7)  8 (44.5)  5 (22.7)    GEO1 (per cent)  GEO2 (per cent)  GEO3 (per cent)  PSY1 (per cent)  PSY2 (per cent)  PSY3 (per cent)  GEO (per cent)  PSY (per cent)  Positive  Overt  5 (45.4)  6 (50)  4 (57.1)  4 (33.3)  2 (11.1)  5 (22.7)  24 (80.6)  30 (58.8)  Covert  3 (27.3)  4 (33.3)  2 (28.6)  5 (41.7)  4 (22.2)  10 (45.5)  Negative  Overt  1 (9.1)  0 (0)  0 (0)  1 (8.3)  4 (22.2)  2 (9.1)  6 (19.4)  22 (41.1)  Covert  2 (18.2)  2 (16.7)  1 (14.3)  2 (16.7)  8 (44.5)  5 (22.7)  Table 5: Positive and negative attitudes toward term variation by discipline   GEO1 (per cent)  GEO2 (per cent)  GEO3 (per cent)  PSY1 (per cent)  PSY2 (per cent)  PSY3 (per cent)  GEO (per cent)  PSY (per cent)  Positive  Overt  5 (45.4)  6 (50)  4 (57.1)  4 (33.3)  2 (11.1)  5 (22.7)  24 (80.6)  30 (58.8)  Covert  3 (27.3)  4 (33.3)  2 (28.6)  5 (41.7)  4 (22.2)  10 (45.5)  Negative  Overt  1 (9.1)  0 (0)  0 (0)  1 (8.3)  4 (22.2)  2 (9.1)  6 (19.4)  22 (41.1)  Covert  2 (18.2)  2 (16.7)  1 (14.3)  2 (16.7)  8 (44.5)  5 (22.7)    GEO1 (per cent)  GEO2 (per cent)  GEO3 (per cent)  PSY1 (per cent)  PSY2 (per cent)  PSY3 (per cent)  GEO (per cent)  PSY (per cent)  Positive  Overt  5 (45.4)  6 (50)  4 (57.1)  4 (33.3)  2 (11.1)  5 (22.7)  24 (80.6)  30 (58.8)  Covert  3 (27.3)  4 (33.3)  2 (28.6)  5 (41.7)  4 (22.2)  10 (45.5)  Negative  Overt  1 (9.1)  0 (0)  0 (0)  1 (8.3)  4 (22.2)  2 (9.1)  6 (19.4)  22 (41.1)  Covert  2 (18.2)  2 (16.7)  1 (14.3)  2 (16.7)  8 (44.5)  5 (22.7)  As can be seen, five of six participants manifested more positive judgments toward term variation than negative ones, both overt and covert. However, the total percentages show a more positive attitude toward variation in Geology (80.6 per cent) than in Psychology (58.8 per cent). This can be explained by the deviant position of psychologist 2, who clearly manifested a negative attitude toward term variation that remained consistent throughout the interview. Even when he analyzed his own writing, he regretted having used terminological variants. Another interesting result is that there is more consistency between overt and covert attitudes among psychologists. This difference can be interpreted as the result of a higher awareness of the process of writing in this group. When they were asked to reflect on their own and others’ writing, answers showed that psychologists had clearer representations about their writing practices. On the contrary, geologist manifested fewer opinions, both positive and negative, about the use of terminology in their own or others texts. A negative view shared by geologists and psychologists is that consistency favors precision and clarity, and that term variation can lead to confusion. Among positive attitudes, there is agreement that the use of several term variants can be useful for explaining a concept and is sometimes necessary for avoiding excessive lexical repetition. However, psychologists adhere to other motivations for variation that are absent in the geologists’ accounts: the necessity to vary the choice of terms depending on the level of expertise of the target audience; and the need to use several term variants because they cannot find the term that exactly represents their ideas. When comparing the results of the corpus analysis with those of the interviews, our results do not support the hypothesis about the relationship between language attitudes and linguistic behavior, at least in quantitative terms. Geologists have shown more positive judgments to term variation than psychologists, but the incidence of term variation is lower in the corpus of Geology (52.8 per cent) than in Psychology (63.3 per cent). However, these results are not conclusive given the small sample of participants in the interviews, and should definitely be tested again with a larger sample. Having said that, an interesting observation is that term variation has a positive consideration among most experts and is seen not only as necessary but sometimes functional resource in expert communication. As a next step, the interviews were analyzed to assess what the perceived functionalities of term variation are according to experts. Table 6 shows the functions that emerged from the analysis and the frequency with which they were reported by each group. Functions have been grouped under a broad category of cognitive function, when they were related to the process of knowledge construction and transfer; or communicative function, when they were related to the operations of text construction or accommodation to the communicative situation. Table 6: Perceived functions of term variation according to experts   PSI (per cent)  GEO (per cent)  Communicative  Avoid repetition (stylistic)  27 (50.94)  13 (35.14)  Accomodate text to intended audience (functional)  4 (7.55)  1 (2.70)  Generalize and/or particularize (rhetorical)  2 (3.77)  2 (5.41)  Cognitive  Provide intracategorial information  9 (16.98)  11 (29.73)  Provide intercategorial information  4 (7.55)  4 (10.81)  Favor conceptual clarity  7 (13.21)  6 (16.22)    53 (100.00)  37 (100.00)    PSI (per cent)  GEO (per cent)  Communicative  Avoid repetition (stylistic)  27 (50.94)  13 (35.14)  Accomodate text to intended audience (functional)  4 (7.55)  1 (2.70)  Generalize and/or particularize (rhetorical)  2 (3.77)  2 (5.41)  Cognitive  Provide intracategorial information  9 (16.98)  11 (29.73)  Provide intercategorial information  4 (7.55)  4 (10.81)  Favor conceptual clarity  7 (13.21)  6 (16.22)    53 (100.00)  37 (100.00)  Table 6: Perceived functions of term variation according to experts   PSI (per cent)  GEO (per cent)  Communicative  Avoid repetition (stylistic)  27 (50.94)  13 (35.14)  Accomodate text to intended audience (functional)  4 (7.55)  1 (2.70)  Generalize and/or particularize (rhetorical)  2 (3.77)  2 (5.41)  Cognitive  Provide intracategorial information  9 (16.98)  11 (29.73)  Provide intercategorial information  4 (7.55)  4 (10.81)  Favor conceptual clarity  7 (13.21)  6 (16.22)    53 (100.00)  37 (100.00)    PSI (per cent)  GEO (per cent)  Communicative  Avoid repetition (stylistic)  27 (50.94)  13 (35.14)  Accomodate text to intended audience (functional)  4 (7.55)  1 (2.70)  Generalize and/or particularize (rhetorical)  2 (3.77)  2 (5.41)  Cognitive  Provide intracategorial information  9 (16.98)  11 (29.73)  Provide intercategorial information  4 (7.55)  4 (10.81)  Favor conceptual clarity  7 (13.21)  6 (16.22)    53 (100.00)  37 (100.00)  As can be seen, avoiding repetition stands as the most frequent function of term variation according to psychologists (50.94) and geologists (35.14), but it is notably higher in Psychology. It was a common view among participants that reiterating the same terms over and over again can be cumbersome. For example, psychologist 3 affirms that it is ‘important to be careful when one writes to avoid repetition. I hate repeating myself’; and psychologist 1 reports that ‘I always tell my students to avoid redundancy when writing a text’. Experts recognized stylistic reasons not only with formal variants (etapa de orogénesis regional vs. etapa orogénica regional [regional orogenesis/orogenic phase]) or reductions (triásico superior vs. triásico [upper Triassic vs. Triassic]); for example, geologist 3 said that he/she used holocenic glacial fluctuations instead of glacial variations ‘for aesthetic reasons, to avoid repeating the same term’. The stylistic function of term variation overshadows other communicative functions, which seem to be far less conscious to experts. For example, the rhetorical function is only reported two times by each group (3.77 per cent in Psychology and 5.41 per cent in Geology). The rhetorical function is used to generalize and/or particularize, making a link between the specific study and the general field or context of the study. For example, regarding Excerpt (1), psychologist 2 explains that he used participants (the individuals who were actually part of the study) instead of 4th–7th-grade schoolchildren (the population sector under study) ‘to situate them as participants of the study’: (1) El objetivo de esta investigación fue identificar factores predictores del inicio en el consumo de drogas lícitas en escolares de 4° a 7° básico [4th to 7th-grade schoolchildren]. Se encuestaron 234 participantes [participants] de dos comunas de Santiago, mediante la metodología de pares. Finally, term variation is also perceived as a device to accommodate the text to the intended audience. However, this communicative function appears almost three times more frequently in interviews with psychologists (7.55 per cent) than with geologists (2.70 per cent). For example, all psychologists and only one geologist agreed with the sentence ‘I don’t use the same terms when I write a text for experts and for students’ in the questionnaire. Furthermore, psychologist 2 commented upon Excerpt (2): ‘I guess that the author used all these terms to approach the text to readers. And I think he/she succeeded’. (2) el apego [bond] consiste en ‘un conjunto de pautas de conducta características […] que tienen por objetivo mantener al niño en una proximidad más o menos estrecha con su figura materna’ […] Hacia el final del primer año, esta conducta [behavior] se organiza como un sistema y se vuelve activa cada vez que se dan ciertas condiciones […] Este conjunto de conductas [set of behaviors] tiene como finalidad la protección. […] La conducta de apego [bonding behavior] se alterna con la de exploración del medio […] The other functions of term variation that showed up in the interviews were classified as cognitive. In these cases, term variation is used for better explaining the concepts, in other words, for constructing and transferring knowledge through linguistic means. Experts have appointed three cognitive functions during the interviews. First, term variation is used to provide intracategorial information, that is information about the characteristics of a given concept. This can involve using a longer and somehow redundant term variant, as can be seen in Excerpt (3). The author declared that she/he used the term variant parenthood of adolescent fathers instead of adolescent parenthood ‘to make clear that in this case it is referred to men. This concept always leads to confusion and is normally attributed to women’: (3) Los principales resultados y conclusiones señalan que la paternidad adolescente [adolescent parenthood] es un tema lleno de contradicciones; […] Como objetivos específicos, el estudio contempla […] describir los posibles obstaculizadores del ejercicio de la paternidad de los padres adolescentes [parenthood of adolescent fathers] y el significado que tienen de ellos. This function holds a relevant position among geologists and is recalled in 29.73 per cent of the cases. For example, in Excerpt (4), one geologist mentioned that he/she switched the term paleo-volcano by the term variant paleo-volcanic edifice ‘so the reader could picture the structure of the volcano’. Later, he/she used pre-collapse edifice because ‘I wanted readers to understand that I was talking about the volcanic structure before it collapsed’: (4) Se propone la estructura del paleovolcán [paleo-volcano] previa a su colapso parcial. […] El análisis de facies ha permitido reconstruir, al menos parcialmente, el flanco sur del paleoedificio volcánico [paleo-volcanic edifice]. […]Debido a la ausencia de material del edificio precolapso [pre-collapse edifice] en los cerrillos de brecha piroclástica […] se propone que los materiales que conforman estos cerrillos se encontraban hacia la parte más lejana del flanco sur del volcán. Term variation can be used as a device to provide intercategorial information. This function appears in the interviews with geologists (10.81 per cent) and psychologists (7.55 per cent). In Excerpt (5) a geologist explained that he/she used flow and avalanche flow instead of debris avalanche because ‘they are a type of flow’, and because ‘it is a dynamic process and I wanted to pass the idea that it is a process’: (5) Las avalanchas de detritos [debris avalanche], asociadas a colapsos parciales de edificios volcánicos, son fenómenos comunes en la evolución de un volcán. Este tipo de flujos [flows] son por inestabilidades, que pueden deberse a factores tales como la existencia de zonas afectadas por alteración hidrotermal […] hacia los flancos del volcán se disponían depósitos […] que fueron en parte incorporados por el flujo de avalancha [avalanche flow] durante su avance. Finally, we regrouped under the label ‘favor conceptual clarity’ different statements in which the experts reflected on term variation as a device for better explaining oneself. Five participants agreed with the statement that ‘using different words for a given concept facilitates text comprehension’. All psychologist and one geologist mentioned that ‘when writing a text I sometimes cannot find the right word to convey my idea, and I use different expressions’; and a geologist said that using different variants was more explanatory because ‘if the writer repeated himself, he could lead to misunderstandings’. In sum, interviews revealed that experts are well aware of the different functions of term variation in texts and that they can recall a variety of communicative and cognitive functions. Interestingly, several cognitive functions of term variation, which were considered as ‘unconscious’ in a previous study (Freixa 2005), were pointed out during the interviews. 7. CONCLUSION This study explored the communicative and cognitive functions of intra-textual term variation in RAs in Psychology and Geology. The corpus analysis and the interviews with experts showed that terminological variation is used to avoid repetition, to accommodate the text to the intended audience or to rhetorically move between the specific research and the general topic under investigation. Moreover, term variation is also a device for knowledge construction because it is used to provide information about the concepts and their relationships within the knowledge structure of the field. Our results seem to indicate that most experts hold a positive attitude toward term variation and are aware of the importance of this resource for the representation and transfer of specialized knowledge. These results show the interest of investigating term variation from a cognitive–functional perspective, taking into account cognitive, linguistic, and sociocommunicative factors involved in specialized language use. It also shows that bi-univocity remains an idealistic principle that is neither possible nor desirable. On the other side, we observed interesting differences between disciplines. In Geology, the incidence of term variation in RA is 30 per cent lower than in Psychology and, according to the experts’ opinions, the cognitive functions are more important in Geology than in Psychology. However, we could not find a relationship between a more positive attitude and a higher incidence of term variation. As a matter of fact, geologists manifested more positive judgments about the use of term variation than psychologists during the interviews, but it is also true that they manifested fewer opinions in general (30 opinions against 52, see Table 5). On the other side, psychologists were more critical to their own term choice and that of others; this could suggest a higher awareness of this group about the writing process and the construction of an appropriate text. In any case, the connection between attitudes toward variation and term use remains a hypothesis that should be explored more in depth with a greater sample of interviews in more disciplines. Finally, we highlight the importance of combining different methods for a cognitive study of terminology, such as corpus analysis and interviews. However, both methods still cannot answer crucial questions, such as the real impact of term variation in the transfer of specialized knowledge. As a direction for future research, it is envisaged to include experimental methodologies to assess the role of term variation in the comprehension of specialized knowledge. Only the convergence of different methodologies will allow understanding the complex relationship between cognitive factors of variation and their linguistic manifestations. Sabela Fernández-Silva holds a BA in Translation and Interpreting from University of Salamanca and a PhD degree in Applied Linguistics from Pompeu Fabra University, Barcelona, Spain. She is an assistant professor at the Pontifical Catholic University of Valparaíso, Chile, and coordinator of the Translation and Interpreting Department. She teaches courses on terminology, lexicology, lexicography, and linguistic theories to undergraduate and graduate students. Her research interests cover terminology, neology, and lexical semantics from a cognitive–functional perspective. Address for correspondence: Sabela Fernández-Silva, Av. El Bosque 1290, ILCL Building, off. 5.6. 25000, Viña del Mar, V Región, Chile. <sabela.fernandez@pucv.cl> NOTES 1 Translation is ours. 2 Translation is ours. 3 The VARTERM project (Fondecyt 11121597) was funded by the Government of Chile’s National Commission for Scientific and Technological Research (CONICYT). Further information can be found on its website: www.varterm.org. 4 Scientific Electronic Library Online (Scielo Chile): http://www.scielo.cl/ 5 Textstat: http://neon.niederlandistik.fu-berlin.de/textstat/ SUPPLEMENTARY DATA Supplementary material is available at Applied Linguistics online. Acknowledgements The author would like to thank Nelson Becerra, Camila Cifuentes, Francisco García, Paula Morgado, Joaquín Moya, Carol Valenzuela, and Grace Wilson, members of the Varterm research project, for their valuable assistance during various stages of this research. The author would also like to thank the three anonymous reviewers for their comments that contributed to improving the final version of the article. Conflict of interest statement. None declared. REFERENCES Barsalou L. 2003. ‘ Situated simulation in the human conceptual system,’ Language and Cognitive Processes  18: 513– 62. Google Scholar CrossRef Search ADS   Cabré M. T. 1999. La Terminología: Representación y Comunicación . IULA-Universitat Pompeu Fabra. Cabré M. T. 2003. ‘ Theories of terminology. Their description, prescription and explanation,’ Terminology  9: 163– 99. Google Scholar CrossRef Search ADS   Ciapuscio G. E. 2003. Textos Especializados y Terminología . IULA-Universitat Pompeu Fabra. Collet T. 2003. ‘ A two-level grammar of the reduction processes of French complex terms in discourse,’ Terminology  9: 1– 27. Google Scholar CrossRef Search ADS   Collet T. 2004. ‘ Esquisse d’une nouvelle microstructure de dictionnaire spécialisé reflétant la variation en discours du terme syntagmatique,’ Meta, Journal of Translators  49: 247– 63. Available at https://www.erudit.org/fr/revues/meta/2004-v49-n2-meta770/009349ar/ Daille B. 2005. ‘ Variations and application-oriented terminology engineering,’ Terminology  11: 181– 97. Google Scholar CrossRef Search ADS   Daille B. 2017. Term Variation in Specialised Corpora: Characterisation, Automatic Discovery and Applications . John Benjamins Publishing Company. Google Scholar CrossRef Search ADS   Daille B., Habert B., Jacquemin C., Royauté J.. 1996. ‘ Empirical observation of term variation and principles for their description,’ Terminology  3: 197– 257. Google Scholar CrossRef Search ADS   Drouin P., Francoeur A., Humbley J., Picton A.. (eds). 2017. Multiple Perspectives on Terminological Variation . John Benjamins Publishing Company. Google Scholar CrossRef Search ADS   Faber P. (ed.). 2012. A Cognitive Linguistics View of Terminology and Specialized Language . De Gruyter. Google Scholar CrossRef Search ADS   Fernández-Silva S. 2011. Variación Terminológica y Cognición: Factores Cognitivos En La Denominación Del Concepto Especializado . IULA-Universitat Pompeu Fabra. Available at http://hdl.handle.net/10803/22638 Fernández-Silva S. 2016. ‘ The cognitive and rhetorical role of term variation and its contribution to knowledge construction in research articles,’ Terminology  22: 52– 79. Google Scholar CrossRef Search ADS   Fernández-Silva S., Freixa J., Cabré M. T.. 2011. ‘ A proposed method for analysing the dynamics of cognition through term variation,’ Terminology  17: 49– 74. Google Scholar CrossRef Search ADS   Freixa J. 2002. La Variació Terminològica: Anàlisi de la Variació Denominativa en Textos de Diferent Grau d’Especialització de l’Àrea de Medi Ambient. IULA-UPF. Available at http://hdl.handle.net/10803/1677 Freixa J. 2005. ‘ Variació terminológica, ¿por qué y para qué?,’ Meta, Journal of Translators  50. Available at https://www.erudit.org/revue/meta/2005/v50/n4/019917ar.pdf Freixa J., Fernández-Silva S.. 2017. ‘Terminological variation and the unsaturability of concepts’ in Drouin P., Francoeur A., Humbley J., Picton A. (eds): Multiple Perspectives on Terminological Variation . John Benjamins Publishing Company, pp. 155– 80. Google Scholar CrossRef Search ADS   Fuentes X. 2006. ‘Contra la sinonimia y la polisemia en los lenguajes de especialidad,’ Panace@ 7: 24. Available at www.tremedica.org/panacea/IndiceGeneral/n24_entremes3-f.arderiu.pdf Gallese V., Lakoff G.. 2005. ‘ The brain’s concepts: The role of the sensory-motor system in conceptual knowledge,’ Cognitive Neuropsychology  22: 455– 79. Google Scholar CrossRef Search ADS PubMed  Garrett P. 2010. Attitudes to Language . Cambridge University Press. Google Scholar CrossRef Search ADS   Gass R., Seiter J.. 2003. Persuasion, Social Influence and Compliance Gaining , 2nd edn Allyn and Bacon. Gaudin F. 1993. Pour Une Socioterminologie: Des Problemes Sémantiques Aux Pratiques Institutionnelles . Publications de l’Université de Rouen. Geeraerts D., Grondelaers S., Bakema P.. 1994. The Structure of Lexical Variation: Meaning, Naming, and Context . De Gruyter. Google Scholar CrossRef Search ADS   Gutiérrez B. 2005. El Lenguaje De Las Ciencias . Editorial Gredos. Halliday M. A. K., Hasan R.. 1976. Cohesion in English . Longman. Hamon T., Nazarenko A.. 2001. ‘Detection of synonymy links between terms: Experiment and results’ in Bourigault D., Jacquemin C., L’Homme M. C. (eds): Recent Advances in Computational Terminology . John Benjamins Publishing Company, pp. 185– 208. Google Scholar CrossRef Search ADS   Hartley J., Kostoff R. N.. 2003. ‘ How useful are keywords in scientific journals?,’ Journal of Information Science  29: 433– 8. Google Scholar CrossRef Search ADS   Hill S. S., Soppelsa B. F., West G. K.. 1982. ‘ Teaching ESL students to read and write experimental-research papers,’ TESOL Quarterly  16: 333– 47. Google Scholar CrossRef Search ADS   Hoffmann L. 1998. Llenguatges D’Especialitat. Selecció De Textos . IULA-Universitat Pompeu Fabra. Holden M. T., Lynch P.. 2004. ‘ Choosing the appropriate methodology: Understanding research philosophy,’ The Marketing Review  4: 397– 409. Google Scholar CrossRef Search ADS   ISO/704. 2000. Principles and Methods of Terminology. International Standardization Office. Jacquemin C. 2001. Spotting and Discovering Terms through Natural Language Processing Techniques . MIT Press. Kageura K. 2002. The Dynamics of Terminology: A Descriptive Theory of Term Formation and Terminological Growth . John Benjamins Publishing Company. Google Scholar CrossRef Search ADS   Kerremans K. 2017. ‘Applying computer-assisted coreferential analysis to a study of terminological variation in multilingual parallel corpora’ in Menzel K., Lapshinova-Koltunski E., Kunz K. (eds): New Perspectives on Cohesion and Coherence: Implications for Translation . Language Science Press, pp. 49– 75. Kocourek R. 1991. La Langue Française De La Technique Et De La Science: Vers Une Linguistique De La Langue Savante . Oscar Brandstetter. Messineo C. 2002. ‘ Variación Conceptual y Formal del Término Educación Bilingüe Intercultural (ebi) en Distintos Tipos de Discursos,’ Terminology  8: 113– 19. Google Scholar CrossRef Search ADS   Oppenheim A. N. 2000. Questionnaire Design, Interviewing and Attitude Measurement , 2nd edn Bloomsbury Academic. Ortony A. (ed.). 1999. Metaphor and Thought . Cambridge University Press. Pecman M. 2012. ‘ Tentativeness in term formation: A study of neology as a rhetorical device in scientific papers,’ Terminology  18: 27– 58. Google Scholar CrossRef Search ADS   Pecman M. 2014. ‘ Variation as a cognitive device: how scientists construct knowledge through term formation,’ Terminology  20: 1– 24. Google Scholar CrossRef Search ADS   Picton A., Dury P.. 2017. ‘Diastratic variation in language for specific purposes: Observations from the analysis of two corpora’ in Drouin P., Francoeur A., Humbley J., Picton A.. (eds): Multiple Perspectives on Terminological Variation . John Benjamins Publishing Company, pp. 57– 80. Google Scholar CrossRef Search ADS   Riggs F. W. 1993. ‘Social science terminology: Basic problems and proposed solutions’ in Sonneveld H. B., Loening K. L. (eds): Terminology. Applications in Interdisciplinary Communication . John Benjamins Publishing Company, pp. 195– 221. Google Scholar CrossRef Search ADS   Rondeau G. 1984. Introduction à La Terminologie . Gaëtan Morin. Sager J. C. 1990. A Practical Course in Terminology Processing . John Benjamins Publishing Company. Google Scholar CrossRef Search ADS   Suárez M. 2004. Análisis Contrastivo De La Variación Denominativa En Textos Especializados: Del Texto Original Al Texto Meta . Universitat Pompeu Fabra. Available at http://www.tdx.cat/TDX-0217105-130025. Swales J. 1990. Genre Analysis: English in Academic and Research Settings . Cambridge University Press. Temmerman R. 2000. Towards New Ways of Terminology Description: The Sociocognitive Approach . John Benjamins Publishing Company. Google Scholar CrossRef Search ADS   Tercedor M. 2011. ‘ The cognitive dynamics of terminological variation,’ Terminology  17: 181– 97. Google Scholar CrossRef Search ADS   Wüster E. 1979. Introducción a La Teoría General De La Terminología y a La Lexicografía Terminológica . IULA-Universitat Pompeu Fabra. © Oxford University Press 2018 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Applied Linguistics Oxford University Press

The Cognitive and Communicative Functions of Term Variation in Research Articles: A Comparative Study in Psychology and Geology

Loading next page...
 
/lp/ou_press/the-cognitive-and-communicative-functions-of-term-variation-in-hfEsKG2QNL
Publisher
Oxford University Press
Copyright
© Oxford University Press 2018
ISSN
0142-6001
eISSN
1477-450X
D.O.I.
10.1093/applin/amy004
Publisher site
See Article on Publisher Site

Abstract

Abstract This article reports a study about the behavior and functions of term variation in research articles (RAs) in Geology and Psychology. The aim of the study was twofold: first, to investigate the role of intra-textual term variation as a device for the representation and transfer of specialized knowledge; second, to explore whether there are disciplinary differences and whether these differences can be chalked up to different perceptions of this phenomenon among subject field experts. Two methods of data collection were combined: corpus-based analysis of 38 RAs in Spanish; and semi-structured interviews with six experts. Results show that the incidence of term variation is higher in Psychology, although both groups manifested a positive attitude toward variation. Corpus analysis and interviews confirm that term variation is used as a cognitive device, to provide information about the concept’s characteristics and relationships with other concepts; and as a communicative device to avoid repetition, to accommodate to the audience and to generalize. In sum, term variation proves to be an important device for constructing and communicating specialized knowledge for both disciplines. 1. INTRODUCTION Terms are key elements in scientific discourse because they represent specialized knowledge in texts. An adequate and consistent use of terminology is one of the most challenging requirements of specialized discourse because terms guarantee precision and accuracy, both characteristics that specialized communication should search for (Sager 1990; Kocourek 1991; Hoffmann 1998). Terminological variation, or the existence of different terms referring to the same concept, was traditionally considered as an obstacle for the effective transfer of specialized knowledge, and its use was discouraged in terminology and language for specific purposes (LSP) manuals (Wüster 1979; Rondeau 1984; Gutiérrez 2005). However, corpus-based descriptions of terminology in specialized texts have shown that term variation is a pervasive phenomenon in specialized discourse. The reasons for this phenomenon are both functional and cognitive in nature. Functionally, term variation is the result of the accommodation of language to the geographic, social, and functional sub-registers of specialized communication. Cognitively, term variation has been studied in relation to the dynamic and flexible nature of specialized knowledge. Experientialist approaches to cognition have shown that human knowledge is not a reflection of how the world really is, but of how it is experienced by a group of people that are situated in a sociotemporal context. This is supported by a growing body of neuroscientific evidence (Barsalou 2003; Gallese and Lakoff 2005) and is assumed in recent approaches to the study of specialized knowledge and terminology (Temmerman 2000; Fernández-Silva 2011; Tercedor 2011; Faber 2012). However, the idea that term variation is unnecessary and an obstacle to the precise communication of scientific thought is well entrenched in some scientific circles, especially those that are rooted in the objectivist tradition. Objectivism relies on the classificatory capacity of the mind to capture the structure of the real world and views language as a tool for the objective representation of reality (Ortony 1999; Holden and Lynch 2004). In this article, we present a study of term variation in two disciplines of the Natural and Social Sciences: Geology and Psychology. The purpose of this study was to describe the behavior and functions of term variation in research articles (RAs) of these two disciplines and explore whether these differences could be chalked up to different perceptions of this phenomenon among subject field experts. For that purpose, we carried out an analysis of intra-textual term variation in a corpus of 38 RAs published in Chilean journals in Spanish. We analyzed the incidence of term variation and the types of term variants according to the semantic change and the semantic distance with respect to the base term, to describe the type of information provided by term variation and its role in specialized knowledge construction and transfer. The corpus-illustrated analysis was complemented with semi-structured interviews to six subject field experts, with the purpose of obtaining their views about term variation and its function(s) in the discourse of their disciplines. After this introduction, the article is followed by a discussion about variation in terminological studies, and the communicative and cognitive functions of intra-textual term variation. Then follows the methodological section, which reports the corpus, the method of term variant detection, the semantic analysis, and the interviews with experts. The results section presents the findings of the incidence and behavior of term variation in RA and the attitudes and perceived functions of term variation according to experts. In the conclusion section, we highlight the main contributions of this study, its limitations, and suggest future research directions. 2. VARIATION IN TERMINOLOGICAL STUDIES The existence of terminological variation has been a subject of much debate in terminological studies because it falls within the wider topic of the relationship between reality, knowledge, and language. Different and opposing positions toward variation and its role in specialized communication have been proposed during the development of the discipline. The first and most influential theory of Terminology, the General Theory of Terminology (GTT) (Wüster 1979), was built under the objectivist paradigm and inherited the tenets of logical positivism concerning reality, knowledge, and language. For the GTT, reality can be objectively apprehended by the human mind, and specialized knowledge is clear-cut and universal. The goal of terminology is to ensure precision and unambiguous representation of specialized knowledge, by means of standardization of terms and concepts: Terminology work takes the concept as its point of departure, with the objective of establishing clear delimitations among them. Terminology views as independent the realm of concepts and denominations (=terms). (Wüster 1979: 21)1 Language is only viewed in its naming capacity and treated as ‘a necessary evil which needs to be constricted’ (Temmerman 2000: 60). In the search for a universal language and uniformity of communication, terms are treated as context-independent labels for concepts, and variation is considered an obstacle to effective communication: Ideally, the objective of term-concept assignment in a given special language is to ensure that a given term is attributed to only one concept and a given concept is represented by only one term, a condition called monosemy. This condition reduces ambiguity while homonymy and synonymy can lead to ambiguity. (ISO/704 2000: 24) The GTT remained the most influential theory of terminology until the 1990s, and its legacy is unquestionable today. The principles and methods of terminology work are still implemented in the technical terminology standardization work that is carried out by the International Organization for Standardization (ISO). However, the limitations and problems emerged as a result of implementing the GTT to other subject fields or activities where terminology management is involved, such as translation, language planning, or science popularization. For example, Riggs (1993) reflects on the contradictions between the urge for unambiguous communication and the accommodation to audiences with different levels of knowledge in the Social Sciences: Social scientists experience two contradictory requirements when writing their research reports. First, they need precise concepts that can be designated unambiguously. However, they also think of themselves as writing about human beings and their relationships to each other, matters which ought to be explained as much as possible in familiar everyday language. (Riggs 1993: 195) Alternative theories of terminology emerged in the 1990s as a reaction to the GTT, claiming the need to take into account the cognitive, linguistic, and sociocommunicative dimensions of terminology (Sager 1990; Gaudin 1993; Daille et al. 1996; Cabré 1999; Temmerman 2000). These theories share the view that variation is present in the way reality is experienced, structured, and expressed linguistically, and therefore, variation is an inherent characteristic of the term (Cabré 1999). Specialized knowledge is dynamic and flexible; it is embedded in the historical and cultural experience of specialized communities and reflects similarities and differences between social groups and individual speakers (Gaudin 1993) as well as their evolution over time (Temmerman 2000). Furthermore, terms are not only units that represent subject-specific concepts but ‘are a means of expression and communication’ (Cabré 2003: 21) when they are used in a variety of written and spoken specialized discourses. When terms are observed in their natural environment (i.e. texts), they show redundancy, conceptual, and synonymic variation, which ‘shatters the idealized view that there can or should be only one designation for a concept and vice-versa’ (Sager 1990: 58–59). From a textual perspective, it has been shown that textual factors such as level of specialization, text genre, textual function, theme, or rhetorical purpose determine the choice, treatment, and formal and conceptual variability of the term (Messineo 2002; Ciapuscio 2003; Collet 2003; Pecman 2012). In sum, term variation is the result of the dynamics of categorization and knowledge structuring (Fernández-Silva 2011; Freixa and Fernández-Silva 2017), and of the accommodation of language to the geographic, social, and functional sub-registers of specialized communication (Freixa 2002; Picton and Dury 2017). Today, term variation is widely recognized in descriptive approaches to terminological theory and is a widespread topic of research (Daille 2017; Drouin et al. 2017). Furthermore, handling term variants ‘contributes to the improvement of several terminology-oriented applications’ (Daille 2005: 194), such as specialized dictionaries (Collet 2004), terminology acquisition (Daille 2017), or automatic text indexing (Jacquemin 2001). However, the idea that term variation is contrary to the precision of scientific language still prevails in some scientific circles and LSP manuals. For example, Gutiérrez (2005: 22) suggests that ‘to ensure precision one can use always the same term for a given concept, no matter how many times it is reiterated’. Fuentes (2006) insists that bi-univocity is the only valid relationship between concepts and terms: Ideally, there should be a unique terminology system that departs from fundamental ‘axiomatic terms’, from which the rest of terms would be derived, forming a structure similar to that of Mathematics. In this terminology, bi-univocity would be the only valid relation between concepts and terms. Fuentes (2006: 241)2 3. INTRA-TEXTUAL TERM VARIATION Terminological variation within the same text cannot be related to differences in conceptualization, dialectal, or functional factors because it is produced by the same author(s) who write for the same audience(s). That is why it has been deemed a stylistic device to avoid repetition and redundancy in order to produce ‘a text that is stylistically acceptable’ (Freixa 2005: 10). From a textual perspective, term variation performs an important function in text construction because it is one of the main devices of lexical cohesion that contribute to the general texture of the text. If we take Halliday and Hasan’s (1976) typology of cohesion, term variation could be identified with reiteration, ‘a form of lexical cohesion which involves the repetition of a lexical item, at one end of the scale; the use of a general word to refer back to a lexical item, at the other end of the scale; and a number of things in between—the use of a synonym, near-synonym, or superordinate’ (Halliday and Hasan 1976: 278). Term variation is seen as a process of reiteration whereby the writer uses different terminological units to express the same concept (Kerremans 2017). As Collet (2003: 4) states, the base term and its variants are equally able to refer to the same concept within the boundaries of the text ‘in light of this potential for co-referentiality’. From a cognitive–semantic perspective, term variation can be functional in shaping understanding and transferring knowledge because the coexistence of term variants allows focusing on different aspects of the concept in different contexts. Several authors have suggested this idea. Temmerman (2000: 227) states that ‘synonymy and polysemy appear to be functional in the process of progress of understanding’; Fernández-Silva (2016) observed how intra-textual term variation allows the author to present different but complementary views of the same concept within the same text by selecting variants with different semantic information; and Pecman (2014: 10) suggested that: The terminological variation could undoubtedly be explained in relation to cognition and the process of knowledge construction in progress: a speaker, a scientist, would thus resort to terminological variation as a device for transferring his or her experience into language, and consequently into knowledge. 4. METHODOLOGY This study was conducted within the framework of the VARTERM research project, which aims to investigate the role of term variation in academic discourse across four disciplines of Natural and Social Sciences and Humanities.3 For this particular study, a two-step methodology was designed which combined corpus-based analysis of term variation in Geology and Psychology RAs and semi-structured interviews with field experts. 4.1 Corpus-based analysis of term variation in RAs The study is based on a small corpus of 38 RAs written in Spanish (358,425 words). These were published in three Chilean peer-reviewed journals indexed in the open-access digital scientific library Scielo Chile4: Andean Geology (Chilean National Service of Geological Sciences and Mining Engineering), Terapia Psicológica (Chilean Society of Clinical Psychology), and Psykhe (Catholic University of Chile). These are the only journals available in Scielo Chile in both fields that publish articles in Spanish, and all three cover a wide range of topics. Andean Geology publishes research on general topics of broad interest concerning the geology of South and Central America and Antarctica, and particularly of the Andes; Psyche covers all topics of scientific Psychology, and Terapia Psicológica focuses on all topics and subareas of Clinical Psychology. Articles were selected at random from issues published between 2004 and 2014. 4.2 Detection of term variants This study adopted an onomasiological and contextual approach to term variation (Geeraerts et al. 1994; Hamon and Nazarenko 2001). The first lexical occurrence of a given concept was considered the base term, and all subsequent lexical instantiations of the same concept were treated as term variants. Under this approach, terminological units with varying degrees of fixedness were included, and conceptual equivalence among them was considered within the boundaries of the text. Table 1 shows the terminological variants of one concept in our corpus. Table 1: Detection of term variants Status  Term  Context  Base term  vínculo inseguro  [abstract] Estudios recientes han sugerido que el TDAH podría darse en el contexto de un vínculo inseguro. [Recent studies suggest that ADHD can arise in the context of an insecure bond]  Term variant [validated]  relación de apego inseguro  La investigación de las interacciones actuales entre padres e hijos [intro] con TDAH revelan patrones similares a los observados en el contexto de relaciones de apego inseguro. [Research into relationships between parents and ADHD children reveal patterns similar to those observed in insecure attachment relationships]  Term variant [rejected]  patrón vincular inseguro  [intro] Los autores sugieren una asociación entre TDAH y vínculo inseguro. El patrón vincular inseguro en este grupo se caracterizaría por una alta expresividad emocional. [Authors suggest an association between ADHD and insecure bond. The insecure bond pattern is characterized by a high emotional expressiveness]  Term variant [validated]  apego inseguro  [results] La distribución del tipo de apego inseguro en el grupo de niños con TDAH es la siguiente: [The distribution of types of insecure attachment in ADHD children is shown below]  Status  Term  Context  Base term  vínculo inseguro  [abstract] Estudios recientes han sugerido que el TDAH podría darse en el contexto de un vínculo inseguro. [Recent studies suggest that ADHD can arise in the context of an insecure bond]  Term variant [validated]  relación de apego inseguro  La investigación de las interacciones actuales entre padres e hijos [intro] con TDAH revelan patrones similares a los observados en el contexto de relaciones de apego inseguro. [Research into relationships between parents and ADHD children reveal patterns similar to those observed in insecure attachment relationships]  Term variant [rejected]  patrón vincular inseguro  [intro] Los autores sugieren una asociación entre TDAH y vínculo inseguro. El patrón vincular inseguro en este grupo se caracterizaría por una alta expresividad emocional. [Authors suggest an association between ADHD and insecure bond. The insecure bond pattern is characterized by a high emotional expressiveness]  Term variant [validated]  apego inseguro  [results] La distribución del tipo de apego inseguro en el grupo de niños con TDAH es la siguiente: [The distribution of types of insecure attachment in ADHD children is shown below]  Table 1: Detection of term variants Status  Term  Context  Base term  vínculo inseguro  [abstract] Estudios recientes han sugerido que el TDAH podría darse en el contexto de un vínculo inseguro. [Recent studies suggest that ADHD can arise in the context of an insecure bond]  Term variant [validated]  relación de apego inseguro  La investigación de las interacciones actuales entre padres e hijos [intro] con TDAH revelan patrones similares a los observados en el contexto de relaciones de apego inseguro. [Research into relationships between parents and ADHD children reveal patterns similar to those observed in insecure attachment relationships]  Term variant [rejected]  patrón vincular inseguro  [intro] Los autores sugieren una asociación entre TDAH y vínculo inseguro. El patrón vincular inseguro en este grupo se caracterizaría por una alta expresividad emocional. [Authors suggest an association between ADHD and insecure bond. The insecure bond pattern is characterized by a high emotional expressiveness]  Term variant [validated]  apego inseguro  [results] La distribución del tipo de apego inseguro en el grupo de niños con TDAH es la siguiente: [The distribution of types of insecure attachment in ADHD children is shown below]  Status  Term  Context  Base term  vínculo inseguro  [abstract] Estudios recientes han sugerido que el TDAH podría darse en el contexto de un vínculo inseguro. [Recent studies suggest that ADHD can arise in the context of an insecure bond]  Term variant [validated]  relación de apego inseguro  La investigación de las interacciones actuales entre padres e hijos [intro] con TDAH revelan patrones similares a los observados en el contexto de relaciones de apego inseguro. [Research into relationships between parents and ADHD children reveal patterns similar to those observed in insecure attachment relationships]  Term variant [rejected]  patrón vincular inseguro  [intro] Los autores sugieren una asociación entre TDAH y vínculo inseguro. El patrón vincular inseguro en este grupo se caracterizaría por una alta expresividad emocional. [Authors suggest an association between ADHD and insecure bond. The insecure bond pattern is characterized by a high emotional expressiveness]  Term variant [validated]  apego inseguro  [results] La distribución del tipo de apego inseguro en el grupo de niños con TDAH es la siguiente: [The distribution of types of insecure attachment in ADHD children is shown below]  The terminological analysis focused on the central concepts of each article, that is those appearing in the title, abstract, and keywords sections. This assumption is supported by extensive research on RA agreeing that these sections summarize the main content and state the central concepts of the article (Swales 1990; Hartley and Kostoff 2003). A combination of manual and semi-automatic searches with concordance software5 was used to identify term variants in texts (Fernández-Silva 2011). Manual identification involved reading the whole text and highlighting the term variant candidates based on formal and/or semantic similarity and co-referential analysis—identification of reformulation procedures (Ciapuscio 2003; Kerremans 2017). Semi-automatic searching was based on discourse markers of synonymy relation—such as known as, also called (Suárez 2004)—and search based on constituent elements of term variants (Hamon and Nazarenko 2001). For example, the base term uso de drogas [drug use], was split into constituent elements and the occurrences of each element—uso and drogas—were used to search for other variants in the corpus. The search based on the head element returned the variants uso de sustancias [substance use] and uso de tabaco, alcohol y drogas ilícitas [tobacco, alcohol and illicit drug use]; a search based on the modifier returned consumo de drogas [drug consumption]. Then, contexts of occurrence of term variants were automatically retrieved and manually validated by members of the research team (n = 5). Finally, all terms and equivalence relations were validated by field experts (n = 42). Expert validation is crucial for confirming conceptual equivalence between terms, since it requires expert field knowledge to ultimately determine whether two expressions refer to the same concept. Each expert was given a document with all the variants found in a text grouped by concepts (on average each Geology RA had 15.42 concepts per text and 8.36 in Psychology). They were asked to confirm the equivalence relation by examining the term variants embedded in their contexts of occurrence (see Table 1). The experts’ answers were analyzed again by the research team and, in case of disagreement between both opinions, the text was handed out to a second validator—that was the case with three texts in Geology and two in Psychology. Based on this procedure, 452 concepts, 1,766 term variants, and 6,566 occurrences were identified (see Table 2). Table 2: Raw frequency of concepts, term variants and contexts by discipline   Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1,766  Contexts  3,253  3,313  6,566    Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1,766  Contexts  3,253  3,313  6,566  Table 2: Raw frequency of concepts, term variants and contexts by discipline   Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1,766  Contexts  3,253  3,313  6,566    Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1,766  Contexts  3,253  3,313  6,566  4.3 Semantic classification of term variants Term variants were classified according to the type and degree of semantic change from the base term to assess a possible cognitive function of term variation. This involved comparing the conceptual information displayed by terms and variants (Kageura 2002; Fernández-Silva et al. 2011). Simple and complex terms were interpreted as onomasiological structures, that is combinations of concepts within the conceptual structure. The head element indicates the general category the concept belongs to (or hypernym), whereas the modifier—whether simple or complex—states the defining feature(s), that is the feature(s) that differentiate(s) the concept from its co-hyponyms. For example, the term theropod footprint is analyzed as a concept belonging to the general category of footprints—as expressed in the head element—whose defining feature is that it belongs to a theropod dinosaur—as stated in the modifier—. For single-word terms, the conceptual information was either deduced from the morphological structure—for example, hypsometry is the measure (general concept) of altitude (defining feature) and erosion is the act (general concept) of eroding (defining feature). Based on this analysis, eight variation types were identified (Fernández-Silva 2016). Then, variation types were ranked along a semantic distance gradient and grouped into three categories (see Table 3): Table 3: Classification of term variation according to semantic distance Semantic distance  Variation type  Example (base terms; term variants)  Maximum semantic distance  Change of conceptual configuration  paternidad tradicional/modelo hegemónico de paternidad [traditional paternity/hegemonic fatherhood model]  Hypernymic/hyponymic selection  embarazadas primigestas/población [primigravidae/population]; relaciones de amistad/díadas de amistad [friendship relations/dyadic friendship]  Change of category  domo/estructura dómica [dome/dome structure]  Medium semantic distance  Change of defining feature  grupo clínico/grupo TDAH [clinical group/ADHD group]  Change of non-defining feature  huella de terópodo/huella de dinosaurio terópodo [Theropod (dinosaur) footprint]  Reduction  nanofósiles calcáreos/nanofósiles [calcareous nannofossils/nannofossils]  Minimum semantic distance  Synonymy  estudio/investigación [study/research]  Formal variation  sistema estuárico distal/DES [distal estuarine system]  Semantic distance  Variation type  Example (base terms; term variants)  Maximum semantic distance  Change of conceptual configuration  paternidad tradicional/modelo hegemónico de paternidad [traditional paternity/hegemonic fatherhood model]  Hypernymic/hyponymic selection  embarazadas primigestas/población [primigravidae/population]; relaciones de amistad/díadas de amistad [friendship relations/dyadic friendship]  Change of category  domo/estructura dómica [dome/dome structure]  Medium semantic distance  Change of defining feature  grupo clínico/grupo TDAH [clinical group/ADHD group]  Change of non-defining feature  huella de terópodo/huella de dinosaurio terópodo [Theropod (dinosaur) footprint]  Reduction  nanofósiles calcáreos/nanofósiles [calcareous nannofossils/nannofossils]  Minimum semantic distance  Synonymy  estudio/investigación [study/research]  Formal variation  sistema estuárico distal/DES [distal estuarine system]  Table 3: Classification of term variation according to semantic distance Semantic distance  Variation type  Example (base terms; term variants)  Maximum semantic distance  Change of conceptual configuration  paternidad tradicional/modelo hegemónico de paternidad [traditional paternity/hegemonic fatherhood model]  Hypernymic/hyponymic selection  embarazadas primigestas/población [primigravidae/population]; relaciones de amistad/díadas de amistad [friendship relations/dyadic friendship]  Change of category  domo/estructura dómica [dome/dome structure]  Medium semantic distance  Change of defining feature  grupo clínico/grupo TDAH [clinical group/ADHD group]  Change of non-defining feature  huella de terópodo/huella de dinosaurio terópodo [Theropod (dinosaur) footprint]  Reduction  nanofósiles calcáreos/nanofósiles [calcareous nannofossils/nannofossils]  Minimum semantic distance  Synonymy  estudio/investigación [study/research]  Formal variation  sistema estuárico distal/DES [distal estuarine system]  Semantic distance  Variation type  Example (base terms; term variants)  Maximum semantic distance  Change of conceptual configuration  paternidad tradicional/modelo hegemónico de paternidad [traditional paternity/hegemonic fatherhood model]  Hypernymic/hyponymic selection  embarazadas primigestas/población [primigravidae/population]; relaciones de amistad/díadas de amistad [friendship relations/dyadic friendship]  Change of category  domo/estructura dómica [dome/dome structure]  Medium semantic distance  Change of defining feature  grupo clínico/grupo TDAH [clinical group/ADHD group]  Change of non-defining feature  huella de terópodo/huella de dinosaurio terópodo [Theropod (dinosaur) footprint]  Reduction  nanofósiles calcáreos/nanofósiles [calcareous nannofossils/nannofossils]  Minimum semantic distance  Synonymy  estudio/investigación [study/research]  Formal variation  sistema estuárico distal/DES [distal estuarine system]  Minimum semantic distance: Variants that do not involve changes in the conceptual content displayed by terms (i.e. semantically equivalent variants) fall into this category. Formal variation can involve graphical changes (sistema estuárico distal/DES [distal estuarine system]), morphological changes (vínculo inseguro/vinculación insegura [insecure attachment]), and morphosyntactic changes (programa de prevención/programa preventivo [prevention/preventive program]). Synonymy—replacing one lexical item with another with a similar meaning in the language system (Freixa 2002: 90)—entails lexical changes to the whole term (estudio/investigación [study/research]) or constituent elements of multiword terms (vínculo inseguro/apego inseguro [insecure attachment/bond]). Medium semantic distance: Variants involving changes in the modifier—where the concept’s defining feature is expressed—were ranked in the category of medium semantic distance. In reductions, the modifier is suppressed and the resulting term variant stands as a hypernym (nanofósiles calcáreos/nanofósiles [(calcareous) nannofossils]); in the next variation type, with greater semantic distance than the last, a non-defining feature is added to (huella de terópodo/huella de dinosaurio terópodo [Theropod (dinosaur) footprint]) or removed (coeficiente de correlación de Pearson/coeficiente de Pearson [Pearson (correlation) coefficient]) from the original head–modifier structure. In the last category, a different defining feature is selected. For example, the defining feature of grupo clínico [clinical group] is the role the group of individuals plays in the study, whereas the defining feature in grupo TDAH [ADHD group] is their pathology, that is attention-deficit/hyperactivity disorder. Maximum semantic distance: Variants under this category involve changes in the term’s head or in both head and modifier. These variations were considered to be cognitively more complex, since they change the concept’s conceptual category. Within this group, four categories with increasing semantic distance were identified. The first variation type involves a change of category. This change in head selection does not place the concept in a different category but rather emphasizes a given perspective of conceptualization. For example, the variant estructura dómica [dome structure] adds the word structure to the base term domo [dome] to emphasize this perspective of conceptualization. The next variation type selects a hypernymic category (embarazadas primigestas/población [primigravidae/population]) or a hyponymic category (relaciones de amistad/díadas de amistad [friendship relations/dyadic friendship]). Finally, the term variants marked by the highest degree of semantic distance entail a change of conceptual configuration: both the main head and modifier represent taxonomically unrelated categories, such as paternidad tradicional [traditional paternity] and modelo hegemónico de paternidad [hegemonic fatherhood model]. 4.4 Semi-structured interviews with field experts The second method for data collection involved semi-structured interviews with field experts. The aim of this interview was to obtain information about experts’ attitudes toward term variation and to dig deep into the functions of term variation in RAs according to their opinion. A three-part questionnaire was designed for the interview (see Supplementary Appendix 1), which combined direct and indirect methods for measuring language attitudes that are common in language attitudes research (Garrett 2010). The interview was structured as follows: Part 1. Experts were asked to read two excerpts from an RA from the corpus (300–500 words) containing a set of terminological variants for a given concept in bold letters. Then, they had to answer a set of open questions aimed at obtaining their opinion about term variation (e.g. in your opinion the quality of the text would have improved if the author had used fewer terms to refer to the same concept?), or the reasons for term variation (e.g. Why did the author use these variants to refer to the same concept?). They had to answer the questions twice, once for each text excerpt. Part 2. Experts had to read two excerpts of an RA from the corpus authored by them with a set of terminological variants for a given concept in bold letters. As in the previous part, they had to answer a series of questions aimed at eliciting their opinion on their use of term variation (e.g. If you could rewrite the text, would you use less different terms for the same concept and why?) and the reasons for choosing specific variants (e.g. Why did you use ‘early parenthood’ instead of repeating ‘adolescent parenthood’?). Again, they had to repeat the same exercise twice, one for each excerpt. Part 3. Experts were given 12 statements expressing different opinions on term variation and had to put a checkmark against those statements they agreed with (e.g. using different term variants for the same concept helps me to better explain its content). The questionnaire combines direct and indirect elicitation of attitudes. In the third part participants are asked to articulate explicitly what their attitudes are toward term variation; in Parts 1 and 2, on the contrary, they are not aware that they are doing an attitude rating task. The use of both methods allows obtaining information about private, unconscious attitudes (covert attitudes) and those that are conscious and overtly declared by respondents (overt attitudes). This helps reduce the social desirability bias—‘the tendency for people to give answers to questions in ways that they believe to be socially appropriate’ (Garrett 2010: 45)—and the acquiescence bias—the tendency to agree with an item as a way of gaining the researcher’s approval. Both are pervasive phenomena in language attitudes research, whose incidence is more significant in face-to-face interviews (Oppenheim 2000; Gass and Seiter 2003). The interview was administered to six experts (three psychologists and three geologists) who were single or first authors of RA from the corpus, and are affiliated to Academic Institutions in Chile. The experts who answered the questionnaire were different from those who validated the variants, to prevent circularity. The informed consent form was read and signed before the interview began. The interviews lasted between 45 and 75 min. In the first and second parts, respondents were asked the questions orally by the interviewers, who wrote the answers and the various comments made by the respondents on the questionnaires. The third questionnaire was answered directly by the participant. The analysis proceeded by identifying the main views on term variation that emerged from the interviews. For instance, comments such as ‘this term expands the concept of virtual reality’, and ‘to emphasize that debris flow is a dynamic process’ were grouped into a function category called ‘explaining the concept’. 5. INCIDENCE AND BEHAVIOR OF TERM VARIATION IN GEOLOGY AND PSYCHOLOGY RAs: CORPUS FINDINGS To compare the incidence of term variation in Psychology and Geology RA, two indexes were calculated: Terminological variability index: It is obtained by dividing the number of term variants per the number of concepts. It indicates the average of term variants each concept has. The higher the terminological variability index, the greater the variability of the corpus. Type/token ratio: It is calculated by dividing the number of term variants per the number of occurrences of these variants, and expressed in percentages. The more term variants (types) there are in comparison to the number of occurrences (tokens), the greater is the terminological variability in the corpus. As can be seen in Table 4, the terminological variability index is higher in Psychology RA, with 4.84 variants per concept. However, the type/token ratio is higher in Geology, with 30.62 per cent. This means that in the Psychology corpus there are more term variants per concept, but each variant is repeated more times than in Geology. This could be related to the fact that the terminological density in Psychology RA is lower—8 concepts per abstract in average, as opposed to 15 concepts per abstract in Geology. This might explain why specialized concepts in Psychology are reiterated more frequently throughout the text. Table 4: Raw frequency of concepts, term variants and contexts and terminological variability measures in Geology and Psychology RA   Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1766  Contexts  3,253  3,313  6,566  Terminological variability index  3.40  4.84  3.91  Type/token ratio (per cent)  30.62  23.24  26.90    Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1766  Contexts  3,253  3,313  6,566  Terminological variability index  3.40  4.84  3.91  Type/token ratio (per cent)  30.62  23.24  26.90  Table 4: Raw frequency of concepts, term variants and contexts and terminological variability measures in Geology and Psychology RA   Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1766  Contexts  3,253  3,313  6,566  Terminological variability index  3.40  4.84  3.91  Type/token ratio (per cent)  30.62  23.24  26.90    Geology  Psychology  Total  Concepts  293  159  452  Term variants  996  770  1766  Contexts  3,253  3,313  6,566  Terminological variability index  3.40  4.84  3.91  Type/token ratio (per cent)  30.62  23.24  26.90  As a next step, the incidence of semantic variation types was examined in the two corpora. Figure 1 shows the percentage frequency of occurrence of each variation type in texts along with the percentage frequency of occurrence of the base term: Figure 1: View largeDownload slide Distribution of variation types in Psychology and Geology Figure 1: View largeDownload slide Distribution of variation types in Psychology and Geology This analysis shows that the incidence of term variation is higher in Psychology than in Geology: in Psychology RA, the authors used the base term consistently in 36.7 per cent of cases, whereas in Geology repetition of the base term amounts to 47.2 per cent. These differences proved statistically highly significant (P < p-value = 0.00 using Pearson’s chi-square test). Our results seem to indicate that the naming function of terminology, which aims at quickly and efficiently identify the concept by its denomination (Sager 1990), is more frequent in Geology than in Psychology, where the repetition of the base term only occurs one-third of the time. However, in none of the cases it can be concluded that the univocity principle prevails because term variation accounts for more than 50 per cent of terminological occurrences in Geology and more than 60 per cent in Psychology RA. Variants with minimum semantic distance have a small impact in both corpora. They account for 7.22 per cent in Geology and 10.42 per cent in Psychology. The function of these variants is merely stylistic because they do not provide additional information to the reader about the concept. For example, replacing a multiword term (sistema estuárico distal [distal estuarine system] or representaciones sociales [social representations]) by its acronym (DES and RS) is a frequent strategy in academic texts that favors verbal economy and avoids the repetitive use of long, cumbersome multiword terms. Formal variants involve graphical changes but also morphological changes—for example, suffix changes such as in vínculo inseguro/vinculación insegura [insecure attachment])—or morphosyntactic changes, mostly concerned with the alternation between the syntagmatic structure N P N and N A, like in conjunto de sierras/conjunto serrano [system of mountain ranges]. This variation has no cognitive consequences because the way the concept is represented does not change; they can have a stylistic function of avoiding repetition or respond to the author’s expressive needs or carelessness in the consistent use of terminology. Synonyms have little impact in the corpus, both in Geology (3.1 per cent) and Psychology (4.3 per cent). They can involve replacement of words having the same meaning in general Spanish language, such as estudio/investigación [study/research]; in other cases, more frequently in Geology, synonymy is the result of the cohabitation of different term formation processes, such as vernacular derivation or compounding, English borrowing (antepaís fracturado/broken foreland), or classical compounding with Greco-Latin root words (mineral arcilloso/argilomineral [clay mineral]). The potential benefits of using these synonyms do not lie in their cognitive value because they have equivalent meanings; their co-occurrence inside a text seems more a matter of stylistic choice or the cohabitation of different practices of term formation in the discipline. Variation types with medium semantic distance, that is involving lexical changes in the term’s modifier, do have cognitive consequences in the way the concept is presented. The term’s modifier reflects the concept’s defining feature(s), so modification of its constituent elements involves adding, deleting, or changing information about the concept’s content. Reductions are the most frequent type of term variation in Geology texts (15.9 per cent) and the second more frequent in Psychology (14.1 per cent), although the differences between both are not statistically significant (p-value = 0.1185). Reduced variants are, conceptually speaking, hypernyms (conglomerado sintectónico/conglomerados [(syntectonic) conglomerates]. They provide intercategorial information about the concept, that is to say, about its position within the conceptual structure (Temmerman 2000). Besides this cognitive function, reduction is also a textual strategy that ensures verbal economy and facilitates cohesion and coherence, both requirements for textuality (Halliday and Hasan 1976; Collet 2003). Term variants reflecting a different defining feature (relleno de la cuenca/relleno volcanosedimentario [basin filling/vocanosedimentary filling]) or selecting an additional, non-defining feature (huella de terópodo/huella de dinosaurio terópodo [theropod (dinosaur) footprint]) provide information about the intracategorial aspects of the concept—its internal characteristics—which are useful for explaining its content (Temmerman 2000). However, their frequency is low in the corpus in both disciplines (3.3 and 3.2 per cent in Geology and 3.1 and 2.1 per cent in Psychology). Finally, the maximum semantic distance category—which includes variation types involving changes in the head element or in the whole term—is the most frequent in both corpora. It accounts for 23.21 per cent of occurrences in Geology and 33.67 per cent in Psychology. However, it is also where more distributional differences are found between corpora. In Geology, variants selecting a different category in the head element of the term are the most frequent after reductions (11.34 per cent). This variation provides different perspectives of categorization of a given concept, and in our opinion, it could reflect the dynamic nature of categorization and the prototypical structure of concepts (Temmerman 2000; Faber 2012). For example, when the term lago somero [shallow lake] is replaced by the term variant sistema lacustre somero [shallow lake system], it emphasizes the complexity of this concept in the field, covering hydrological, geomorphological, physical, chemical, and biological conditions; when the term variant cuerpo lacustre somero [shallow lake body] is selected, the lake is categorized as a body of water. As can be seen, these three categorizations cross-cut the conceptualization of lake in the field. Hypernymic and hyponymic variants represent 8.7 per cent in Geology and 18.99 per cent in Psychology, where it stands as the most frequent variation type. These variants fulfill a cognitive function at the intercategorial level because they specify the relationships the concept bears with others within the domain’s conceptual structure. For example, when the term afectividad [affection] is replaced by its hypernym interacción emocional [emotional interaction], it informs about the classification of this concept in Psychology. However, it is also often the case that hypernymic variants correspond to generic terms, such as phenomenon as a variant of domestic violence, case for HIV patients, or subjects for university students. In those cases, it seems that the cognitive function of term variation is not so obvious, and one could argue that this variation functions as a textual strategy for establishing co-referential relations within the text (Halliday and Hasan 1976; Collet 2003). Finally, variants with a change in the conceptual configuration, that is variants involving a change in the term’s head and modifier, are not so common in Geology RA (3.17 per cent) but stand in the third position in the Psychology texts, with 9.03 per cent of occurrences. The equivalence of these variants with the base term is limited to the scope of the text. In a previous study (Fernández-Silva 2016), it was observed that this kind of variation fulfills a rhetorical function in Psychology RA. It appears mostly in the method and results sections of RA, and it is used to link the theoretical concepts identified in the introduction with the specific sample investigated in a given study, which is representative of the general concept. For example, in an RA on the relationship between quality of life and health care for patients with HIV/AIDS, the base term pacientes diagnosticados con VIH/Sida [patients diagnosed with HIV/AIDS] is used in the title, introduction, and conclusion sections and is replaced by the term variants sujetos del estudio [subjects of the study], participantes del estudio [participants], or pacientes de nuestro estudio [patients of our study] in the method and results sections. This shows the ‘particularizing function of term variation’ (Fernández-Silva 2016: 73), which is used to ‘making transition from the general field or context of the experiment to the specific experiment’ (Hill et al. 1982, in Swales 1990: 134). In sum, the corpus-illustrated analysis has revealed differences in the use of term variation between disciplines. The incidence of term variation is quantitatively higher in Psychology RA than in Geology, and differences in the distribution of variation types proved statistically significant for all categories with the exception of reductions and variants with changes in non-defining feature(s). Furthermore, the semantic analysis shows that variants with medium and maximum semantic distance prevail in both corpora. The use of variants with semantic changes provide additional information about the concept, either selecting different defining and non-defining feature(s), selecting a different head category, or indicating its position in the domain’s conceptual structure—when the base term is replaced by hypernyms, hyponyms, or reductions. Term variation is used to explain the concept’s content throughout the text, and therefore, it could be claimed that it fulfills a cognitive function. In addition to this, variation is also the result of accommodating technical vocabulary to the textual environment, where it establishes relations with other linguistic elements that may cause transformations in their form and structure—such as reductions—or where the representative function of terms can be overrun by other rhetorical purposes of the textual genre, such as generalizing or particularizing the specific study to the general research field. In sum, these results reveal that term variation is not only a stylistic resource for avoiding repetition but it also fulfills a cognitive and communicative function in both corpora. However, some additional questions remain that the corpus analysis cannot answer. For example, does the higher incidence of term variation in Psychology correlate to a more positive attitude toward this phenomenon in the disciplinary community, as compared to Geology? Are the cognitive and communicative functions of term variation consciously handled by the members of the disciplinary community? The results of the questionnaire will shed light on these questions, helping to better assess the role of term variation in both disciplines. 6. THE PERCEIVED FUNCTIONS OF TERM VARIATION IN SPECIALIZED COMMUNICATION: RESULTS OF THE INTERVIEWS WITH EXPERTS The content of the interviews was analyzed in relation to two aspects: experts’ attitudes toward term variation and the perceived functions of intra-textual term variation according to their opinion. Overt attitudes toward term variation were retrieved from the questionnaire (Part 3), where 10 of the 12 assessments reflected either positive or negative attitudes toward this resource. Covert language attitudes were extracted by analyzing the participants’ reflections, as they were doing Parts 1 and 2. Table 5 shows, for each participant, the number of positive and negative judgments selected in the questionnaire (overt) and formulated during the interview (covert). It is also presented in percentages to compare the proportion of positive and negative statements expressed by each participant. The right column provides the percentages for each discipline. Table 5: Positive and negative attitudes toward term variation by discipline   GEO1 (per cent)  GEO2 (per cent)  GEO3 (per cent)  PSY1 (per cent)  PSY2 (per cent)  PSY3 (per cent)  GEO (per cent)  PSY (per cent)  Positive  Overt  5 (45.4)  6 (50)  4 (57.1)  4 (33.3)  2 (11.1)  5 (22.7)  24 (80.6)  30 (58.8)  Covert  3 (27.3)  4 (33.3)  2 (28.6)  5 (41.7)  4 (22.2)  10 (45.5)  Negative  Overt  1 (9.1)  0 (0)  0 (0)  1 (8.3)  4 (22.2)  2 (9.1)  6 (19.4)  22 (41.1)  Covert  2 (18.2)  2 (16.7)  1 (14.3)  2 (16.7)  8 (44.5)  5 (22.7)    GEO1 (per cent)  GEO2 (per cent)  GEO3 (per cent)  PSY1 (per cent)  PSY2 (per cent)  PSY3 (per cent)  GEO (per cent)  PSY (per cent)  Positive  Overt  5 (45.4)  6 (50)  4 (57.1)  4 (33.3)  2 (11.1)  5 (22.7)  24 (80.6)  30 (58.8)  Covert  3 (27.3)  4 (33.3)  2 (28.6)  5 (41.7)  4 (22.2)  10 (45.5)  Negative  Overt  1 (9.1)  0 (0)  0 (0)  1 (8.3)  4 (22.2)  2 (9.1)  6 (19.4)  22 (41.1)  Covert  2 (18.2)  2 (16.7)  1 (14.3)  2 (16.7)  8 (44.5)  5 (22.7)  Table 5: Positive and negative attitudes toward term variation by discipline   GEO1 (per cent)  GEO2 (per cent)  GEO3 (per cent)  PSY1 (per cent)  PSY2 (per cent)  PSY3 (per cent)  GEO (per cent)  PSY (per cent)  Positive  Overt  5 (45.4)  6 (50)  4 (57.1)  4 (33.3)  2 (11.1)  5 (22.7)  24 (80.6)  30 (58.8)  Covert  3 (27.3)  4 (33.3)  2 (28.6)  5 (41.7)  4 (22.2)  10 (45.5)  Negative  Overt  1 (9.1)  0 (0)  0 (0)  1 (8.3)  4 (22.2)  2 (9.1)  6 (19.4)  22 (41.1)  Covert  2 (18.2)  2 (16.7)  1 (14.3)  2 (16.7)  8 (44.5)  5 (22.7)    GEO1 (per cent)  GEO2 (per cent)  GEO3 (per cent)  PSY1 (per cent)  PSY2 (per cent)  PSY3 (per cent)  GEO (per cent)  PSY (per cent)  Positive  Overt  5 (45.4)  6 (50)  4 (57.1)  4 (33.3)  2 (11.1)  5 (22.7)  24 (80.6)  30 (58.8)  Covert  3 (27.3)  4 (33.3)  2 (28.6)  5 (41.7)  4 (22.2)  10 (45.5)  Negative  Overt  1 (9.1)  0 (0)  0 (0)  1 (8.3)  4 (22.2)  2 (9.1)  6 (19.4)  22 (41.1)  Covert  2 (18.2)  2 (16.7)  1 (14.3)  2 (16.7)  8 (44.5)  5 (22.7)  As can be seen, five of six participants manifested more positive judgments toward term variation than negative ones, both overt and covert. However, the total percentages show a more positive attitude toward variation in Geology (80.6 per cent) than in Psychology (58.8 per cent). This can be explained by the deviant position of psychologist 2, who clearly manifested a negative attitude toward term variation that remained consistent throughout the interview. Even when he analyzed his own writing, he regretted having used terminological variants. Another interesting result is that there is more consistency between overt and covert attitudes among psychologists. This difference can be interpreted as the result of a higher awareness of the process of writing in this group. When they were asked to reflect on their own and others’ writing, answers showed that psychologists had clearer representations about their writing practices. On the contrary, geologist manifested fewer opinions, both positive and negative, about the use of terminology in their own or others texts. A negative view shared by geologists and psychologists is that consistency favors precision and clarity, and that term variation can lead to confusion. Among positive attitudes, there is agreement that the use of several term variants can be useful for explaining a concept and is sometimes necessary for avoiding excessive lexical repetition. However, psychologists adhere to other motivations for variation that are absent in the geologists’ accounts: the necessity to vary the choice of terms depending on the level of expertise of the target audience; and the need to use several term variants because they cannot find the term that exactly represents their ideas. When comparing the results of the corpus analysis with those of the interviews, our results do not support the hypothesis about the relationship between language attitudes and linguistic behavior, at least in quantitative terms. Geologists have shown more positive judgments to term variation than psychologists, but the incidence of term variation is lower in the corpus of Geology (52.8 per cent) than in Psychology (63.3 per cent). However, these results are not conclusive given the small sample of participants in the interviews, and should definitely be tested again with a larger sample. Having said that, an interesting observation is that term variation has a positive consideration among most experts and is seen not only as necessary but sometimes functional resource in expert communication. As a next step, the interviews were analyzed to assess what the perceived functionalities of term variation are according to experts. Table 6 shows the functions that emerged from the analysis and the frequency with which they were reported by each group. Functions have been grouped under a broad category of cognitive function, when they were related to the process of knowledge construction and transfer; or communicative function, when they were related to the operations of text construction or accommodation to the communicative situation. Table 6: Perceived functions of term variation according to experts   PSI (per cent)  GEO (per cent)  Communicative  Avoid repetition (stylistic)  27 (50.94)  13 (35.14)  Accomodate text to intended audience (functional)  4 (7.55)  1 (2.70)  Generalize and/or particularize (rhetorical)  2 (3.77)  2 (5.41)  Cognitive  Provide intracategorial information  9 (16.98)  11 (29.73)  Provide intercategorial information  4 (7.55)  4 (10.81)  Favor conceptual clarity  7 (13.21)  6 (16.22)    53 (100.00)  37 (100.00)    PSI (per cent)  GEO (per cent)  Communicative  Avoid repetition (stylistic)  27 (50.94)  13 (35.14)  Accomodate text to intended audience (functional)  4 (7.55)  1 (2.70)  Generalize and/or particularize (rhetorical)  2 (3.77)  2 (5.41)  Cognitive  Provide intracategorial information  9 (16.98)  11 (29.73)  Provide intercategorial information  4 (7.55)  4 (10.81)  Favor conceptual clarity  7 (13.21)  6 (16.22)    53 (100.00)  37 (100.00)  Table 6: Perceived functions of term variation according to experts   PSI (per cent)  GEO (per cent)  Communicative  Avoid repetition (stylistic)  27 (50.94)  13 (35.14)  Accomodate text to intended audience (functional)  4 (7.55)  1 (2.70)  Generalize and/or particularize (rhetorical)  2 (3.77)  2 (5.41)  Cognitive  Provide intracategorial information  9 (16.98)  11 (29.73)  Provide intercategorial information  4 (7.55)  4 (10.81)  Favor conceptual clarity  7 (13.21)  6 (16.22)    53 (100.00)  37 (100.00)    PSI (per cent)  GEO (per cent)  Communicative  Avoid repetition (stylistic)  27 (50.94)  13 (35.14)  Accomodate text to intended audience (functional)  4 (7.55)  1 (2.70)  Generalize and/or particularize (rhetorical)  2 (3.77)  2 (5.41)  Cognitive  Provide intracategorial information  9 (16.98)  11 (29.73)  Provide intercategorial information  4 (7.55)  4 (10.81)  Favor conceptual clarity  7 (13.21)  6 (16.22)    53 (100.00)  37 (100.00)  As can be seen, avoiding repetition stands as the most frequent function of term variation according to psychologists (50.94) and geologists (35.14), but it is notably higher in Psychology. It was a common view among participants that reiterating the same terms over and over again can be cumbersome. For example, psychologist 3 affirms that it is ‘important to be careful when one writes to avoid repetition. I hate repeating myself’; and psychologist 1 reports that ‘I always tell my students to avoid redundancy when writing a text’. Experts recognized stylistic reasons not only with formal variants (etapa de orogénesis regional vs. etapa orogénica regional [regional orogenesis/orogenic phase]) or reductions (triásico superior vs. triásico [upper Triassic vs. Triassic]); for example, geologist 3 said that he/she used holocenic glacial fluctuations instead of glacial variations ‘for aesthetic reasons, to avoid repeating the same term’. The stylistic function of term variation overshadows other communicative functions, which seem to be far less conscious to experts. For example, the rhetorical function is only reported two times by each group (3.77 per cent in Psychology and 5.41 per cent in Geology). The rhetorical function is used to generalize and/or particularize, making a link between the specific study and the general field or context of the study. For example, regarding Excerpt (1), psychologist 2 explains that he used participants (the individuals who were actually part of the study) instead of 4th–7th-grade schoolchildren (the population sector under study) ‘to situate them as participants of the study’: (1) El objetivo de esta investigación fue identificar factores predictores del inicio en el consumo de drogas lícitas en escolares de 4° a 7° básico [4th to 7th-grade schoolchildren]. Se encuestaron 234 participantes [participants] de dos comunas de Santiago, mediante la metodología de pares. Finally, term variation is also perceived as a device to accommodate the text to the intended audience. However, this communicative function appears almost three times more frequently in interviews with psychologists (7.55 per cent) than with geologists (2.70 per cent). For example, all psychologists and only one geologist agreed with the sentence ‘I don’t use the same terms when I write a text for experts and for students’ in the questionnaire. Furthermore, psychologist 2 commented upon Excerpt (2): ‘I guess that the author used all these terms to approach the text to readers. And I think he/she succeeded’. (2) el apego [bond] consiste en ‘un conjunto de pautas de conducta características […] que tienen por objetivo mantener al niño en una proximidad más o menos estrecha con su figura materna’ […] Hacia el final del primer año, esta conducta [behavior] se organiza como un sistema y se vuelve activa cada vez que se dan ciertas condiciones […] Este conjunto de conductas [set of behaviors] tiene como finalidad la protección. […] La conducta de apego [bonding behavior] se alterna con la de exploración del medio […] The other functions of term variation that showed up in the interviews were classified as cognitive. In these cases, term variation is used for better explaining the concepts, in other words, for constructing and transferring knowledge through linguistic means. Experts have appointed three cognitive functions during the interviews. First, term variation is used to provide intracategorial information, that is information about the characteristics of a given concept. This can involve using a longer and somehow redundant term variant, as can be seen in Excerpt (3). The author declared that she/he used the term variant parenthood of adolescent fathers instead of adolescent parenthood ‘to make clear that in this case it is referred to men. This concept always leads to confusion and is normally attributed to women’: (3) Los principales resultados y conclusiones señalan que la paternidad adolescente [adolescent parenthood] es un tema lleno de contradicciones; […] Como objetivos específicos, el estudio contempla […] describir los posibles obstaculizadores del ejercicio de la paternidad de los padres adolescentes [parenthood of adolescent fathers] y el significado que tienen de ellos. This function holds a relevant position among geologists and is recalled in 29.73 per cent of the cases. For example, in Excerpt (4), one geologist mentioned that he/she switched the term paleo-volcano by the term variant paleo-volcanic edifice ‘so the reader could picture the structure of the volcano’. Later, he/she used pre-collapse edifice because ‘I wanted readers to understand that I was talking about the volcanic structure before it collapsed’: (4) Se propone la estructura del paleovolcán [paleo-volcano] previa a su colapso parcial. […] El análisis de facies ha permitido reconstruir, al menos parcialmente, el flanco sur del paleoedificio volcánico [paleo-volcanic edifice]. […]Debido a la ausencia de material del edificio precolapso [pre-collapse edifice] en los cerrillos de brecha piroclástica […] se propone que los materiales que conforman estos cerrillos se encontraban hacia la parte más lejana del flanco sur del volcán. Term variation can be used as a device to provide intercategorial information. This function appears in the interviews with geologists (10.81 per cent) and psychologists (7.55 per cent). In Excerpt (5) a geologist explained that he/she used flow and avalanche flow instead of debris avalanche because ‘they are a type of flow’, and because ‘it is a dynamic process and I wanted to pass the idea that it is a process’: (5) Las avalanchas de detritos [debris avalanche], asociadas a colapsos parciales de edificios volcánicos, son fenómenos comunes en la evolución de un volcán. Este tipo de flujos [flows] son por inestabilidades, que pueden deberse a factores tales como la existencia de zonas afectadas por alteración hidrotermal […] hacia los flancos del volcán se disponían depósitos […] que fueron en parte incorporados por el flujo de avalancha [avalanche flow] durante su avance. Finally, we regrouped under the label ‘favor conceptual clarity’ different statements in which the experts reflected on term variation as a device for better explaining oneself. Five participants agreed with the statement that ‘using different words for a given concept facilitates text comprehension’. All psychologist and one geologist mentioned that ‘when writing a text I sometimes cannot find the right word to convey my idea, and I use different expressions’; and a geologist said that using different variants was more explanatory because ‘if the writer repeated himself, he could lead to misunderstandings’. In sum, interviews revealed that experts are well aware of the different functions of term variation in texts and that they can recall a variety of communicative and cognitive functions. Interestingly, several cognitive functions of term variation, which were considered as ‘unconscious’ in a previous study (Freixa 2005), were pointed out during the interviews. 7. CONCLUSION This study explored the communicative and cognitive functions of intra-textual term variation in RAs in Psychology and Geology. The corpus analysis and the interviews with experts showed that terminological variation is used to avoid repetition, to accommodate the text to the intended audience or to rhetorically move between the specific research and the general topic under investigation. Moreover, term variation is also a device for knowledge construction because it is used to provide information about the concepts and their relationships within the knowledge structure of the field. Our results seem to indicate that most experts hold a positive attitude toward term variation and are aware of the importance of this resource for the representation and transfer of specialized knowledge. These results show the interest of investigating term variation from a cognitive–functional perspective, taking into account cognitive, linguistic, and sociocommunicative factors involved in specialized language use. It also shows that bi-univocity remains an idealistic principle that is neither possible nor desirable. On the other side, we observed interesting differences between disciplines. In Geology, the incidence of term variation in RA is 30 per cent lower than in Psychology and, according to the experts’ opinions, the cognitive functions are more important in Geology than in Psychology. However, we could not find a relationship between a more positive attitude and a higher incidence of term variation. As a matter of fact, geologists manifested more positive judgments about the use of term variation than psychologists during the interviews, but it is also true that they manifested fewer opinions in general (30 opinions against 52, see Table 5). On the other side, psychologists were more critical to their own term choice and that of others; this could suggest a higher awareness of this group about the writing process and the construction of an appropriate text. In any case, the connection between attitudes toward variation and term use remains a hypothesis that should be explored more in depth with a greater sample of interviews in more disciplines. Finally, we highlight the importance of combining different methods for a cognitive study of terminology, such as corpus analysis and interviews. However, both methods still cannot answer crucial questions, such as the real impact of term variation in the transfer of specialized knowledge. As a direction for future research, it is envisaged to include experimental methodologies to assess the role of term variation in the comprehension of specialized knowledge. Only the convergence of different methodologies will allow understanding the complex relationship between cognitive factors of variation and their linguistic manifestations. Sabela Fernández-Silva holds a BA in Translation and Interpreting from University of Salamanca and a PhD degree in Applied Linguistics from Pompeu Fabra University, Barcelona, Spain. She is an assistant professor at the Pontifical Catholic University of Valparaíso, Chile, and coordinator of the Translation and Interpreting Department. She teaches courses on terminology, lexicology, lexicography, and linguistic theories to undergraduate and graduate students. Her research interests cover terminology, neology, and lexical semantics from a cognitive–functional perspective. Address for correspondence: Sabela Fernández-Silva, Av. El Bosque 1290, ILCL Building, off. 5.6. 25000, Viña del Mar, V Región, Chile. <sabela.fernandez@pucv.cl> NOTES 1 Translation is ours. 2 Translation is ours. 3 The VARTERM project (Fondecyt 11121597) was funded by the Government of Chile’s National Commission for Scientific and Technological Research (CONICYT). Further information can be found on its website: www.varterm.org. 4 Scientific Electronic Library Online (Scielo Chile): http://www.scielo.cl/ 5 Textstat: http://neon.niederlandistik.fu-berlin.de/textstat/ SUPPLEMENTARY DATA Supplementary material is available at Applied Linguistics online. Acknowledgements The author would like to thank Nelson Becerra, Camila Cifuentes, Francisco García, Paula Morgado, Joaquín Moya, Carol Valenzuela, and Grace Wilson, members of the Varterm research project, for their valuable assistance during various stages of this research. The author would also like to thank the three anonymous reviewers for their comments that contributed to improving the final version of the article. Conflict of interest statement. None declared. REFERENCES Barsalou L. 2003. ‘ Situated simulation in the human conceptual system,’ Language and Cognitive Processes  18: 513– 62. Google Scholar CrossRef Search ADS   Cabré M. T. 1999. La Terminología: Representación y Comunicación . IULA-Universitat Pompeu Fabra. Cabré M. T. 2003. ‘ Theories of terminology. Their description, prescription and explanation,’ Terminology  9: 163– 99. Google Scholar CrossRef Search ADS   Ciapuscio G. E. 2003. Textos Especializados y Terminología . IULA-Universitat Pompeu Fabra. Collet T. 2003. ‘ A two-level grammar of the reduction processes of French complex terms in discourse,’ Terminology  9: 1– 27. Google Scholar CrossRef Search ADS   Collet T. 2004. ‘ Esquisse d’une nouvelle microstructure de dictionnaire spécialisé reflétant la variation en discours du terme syntagmatique,’ Meta, Journal of Translators  49: 247– 63. Available at https://www.erudit.org/fr/revues/meta/2004-v49-n2-meta770/009349ar/ Daille B. 2005. ‘ Variations and application-oriented terminology engineering,’ Terminology  11: 181– 97. Google Scholar CrossRef Search ADS   Daille B. 2017. Term Variation in Specialised Corpora: Characterisation, Automatic Discovery and Applications . John Benjamins Publishing Company. Google Scholar CrossRef Search ADS   Daille B., Habert B., Jacquemin C., Royauté J.. 1996. ‘ Empirical observation of term variation and principles for their description,’ Terminology  3: 197– 257. Google Scholar CrossRef Search ADS   Drouin P., Francoeur A., Humbley J., Picton A.. (eds). 2017. Multiple Perspectives on Terminological Variation . John Benjamins Publishing Company. Google Scholar CrossRef Search ADS   Faber P. (ed.). 2012. A Cognitive Linguistics View of Terminology and Specialized Language . De Gruyter. Google Scholar CrossRef Search ADS   Fernández-Silva S. 2011. Variación Terminológica y Cognición: Factores Cognitivos En La Denominación Del Concepto Especializado . IULA-Universitat Pompeu Fabra. Available at http://hdl.handle.net/10803/22638 Fernández-Silva S. 2016. ‘ The cognitive and rhetorical role of term variation and its contribution to knowledge construction in research articles,’ Terminology  22: 52– 79. Google Scholar CrossRef Search ADS   Fernández-Silva S., Freixa J., Cabré M. T.. 2011. ‘ A proposed method for analysing the dynamics of cognition through term variation,’ Terminology  17: 49– 74. Google Scholar CrossRef Search ADS   Freixa J. 2002. La Variació Terminològica: Anàlisi de la Variació Denominativa en Textos de Diferent Grau d’Especialització de l’Àrea de Medi Ambient. IULA-UPF. Available at http://hdl.handle.net/10803/1677 Freixa J. 2005. ‘ Variació terminológica, ¿por qué y para qué?,’ Meta, Journal of Translators  50. Available at https://www.erudit.org/revue/meta/2005/v50/n4/019917ar.pdf Freixa J., Fernández-Silva S.. 2017. ‘Terminological variation and the unsaturability of concepts’ in Drouin P., Francoeur A., Humbley J., Picton A. (eds): Multiple Perspectives on Terminological Variation . John Benjamins Publishing Company, pp. 155– 80. Google Scholar CrossRef Search ADS   Fuentes X. 2006. ‘Contra la sinonimia y la polisemia en los lenguajes de especialidad,’ Panace@ 7: 24. Available at www.tremedica.org/panacea/IndiceGeneral/n24_entremes3-f.arderiu.pdf Gallese V., Lakoff G.. 2005. ‘ The brain’s concepts: The role of the sensory-motor system in conceptual knowledge,’ Cognitive Neuropsychology  22: 455– 79. Google Scholar CrossRef Search ADS PubMed  Garrett P. 2010. Attitudes to Language . Cambridge University Press. Google Scholar CrossRef Search ADS   Gass R., Seiter J.. 2003. Persuasion, Social Influence and Compliance Gaining , 2nd edn Allyn and Bacon. Gaudin F. 1993. Pour Une Socioterminologie: Des Problemes Sémantiques Aux Pratiques Institutionnelles . Publications de l’Université de Rouen. Geeraerts D., Grondelaers S., Bakema P.. 1994. The Structure of Lexical Variation: Meaning, Naming, and Context . De Gruyter. Google Scholar CrossRef Search ADS   Gutiérrez B. 2005. El Lenguaje De Las Ciencias . Editorial Gredos. Halliday M. A. K., Hasan R.. 1976. Cohesion in English . Longman. Hamon T., Nazarenko A.. 2001. ‘Detection of synonymy links between terms: Experiment and results’ in Bourigault D., Jacquemin C., L’Homme M. C. (eds): Recent Advances in Computational Terminology . John Benjamins Publishing Company, pp. 185– 208. Google Scholar CrossRef Search ADS   Hartley J., Kostoff R. N.. 2003. ‘ How useful are keywords in scientific journals?,’ Journal of Information Science  29: 433– 8. Google Scholar CrossRef Search ADS   Hill S. S., Soppelsa B. F., West G. K.. 1982. ‘ Teaching ESL students to read and write experimental-research papers,’ TESOL Quarterly  16: 333– 47. Google Scholar CrossRef Search ADS   Hoffmann L. 1998. Llenguatges D’Especialitat. Selecció De Textos . IULA-Universitat Pompeu Fabra. Holden M. T., Lynch P.. 2004. ‘ Choosing the appropriate methodology: Understanding research philosophy,’ The Marketing Review  4: 397– 409. Google Scholar CrossRef Search ADS   ISO/704. 2000. Principles and Methods of Terminology. International Standardization Office. Jacquemin C. 2001. Spotting and Discovering Terms through Natural Language Processing Techniques . MIT Press. Kageura K. 2002. The Dynamics of Terminology: A Descriptive Theory of Term Formation and Terminological Growth . John Benjamins Publishing Company. Google Scholar CrossRef Search ADS   Kerremans K. 2017. ‘Applying computer-assisted coreferential analysis to a study of terminological variation in multilingual parallel corpora’ in Menzel K., Lapshinova-Koltunski E., Kunz K. (eds): New Perspectives on Cohesion and Coherence: Implications for Translation . Language Science Press, pp. 49– 75. Kocourek R. 1991. La Langue Française De La Technique Et De La Science: Vers Une Linguistique De La Langue Savante . Oscar Brandstetter. Messineo C. 2002. ‘ Variación Conceptual y Formal del Término Educación Bilingüe Intercultural (ebi) en Distintos Tipos de Discursos,’ Terminology  8: 113– 19. Google Scholar CrossRef Search ADS   Oppenheim A. N. 2000. Questionnaire Design, Interviewing and Attitude Measurement , 2nd edn Bloomsbury Academic. Ortony A. (ed.). 1999. Metaphor and Thought . Cambridge University Press. Pecman M. 2012. ‘ Tentativeness in term formation: A study of neology as a rhetorical device in scientific papers,’ Terminology  18: 27– 58. Google Scholar CrossRef Search ADS   Pecman M. 2014. ‘ Variation as a cognitive device: how scientists construct knowledge through term formation,’ Terminology  20: 1– 24. Google Scholar CrossRef Search ADS   Picton A., Dury P.. 2017. ‘Diastratic variation in language for specific purposes: Observations from the analysis of two corpora’ in Drouin P., Francoeur A., Humbley J., Picton A.. (eds): Multiple Perspectives on Terminological Variation . John Benjamins Publishing Company, pp. 57– 80. Google Scholar CrossRef Search ADS   Riggs F. W. 1993. ‘Social science terminology: Basic problems and proposed solutions’ in Sonneveld H. B., Loening K. L. (eds): Terminology. Applications in Interdisciplinary Communication . John Benjamins Publishing Company, pp. 195– 221. Google Scholar CrossRef Search ADS   Rondeau G. 1984. Introduction à La Terminologie . Gaëtan Morin. Sager J. C. 1990. A Practical Course in Terminology Processing . John Benjamins Publishing Company. Google Scholar CrossRef Search ADS   Suárez M. 2004. Análisis Contrastivo De La Variación Denominativa En Textos Especializados: Del Texto Original Al Texto Meta . Universitat Pompeu Fabra. Available at http://www.tdx.cat/TDX-0217105-130025. Swales J. 1990. Genre Analysis: English in Academic and Research Settings . Cambridge University Press. Temmerman R. 2000. Towards New Ways of Terminology Description: The Sociocognitive Approach . John Benjamins Publishing Company. Google Scholar CrossRef Search ADS   Tercedor M. 2011. ‘ The cognitive dynamics of terminological variation,’ Terminology  17: 181– 97. Google Scholar CrossRef Search ADS   Wüster E. 1979. Introducción a La Teoría General De La Terminología y a La Lexicografía Terminológica . IULA-Universitat Pompeu Fabra. © Oxford University Press 2018

Journal

Applied LinguisticsOxford University Press

Published: Mar 5, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off