Iberica 13 Ibérica 23 (2012): 47-64 ISSN 1139-7241 Abstract This paper aims at analysing the role of pragmatic transfer in the use of the rhetorical strategy of hedging in academic writing in English as a second language (henceforth, L2). Two groups of Spanish researchers (n=30) took part in the study, performing two different experiments. The first one consisted of reading two versions of the same passage, one hedged and the other unhedged, and they were asked to decide which text they considered more academic and why. The second experiment consisted of identifying hedges in a passage. The results indicate that Spanish researchers apply pragmatic transfer, either frequently failing to identify hedges in the L2 or considering them as negative evasive concepts. Keywords: English for Academic Purposes, hedging, nonnative English readers, pragmatic transfer. Resumen Los matizadores: un estudio experimental sobre la transferencia pragmática en las preferencias retóricas de lectores no nativos de inglés Este artículo analiza el papel de la transferencia pragmática en el uso de los matizadores discursivos como estrategia retórica en el lenguaje académico del inglés como segunda lengua (en adelante, L2). Dos grupos de investigadores españoles (n=30) han tomado parte en este estudio, habiendo realizado dos experimentos diferentes. El primero consistía en leer dos versiones del mismo fragmento, una en la que se habían incluido matizadores y otra en la que no, para que decidiesen qué texto consideraban más académico y por qué. El segundo experimento consistía en identificar los matizadores de un fragmento. Los Hedging: An exploratory study of pragmatic transfer in nonnative English readers’ rhetorical preferences Rosa Alonso Alonso, María Alonso Alonso and Laura Torrado Mariñas Universidade de Vigo (Spain) ralonso@uvigo.es, malonsoalonso@uvigo.es, ltorrado@uvigo.es 47 Ibérica 23 (2012): 47-64 R. ALOnSO ALOnSO, M. ALOnSO ALOnSO & L. TORRADO MARIñAS resultados obtenidos indican que los investigadores españoles llevan a cabo una transferencia pragmática al usar dichos elementos, bien al no ser capaces de identificar los matizadores en la L2 o bien al considerarlos como conceptos negativos de evasión en el discurso académico. Palabras clave: Inglés para Fines Académicos, matizadores discursivos, lectores no nativos de inglés, transferencia pragmática. 1. Introduction Scholars around the world are increasingly forced to publish their research papers in English in order to become part of an international community which is, in fact, governed by a number of pragmatic and content-dependent conventions that can sometimes be difficult to identify and respect. Academic productions usually contain a series of elements that need to be respected for the text to be recognised as “scholarly” or “academic”. It cannot be forgotten that hedging is not the only rhetorical device that identifies a text as academic or non-academic, other elements such as lexis also provide academic nature to a text, but one cannot overlook the importance of hedging when determining whether a given text is academic or not. This rhetorical device is widely used in English academic writing (Hyland, 1998; Hyland & Bondi, 2006). The present paper looks into the rhetorical preferences of two groups of nonnative English researchers in order to determine whether pragmatic transfer affects the perception of hedging in their academic reading. The study is divided into the following sections. Section 2 deals with the connection between pragmatic transfer and hedging, section 3 focuses on hedging in academic English. The following section contains the statement of purpose. The study is detailed in section 5, it includes two experiments, one in which the subjects had to choose between a hedged and an unhedged passage stating the reasons for their choice and a second one in which they were asked to identify hedges in a passage. Both experiments aimed at analysing whether pragmatic transfer affects nonnative writers’ academic reading. Finally, the conclusions are included in section 6. 2. Pragmatic transfer and hedging Various aspects of pragmatic transfer have been particularly studied since the late 1980s. Odlin (1989) referred to it as “discourse transfer”, 48 highlighting the rhetoric implication of the term; whereas Wolfson (in Bou, 1998) and Kasper (1992) put the emphasis on the sociolinguistic and sociocultural significance of this concept. There is a series of factors that can favour pragmatic transfer in certain contexts. For example, the influence of the source language on the target language will depend on the level of proficiency and cultural knowledge that the speaker of the latter has. Moreover, it should also be taken into account that the relationship that the interlocutors (writer and readers in this case) establish, will also determine the kind of register and the strategies for assisting the communication. Consequently, as Bou (1998) indicates, pragmatic transfer implies an interdisciplinary and empirical approach depending on both, the “pragmalinguistic” and “sociopragmatic” aspects of the variation. This distinction is based on the inseparable relationship that Kasper (1992) points out between language and culture. On the one hand, “pragmalinguistic transfer” makes reference to the transfer of particular politeness values from a first language (henceforth L1) into an L2. In this case, the linguistic proficiency that the user has of the target language will determine the communicative strategies that might be influenced by pragmatic transfer from the native language. On the other hand, “sociopragmatic transfer” will depend on the context and the relationship that exists between the participants. Hence, the lack of cultural information related to the target language might work as a constraint due to the fact that the “performance of linguistic actions in L2 [is] influenced by [the] assessment of subjectively equivalent L1 contexts” (Kasper, 1992: 209). Since hedges are specifically considered as an interpersonal rhetorical strategy, transfer can occur both at a pragmalinguistic and sociopragmatic level according to the knowledge that the participant has of the target language. For instance, politeness is probably a universal notion but the communicative strategies that might favour the transfer of hedges will vary depending on the motivating factors. Indeed, cross-cultural studies on the phenomenon of hedging have opened new avenues of research. For example, german scholars use hedges when they write in English more often than native users due to the fact that hedging is a common communicative strategy in german (Clyne, 1991). Similarly, Bulgarian writers, as Vassileva (1997) acknowledges, transfer the distribution of hedges among the paragraphs according to the convention in their native language. According to Hu, Brown and Brown (1982), Chinese L2 writers tend to be authoritative and use stronger modals than native speakers of English. HEDgIng: An ExPLORATORy STuDy Ibérica 23 (2012): 47-64 49 Indeed, the use of hedging is conditioned by the subjectivity of the individuals, on the particular contexts where the communication is established and on the relative knowledge that the person has of the target language. Comparative analysis has shown that pragmatic transfer regarding the use of hedges certainly takes place (Clyne, 1991; Vassileva, 1997; Martín, 2008). This is probably due to the fact that writers use similar rhetorical strategies that favour language transfer. Actually, as Odlin (1989) suggests, there are differences from most languages in relation to politeness and discourse which might affect transfer. This will occur “[i]f native language patterns influence learners [of a target language] in inappropriate ways” (Odlin, 1989: 48) and, therefore, the language that a person uses may seem impolite or incoherent. As a matter of fact, cross- linguistic differences in these cases might affect not only production but also perception. The use of hedges in Spanish is usually connected with politeness structures also known as cortesía asertiva. The aim of these devices is to attenuate an assertion mainly by the employment of verbs or subordinate phrases that imply a certain lack of commitment that the individual has with the manifested proposition. Moreover, there are a number of sociological factors also in the Spanish context that determine the use of politeness structures. For instance, Mulder (1991) highlights the principles of authority, social distance and culture as the main ground for hedging in Spanish. Accordingly, it appears that the communicative strategies regarding the use of hedges in English and Spanish are similar. Even so, and as Haverkate (2002) indicates, the pragmatic connotations in the use of hedges in the two languages differ in their implications. The pragmatic content of hedges in Spanish – also referred to as doxastic predicates – is usually aimed to set aside a statement that is subject to question. Therefore, leaving politeness structures aside, it seems that hedges in English and in Spanish have different functions. Martín (2003a) has noticed that English scholars use the strategy of indeterminacy in their works to a much greater extent than their Spanish counterparts. It seems that on the one hand, English scholars are more careful about stating their claims using hedging devices in order to avoid the audience’s rejection, to mitigate their critical speech acts or to maintain a social distance between readers and writers. On the other hand, Spanish scholars tend to include a higher degree of authorial presence by using strategies, such as, for instance, direct claims. Hence, hedges are R. ALOnSO ALOnSO, M. ALOnSO ALOnSO & L. TORRADO MARIñAS Ibérica 23 (2012): 47-6450 considered to be used in English as part of a formal convention whereas in Spanish, they possibly imply that the claims made are tentative or indeterminate. This might be the reason why Spanish native speakers appear to be reluctant to use hedges as part of their discourse when they use English as the target language. 3. Hedging in academic English Ever since Lakoff first introduced the term “hedge” in 1972 to refer to expressions that modify the category membership of a predicate or noun phrase in their ability to “make things fuzzier or less fuzzy” (Lakoff, 1972: 195), the term has undergone a number of changes, widening its initial definition. Hence, the communicative value of hedges has been increasingly explored insofar as their position as modifiers of the speaker’s commitment to the truth-value of a whole proposition and their actual use as politeness strategies. nonetheless, the wide number of converging theories can sometimes make the study of such devices quite vague due to the ample number of linguistic aspects and theories involved in their categorisation. In fact, Markkanen and Schröder wrote (1997: 15): The concept [of hedges] has lost some of its clarity and sometimes seems to have reached a state of definitional chaos, as it overlaps with several other concepts. This problem concerns many other linguistic concepts and their definitions, beginning with the concept of ‘language’ itself. Broadly speaking, hedging studies have come to the conclusion that such rhetorical devices can be used for blurring the speaker’s commitment to the truth of the preposition conveyed (Prince, Frader & Bosk, 1982). They can also be used to express indetermination (Hübler, 1983). More importantly, and as far as academic writing is concerned, they can be used either as a strategy of negative politeness with the function of avoiding disagreement (Brown & Levinson, 1987) or as a way of avoiding being proved wrong later on, thus leaving researchers the door open to assert that their previous claims were only tentative, as Markkanen and Schröder (1997: 6) acknowledge when they claim that “hedges offer a possibility for textual manipulation in the sense that the reader is left in the dark as to who is responsible for the truth value of what is being expressed”. Some linguists, dissatisfied with the lack of consensus on what the term “hedge” denotes, have attempted to clarify the phenomenon of hedging by HEDgIng: An ExPLORATORy STuDy Ibérica 23 (2012): 47-64 51 proposing a test or a “functional definition” that would allow the proper identification of hedged propositions (Crompton, 1997 & 1998), thus overlooking the importance of the mental or psychological essence of the linguistic concept. However, such approaches prove to be essentially reductionist, seeing hedges as expressions that have the exclusive function of avoiding commitment. In recent years researchers have started to be aware of the multiplicity of functions of these devices and, more importantly, of their actual dependency on context to work or not, as such. As Clemen (1997: 237) puts forth: Hedges are determined by context, the colloquial situation and the speaker’s/writer’s intention, plus the background knowledge of the interlocutors. Hedging cannot be deduced only from the combination of the individual clausal elements plus the relevant illocution. Hedges function in a particular context. (italics added) In line with the statement mentioned above, the classification provided in this article will be strongly dependent on the socio-pragmatic context in which hedges occur, because, as Martín (2003b: 66) acknowledges, “it appears that it is virtually impossible to attribute a function to a hedge without considering both the linguistic and situational context”. According to Martín (2003b), the linguistic devices that can work as hedges at a lexico-grammatical and syntactic level can be categorised as performing the following basic strategies: 1. Strategy of indetermination, thus endowing the proposition with a certain shade of lesser qualitative and quantitative explicitness, vagueness and uncertainty. This strategy includes performers of epistemic modality, such as modal verbs expressing possibility (“may”/ “might”/ “can”), verbs of cognition, like “seem to”, “or appear to”; epistemic verbs drawing on the probability of the proposition or hypothesis expressed being true, like “to assume” or “to suggest”; modal adverbs (“probably”, “possibly”), modal nouns (“suggestion”, “possibility”) and, finally, modal adjectives (“probable”, “possible”). It also includes approximators of frequency, quantity, degree and time, indicating an unwillingness to clarify the writer’s actual commitment to the proposition, such as “generally”, “approximately”, “frequently”. 2. Strategy of camouflage hedging (as proposed by namsaraev, R. ALOnSO ALOnSO, M. ALOnSO ALOnSO & L. TORRADO MARIñAS Ibérica 23 (2012): 47-6452 1997), which comprises metalinguistic operators, which provoke the displacement of the focus of the reader’s attention and negative reaction from the proposition. Among these we can highlight: “really”, “actually”, “in fact”, “generally speaking”, etc. 3. Strategy of subjectivisation, consisting of the use of first personal pronouns followed by verbs of cognition as a means of highlighting the subjective nature of the given propositions, thus inviting the reader to agree or disagree with them from a neutral position. It also includes expressions constituted by first person pronouns (as proposed by Salager-Meyer, 1994), because they express the author’s personal doubt and direct involvement, like “to our knowledge”, “in our view”, “in my experience”. A third group consisted of quality-emphasising adjectival and adverbial expressions (labelled by Salager-Meyer as “emotionally-charged intensifiers”), used to convince the reader of the importance of the expressed propositions by exposing the writer’s emotional state. These include “extremely interesting”, “particularly important”, etc. Martín is careful to point out here that Myers (1989) considers such intensifiers as markers of positive politeness because “they can be seen as showing solidarity with the discourse community by exhibiting responses that assume shared knowledge and desires” (Martín, 2003b: 69). 4. Strategy of depersonalisation, where writers attempt to blur their presence by using a variety of impersonal passive constructions without obvious agents, hence relieving themselves of the inherent responsibility contained in the propositions expressed. Such depersonalisation can be achieved by means of agentless passive and impersonal constructions, for instance: “In this research data were analysed to find out” instead of “In this research, I analysed a number of data to find out”. It also includes impersonal active constructions, in which a non-human entity is used as subject so as to detach the writers from their findings, such as: “for instance”, “the results suggest”, etc. In spite of the clear classification provided by Martín (2003b), we cannot forget that hedging devices have a polysemous and polypragmatic nature, as Hyland (1998: 158) proposed: HEDgIng: An ExPLORATORy STuDy Ibérica 23 (2012): 47-64 53 While the functions of hedges are important, they are not always apparent to insiders at a conscious level of awareness. The view taken here therefore predicts that there will invariably be some indeterminacy between hedging functions, so cases assigned with one category will include meanings associated with another. Therefore one must be cautious in the identification of hedges since hedging functions can be indeterminate. 4. Statement of purpose The results found out in previous studies regarding hedges in English as a Second Language (Odlin, 1989; Clyne, 1991; Clemen, 1997; Vassileva, 1997; Hyland, 1998; and Martín, 2008) have led us to propose the following research hypotheses: 1. native speakers of Spanish will be influenced by their native language in the identification of hedges in an English-language text. 2. native speakers of Spanish will reject the use of hedges as negative evasive concepts due to pragmatic transfer. 5. The study 5.1. Subjects Two groups of participants took part in this study. The purpose of including two groups of the same level in the study was twofold: first to determine that the results obtained in the first group could be replicated since no significant differences were expected between both of them and secondly to have a wider range of disciplines represented in the study. They were all members of the scientific community and worked as teachers/researchers at the universidade de Vigo. These groups were selected according to their level, they all showed an upper-intermediate level of English. The first group consisted of 14 teachers from the following disciplines: marine sciences (n=3), chemistry (n=1), law (n=1), biology (n=3), economics (n=3), translation (n=1), engineering (n=1), computer science (n=1). Their native language is Spanish and they have been learning English for more than fifteen years (90%) and for five years (10%) and 5% of them have attended R. ALOnSO ALOnSO, M. ALOnSO ALOnSO & L. TORRADO MARIñAS Ibérica 23 (2012): 47-6454 other academic writing courses. The second group consisted of 16 subjects belonging to the following research fields: law (n=2), science (n=9), business administration (n=3), computer science (n=1), biology (n=1). Their native language is Spanish (80%) and galician (10%). They have been learning English for more than fifteen years, since they began secondary education and they have not attended any other academic writing courses. none of the subjects in both groups had been introduced to the concept of hedging before participating in the experiment. 5.2. Materials Two research materials were used: 1. Experiment 1: The subjects were asked to read two versions of the same text, one hedged and the other unhedged. They were also asked to state the reasons for the choice of one version or the other (see Appendix) by asking them which version they found more academic and why. The purpose of this experiment was to find out whether they identified hedges as an important part of academic writing style in English or whether they rejected them as uncertain and indirect due to pragmatic transfer. In the first research instrument, the hedged passage contained the following rhetorical devices: a metalinguistic operator, verbs of cognition and modal verbs, according to Martín’s (2003b) classification. 2. Experiment 2: In the second experiment, they were asked to underline the words in a text which made it show an academic writing style – that is, to identify hedges. The aim of this exercise was to analyse whether they identified hedges as an important part of writing style or whether they did not perceive them. For both instruments (see Appendix), the texts were selected passages adapted from a paper by Fitzpatrick and Meara (2004). This second research instrument contained verbs of cognition, impersonal active constructions and modal verbs. Martín’s (2003b) classification was applied, as explained in section 3. 5.3. Procedure and data analysis The tests were given to the participants during a 12 hour-course on academic writing which took place for four days at the universidade de Vigo. They HEDgIng: An ExPLORATORy STuDy Ibérica 23 (2012): 47-64 55 were told that these were classroom exercises that would be corrected at the end of the course. It took them fifteen minutes to do the first experiment and twenty minutes to do the second. Both experiments were conducted on the third class day. One of the researchers collected the data. Copies of both experiments were given back to the students on the last day and both texts were corrected in order to improve the quality of their academic reading. As regards the statistical analysis, a percentage analysis was carried out and Fisher’s exact test was used for the statistical analysis of the data. The results obtained are included and discussed in the following section. 5.4. Results and discussion The first group seemed to deal worse with the identification of the hedged passage. Only 42.8% of them chose the hedged text as the most academic while a higher percentage (62.5%) did so in the second group. It may be the case that the subjects in the second group exhibit a better proficiency level in the second language. However, the difference between both groups is not statistically significant, as can be seen in Table 1. The main issue regarding pragmatic transfer lies in the fact that none of the groups obtained total success in the recognition of the hedged text as the closest to academic style. From the results obtained in the second experiment, it can be seen that the second group identifies more hedges, which is also supported by the fact that this group outperformed the first in the identification of the hedged passage. As regards the type of hedges identified, in both groups the verb of cognition “seem to” to and the modal verb “might” constitute the most clearly identified, followed by the impersonal active constructions. However, the differences between both groups are not significant. What matters is that both groups have difficulties in identifying the hedges in the passage. This can be seen in Table 2, which includes the percentages with raw figures in parentheses and p-values of each of the examples analysed. Hedge Group 1 Group 2 R. ALOnSO ALOnSO, M. ALOnSO ALOnSO & L. TORRADO MARIñAS Ibérica 23 (2012): 47-6456 Text 1 Text 2 First group 42.8% 57.2% Second group 62.5% 37.5% Table 1. Results of the first experiment. F The final results obtained from the analysis of both groups confirm that the hedges have not been detected totally, as Table 3 shows. In statistical terms the lack of identification of the rhetorical devices in the hedged passage is significant. Most cases obtain a p-value 0.000 while only the two instances of the verb of cognition “seem to” and the modal verb “might” obtain a p-value 0.006 and 0.002. The statistical results obtained from the application of Fisher’s exact test show that the percentage of success in the identification of hedges constitutes less than 70% – that is, all percentages obtained are significantly smaller than this. The subjects fail to identify all the hedges HEDgIng: An ExPLORATORy STuDy Ibérica 23 (2012): 47-64 57 Table 2. Results of the second experiment. Hedge Group 1 Group 2 Comparison of groups 1 & 2 Verb of cognition (“seemed to be”) 21% (3) p-value: 0.000 68% (11) p-value: 0.550 p-value: 0.014 Impersonal active (“the results indicate”) 35% (5) p-value: 0.008 31% (5) p-value: 0.002 p-value: 1.000 Impersonal active (“we suggest”) 21% (3) p-value: 0.000 50% (8) p-value: 0.074 p-value: 0.142 Modal verb (“might”) 35% (5) p-value: 0.008 37% (6) p-value: 0.007 p-value: 1.000 Verb of cognition (“seems to”) 21% (3) p-value: 0.000 62% (10) p-value: 0.340 p-value: 0.033 Impersonal active (“assumes”) 21% (3) p-value: 0.000 43% (7) p-value: 0.026 p-value: 0.260 Impersonal active (“indicates”) 14% (2) p-value: 0.000 31% (5) p-value: 0.002 p-value: 0.399 Modal verb (“might”) 35% (5) p-value: 0.008 50% (8) p-value: 0.074 p-value: 0.484 Verb of cognition (“appears”) 35% (5) p-value: 0.008 37% (6) p-value: 0.007 p-value: 1.000 Hedge Total g ( pp ) 3 Table 3. Final results for both groups. Verb of cognition (“seemed to be”) 44.5% (14) p-value: 0.006 Impersonal active (“the results indicate”) 23% (10) p-value: 0.000 Impersonal active (“we suggest”) 35.5% (11) p-value: 0.000 Modal verb (“might”) 36% (11) p-value: 0.000 Verb of cognition (“seems to”) 41.5% (13) p-value: 0.002 Impersonal active (“assumes”) 32% (10) p-value: 0.000 Impersonal active (“indicates”) 22.5% (7) p-value: 0.000 Modal verb (“might”) 42.5% (13) p-value: 0.002 Verb of cognition (“appears”) 36% (11) p-value: 0.000 occurring in the passage. Pragmatic transfer may be at work and the subjects in both groups cannot identify the hedges in the passage as key characteristics of academic writing style, which confirms our first hypothesis. A similar result was found by Vassileva (1997), as mentioned in section 2. This author found out that L1 Bulgarian made subjects transfer the distribution of hedges among the paragraphs not according to the conventions of the L2 but based on their use in the L1. The pragmatic content of hedges in Spanish is not tentative or indeterminate; it tends to set aside a statement that is subject to question. Therefore the use of hedges that imply indetermination or depersonalization is associated with lack of clarity, insecurity and lack of validity of the proposal being expressed – as the opinions of the subjects who chose the unhedged text confirm (see Table 4) –, while in English politeness and respect for the scientific community are embedded in the meaning of the hedge. Both scientific communities differ in the functions they assign to the hedge making transfer operate across the two discourse communities. While as Brown and Levinson (1987) mention (see section 3) that hedges can be used to avoid disagreement, our subjects’ L1 influence makes them perceive hedges as elements implying lack of clarity or insecurity. This can be clearly observed in the reasons the subjects of both groups provide for their choice of the unhedged passage (see Table 4). Besides, hedges in English somehow imply the judgement of readers, the assessment they make of the proposal conveyed and hedged by the writer. The evasive concept they imply is not shared by Spanish academic readers since it is perceived as a negative lack of commitment, as Table 4 indicates when they consider the unhedged passage more conclusive, more clear, more categorical and less uncertain, confirming our second hypothesis. When pragmatic transfer takes place, these rhetorical devices may not be perceived as markers of academic writing style by readers or they may be understood as R. ALOnSO ALOnSO, M. ALOnSO ALOnSO & L. TORRADO MARIñAS Ibérica 23 (2012): 47-6458 First group Second group It is more clear and direct (14.5%) It is less uncertain (14.5%) It does not have unnecessary redundancies (14.5%) It is simpler (14.5%) It is more conclusive (14.5%) It shows more security in the statements (14.5%) I don´t know (13%) It is more assertive (33.3%) It is more categorical (33.3%) It is clear and direct (33.3%) Table 4. Reasons for the choice of the unhedged passage as more academic. markers of insecurity in the discourse, as is shown in the assertions made by our subjects. L2 readers who are unaware of the presence and/or function of hedges do not perceive the connotations and formal conventions conveyed by these rhetorical devices; therefore, they do not share the principles of writing of the L2 scientific community. As Markkanen and Schröder (1997) mention, a hedge only exists if it is perceived by author and reader. Apparently, the subjects who selected the unhedged passage in this study perceive hedges but they consider that the lack of them makes the passage sound more direct or clear. This can be attributed to the perception of hedges as markers of insecurity. In other words, L2 readers who understand hedges do not seem to perceive them as markers of commitment to the truth of a proposition but as elements which avoid the categorical assertions typical of assertive statements in their native language. The strategy of indetermination and that of depersonalisation, mentioned by Martín (2003b), were identified in less than 50% of the cases – that is, hedges were identified only partially. The subjects did have trouble in identifying these devices and in only in 44.5% and 41.5 % of the cases did they identify the two occurrences of the verb of cognition as a hedge. These were the highest percentages of identification of a hedge. In 36% and 42.5% of the cases the modal verb was also perceived as a hedge. The verb of cognition “appear” was seen as a difficult example since only 36% of the subjects identified it. As Wishnoff (2000) states, direct writing typically distinguishes nonnative writers from their native counterparts, in other words pragmatic transfer inhibits the use of rhetorical devices that are not common to the native language. However, we cannot overlook the possibility that the deficiency in the identification of hedges may be related to the academic orientations of the readers or to the individual preferences related to their disciplinary or institutional backgrounds. Further studies should be carried out to determine the influence of other variables on the identification of hedges. 6. Conclusion Pragmatic transfer seems to operate in the identification of hedges. Spanish academic readers consider these rhetorical devices as elements leading to confusion and evasive statements which should not be used in academic writing, instead of perceiving them as downtoners in academic discourse. As HEDgIng: An ExPLORATORy STuDy Ibérica 23 (2012): 47-64 59 Wishnoff (2000) mentions, nonnative writers are easily distinguishable by their direct writing when compared with their native counterparts. The subjects are influenced by the rhetorical structure of their L1 where hedges are considered tentative and unclear, lacking the clarity that academic proposals require. The use of rhetorical devices is identified with difficulty and the subjects fail to distinguish hedges. It is quite obvious that pragmatic transfer makes the choice of rhetorical devices differ in both discourse communities (Vassileva, 1997). However, other possible factors contributing to the identification of hedges should be analysed such as the disciplinary or institutional background of the subjects or the academic orientations of the readers since more than the pragmatic transfer variable may be involved. Our study seems to indicate that hedges in academic English are perceived as indicators of a negative lack of commitment by native speakers of Spanish and pragmatic transfer appears to inhibit the use of hedges that are not common to the native language. However, we cannot forget that this is an exploratory study with a limited number of subjects, further studies with larger groups should be carried out to determine whether the results obtained in this study are replicated and can be extended to other L2 learners. It would be interesting to carry out a study comparing groups with different proficiency levels in order to test whether pragmatic transfer is inhibited in the perception of rhetorical devices as more advanced learners approach the L2 norm. A comparative study on the disciplinary and/or background of different groups of subjects should also be undertaken to check the influence of other variables in the identification of hedges. Moreover, the increasing demand for courses in English for research publication purposes seems to show an increasing interest to apply a cross- cultural perspective into academic writing. Accordingly, Spanish scholars may need to be encouraged to modify their critical and stylistic habits in order to communicate accurately and engage with an international audience. Acknowledgements We would like to acknowledge Juan Carlos Pardo-Fernández (Department of Statistics, universidade de Vigo) for the statistical analysis of the data and Professor Terence Odlin (Ohio State university) for his comments on a previous draft of this paper. We would also like to thank two anonymous reviewers for their comments on an earlier version of this article. All errors R. ALOnSO ALOnSO, M. ALOnSO ALOnSO & L. TORRADO MARIñAS Ibérica 23 (2012): 47-6460 remain our responsibility. This piece of research has been supported by xunta de galicia, research project InCITE08PxIB204033PR. [Paper received 21 September 2010] [Revised paper accepted 17 February 2011] References HEDgIng: An ExPLORATORy STuDy Ibérica 23 (2012): 47-64 61 Bou Franch, P. (1998). “On pragmatic transfer”. Studies in English Language and Linguistics 0: 5- 20. Brown, P. & S. Levinson (1987). Politeness: Some Universals in Language Usage. Cambridge: Cambridge University Press. Clemen, G. (1997). “The concept of hedging: origins, approaches, definitions” in R. Markkanen & H.Schröder (eds.), 235-249. Clyne, M. (1991). “The sociocultural dimension: The dilemma of the German-speaking scholar” in H. Schröder (ed.), Subject-Oriented Texts, 49-68. Berlin: Mouton de Gruyter. Crompton, P. (1997). “Hedging in academic writing: Some theoretical aspects”. English for Specific Purposes 16: 271-289. Crompton, P. (1998). “Identifying hedges: definitions or divination”. English for Specific Purposes 17: 303-313. Fitzpatrick, T. & P. Meara (2004). “Exploring the validity of a test of productive vocabulary”. Vigo International Journal of Applied Linguistics 1: 55- 74. Haverkate, H. (2002). The Syntax, Semantics and Pragmatics of Spanish Mood. Amsterdam: John Benjamins. Hu, Z., D.F. Brown & L.B. Brown (1982). “Some linguistic differences in the written English of Chinese and Australian students”. Language Learning and Communication 1: 39-49. Hübler, A. (1983). Understatements and Hedges in English. Amsterdam: John Benjamins. Hyland, K. (1998). Hedging in Scientific Research Articles. Amsterdam: John Benjamins. Hyland, K. & M. Bondi (2006). Academic Discourse across Disciplines. Bern: Peter Lang. Kasper, G. (1992). “Pragmatic transfer”. Second Language Research 8: 203-231. Lakoff, G. (1972). “Hedges: A study in meaning criteria and the logic of fuzzy forms”. Journal of Philosophical Logic 2: 458-508. Markkanen, R. & H. Schröder (1997). Hedging and Discourse. Approaches to the Analysis of a Pragmatic Phenomenon in Academic Texts. Berlin: Mouton de Gruyter. Martín Martín, P. (2003a). “A genre analysis study of English and Spanish research article abstracts in the experimental social sciences”. Ph.D. thesis. URL: ftp://tesis.bbtk.ull.es/ccssyhum/cs143.pdf [10/10/09]. Martín Martín, P. (2003b). “The pragmatic rhetorical strategy of hedging in academic writing”. Vigo International Journal of Applied Linguistics 0: 57-72. Martín Martín, P. (2008). “The mitigation of scientific claims in research papers: A comparative study”. International Journal of English Studies 8: 133-152. Mulder, G. (1991). “Aproximarse y evitar: estrategias de cortesía en castellano”. Foro Hispánico 2. Exploraciones semánticas y pragmáticas del español 2: 69-80. Myers, G. (1989). “The pragmatics of politeness in scientific articles”. Linguistics 10: 1–35. Namsaraev. V. (1997). “Hedging in Russian academic writing in sociological texts” in R. Markkanen & H. Schröder (eds.), 64-79. Odlin, T. (1989). Language Transfer. Cambridge: Cambridge University Press. Prince, E.F., J. Frader & C. Bosk (1982). “On hedging in physician-physician discourse” in R. di Pietro (ed.), Proceedings of the Second Annual Delaware Symposium on Language Studies, 83- 97. Norwood: Ablex Publishing Corporation. Salager-Meyer, F. (1994). “Hedges and textual communicative function in medical English written discourse”. English for Specific Purposes 13: 149- 170. Vassileva, I. (1997). “Hedging in English and Bulgarian academic writing” in A. Duszak (ed.), Culture and Styles in Academic Discourse, 141- 157. Berlin: Mouton de Gruyter. Rosa Alonso Alonso is a Senior Lecturer at the Department of English at the universidade de Vigo. She has published a number of articles on Second Language Acquisition. She has collaborated in many research projects and has attended national and international conferences. She is currently a member of the SPERTuS research team and the editor of the Vigo International Journal of Applied Linguistics (VIAL). Maria Alonso Alonso holds a BA in Engish Literature and another BA in Hispanic Philology, both earned at the universidade de Vigo. She is currently towards her PhD, after having completed an MA in English Studies and an MA in Applied Linguistics. She has published articles on language and literature in national and international journals while combining her research with teaching activities within one of the I2C predoctoral contracts awarded by the xunta de galicia. After having earned her BA in English Studies and her MA in Advanced English Studies at the universidade de Vigo, Laura Torrado Mariñas enjoys a predoctoral contract and is currently working in her PhD. thesis. Apart from having attended a number of conferences, both national and international, and having published a variety of articles in academic journals, she has taught graduate and postgraduate courses on English as a Foreign Language and English Literature. R. ALOnSO ALOnSO, M. ALOnSO ALOnSO & L. TORRADO MARIñAS Ibérica 23 (2012): 47-6462 Wishnoff, J.R. (2000). “Hedging your bets: L2 learners’ acquisition of pragmatic devices in academic writing and computer-mediated discourse”. Second Language Acquisition Studies 19: 119-148. Appendix HEDgIng: An ExPLORATORy STuDy Ibérica 23 (2012): 47-64 63 M Experiment 1: 1. Which text do you consider more academic: text 1 or text 2? ………………………………………………… 2. Why? ………………………………………………………………………………………………………………… Text 1 A closer look at the statistics, though, shows us that the difference between the scores of the two groups is not an absolute one. As illustrated in Figure 6, which shows the number of native speaker and non-native speaker cases falling within each band of scores, there seems to be a good deal of overlap between the scores of the groups. These results raise two important issues about the way Lex30 measures the productive lexicon. Firstly, there appears to be a broad but distinct difference between the scores achieved by native and non-native speakers. Secondly, there is a considerable degree of overlap between the scores of the two groups. We should perhaps not be surprised about the variation in native speaker scores; while in theory Bachman believes native speakers should provide us with an effective control group, the complexities of their language use can make this a problematic choice in reality. Text 2 A closer look at the statistics, though, shows us that the difference between the scores of the two groups is not an absolute one. As illustrated in Figure 6, which shows the number of native speaker and non-native speaker cases falling within each band of scores, there is a good deal of overlap between the scores of the groups. These results raise two important issues about the way Lex30 measures the productive lexicon. Firstly, there is a broad but distinct difference between the scores achieved by native and non-native speakers. Secondly, there is a considerable degree of overlap between the scores of the two groups. We should perhaps not be surprised about the variation in native speaker scores; while in theory Bachman believes native speakers provide us with an effective control group, the complexities of their language use make this a problematic choice in reality. Experiment 2: Underline the words that you consider make this text show an academic writing style: We began our exploration of the Lex30 test by identifying a need for an effective test of productive vocabulary. The design of the Lex30 test seemed to be an attractively simple way of meeting this need. The results also indicate that the elicited vocabulary is being measured with some accuracy too. We still need to explain, though, the lack of a strong correlation between Lex30 and the other two tests, and we suggest that this is due to the fact that the tests are measuring different aspects of vocabulary knowledge. However, we believe that using a more up-to-date set of frequency bands might improve the accuracy of the Lex30 measure. The test seems to demand knowledge of form, meaning and collocation of target words, as well as understanding of the contextual cue sentence. Producing a word in response to the Lex30 task implies a minimal level of productive knowledge .There is a need among teachers, learners and researchers for an effective battery of test tools. An expectation of high correlations between the tests assumes that all three tests measure productive vocabulary knowledge exclusively and completely. The table indicates that despite superficial similarities we might expect correlations between the three tests. The test Lex30 appears to be tapping into different aspects of productive vocabulary knowledge than other tests.