01 Editorial.qxd Ibérica 35 (2018): 13-40 ISSN: 1139-7241 / e-ISSN: 2340-2784 Abstract While vast research efforts have been directed to the identification of moves and their constituent steps in research articles (RA), less attention has been paid to the social negotiation of knowledge, in particular in the Conclusion section of RAs. In this paper, I examine the Conclusion sections of RAs in English and Spanish, including RA Conclusions written in English by Spanish-background speakers in the field of applied linguistics. This study brings together two complementary frameworks, genre-based knowledge and evaluative stance, drawing on Swales’s (1990, 2004) move analysis framework and on the engagement system in Martin and White’s (2005) Appraisal framework. The results indicate that the English L1 group negotiates a consistent space for readers to approve or disapprove the writers’ propositions. However, the Spanish L1 group aligns with readers, using a limited space through contracting resources, which may be because this group addresses a smaller audience in comparison to the English L1 group which addresses an international readership. On the other hand, the English L2 group tends to move towards English rhetorical international practice, but without fully abandoning their SpL1. These results contribute to gaining a better understanding of how successful scholarly writing in English is achieved, and offers important insights for teaching multilingual researchers. Keywords: functional rhetoric moves, evaluative stances, specific rhetorical effects in English and Spanish, multilingual writers, wider audience in English. Dialogic spaces of knowledge construction in research article Conclusion sections written by English L1, English L2 and Spanish L1 writers Elena Sheldon University of Technology Sydney (Australia) Elena.Sheldon@uts.edu.au 13 Ibérica 35 (2018): 13-40 ELEnA SHELDOn Resumen E s p ac io s d ia l óg ic o s e n l a c on s tru c c i ón d e c ono c i mie n to s d e l a s e c c ió n de l a C on c l us i ón e n ar tí c ul o s de in v e s ti g ac ión e s c r it os e n in g l é s L 1 , ing l é s L 2 y e n e s pa ño l L 1 Si bien se han dirigido vastos esfuerzos de investigación a la identificación de los movimientos retóricos y sus pasos en artículos académicos, menos atención se ha puesto en la negociación social de conocimiento, en particular en la sección de la Conclusión. En este artículo, examino la sección de la Conclusión de artículos de investigación en inglés y español, incluyendo las conclusiones de investigaciones escritas en inglés tomando en cuenta distintos enfoques, el conocimiento basado en la teoría del género y el proceso de evaluación, basándome en el marco de análisis de movimientos retóricos propuesto por Swales (1990, 2004) y en el sistema de compromiso propuesto por Martin y White (2005) dentro del marco de la evaluación. Los resultados indican que el grupo de inglés como primera lengua (L1) negocia un espacio consistente para que los lectores aprueben o desaprueben las propuestas de los escritores. Sin embargo, el grupo de español L1 se alinea con los lectores, utilizando un espacio limitado a través de recursos de contracción, lo cual puede atribuirse a que este grupo se dirige a una audiencia más pequeña en comparación con el grupo inglés L1 que se dirige a lectores internacionales. Por otro lado, el grupo de escritores en inglés como segunda lengua (L2) tiende a adecuarse a la práctica de retórica internacional del inglés, pero sin abandonar totalmente su español como primera lengua (L1). Estos resultados contribuyen a lograr una mejor comprensión de cómo lograr una escritura académica en inglés y brindan un importante conocimiento a los profesores investigadores multilingües. Palabras clave: movimientos retóricos funcionales, posturas evaluativas, efectos retóricos específicos en inglés y español, escritores multilingües, audiencia más amplia en inglés. 1. Introduction In recent decades, numerous studies have paid attention to the research article (RA), which is undoubtedly “the principal site of knowledge-making” (Hyland, 2009: 67) in academia globally. Some studies have explored the textual organization of the four main sections of research articles (RAs) in English: Introduction, Methodology, Results and Discussion [IMRD] (Brett, 1994; Holmes, 2001; Samraj, 2002; Yang & Allison, 2003). There has also been substantial interest in the ways writers negotiate and construct interactive relations through the functions of evaluative features in RAs in 14 English (Hunston & Thompson, 2000; Hyland, 2001, 2005). This interactive negotiation “locates participants’ relationships at the heart of academic writing, assuming that every successful text must display the writer’s awareness of both its readers and its consequences” (Hyland, 2001: 549- 550). An important perspective of evaluation is the Appraisal framework, which has been developed within the SfL paradigm (Halliday, 1994; Matthiessen, 1995), and which makes transparent the relationship of language choices to semantic functions, and accommodates the analysis of stance and how it is achieved linguistically. It stands apart from numerous other studies that have addressed evaluative features in discourse by focusing on hedging, intensification, evidentiality, attitude markers and voice as self-representation (Salager-Meyer, 1994; Tang & John, 1999; Hyland, 2005, to name a few). Drawing on the engagement system in appraisal theory, Chang and Schleppegrell (2011) explored authorial stance connected with functional moves; and more recently, Cheng and unsworth (2016) focussed on academic conflict in the Discussion section of RAs in applied linguistics. Both studies have pedagogical implications for novice researchers. However, as English has become the global language for research and scholarship over the past decades, multilingual scholars are disadvantaged in that they have to compete for academic recognition in a language other than their own. Thus, the mastering of such aspects of RAs in English is essential. Consequently, cross-cultural and multilingual studies have explored the dimension of evaluation, producing valuable findings. Studies have compared RAs in English and Spanish in various disciplines in terms of attitude markers (Mur-Dueñas, 2010), modality (ferrari & gallardo, 2006) and engagement markers (Mur-Dueñas, 2009). Others have explored hedging and boosters (Mendiluce Cabrera & Hernández Bartolomé, 2005), first-person markers (Martín-Martín, 2005; Sheldon, 2009; Burgess & Martín-Martín, 2010); author voice (Lorés-Sanz, 2011), stance (Resinger, 2010), epistemic commitment, amplified attitude, self-mention and periphrastic expressions (Perales-Escudero & Swales, 2011), citation (fortanet, 1997) and metadiscourse features (Mur-Dueñas, 2011). Although these studies have demonstrated that texts are filled with rhetorical choices that carry evaluative stance contributing to the social negotiation of knowledge, they have focused “predominantly on identifying language choices at the level of grammar” (Hood, 2010: 17), with the exception of the study by Pérez-Llantada (2011). DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 15 While the findings from such studies in English-language RAs as well as contrastive studies in English and Spanish have been especially beneficial for newcomers in helping them to meet the expectations of international scientific academia, few studies have explored the Conclusion section independently (Ciapuscio & Otañi, 2002; Williams, 2005). This section has been more generally conflated with the Discussion section of RAs, although the Conclusion section offers specific information that typically is not included in the Discussion section. In fact, in the Conclusion section writers present their contributions as valid and/or offer new insights as well as provide “important elements, such as implications and recommendations” (Amnuai & Wannaruk, 2013: 54). The interpretation and justification of the results have been shown to be problematic for multilingual scholars, as they have to establish themselves as competent and credible members in their discipline (Hyland, 2001, 2005; Swales, 2004; Mauranen at al., 2010). In light of the above, the present study aims to fill the gap in the literature. no studies have focused on evaluative resources across the moves of the Conclusion section of RAs in the field of applied linguistics, contrasting English and Spanish and three groups of writers, native speakers of English (Eng L1), native speakers of Spanish writing in their own language (Sp L1), and native speakers of Spanish writing in English (Eng L2). The present study combines genre-based knowledge and evaluative stance, drawing on the move analysis framework (Swales, 1990, 2004), and on the engagement system in the Appraisal framework (Martin & White, 2005). The engagement system, as theorized within the Appraisal framework, is found to complement move analysis, as it addresses text evaluation from a semantic perspective. Although Appraisal theory is concerned with three main evaluative systems, i.e. engagement, attitude and graduation, I focus on one aspect for closer examination, namely the engagement system. This is because the engagement system deals with the resources of inter-subjective positioning in the discourse community and examines the way in which writers include readers in their discussion. The combination of move analysis and the engagement system allows me to investigate how the Conclusion section is rhetorically organized and how patterns of lexicogrammatical choices within moves and steps craft different types of authorial personae and ideal readerships in English and Spanish. It is hypothesized that the analysis of the Conclusion sections may enfold discursive peculiarities, which may be explained in terms of the potentially different conventions or traditional views of national culture as compared to ELEnA SHELDOn Ibérica 35 (2018): 13-4016 those of the “big culture” (Atkinson, 2004; Sheldon, 2011, 2018) of English as the dominant global language in academia. Therefore, this study aims to provide multilingual scholars with a more comprehensive representation of evaluative stance in both English and Spanish. In particular, the analysis aims to illuminate the ways in which writers and readers interact in a dialogue in RAs and reveal whether evaluative realizations make meaning in similar or different ways in each language. The results of this study, therefore, have pedagogical implications for non-native speakers of English in the context of international publications. I ask three questions of my data to gain an understanding of L1 and L2 scholarly writing considering the multiple aspects of evaluation in RAs: (i) What are the similarities and differences between the English L1, English L2 and Spanish L1 Conclusion sections of RAs in terms of the taxonomy of the genre structure of moves and steps? (ii) What are the similarities and differences in the use of the evaluative stance in the identified rhetorical moves in the three sets of Conclusion sections? (iii)Are the English L2 Conclusion sections influenced by the writers’ native or L1 (i.e. Spanish) written academic culture? 2. Methodology 2.1. Corpus selection The corpora were selected from the field of applied linguistics, with thirty Conclusion sections of RAs, written in English and Spanish, divided into three groups. The first group comprises ten RAs written by native speakers of English, of which six are from the Journal of English for Specific Purposes and four from TESOL Quarterly. The second group comprises ten RAs written in Spanish by native speakers, five from Revista Española de Lingüistica Aplicada (RESLA) and five from Ibérica. The third group of ten RAs was written by native Spanish speakers in English, from Ibérica, and six from RESLA. Scholars from the first group have their home institutions in Anglophone countries, while the Spanish L1 Conclusion sections were written by scholars working in Spanish institutions. I selected only RAs written by scholars from universities in Spain, and so the corpus is limited to Castilian Spanish. This provides a homogeneity that not only facilitates the analysis but also makes DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 17 it more reliable, for example, by not having to account for linguistic differences with RAs written by scholars in Latin America. The third group of Conclusion sections, English L2 texts, comprises articles by Spanish authors who were educated at Spanish universities, increasing the likelihood that they were writing in English as L2. However, it is possible that native speakers of English have edited the articles in the English L2 group. The English L1 corpus comprises 7,650 words with an average of 765 words per RA Conclusion, which is longer than those in the Spanish L1 corpus. The Spanish L1 corpus comprises 6,950 words with an average of 695 words per RA conclusion, making the English L2 Conclusions shorter than the English L1 and Spanish L1 ones. The English L1 group corpus comprises 6,650 words with an average of 665 words per RA Conclusion. Although the audience (national vs. international) will vary, the three data sets are comparable in their main contextual features (field of study, text form, genre, mode, participants, peer review system), as recommended by Moreno (2008). 2.2. Analysis framework As noted in the Introduction above, the present study analyses evaluative stance by drawing on the engagement system (Martin & White, 2005), integrated with the framework of rhetorical moves (Swales, 1990, 2004). According to the framework of rhetorical moves, the Conclusion section consists of three moves: Move 1 Consolidation of results; Move 2 Limitations of the study; and Move 3 further research suggested and their corresponding steps. The current study adopts the concepts of moves and steps in the coding procedures. A move is defined as a segment of text that performs a particular communicative purpose and that contributes to the overall communicative purpose of the text while a step is a smaller functional text that forms part of a move (Sheldon, 2018). Move 1 justifies the research in question through five steps: Step 1: Restating methodology (purposes, research questions, hypotheses restated, and procedures); Step 2: Stating selected findings; Step 3: Making overt claims or generalizations (deduction, speculation, and possibility); Step 4: Recommending; and Step 5: Exemplifying. Move 2 is realized through three steps: Step 1: Limitations of the findings; Step 2: Limitation of the methodology; and Step 3: Limitations of the claims made. Move 3, however, does not have any steps (see Table 1 below). ELEnA SHELDOn Ibérica 35 (2018): 13-4018 Because the application of move analysis has been hampered by the lack of objective ways of identifying of boundaries between moves, the study uses inter-coder reliability to demonstrate that a move can be identified with a high degree of accuracy by trained coders. Three coders practised coding of the data before applying the coding scheme to the data in English and Spanish, allowing them to develop a consistent approach. One of the three coders and myself demarcated each group of the Conclusion section individually. This approach was followed by an evaluation where a coder and I together verified the labelling of move/steps to enhance the reliability and empirical validity of the analysis. I follow Soler-Monreal et al.’s (2011) criteria, which recognize that ninety percent of occurrences in each move are deemed obligatory, but if the move occurrences reach less than ninety per cent, it is deemed optional. In the examples below some words are marked in bold to show the coding of moves: EngL1 (7) In this study, I set out to examine where writers (Move 1 Step 1) playfully depart from convention in the occluded RPT report genre as it is realized in this corpus. SpL1 (18) En este trabajo h e mo s a na l iz a do l a f unc ió n e v al u at iv a ( M ov e 1 S te p 1 ) de las reseñas de libros en su vertiente positiva. Hemos interpretado los resultados obtenidos desde una óptica sociopragmática, pues hemos considerado que esta perspectiva era necesaria DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 19 ELENA SHELDON Ibérica 35 (2018): …-… Move 1: Consolidation of results (i) Step 1: Restating methodology (purposes, research questions, hypotheses restated, and procedures) (ii) Step 2: Stating selected findings (iii) Step 3: Making overt claims or generalizations (deduction, speculation, and possibility) (iv) Step 4: Recommending (v) Step 5: Exemplifying Move 2: Limitations of the study (i) Step 1: Limitations of the findings (ii) Step 2: Limitations of the methodology (iv) Step 3: Limitations of the claims made Move 3: Further research suggested Table 1. Move-structure model of the conclusion section of an RA. Because the application of move analysis has been hampered by the lack of objective ways of identifying of boundaries between moves, the study uses inter- coder reliability to demonstrate that a move can be identified with a high degree of accuracy by trained coders. Three coders practised coding of the data before applying the coding scheme to the data in English and Spanish, allowing them to develop a consistent approach. One of the three coders and myself demarcated each group of the Conclusion section individually. This approach was followed by an evaluation where a coder and I together verified the labelling of move/steps to enhance the reliability and empirical validity of the analysis. I follow Soler-Monreal et al.’s (2011) criteria, which recognize that ninety percent of occurrences in each move are deemed obligatory, but if the move occurrences reach less than ninety per cent, it is deemed optional. In the examples below some words are marked in bold to show the coding of moves: EngL1 (7) In this study, I set out to examine where writers (Move 1 Step 1) playfully depart from convention in the occluded RPT report genre as it is realized in this corpus. SpL1 (18) En este trabajo hemos analizado la función evaluativa (Move 1 Step 1) de las reseñas de libros en su vertiente positiva. Hemos interpretado los resultados obtenidos desde una óptica sociopragmática, pues hemos considerado que esta perspectiva era necesaria para poder llegar a interpretaciones fiables. [In this article, we have analysed the evaluative function (Move 1 Step 1) of book reviews, and considered their positive slant. We have interpreted the results obtained from a social pragmatics standpoint, since we considered this perspective necessary to reach reliable interpretations.] para poder llegar a interpretaciones fiables. [In this article, we have analysed the evaluative function (Move 1 Step 1) of book reviews, and considered their positive slant. We have interpreted the results obtained from a social pragmatics standpoint, since we considered this perspective necessary to reach reliable interpretations.] EngL2 (14) In the present study we have carried out a quantitative analysis (Move 1 Step 1) of the vocabulary input in four textbooks from two educational levels. furthermore, the quantification of moves followed in the present study is recognised as common practice in move analysis (see Martin-Martin, 2005; Adnan, 2008; Sheldon, 2011). However, it should be noted that while these figures are used to identify trends across the three groups no claim is made for statistical significance, as the number of texts analysed it is not sufficient to carry out statistical analysis. In stage two of the analysis, the study adopted the analysis framework of the engagement system. This is constituted by two main categories, “monoglossic” and “heteroglossic” options. Monoglossic options lack dialogic functionality, i.e. they are construed by a single voice and are generally characterized as denoting objectivity and neutrality. On the other hand, heteroglossic choices include either the writer’s point of view or other points of view, via projection, modality, negation and concession. Heteroglossic resources are grouped as having either dialogically expansive or dialogically contracting positions (Martin & White, 2005, after Bakhtin, 1981) to explore the intersubjective functionality of texts. Dialogic contractions are produced when an utterance challenges or restricts contrary positions, aiming to align the reader with the argument being advanced. In addition, contractions are further sub-divided into “disclaim” and “proclaim”. The disclaim feature deals with textual voices or rejects contradictory opinions, for example “deny”, “counter”, “proclaim, “concur”, “pronounce” and “endorse”. Expansion, by contrast, denotes that the text has included other voices as claims are still open to question, and it is also further subdivided into “entertain” and “attribute”. The attribute resources open up dialogic space by referencing an external source, acknowledging that source. The “Acknowledge” feature is a locution where the authorial voice stands with respect to the proposition, which is usually represented through reporting ELEnA SHELDOn Ibérica 35 (2018): 13-4020 verbs such as say, report, state, declare, announce, believe and think. The entertain options include “wordings by which the authorial voice indicates that its position is but one of a number of possible positions and thereby, to greater or lesser degrees, makes dialogic space for those possibilities” (Martin & White, 2005: 104). figure 2 presents the options “contract” or “expand” and their sub-categories with patterns of their linguistic realizations. On the basis that cross-cultural studies must have equivalent corpora, all the Conclusion sections for the analysis have three moves (M1, M2, M3). The analysis of engagement features can be subjective (Mei, 2007), thus double coding at an interval of two months was used to give me an opportunity to identify any bias in the analytical process. Mei (2007), who analysed high- and low-rated undergraduate geography essays, followed a similar process. Mei was able to reflect on her own subjectivity and assess the coding process in light of her own interpretations. furthermore, because I am bilingual, having DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 21 ELENA SHELDON Ibérica 35 (2018): …-… Figure 1. The engagement system – contract and expand (Martin & White, 2005: 134). On the basis that cross-cultural studies must have equivalent corpora, all the Conclusion sections for the analysis have three moves (M1, M2, M3). The analysis of engagement features can be subjective (Mei, 2007), thus double coding at an interval of two months was used to give me an opportunity to identify any bias in the analytical process. Mei (2007), who analysed high- and low-rated undergraduate geography essays, followed a similar process. Mei was able to reflect on her own subjectivity and assess the coding process in light of her own interpretations. Furthermore, because I am bilingual, having Spanish as my mother tongue and English as a second language, I was able to analyse the three groups of writers and supervise the coders. The following texts provide instantiations of engagement (contraction/expansions) and these are bolded. EngL1 (5) NDOSHELENA EL 1e gurFi . sentengagemThe ssorctahtsisabehtOn - utlcu ehtrsnoitcesnoisulcnCo tenemaggenfosisyalan owtfoalvertnianatgnidco lyanaethinsiabynaneid lo tdddt emtsys – W&n itarM(expand and actrcont alviueqe avhtsumesidutsalru evahsisylana (sevomeerth ,eiM(e viectuse bcanesrueat me vigotedsuaswshtnomo ohw, )7002(i eM. ssecorpl aticly iidllfh 134)2005:e,thi . e htlala,roprcotenal ehT. )3M, 2M, 1M( e lbuodsuht,)7002 otytinutroppoane m highdeslyanao - dan iMli 5 3aicréIb 810(2 ): …-…8 wlo - goegetaudrargednudetra usnwoerhnoectrote lab heturF.onsitateprrentin owrhe otrehtmomy inglEnd angue dnasriterwfospuorgeerth t (nemegagn osntiotiantasin )5(L1gEn imisadewoll,syasseyhprag nidcoe htsessasdanytiviectu nguailbimaIeusacbe,eormrhe awI,genguaalond cesasah s lloehT. sredocetheisvrepus thdn a)sniosnapx/entiocatrnoct ( sawieM.ssecropr ali fothgilniscesorpgn sah snipaSng viha,lngua heteyslnaao teblasa eidvorptsxteginwllo deldo ber aese th . ELENA SHELDON Ibérica 35 (2018): …-… Figure 1. The engagement system – contract and expand (Martin & White, 2005: 134). On the basis that cross-cultural studies must have equivalent corpora, all the Conclusion sections for the analysis have three moves (M1, M2, M3). The analysis of engagement features can be subjective (Mei, 2007), thus double coding at an interval of two months was used to give me an opportunity to identify any bias in the analytical process. Mei (2007), who analysed high- and low-rated undergraduate geography essays, followed a similar process. Mei was able to reflect on her own subjectivity and assess the coding process in light of her own interpretations. Furthermore, because I am bilingual, having Spanish as my mother tongue and English as a second language, I was able to analyse the three groups of writers and supervise the coders. The following texts provide instantiations of engagement (contraction/expansions) and these are bolded. EngL1 (5) NDOSHELENA EL 1e gurFi . sentengagemThe ssorctahtsisabehtOn - utlcu ehtrsnoitcesnoisulcnCo tenemaggenfosisyalan owtfoalvertnianatgnidco lyanaethinsiabynaneid lo tdddt emtsys – W&n itarM(expand and actrcont alviueqe avhtsumesidutsalru evahsisylana (sevomeerth ,eiM(e viectuse bcanesrueat me vigotedsuaswshtnomo ohw, )7002(i eM. ssecorpl aticly iidllfh 134)2005:e,thi . e htlala,roprcotenal ehT. )3M, 2M, 1M( e lbuodsuht,)7002 otytinutroppoane m highdeslyanao - dan iMli 5 3aicréIb 810(2 ): …-…8 wlo - goegetaudrargednudetra usnwoerhnoectrote lab heturF.onsitateprrentin owrhe otrehtmomy inglEnd angue dnasriterwfospuorgeerth t (nemegagn osntiotiantasin )5(L1gEn imisadewoll,syasseyhprag nidcoe htsessasdanytiviectu nguailbimaIeusacbe,eormrhe awI,genguaalond cesasah s lloehT. sredocetheisvrepus thdn a)sniosnapx/entiocatrnoct ( sawieM.ssecropr ali fothgilniscesorpgn sah snipaSng viha,lngua heteyslnaao teblasa eidvorptsxteginwllo deldo ber aese th . Spanish as my mother tongue and English as a second language, I was able to analyse the three groups of writers and supervise the coders. The following texts provide instantiations of engagement (contractions/ expansions) and these are bolded. EngL1 (5) This examination of the Noun that pattern has shown [contract: proclaim: endorse] clear evidence of disciplinary variation. SpL1 (10) En los resultados obtenidos e n e s te tra ba j o h e m os p od id o m os t ra r [contract: proclaim: endorse] que existe una tendencia… (In the results obtained from this work we have been able to show [contract: proclaim: endorse] that there exists a fairly generalized tendency…) EngL2 (10) Results from our study showed [contract: proclaim: endorse] that the learners’ proficiency level affected both the amount of appropriate advice acts… The results of the coding were subjected to quantitative analysis, which included frequency of engagement features (contract and expansion) indicated in bold and also non-quantitative analyses. The first stage of the analysis asked whether the Conclusion sections written by the three groups of writers show the presence of the three moves. The second stage of the analysis combined the linguistic resources identified in the framework of moves with the evaluative language encoded in the Conclusion sections, and looked at how the three groups of writers manage their interpersonal positions in the Conclusion sections and asked whether there are convergences and divergences in the use of evaluative stance in the identified rhetorical moves and steps. I discuss the two main categories of the engagement values, monoglossic and heteroglossic. With regards to the heteroglossic values, I include examples of texts from the three groups of writers, as noted in the Introduction section. As evaluative stance in the engagement system provides writers with the means to represent themselves as holding different positions in their arguments, it is of interest to explore whether a similar stance represents the same meaning in both languages. ELEnA SHELDOn Ibérica 35 (2018): 13-4022 3. Results and Discussion The first stage of the analysis reveals that the English and Spanish RA Conclusion sections display three moves and their corresponding steps; and it is noted that each move occurs cyclically rather than linearly, supporting the finding of past studies that the Conclusion section is highly cyclical (Swales, 1990, 2004; Yang & Allison, 2003; Loi at al., 2016). However, this aspect is beyond the scope of this article. In regard to Move 1, the three groups of writers used a larger number of steps compared to Move 2 and Move 3, which strongly indicates that their communicative focus is on Move 1 Consolidation of results. This can thus be characterized as conventional, as found in past studies (Swales, 1990, 2004; Posteguillo, 1999; Yang & Allison, 2003; Loi et al., 2016; Sheldon, 2018). However, despite using this move more often, the English L1 group displayed seventy-four instances of Move 1, compared to the fifty instances displayed by the Spanish L1 group and the forty-nine by the English L2 group. These results suggest that writers have a purposefully defined range of textual choices at their disposal to craft the consolidation of their results effectively, which is testimony of the complexity of this move. The English L1 group has engaged the five steps to consolidate their results, with a delineated path of making claims being foregrounded. On the other hand, the Spanish L1 group does not use the five steps as often as does the English L1 group. It appears that the demands of the discourse community of each language have contributed to their framing of their texts. With this move, the English L2 group is positioned more towards the Spanish L1 group than to the English L1 group, as shown in figure 2. DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 23 ELENA SHELDON Ibérica 35 (2018): …-… However, despite using this move more often, the English L1 group displayed seventy-four instances of Move 1, compared to the fifty instances displayed by the Spanish L1 group and the forty-nine by the English L2 group. These results suggest that writers have a purposefully defined range of textual choices at their disposal to craft the consolidation of their results effectively, which is testimony of the complexity of this move. The English L1 group has engaged the five steps to consolidate their results, with a delineated path of making claims being foregrounded. On the other hand, the Spanish L1 group does not use the five steps as often as does the English L1 group. It appears that the demands of the discourse community of each language have contributed to their framing of their texts. With this move, the English L2 group is positioned more towards the Spanish L1 group than to the English L1 group, as shown in Figure 2. Figure 2. English L1, Spanish L1 and English L2 RA Conclusions employing moves. Move 2 reveals limitations or shortcomings of the writers’ research. This move received some attention from the Spanish L1 group but less from the English L1 and English L2 groups. However, there is no expectation that every research publication needs to contribute to a discussion of limitations of findings, methodology or claims. Neither does Move 2 figure in comparative studies of German and Anglo-American Introductions and Conclusions in linguistics by Gnutzmann and Oldenburg (1991), and Spanish, German and English by Ciapuscio and Otañi (2002). Although generalizations cannot be drawn due to the small corpora the results here may raise awareness of the consistent presence of Move 2 in the two groups writing in English L1 and Spanish L1. This seems to indicate that the inclusion of recommendations may promote knowledge growth as it strengthens everyone’s case for research grants. Due to the smaller number of instances of Move 2, it is deemed optional. NDOSHELENA EL sihtgnisuetipsed,revweHo ytnevse - oMf osecnatsnirufo ethdnapuorg1LhisnapSeth upaevahs retirwtahtsteggsu daiolonschetarco tlapossdi ovemshitofy txieplomchetof ltsirthlid hsilgnEeht,neoeromevom ,1evo niye htotedarpmco e - 2 Lh singlEhetby neni etfoegnardenedyllseopru userrihetofon itda lyetivcelts eshaoup gr1 Lh singlEheT.ove fthdlindith deyalpsidpuorg1Lh ybedaylpsidcesantsn stulseresheToup.gr riehttas eciohclautxe ynotimsteishichw, ly psetsvehetd genga inbimlaink , ltsuserirethtealidosnocto hrehtoehtnO.dednuorgerfo nEehts eods aneos as pest h caeofy tunimomcesourcsdi Eeth, evomisthithW. tsxte LhsinaSp heto tn hatoup gr1 mfothapdteaelinedaithw dpuorg1LhsinapSeht,dnah ahts raeppatI.puorg1Lshilgn to td ebutirontcvehagenguaalh enitiosopispuorg2LhlisgnE n in howssaoup,gr1 Lh singlE ginebsimlacginkam utonseod evehtse ehtfos dnamedehtta rihetofng imarihet ethsdrawtoeromde guiF 2re . 2e gurFi . naSp,1LhsilgEn rosnoitatimilslaever2evMo mnoitnettaemosdeviecre woH.spuorg2LhsilgnEdan butirontco tdseneon itacipubl tieN.msialcroygolodohtme olgAndnanaGe - acireAm grubnedOldnannamztuGn mesnoisulcnCoRA 2LhsilgEndna1Lhsi ’sretirwehtfosgnimoctrohsr Seht sselbutoup gr1 Lh snipa noiatectpexonsie erht,erevw imilofon isuscsdiao tebut ocnierug2evoMseodreht oisulcnoCdnasnoitcudortnIn mrGe,hsinapSdna,)1991( sevomgniyolp . evomsihT.hcraeser 1 Lh singlEhetom chearesryerevathtn ,ngsndiofonsitati rampo foesidutse viat ybscitsiugnilnisno ybhsilgnEdnanam 5 3aicréIb 810(2 ): …-…10 g .)2002(iñatOdnaoicsupaCi etharoprocll amseth hstlusre oupsgro wthetn i2 oveMof ioslucinetht athteaicdinto veenshengtertstisah towgr oveMofsencatnsiofrbenum ,p,)( nacsnoitazilareneghguohtlA htfossenreawaesirayamreeh paSnd a1 Lh singlEn ing itirw yamsntioadnemmocerfonio eresacs’yone tnarghcrase .lonaioptd emedesiti2, yg oteudnwardebton ecnesreptnetsisnoce smeesshiT1.Lh snipa egdlewonkteomorp rellasmehtoteuDs. Move 2 reveals limitations or shortcomings of the writers’ research. This move received some attention from the Spanish L1 group but less from the English L1 and English L2 groups. However, there is no expectation that every research publication needs to contribute to a discussion of limitations of findings, methodology or claims. neither does Move 2 figure in comparative studies of german and Anglo-American Introductions and Conclusions in linguistics by gnutzmann and Oldenburg (1991), and Spanish, german and English by Ciapuscio and Otañi (2002). Although generalizations cannot be drawn due to the small corpora the results here may raise awareness of the consistent presence of Move 2 in the two groups writing in English L1 and Spanish L1. This seems to indicate that the inclusion of recommendations may promote knowledge growth as it strengthens everyone’s case for research grants. Due to the smaller number of instances of Move 2, it is deemed optional. Move 3 proposes further research whose marked use by the English L1 and L2 groups may be explained by the internationalization of English discourse, which functions as a norm encouraging universal application (Swales, 1990, 2004; Salager-Meyer, 1997). On the other hand, the smaller number of instances of Move 3 by the Spanish L1 group may suggest that scholars in Spain have been discouraged from proposing future research. However, the number of instances by the three groups is below the ninety percent threshold, thus Move 3 is also optional. Overall, the analysis of moves in the thirty Conclusion sections written by three groups of writers in English and Spanish reveals the complexity of the Conclusion section, suggesting that these texts enfold discursive peculiarities of their specific cultural discourse communities. In the analysis of stage two, it is noted that heteroglossic resources are chosen more frequently by the writers than monoglossic ones to position themselves in the discourse community to make knowledge claims. Within the domain of monoglossic resources, the English L1 group displays 20 instances, equivalent to 7% of the total number of monoglossic and heteroglossic resources which total 249 instances, 93%. The Spanish L1 group exhibits 21 instances, equivalent to 9% of the total number of monoglossic and heteroglossic resources with a total of 211, 91%, while the English L2 group displays 17 instances, equivalent to 8% of the total number of monoglossic and heteroglossic resources which total of 174, 92 %; as shown in figure 3. ELEnA SHELDOn Ibérica 35 (2018): 13-4024 The fact that the writers in all three groups overwhelmingly choose heteroglossic categories rather than monoglossic ones, as shown in figure 3, indicates that the Conclusion section is filled with alternative views. This appears to confirm that texts are multi-vocal in academic writing. Although monoglossic statements are dialogically inert, they nevertheless contribute to achieving the communicative objectives of a text as the writers are construing a reader with whom they assume to share a similar position. A closer examination of the heteroglossic resources further confirms that the Conclusion section is highly engaging and loaded with evaluative language in both English and Spanish. However, variations in expressing expanding and contracting resources are evident in the three groups of writers. The English L1 group favoured expanding resources, with 141 instances (52.50%), rather than contracting resources, with 108 instances (40%). On the other hand, the Spanish L1 group favoured contracting resources, with 115 instances (50%), rather than expanding resources, with nineteen instances (41%). As with the English L1 group, the English L2 group favoured expanding resources, with 98 instances (51.30%), over contracting resources, with 17 instances (40%). figure 4 presents the frequency of contracting evaluative language observed in the Conclusion section, including “deny”, “counter”, “concur”, “concede”, “affirm”, “pronounce” and “proclaim”. DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 25 DIALOGIC SPACES OF KNOWLEDGE CONSTRUCTION Ibérica 35 (2018): …-… Move 3 proposes further research whose marked use by the English L1 and L2 groups may be explained by the internationalization of English discourse, which functions as a norm encouraging universal application (Swales, 1990, 2004; Salager-Meyer, 1997). On the other hand, the smaller number of instances of Move 3 by the Spanish L1 group may suggest that scholars in Spain have been discouraged from proposing future research. However, the number of instances by the three groups is below the ninety percent threshold, thus Move 3 is also optional. Overall, the analysis of moves in the thirty Conclusion sections written by three groups of writers in English and Spanish reveals the complexity of the Conclusion section, suggesting that these texts enfold discursive peculiarities of their specific cultural discourse communities. In the analysis of stage two, it is noted that heteroglossic resources are chosen more frequently by the writers than monoglossic ones to position themselves in the discourse community to make knowledge claims. Within the domain of monoglossic resources, the English L1 group displays 20 instances, equivalent to 7% of the total number of monoglossic and heteroglossic resources which total 249 instances, 93%. The Spanish L1 group exhibits 21 instances, equivalent to 9% of the total number of monoglossic and heteroglossic resources with a total of 211, 91%, while the English L2 group displays 17 instances, equivalent to 8% of the total number of monoglossic and heteroglossic resources which total of 174, 92 %; as shown in Figure 3. Figure 3. Instances of monoglossic and heteroglossic categories in the Conclusion section of three groups of writers. The fact that the writers in all three groups overwhelmingly choose heteroglossic categories rather than monoglossic ones, as shown in Figure 3, indicates that the Conclusion section is filled with alternative views. This appears to confirm that texts are multi-vocal in academic writing. Although monoglossic statements are eserrehtrsesoporp3evMo tby d neiaxplebey amoupsgr sasnoitcnfu rucoenmrona regalSa - htnO.)7991,reyMe rg1LhsinapSehtyb3evMo ng iopospromd geaourcsdi lbeihrheby KNOFOSPACES C IGOALDI htybesudeamesohwhcrae nglEofon itazilonaitnarentihet (noicatilpapalservinugniagr Sw munrellamseht,dnahrehtoeh ralohcstahttseggusyampuor hcraesereur eth, revewoH. d,holhrnihe NOIRUCTNSTCOE EDGLW 2Ldna1LhsilgnEe h chiw,esourcsdih singl 4002,0991,selaSw ; fosecnatsniforebm neebevahniapSnis secntasinforebmun lihus owlbesioupsgreehrthetby siyslnaahet,llarveO.lonaiopt n isretirwofoupsgreehrtby nitseggus,noitcesnoisulcnCo ccep sireth sourcsdilaurtul ,owtegatsfosisylanaehtIn ehtybyltneeermo retiwr toitynummocesruocisdeth nEeht,secruosercissolgonmo mofrbenumlaotthetof7% heT93%ti249 Sp d,holsehrtntecrpey tnenihetow usloncCy hithetn isovemofs tslaveerh snipaSnd ah singlE ucsiddlnestxetesehttahtgn .seitunimomces cissolgroetehtahtdetonsiti opotsenocissolgonomnahtsr W. simlacegdlewonkekam sni02syalpsidpuorg1Lhsilgn rethend acisosonoglm rcisosogl tibih1Lhi i21 o slasi3 oveMhust n ettirwonsitceson ius hetofy txieplomche foseitirailucepevisru nesohcreasercuosre nisevlesmehtnoitiso foinamodethinithW ottnelaviuqe,secnats hchiwsecourser l tato tntlquit heT.93%,sencatnsi249 aSp mofrbenumlaotthetof9% singlEhetelhiw,91%211,of onoglmofrbenumlaotthetof guriFn in howssa;%92 174, e stibihxepuorg1Lhsina nsi21 rcisosoglrethend acisosonogl ncatnsi17 ysaplsdioup gr2 Lh s oursercisosoglrethend acisosonogl .3e o tntelvaquie,sencatns laottah tiwsecourser quie,senc 8%o tntelva oflaotth chiwsecour 3e gurFi . dn aicsslogonof m osecntasIn llanisretirwehttahttcafeTh lgonomanhterhatresiroegcat wdellsinoitcesnoisulcnCo eratsxte itlmu - deacan ilavoc Ibér ioslucno Ce th insrieogtea cicsslogrotee hd sretiwr . ylgnimlehwrevospuorgeerhtl e rugiFninwohsas,esnoc issol ppasihT.sweivevitanretlahtiw onoglmhough tlAng.itirwcimde 35ca ibér 810(2 ): …-… 11 f ospuro geref th ontioce sn cissolgoretehesoohc e htathtescatidni,3e tahtmrnocotsraep erasntemetatscisosonogl In general terms, the Spanish L1 group favours the “deny” feature by displaying 39 instances (17%) of it, while the English L1 group displays 19 instances (7.06%) and the English L2 group, 24 instances (12.56%). Examples are as follows: EngL1 (2) I would [expand: entertain] argue that the identification of such patterning not [contract: disclaim: deny] only contributes to the understanding and description of disciplinary language use, (…) SpL1 (4) Estos hallazgos resultan [contract: proclaim: endorse] de gran utilidad no [contract: disclaim: deny] solo a la Didáctica del Español como L2, sino a la Didáctica de las respectivas asignaturas. (The results are very useful not only to the Spanish Didactic as L2 but to the Didactic to the respective subjects.) EngL2 (2) The frequency of the passive voice is not [contract: disclaim: deny] meaningful and the use of active voice is twice that of the passive; the simple present tense stands out over the rest. The three examples above show a similar execution of the “deny” resource. The negative voice simultaneously includes the positive voice, and it is assumed that the reader will take the alternative perspective, thus preventing readers from gaining incorrect information. By guiding the reader away from any potential misunderstanding, a closer relationship has been established between writer and reader, thus enhancing solidarity between them. ELEnA SHELDOn Ibérica 35 (2018): 13-4026 ELENA SHELDON Ibérica 35 (2018): …-… dialogically inert, they nevertheless contribute to achieving the communicative objectives of a text as the writers are construing a reader with whom they assume to share a similar position. A closer examination of the heteroglossic resources further confirms that the Conclusion section is highly engaging and loaded with evaluative language in both English and Spanish. However, variations in expressing expanding and contracting resources are evident in the three groups of writers. The English L1 group favoured expanding resources, with 141 instances (52.50%), rather than contracting resources, with 108 instances (40%). On the other hand, the Spanish L1 group favoured contracting resources, with 115 instances (50%), rather than expanding resources, with nineteen instances (41%). As with the English L1 group, the English L2 group favoured expanding resources, with 98 instances (51.30%), over contracting resources, with 17 instances (40%). Figure 4 presents the frequency of contracting evaluative language observed in the Conclusion section, including “deny”, “counter”, “concur”, “concede”, “affirm”, “pronounce” and “proclaim”. Figure 4. Instances of contracting resources in the Conclusion section of three groups of writers. In general terms, the Spanish L1 group favours the “deny” feature by displaying 39 instances (17%) of it, while the English L1 group displays 19 instances (7.06%) and the English L2 group, 24 instances (12.56%). Examples are as follows: EngL1 (2) I would [expand: entertain] argue that the identification of such patterning not [contract: disclaim: deny] only contributes to the understanding and description of disciplinary language use, (…) NDOSHELENA EL rveney het,trneiy llacogiladi tirwhetsaxtetaofsveitceobj itiso prilaim s aerah sto on. ehtfonoitanimaxeresolcA ylhgihsinoitcesnoisulcnCo Hh.snipaSnd ah singlEh bot ng viehicao tebutirontcsselhetr tiwrdeaerang tonscerasret htrsecruosercissolgoretehe vehtiwdedaoldnagnigagne sexpren ionsitairvarveeowH veitacunimomchetng eumssay hethomwh t ehttahtsmrnocreh niegaugnalevitaulav nd ang ndixpaeng iss Hh.snipaSnd ah singlEh bot cesruoesrgniactrtnco dieve ar erng ndixpaed evourafoup gr 01htiw,cesruoesrgniactrtnco nitcartnocderuovafpuorgL1 nhtiw,cesruoesrgnidanpex oup gr2 Lh singlEhetoup,gr regnitcratnocr evo),%03.1(5 gtincatrnocfoycneuqeeth ”yned“gnidulcni,noitcse alcor“pdna”ecnuonor“p .”im sexpren ionsitairva,rveeowH irwfospuorgee rhte htnitend (sencatnsi141 h tiw,secourse htoe htnO.)%04(cesantsni80 ,secruosergn ecnatsni511htwi wsA.)%14(cesantsnieenetni ecourserng ndixpaed evourafoup %0(4secnatsni71htiw,sercuosre evresboegaugnlaetivaluaveg c“,”rucnoc“,”retnuoc“,” nd ang ndixpaeng iss 1LhsilgnEe hT.serti n hatrhetar,)50%52. hsianpSe ht,danherh nahtrehtar,)%05(se 1LhsilgnEe hthtiw sencatnsi98 h tiw,se giF).% sntesepr4 eur nioslucnoCethinde ,”mra“,”edecnoc 4e gurFi . gtinctranof c osecntasIn hsinapSeht,srmetlraenegIn w,tiof)17%(sencatnsi39 e snioslucno Ce th insercuos reg hrtofon ict f”yned“ehtrsuovafpurog1L dioup gr1 Lh singlEhetelhiw sertirwofoups gree hr . gniyalpsidybreuta sencatnsi19 ysaplsdi 5 3aicréIb 810(2 ): …-…12 2LhsilgnEehtdna) %60.(7 :swollfo )2(L1gEn :dnapx[edluowI aiterten :tcratno[c :mialcsdi ny]de eusgenguaaly rnaiplicsdi %65.2(1secnatsni42,purog2 noicatitendie htathte ugar]nai ndiatsrundeheto tsebutirontcy onlny] ,e (…) selpmaxE).% ase ar tongninertatpchsufo ofon iptircsdend ang ndi Another evaluative feature that merits attention is endorsement. Endorsement choices close down the space for dialogic alternatives where the authorial voice presents the proposition as true. These formulations are realized by reporting verbs such as “show” and “demonstrate”, and although these features limit the room for negotiation, they deploy objectivity. In the present study, the three groups of writers have endorsed their findings by placing other sources in Theme position, so their interpolation in the text is positioned as objective. The Conclusion section achieves its effect through the negotiation of contracting resources (endorse/proclaim/affirm). On the basis that formulations of pronouncement imply consensus or appeals to general knowledge (the truth of the matter…), known as “evidentials” (Chafe, 1986; Hyland, 2005), they are appropriate for discussing results and findings. With expressions of certainty, or “evaluations of pronouncements” in Appraisal terminology (Martin & White, 2005), writers address readers as being as knowledgeable as themselves. The endorse category, which represents the communicative purpose of Move 1, Step 3, has been favoured by the Spanish group, displaying 38 instances (16.37%), while the English L1 group displayed 29 (11%) and the English L2 group 24 instances (12.56%). Examples are as follows: EngL1 (7) Another important finding of this study is that most dimensions show [contract: proclaim: endorse] a strong polarization between spoken and written registers. SpL1 (2) En la segunda parte de este trabajo, s e h a c o mp rob a do [contract: proclaim: endorse] estadísticamente el nivel de discriminación de estas medidas para ambos grupos. (In the second part of this work it has been proved [contract: proclaim: endorse] statistically the level of discrimination in both groups…) EngL2 (3) The analysis also showed [contract: proclaim: endorse] how these units cluster at moments making bundles to attain their purposes more effectively. The repetitive use of contractive resources such as endorse pushes the examples above to a different level. The exclusion of other voices and dialogic alternatives achieves its effect through the proclamation of new knowledge, for example in “show a strong polarization between”, se ha comprobado (it has been proved), “show how these units cluster at moments DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 27 making bundles”. With these features, writers have developed a prosody of assertive claims1, as the accumulation of contractive resources (endorse, pronounce and affirm), distributed through the clause and across the clause, resonate with the values of graduation2. The English L1 Text 7 uses “strong polarization”, the Spanish L1 Text 8 estadísticamente el nivel de discriminación (statistically the level of discrimination), and the English L2 Text 3 “more effectively”, all of which allow them to advance their argument with an explicit authoritative voice. As the Spanish L1 group produced almost triple the number of instances of endorsement patterns as the English L1 group (16% Spanish L1 and 11% English L1), we may say that the register of the texts of the Spanish L1 group has been constructed with formulations characterized as correct and valid. Based on the quantitative analysis, the sub-categories of contracting resources, such as “deny” and “endorsement”, have contributed to construing the Spanish L1 register with a relatively narrow space for alternative views. In other words, the Spanish L1 texts appear more direct than their English counterparts, as these resources tend to close down dialogic space. The notion that the Spanish writers are “more comfortable with higher degrees of epistemic commitment” (Perales- Escudero & Swales, 2011: 66) appears to correlate with the results of the present study. Evaluative resources denoting contributions are a typical feature of the Conclusion section, and the reader expects this type of contribution. Another popular contracting feature employed by the three groups of writers is “counter”. Counter invokes a particular proposition but is replaced by a proposition that would have been expected. It is usually realized through conjunctions and connectives such as “even though”, “however”, “yet” and “but”. These locutions have also been categorized in the traditional literature as evidentials of contrast (Swales, 1990, 2004). The English L1 group favours this feature, displaying 31 instances (13%) compared to the Spanish L1 group with 19 instances (8.18%) and the lesser usage of it by the English L2 group, displaying only 12 instances (6.28%). Examples are as follows: EngL1 (7) Although [contract: disclaim: counter] many questions about academic language remain, this study has made a substantial contribution to the description of academic discourse, providing a relatively comprehensive analysis of language use in the university. ELEnA SHELDOn Ibérica 35 (2018): 13-4028 SpL1 (2) A un qu e [contract: disclaim: counter] en este trabajo no [contract: disclaim: deny] nos hemos planteado determinar como objetivo la evaluación interna de los materiales - en la cual, se analiza la adecuación de los mismos a la edad-este primer análisis ya nos indica que todas las series presentan libros de texto apropiados a las características y necesidades del niño de tres a cinco años. (Although [contract: disclaim: counter] in this work we have not [contract: disclaim: deny] aimed to determine the evaluation of resources with respect to their suitability for the age; the first analysis indicates that all the series of books present appropriate texts to the needs of children from three to five years old.) EngL2 (5) Quantitative Economy prefers to start with the problem to solve but [contract: disclaim: counter] Management and financial is not [contract: disclaim: deny] very strict in the moves included in the introduction. The countering expectation here allows these writers to put forward their arguments. Besides associating the writers’ own position with what they are reporting, the focus of the message is on the clause that puts the concession first, while the main clause provides a general statement that “this study has made a substantial”, este primer análisis ya nos indica que todas… (“the prime analysis points that all the series”) and “Management and financial is not very strict in the moves included in the introduction”. It appears that the category of disclaiming through countering choices takes an authoritative position and prevents readers from gaining the wrong information, which is more noticeable in the English L1 than in the other two groups of writers. In regard to expanding resources, across the moves of the three groups of Conclusion sections, expanding resources realized through “entertain” features assist the writers to moderate their expressions of certainty by providing a dialogic space for the diverse opinions held by readers. It has been hypothesized in the literature that published material in languages other than English displays a distinct interpersonal negotiation of evaluative resources (Mur-Dueñas, 2009, 2010; Sheldon, 2009), which can be conditioned not only by linguistic factors but also by communal disciplinary knowledge. The English L1 group favours this evaluative feature, while the Spanish L1 group as well as the English L2 group use it less, as shown in figure 5. DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 29 As the Spanish L1 group tends to narrow the space between writers and readers, this may confirm the finding that interpersonal engagement choices may “derive from national academic traditions” (Mauranen et al., 2010). Therefore, I would argue that the English L1 group enacted the writer-reader relationship with a more dynamic approach to suit their international readership. for example: EngL1 (5) Researchers investigating academic lectures could [expand: entertain] interview lecturers to determine whether they are aware of their discursive practices when lecturing to different audience size. SpL1 (7) Los resultados del análisis evidencian que s e pu e de [expand: entertain] establecer un paralelismo entre AIDA y la macroestructura de los folletos bancarios… (Results of the analysis show that can [expand: entertain] draws parallels between AIDA and macrostructure of bank brochures…) EngL2 (8) Research on actual texts provides realistic information that can [expand: entertain] be used for academic writing courses allowing graduate students to appreciate the complexity and variation that is involved in the process of writing PhD thesis introductions. Through the range of entertain features, these three writers dynamically construe a heteroglossic backdrop for the text, with modal auxiliaries such as ELEnA SHELDOn Ibérica 35 (2018): 13-4030 ELENA SHELDON Ibérica 35 (2018): …-… Figure 5. Instances of expanding resources used by three groups of writers. As the Spanish L1 group tends to narrow the space between writers and readers, this may confirm the finding that interpersonal engagement choices may “derive from national academic traditions” (Mauranen et al., 2010). Therefore, I would argue that the English L1 group enacted the writer-reader relationship with a more dynamic approach to suit their international readership. For example: EngL1 (5) Researchers investigating academic lectures could [expand: entertain] interview lecturers to determine whether they are aware of their discursive practices when lecturing to different audience size. SpL1 (7) Los resultados del análisis evidencian que se puede [expand: entertain] establecer un paralelismo entre AIDA y la macroestructura de los folletos bancarios… (Results of the analysis show that can [expand: entertain] draws parallels between AIDA and macrostructure of bank brochures…) EngL2 (8) Research on actual texts provides realistic information that can [expand: entertain] be used for academic writing courses allowing graduate students to appreciate the complexity and variation that is involved in the process of writing PhD thesis introductions. Through the range of entertain features, these three writers dynamically construe a heteroglossic backdrop for the text, with modal auxiliaries such as “could” and “can”. However, these modals of probability construct the challenges of the research tentatively, suggesting that a more direct argument is not strategic in persuading English readers at this point. In this instance, persuasion is centered NDOSHELENA EL 5e gurFi . f osecntasIn dnetpuorg1LhsinapSehtAs gindethmnocyamisth idartcimedacalanoitanmofr rg1LhsilgnEe htathte ugar usothcaorppacmianydermo 5(L1gEn ) R gniatgiestvnis erchesear weinmrteedtosrertucle oups gree hrtby used ces esourrng expandi newetebecapsehtw orranotsd t nemegagnel anosreteint ath enaruaM(”snoit 2010),.laten ertirwe htedactenpuor - r edare pihsredaerlanoitanretniriehtti es ruectlc iemacadg dluco apx[e nd: isdirethfoerawaerayethrethehw sertirwofoups . ,sredaerdnasretiwrn eivred“yamseicohc d oulwI,eerheT.2010) ahtiwpihsnoitalrer elmpaxeroF.p : wevirentin]iatrentend: nehwseticcarpeivsrucis dut aner d toginrtucle )7(L1Sp nalsatlusersLo smiellraapnuecer lbaest osiarbanc … fostluse(R ADIAn eewtbeslellarpa )8(L2gEn xetlautcanohcraesRe acadrsedue b]naiterten tiexlpmcoe hte atecirpap .eiz secnie euqnaicnivesisilá eedupse ruestcroamaly enosm tahtwohssisylanaehtf nca x[e camnd a uhcrobknabforeutcrutsro divorpstx oiatmrnic istiealres niwolalses rucognitirwc iemacad niedvlovnis iathtnoiatiarvdany e iatretne:dna[e s oetlls olrauctru swrad] niartetne:dnapx )…sreu athtno nca :dnapx[e ots tenduste atuadrggn gnitirwfocess orpe htn 5 3aicréIb 810(2 ): …-…16 tiexlpmcoe hte atecirpap .snoitcudortnisisehtDPh etnefoegnarehthguorTh iart trpordackbc issolgoeretha adomeseht,revewoH.”na“c itseggus,ylevitatnethrcaesre tasrdeaerh singlEng diuasrpe niedvlovnis iathtnoiatiarvdany . drsetriwerehteseht,sreutan esiarilixaualdomhtiw,texte ht httctsnocytilibaborpfosla matahtgni egumratcerdieor pe,encatnsishitn I.ntpoishitt gnitirwfocess orpe htn erutsnocyllacimanyd dan” dluco“aschuses ehtfosegnellahceh n icgietartsnotsint d erentecsion isuasrpe “could” and “can”. However, these modals of probability construct the challenges of the research tentatively, suggesting that a more direct argument is not strategic in persuading English readers at this point. In this instance, persuasion is centered on entertaining various options, through modal auxiliaries used to create more space for refutation by or debate between writers and readers. This suggests that, although the texts display contractive resources, part of the logic of the argument rests on expanding resources in the form of “entertain”, as they discuss several challenges that need to be acknowledged, such as presenting findings and limitations without a commitment to support the conclusions. Acknowledge resources also contribute to shaping the Conclusion section, particularly Move 1, which opens up the dialogic space and acknowledges the scholarly contributions of other voices that are external to the texts. Other voices are included in the texts through the reporting verbs where the writer attributes the proposition to the literature and shows impartiality towards the proposition being advanced. With this in mind, acknowledgement choices are anticipatory, and through this mechanism, the external voice in the text itself engages with other voices. I explored the evaluative potential of reporting verbs from the semantic perspective of negotiating meaning. from this perspective, I was able to investigate how attributions in the form of reporting verbs dynamically unfold across the text to position the writer’s own research in relation to other contributions. for example: EngL1 (3) Finegan (1999) stated [expand: attribute: acknowledge] that such nouns are one of the primary devices used to mark stance in academic prose... SpL1 (2) Co mo s e ña l a Cr ai g ( 1 9 8 9 ) , [expand: attribute: acknowledge] aunque [contract: disclaim: counter] existen diferencias individuales,… (As Craig (1989) points out [expand: attribute: acknowledge] although [contract: disclaim: counter] there exist individual differences…) EngL2 (10) Cotterall (1999:497) pleads [expand: attribute: acknowledge] for studies on beliefs to aim at the greatest possible conceptual, methodological and psychometric rigor, … DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 31 These three examples reveal expanding resources through attribution, allowing the writers to position their studies in the field to demonstrate knowledge. finegan (1999), Craig (1989) and Cotterall (1999: 497) have acknowledged that, through verbal reporting verbs, an explicit detachment from the message is observed here because the writers have attributed a position to the original authors. In this case, these writers have created a dialogic space to make public the contributions from other researchers. So far, the analysis using the engagement system identified what repertoire of linguistic resources construe Move 1 in the Conclusion section written by three groups of writers. These writers manage their interpersonal positions by opening up or closing down potential negotiations to enact the writer- reader relationship to suit their discourse of practice. Move 2 Limitations of the study emerges as writers acknowledge that their outcomes are not what they were expecting due, for example, to the narrow analysis of data: EngL1 (10) (M2/S1) finally, we recognize [contract: proclaim: pronounce] that the narrow scope of this study leaves many questions unanswered regarding PVs:…[monogloss] SpL1 (4) (M2/S1) Para finalizar, qu is i é ra mo s p on e r [expand: entertain] de manifiesto algunas limitaciones del presente estudio a las que h a b rí a q ue pr e s tar [expand: entertain] más atención en futuras investigaciones. (To finish, we would like to put [expand: entertain] forward some limitations from the present study to which must pay [expand: entertain] attention in future investigations.) EngL1 (4) (M2/S1) Although [contract: disclaim: counter] we are aware of the limited size of the sample of texts taken from the CTC, [monogloss] it is interesting to note that this meaning … The English L1 and L2 writers above assigned functions to segments of information, for example, the monoglossic statement of the English L1 Text 10 “the narrow scope of this study”, and the English L2 Text 4, “we are aware of the limited size of the sample of texts”, to make apparent the limitations of their studies. These texts are constructed with one single voice to present neutrality and objectivity. Although monoglossic locutions are ELEnA SHELDOn Ibérica 35 (2018): 13-4032 minimal in number in the texts of the present study, their presence in publications in the field of linguistics is essential for the writers’ arguments. On the other hand, the Spanish L1 Text 4, quisiéramos poner de manifiesto algunas limitations del presente estudio (we would like to put forward some limitations from the present study), draws on the entertain resources which allow this writer to expand his dialogic space. Overall, in using contracting or expanding resources, writers justify their work in a competitive academic community by acknowledging their research limitations. Moreover, as a kind of motivational force, the three groups of writers advocate the need for further research, thus enacting Move 3 further research suggested. Move 3 reinforces disciplinary practice in general, indicating that these writers are not avoiding competitiveness among research members and institutions but are encouraging further research. for example: EngL1 (6) (M3) future research will [expand: entertain] need to address the issue of how far we can lengthen collocations while retaining cross-disciplinary usefulness. SpL1 (9) (M3) Con respecto a una posible continuidad de este estudio, p e ns a mo s q ue po dr ía [expand: entertain] servir de punto de partida para investigaciones futuras… (With respect to a possible continuation of this study, we thought it would [expand: entertain] serve as a starting point for future research...) EngL2 (6) (M3) To sum up, despite [contract: disclaim: counter] some limitations that might [expand: entertain] be attributed to our study, … These writers dynamically construe a heteroglossic backdrop for their text, with modal auxiliaries such as “will”, “can”, “would” and “might”. However, these modals of probability construe the challenges of the research tentatively, suggesting that a more direct argument is not strategic in persuading their readers at this point. In this instance, persuasion is centered on entertaining various options, through modal auxiliaries used to create more space for refutation by or debate between writers and readers. nevertheless, readers always have the option to refute what has been proposed, so it is important that multilingual scholars are aware of how knowledge is negotiated. DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 33 Overall, it is worth noting that the comparison of the Conclusion sections written in English and Spanish has highlighted how these writers align themselves with other voices to justify their own research and how their engagement with readers operates at different levels. The combination of the engagement system and the move-based framework has provided more contextual data and has thrown light on the scholarly discourse and rhetorical traditions in the written cultures of the two L1 and the L2 groups of writers. Interestingly, the analysis indicates that arguments were created around the three moves, but Move 1 Consolidation of results was used more often because writers purposely highlight the significance of their studies. The differences in the frequency distributions of the engagement resources may be attributable to different ideological positioning, disciplinary norms and the size of the audience in each language. The English L1 writers constructed the international reader/audience by framing their texts using a rich research contextualization. These writers moderated the forcefulness of their claims to reduce the distance between writers and readers, as this approach allowed them to do their “selling job” (Yakhontova, 2002: 231) and persuade their readers/audience of the legitimacy of their claims. Therefore, it appears that the international English-language discourse community has nurtured an informed reader. The Spanish L1 writers constructed the national reader/audience in line with their particular social and cultural conditions and discourse pressures. Their texts were targeted to be read by a national applied linguistics audience in Spain, which is known to be a relatively closed community. It is not surprising that the English L1 and Spanish L1 groups have “instantiated two different culture- specific textual responses to different audience construals” (Pérez Llantada, 2011: 27). The L2 writers validated their contributions using similar resources as the English L1 writers. However, they provided reduced space for readers with opposing views, as did the Spanish L1 writers, suggesting that the English L2 group has transferred some of their L1 rhetorical patterns into L2. A similar conclusion was reached by Amornrattanasirichok and Jaroongkhongdach (2017), who explored RAs in Thai and English in applied linguistics, using the engagement system, noting that “particular differences in the deployment of engagement resources might be attributed to factors such as readers’ expectation, norms and conventions of the academic discourse community and disciplinary culture” (page 325). Thus, these outcomes reveal ELEnA SHELDOn Ibérica 35 (2018): 13-4034 the struggles encountered by second language writers who are forced to enter and sustain a career in a globalized academia, and how any manifestation of rhetorical differences from Anglophone normative rules may reduce those researchers’ opportunities to publish internationally. nevertheless, these texts have gained the final endorsement of editors and expert reviewers for publication, so the variation in the results of each group provides solid evidence that the international and national applied linguistics discourse communities accept a variety of rhetorical organization styles. I argue that disciplinary practices and expectations have a bearing on L1 writers’ ability to adopt a more or less authoritative stance and to facilitate an open discussion with readers. According to Hyland (2006), in the field of the social sciences greater importance is placed on explicit interpretation of results, which in itself is a difficult goal to achieve for multilingual scholars. Surprisingly, the comparison of the writer-reader/audience relationship across English and Spanish has not been the subject of much empirical research using a moves perspective and Appraisal framework in relation to the Conclusion section. 4. Conclusion The Conclusion sections written in English and Spanish in the field of applied linguistics make visible the validation of knowledge, which was noted with the endorse strategy embodying Move 1 in the present study, as these writers contextualized their results using similar evaluative features. The English L1 group mobilized heteroglossic resources in a specific way to instantiate the values of their discourse community. The deployments of contracting resources such as endorsements and of expanding resources create more space for refutation, construing a writer-reader relationship for a “big culture” (Atkinson, 2004). On the other hand, the Spanish L1 group restricted other voices through contracting resources, taking sole responsibility for the propositions made. Their Conclusion sections did not show a rich arrangement of expanding resources, which would have opened a dialogic space through entertain features. This finding supports the study by Mur-Dueñas (2011) which compared metadiscourse features in RAs in business management, written in English and Spanish. Mur-Dueñas claims that Spanish writers tend to use less interactional metadiscourse than English writers who address an international discourse community. This may suggest DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 35 that the discourse produced by the Spanish L1 group, unlike that by the English L1 group, is oriented to be read by a “national” audience. The English L2 group also limited the space for negotiation, but not to the extent of the Spanish L1 group. Thus the discourse of the English L2 group produced a hybrid dialogic space for writer/reader interaction, corroborating the findings of Pérez-Llantada (2011) who argued that the English L2 group “transfer some of their local L1 rhetorical traditions to their text in English” (page 43). The comprehensive explanation of evaluative stance developed in the present study showed how the three groups of writers used the resources of the engagement system, which would be a valuable addition in the teaching of advanced literacy. As can be seen, the engagement system provides a good platform for different options of evaluative resources in arguing the case for a writer’s own research. Therefore, understanding the dialogism in RAs represents a step forward in terms of interpersonal literacy in English. Recognizing that multilingual scholars find difficulty in persuading readers of the validity of their propositions, the engagement framework can assist them in construing effective scholarly writing. This approach “goes beyond reviewing grammatical rules and offering key phrases or new vocabulary” (Chang & Schleppregrell, 2011: 148). My findings may assist multilingual writers to better understand the construction of an assertive stance in academic writing, and these writers may thus be more successful in meeting gatekeepers’ and readers’ expectations in a new broader cultural context. Article history: Received 28 January 2018 Accepted 1 February 2018 References ELEnA SHELDOn Ibérica 35 (2018): 13-4036 Ahmad, U. (1997). “Research articles in Malay: Rhetoric in an emerging research community” in A. Duszak (ed.), Culture and Styles of Academic Discourse, 275-303. Berlin: Mouton de Gruyter. Amornrattanasirichok, S. & W. Jaroongkhongdach (2017). “Engagement in literature reviews of Thai and international research articles in applied linguistics” in Proceedings of The International Conference: DRAL 3/19th ESEA. Amnuai, W. & A. Wannaruk (2013). “A move-based analysis of the conclusion sections of research articles published in International and Thai Journals”. The Southeast Asian Journal of English Language Studies 19,2: 53-63. Atkinson, D. (2004). “Contrasting rhetoric/contrasting cultures: Why contrastive rhetoric needs a better conceptualization of culture”. Journal of English for Academic Purposes 3: 277-289. Bakhtin, M. (1981). The Dialogic Imagination (C. Emerson & M. Holquist, Trans.). Austin: University of Texas Press. Brett, P. (1994). “A Genre analysis of the results section of sociology articles”. English for Specific Purposes 13: 47-59. DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 37 Burgess, S. & P. Martín-Martín (2010). “Interpersonal features of Spanish Social Sciences journal abstracts: A diachronic study” in R. Lorés- Sanz, P. Mur-Dueñas & E. Lafuente-Millán (eds.), Constructing Interpersonality: Multiple Perspectives on Written Academic Genres, 99- 115. Newcastle upon Tyne: Cambridge Scholars Publishing. Chafe, W. (1986). “Evidentiality in English conversation and academic writing” in W. Chafe & J. Nichols (eds.), Evidentiality: The Linguistic Coding of Epistemology, 261-272. Norwood, NJ: Ablex. Chang, P. & M. Schleppegrell (2011). “Taking an effective authorial stance in academic writing: Making the linguistic resources explicit for 12 writers in the social sciences”. Journal of English for Academic Purposes 10: 140-151. Cheng, F. & L. Unsworth (2016). “Stance-taking as negotiating academic conflict in applied linguistics research articles discussion sections”. Journal of English for Academic Purposes 24: 43-57. Ciapuscio, G. & I. Otañi (2002). “Las conclusiones de los artículos de investigación desde una perspectiva contrastive”. RILL 15: 117-132. Duszak, A. (1994). “Academic discourse and intellectual styles”. Journal of Pragmatics 21: 291- 313. Ferrari, S. & S. Gallardo (2006). “Estudio diacrónico de la evaluación en las introducciones de artículos científicos de medicina”. Revista Signos 39,61: 161-180. Fortanet, I. (1997). “Verb usage in academic writing: Reporting verbs in economics research articles in English and in Spanish” in Proceedings of the 11th European Symposium on Language for Special Purposes: Identity and Interface Research, Knowledge and Society (Vol. 1), 231- 240. Copenhagen: Copenhagen Business School. Gnutzmann, C. & H. Oldenburg (1991). “Contrastive text linguistics in LSP-research: Theoretical considerations and some preliminary findings” in H. Schroder (ed.), Subject Oriented Gexts: Languages for Special Purposes and Text Theory, 103-136. Berlin: de Gruyter. Halliday, M.A.K. (1994). An Introduction to Functional Grammar. London: Edward Arnold. Hirano, E. (2009). “Research article introductions in English for Specific Purposes: A comparison between Brazilian Portuguese and English”. English for Specific Purposes 28: 240-250. Holmes, R. (2001). “Variation and text structure: The Discussion section in economics research articles”. International Journal of Applied Linguistics 131-132: 107-137. Hood, S. (2010). Appraising Research: Evaluation in Academic Writing. London: Palgrave Macmillan. Hunston, S. & G. Thompson (2000). “Evaluation: An introduction” in S. Hunston & G. Thompson (eds.), Evaluation in Text: Authorial Stance and the Construction of Discourse, 1-27. Oxford: Oxford University Press. Hyland, K. (2000). Disciplinary Discourses: Social Interactions in Academic Writing. London: Longman. Hyland, K. (2001). “Bringing in the reader: Addressee features in academic articles”. Written Communication, 18/4: 549-574. Hyland, K. (2005). “Stance and engagement: A model of interaction in academic discourse”. Discourse Studies 7,2: 173-192. Hyland, K. (2006). English for Academic Purposes: An Advanced Resource Book. London: Routledge. Hyland, K. (2009). Academic Discourse: English in a Global Context. London: Bloomsbury. Lemke, J. (1992). “Interpersonal meaning in discourse: Value orientations” in M. Davies & L. Ravelli (eds.), Advances in systemic linguistics, 82-104. London: Pinter. Loi, C., Lim, J & S. Wharton (2016). “Expressing and evaluative stance in English and Malay research article conclusions: International publications versus local publications”. Journal of English for Academic Purposes 21: 1-16. Lorés-Sanz, R. (2011). “The construction of the author’s voice in academic writing: The interplay of cultural and disciplinary factors”. Text & Talk 31,2: 173-193. Martin, J. & P. White (2005). Appraisal in English: The Language of Evaluation. London: Palgrave. Martín-Martín, P. (2005). The Rhetoric of the Abstract in English and Spanish Scientific Discourse. Hochfeldstrasse: Peter Lang. Matthiessen, C. (1995). Lexicogrammatical Cartography: English Systems. Tokyo: International Language Sciences Publishers. Mauranen, A. (1993). “Contrastive ESP rhetoric: Metatext in Finnish-English economics texts”. English for Specific Purposes 12: 3-22. Mauranen, A., Pérez-Llantada, C. & J.M. Swales (2010). “Academic Englishes: A standardised knowledge?” in A. Kirkpatrick (ed.), The Routledge Handbook of World Englishes, 634-652. London: Routledge. ELEnA SHELDOn Ibérica 35 (2018): 13-4038 Mei, W.S. (2007). “The use of engagement resources in high- and low-rated undergraduate geography essays”. Journal of English for Academic Purposes 6: 254-272. Mendiluce Cabrera, G. & A. Hernández Bartolomé (2005). “La matización asertiva en el estudio biomédico: Una propuesta de clasificación para los estudios contrastivos inglés-español”. Ibérica 10: 63-90. Moreno, A. (2008). “The importance of comparable corpora in cross-cultural studies” in U. Connor, E. Nagelhout & W. Rozycki (eds.), Contrastive Rhetoric: Reaching to Intercultural Rhetoric, 25- 41. Amsterdam: John Benjamins Publishing. Mur-Dueñas, P. (2009). “Logical markers in L1 (Spanish and English) and L2 (English) business research articles”. English Text Construction 2,2: 246-264. Mur-Dueñas, P. (2010). “Attitude markers in business management research articles: A cross- cultural corpus-driven approach”. International Journal of Applied Linguistics, 20,1: 50-72. Mur-Dueñas, P. (2011). “An intercultural analysis of metadiscourse features in research articles”. Journal of Pragmatics 43: 3068-3079. Perales-Escudero, M. & J. Swales (2011). “Tracing convergence and divergence in pairs of Spanish and English research article abstracts: The case of Ibérica”. Ibérica, Journal of the European Association of Languages for Specific Purposes (AELFE) 21: 49-70. Pérez-Llantada, C. (2011). “Heteroglossic (dis)engagement and construal of the ideal readership: Dialogic spaces in academic texts” in V. Bhatia, P. Sánchez-Hernández & P. Pérez- Paredes (eds.), Researching Specialized Languages, 25-45. Amsterdam: John Benjamins. Posteguillo, S. (1999). “The schematic structure of computer science research articles”. English for Specific Purposes 18: 139-160. Resinger, H. (2010). “Same science, same stance? A cross-cultural view on academic positioning in research articles” in R. Lóres-Sanz, P. Mur-Dueñas & E. Lafuente-Millán (eds.), Constructing Interpersonality: Multiple Perspectives on Written Academic Genres, 205- 218. Newcastle upon Tyne: Cambridge Scholars Publishing. Salager-Meyer, F. (1994). “Hedges and textual communicative function in Medical English written discourse”. English for Specific Purposes 13: 149- 170. Salager-Meyer, F. (1997). “Scientific multilingualism and ‘lesser languages’”. Interciencia, 22/4: 197-201. Samraj, B. (2002). “Introductions in research articles: Variation across disciplines”. English for Specific Purposes 21: 1-17. Sheldon, E. (2009). “From one I to another: Discursive construction of self-representation in English and Castillian Spanish research articles”. English for Specific Purposes 28: 251-265. Sheldon, E. (2011). “Rhetorical differences in RA introductions written by English L1 and L2 and Castilian Spanish L1 writers”. Journal of English for Academic Purposes 10: 238-251. Sheldon, E. (2018). Knowledge Construction in Academia: A Challenge for Multilingual Scholars. Oxford: Peter Lang. Soler-Monreal, C., Carbonell-Olivares, M. & L. Gil- Salom (2011). “A contrastive study of the rhetorical organization of English and Spanish PhD Thesis Introductions”. English for Specific Purposes 30: 4- 17. Swales, J. M. (1990). Genre Analysis: English in Academic and Research Settings. Cambridge: Cambridge University Press. Swales, J. N, (2004). Research Genres: Explorations and Applications. New York, NY: Cambridge University Press. Tang, R. & S. John (1999). “The ‘I’ in identity: Exploring writer identity in student academic writing through the first person”. English for Specific Purposes 18: S23-S39. Taylor, G. & T. Chen (1991). “Linguistics, cultural, and sub-cultural issues in contrastive discourse analysis: Anglo-American and Chinese scientific texts”. Applied Linguistics 12: 319-336. William, I. (2005). “Thematic items referring to research and researchers in the discussion section of Spanish biomedical articles and English- Spanish translations”. Babel 51,2: 124-160. Yang, R. & D. Allison (2003). “Research articles in applied linguistics: Moving from results to conclusions”. English for Specific Purposes 22: 365-385. Yakhontova, T. (2002). “Selling or telling? Towards understanding cultural variation in research genres” in J. Flowerdew (ed.), Academic Discourse, 216-232. London: Longman. DIALOgIC SPACES Of knOWLEDgE COnSTRuCTIOn Ibérica 35 (2018): 13-40 39 Dr Elena Sheldon is a lecturer at the faculty of Arts and Social Sciences, at the university of Technology Sydney (uTS). Her research interests focus on systemic functional grammar, academic writing, and contrastive rhetoric applied to research. She has published articles in international journals contrasting English L1, English L2 and Spanish L1 and a book titled: Knowledge Construction in Academia: A Challenge for Multilingual Scholars. NoTES 1 Prosody is viewed by Lemke (1992: 47) as any evaluative feature in texts, a feature that “is not restricted to a particular meaning distributed through the clause, and across the clause and sentence boundaries but evaluations spread throughout the text”. Evaluative stances are thus not unrelated words or clauses but choices that communicate with each other throughout the text in making meaning (Chang & Schleppegrell, 2011). 2 graduation provides a key dialogistic effect associated with scaling-up/scaling-down the preciseness of statements. These resources can assist the engagement system to explore its dialogic functionality more intensively by considering their potential effects in the construal of agreement and solidarity (Martin & White, 2005). 3 Drawing on the modality of possibilities, these texts employ “might”, “shall”, “may”, “can”, “should”, “could” and “seems”, followed by a verb, which indicates stronger or weaker investment in the propositions.