The discourse-pragmatic functions of if/si-constructions in English and Spanish spoken academic discourse

ASICE / EDICE SOPRAG 2023, 11(1): 42-56 Cris%na Lastres-López* The discourse-pragma1c func1ons of if/si-construc1ons in English and Spanish spoken academic discourse 1 Las funciones pragmá1co-discursivas de las construcciones introducidas por if/si en el discurso académico oral en inglés y español https://doi.org/10.17710/soprag.2023.11.1.lastreslopez3 Publicado en línea el 30 de junio 2023 Abstract: Although if/si-construc.ons are usually defined as condi.onal construc.ons which exhibit cause-consequence pa:erns, prior research has evidenced that these construc.ons may fulfil a wider range of discourse-pragma.c func.ons. In addi.on, research delving into the uses and func.ons of these construc.ons in specific registers is scarce and, moreover, few studies have adopted a contras.ve perspec.ve. This paper intends to fill this gap by examining if/si-construc.ons in English and Spanish in a comparable corpus of spoken academic discourse, a register in which condi.onals are frequent (Biber, Johansson, Leech, Conrad and Finegan, 1999, p. 824-825). The theore.cal framework is based on the three metafunc.ons proposed in Systemic Func.onal Linguis.cs (Halliday and Ma:hiessen, 2014), allowing us to dis.nguish if/siconstruc.ons at the idea.onal, interpersonal and textual levels. Data are drawn from the spoken academic subcorpora of the Bri.sh component of the Interna+onal Corpus of English (ICE-GB) (Nelson, Wallis and Aarts, 2002) and of the Spanish component of the Integrated Reference Corpora for Spoken Romance Languages (C-ORAL-ROM) (Cres. and Moneglia, 2005). Corpus data throw light on the use of these construc.ons in English and Spanish. Results show that condi.onals are used differently in colloquial conversa.on and in spoken academic discourse. Keywords: condi.onal construc.on, academic discourse, corpus linguis.cs, idea.onal, interpersonal Resumen: Aunque las construcciones introducidas por if y si se definen normalmente como construcciones condicionales que exhiben un patrón de causa-consecuencia, * Correspondencia con la autora: Cris&na Lastres-López, Department of English Language and Linguis@cs, University of Seville, Palos de la Frontera s/n, 41004, Seville, Spain, e-mail: [email protected] 1 Acknowledgements: I would like to thank the financial support provided by the Spanish Ministry of Science and Innova@on (grant PID2020-114604GB-100). C. Lastres-López The discourse-pragma@c func@ons of if/si construc,ons - 43 inves.gaciones previas han evidenciado que estas construcciones pueden tener un rango más amplio de funciones pragmá.co-discursivas. Además, la inves.gación que ahonda en los usos y funciones de estas construcciones en registros específicos es escasa y, adicionalmente, pocos estudios han adoptado una perspec.va contras.va. Este ar_culo .ene como obje.vo contribuir a llenar este vacío examinando construcciones introducidas por if y si en inglés y español en un corpus comparable de discurso académico oral, un registro en el que las condicionales son frecuentes (Biber, Johansson, Leech, Conrad y Finegan, 1999, p. 824-825). El marco teórico se basa en las tres metafunciones propuestas en la Lingüís.ca Sistémico-Funcional (Halliday y Ma:hiessen, 2014), lo que nos permite dis.nguir construcciones introducidas por if y si en los niveles ideacional, interpersonal y textual. Los datos se extraen de los subcorpus académicos orales del componente británico del Interna+onal Corpus of English (ICE-GB) (Nelson, Wallis y Aarts, 2002) y del componente español del Integrated Reference Corpora for Spoken Romance Languages (C-ORAL-ROM) (Cres. y Moneglia, 2005). Los datos de corpus arrojan luz sobre el uso de estas construcciones en inglés y español. Los resultados muestran que las construcciones condicionales se emplean de forma diferente en la conversación coloquial y en el discurso académico oral. Palabras clave: construcción condicional, discurso académico, lingüís.ca de corpus, ideacional, interpersonal 1. Introduc,on Condi.onal construc.ons –most frequently introduced by if in English, as in (1), and si in Spanish, as in (2) – typically express a rela.on of con.ngency between two clauses. They usually follow a cause-consequence pa:ern, in which the protasis –marked in bold type– indicates the cause, and the apodosis –in italics– expresses the consequence. In these construc.ons “the truth of the proposi.on in the matrix clause is a consequence of the fulfilment of the condi.on in the condi.onal clause” (Quirk, Greenbaum, Leech and Svartvik, 1985, p. 1088). (1) If the weather is fine, (then) we will have a barbecue. (Quirk et al., 1985, p. 1088) (2) Si invierte en nuestra compañía, obtendrá importantes ganancias. (RAEASALE, 2009, p. 3551) ‘If you invest in our company, you will get significant profits’ However, as Declerck and Reed (2001, p. 1) argue, “condi.onal clauses seem to have many more seman.c and pragma.c func.ons than has hitherto been evident”. Prior research has explored the pragma.c richness of condi.onal construc.ons (Ford & Thompson, 1986; Ford, 1997; Lavid, 1998; Montolío Durán, 1999a; Warchal, 2010; Brinton, 2019, among others), showing that these construc.ons also fulfil various interpersonal and textual func.ons in discourse. Illustra.ons of such uses are provided in examples (3) and (4). (3) Indeed, the only real drawback, if you can call it that, is that people are conXnually coming up and congratulaXng us on our victory over England. (Biber et al., 1999, p. 856) C. Lastres-López The discourse-pragma@c func@ons of if/si construc,ons - 44 (4) It is a fact that you have refused to take any fee for the work you are doing, if you don’t mind my asking? (Biber et al., 1999, p. 857) Research on condi.onal construc.ons from a contras.ve perspec.ve is notably scarce. Excep.ons are Carter-Thomas (2007), on English and French; Hasselgård (2014), on English and Norwegian; and Lastres-López (2019, 2021), on English, French and Spanish. Similarly, there is also a gap in the literature with regard to the study of these construc.ons across registers. 2 While medical discourse (Ferguson, 2001; Carter- Thomas, 2007; Carter-Thomas and Rowley-Jolivet, 2008; Rowley-Jolivet and Carter- Thomas, 2008) or legal discourse, (Mazzi, 2013; Lastres-López, 2019) have received some scholarly a:en.on, other registers –especially spoken ones– remain largely understudied with respect to condi.onal construc.ons. This paper, therefore, intends to contribute to fill this double gap by exploring the discourse-pragma.c func.ons of if/si-construc.ons in English and Spanish spoken academic discourse. We restrict our analysis to the two prototypical markers of condi.onality –if and si– given that prior research has shown that the large majority of condi.onals in both languages are introduced by such markers (Gabrielatos, 2019, p. 308; Lastres-López, 2019, p. 60). Regarding the selec.on of the register under examina.on –spoken academic discourse– the choice is mo.vated by the relevance of condi.onals in this text type. In this respect, Biber et al. (1999, p. 824-825) argue that condi.onals are frequent in this register, given that they serve to introduce or develop arguments. Along the same lines, Carter-Thomas and Rowley-Jolivet (2008, p. 191) also show that condi.onals are a “highly valuable resource in academic discourses, whether spoken or wri:en, as they can be used to hypothesize, hedge, manage interac.on with the addressee, and promote or on the contrary circumscribe the scope of research claims”, hence the interest in exploring their discourse-pragma.c func.ons in this register in par.cular. With respect to register varia.on, Biber (2006, p. 4) also argues that spoken academic discourse tends to be colloquial, as opposed to wri:en academic discourse, which is generally informa.onal. Therefore, a second objec.ve of this paper is to examine to what extent spoken academic discourse is similar to colloquial face-to-face conversa.on in English and Spanish with regard to condi.onal construc.ons. For this purpose, the results provided in this paper for spoken academic discourse will be compared to those discussed in a prior study on face-to-face conversa.on (Lastres- López, 2021). The data analysed are extracted from the spoken academic subcorpora of the Bri.sh component of the Interna+onal Corpus of English (ICE-GB) (Nelson et al., 2002) and of the Spanish component of the Integrated Reference Corpora for Spoken Romance Languages (C-ORAL-ROM) (Cres. and Moneglia, 2005). Although these corpora have not been compiled using the same sampling frame, their spoken academic subcorpora in par.cular are largely comparable. Aqer this introduc.on, the paper is structured as follows: Sec.on 2 offers a brief review of the literature on the discourse-pragma.c func.ons of condi.onals and 2 Following Conrad and Biber (2001) and Biber (2006), among others, the term register is employed here as a synonym for genre and text type. It is used to refer to “situationally-defined varieties described for their characteristic lexico-grammatical features” (Biber, 2006, p. 11). C. Lastres-López The discourse-pragma@c func@ons of if/si construc,ons - 45 establishes the theore.cal framework adopted. Sec.on 3 explains the corpora analysed and the methodology followed. Sec.on 4 discusses the corpus-based findings. This sec.on is divided into three subsec.ons: 4.1 shows the frequency of these construc.ons, 4.2 examines their discourse-pragma.c func.ons, and 4.3 offers a brief register comparison on spoken academic discourse and colloquial face-to-face conversa.on. Finally, Sec.on 5 presents some concluding remarks. 2. The discourse-pragma,c func,ons of if/si-construc,ons As discussed in Sec.on 1, prototypical condi.onal construc.ons respond to a causeconsequence pa:ern. However, previous research has shown that these construc.ons can fulfil a wide range of discourse-pragma.c func.ons. Indeed, as Traugo:, ter Meulen, Reilly, and Ferguson (1986, p. 1) note, “some sentences with the formal markers of condi.onality are seman.cally and pragma.cally only marginally condi.onal or not condi.onal at all”. With respect to English, Quirk et al. (1985, p. 1088-1089) establish an important first dis.nc.on between direct and indirect condi+ons. Whereas the first type includes prototypical cause-consequence pa:erns, as in (1) and (2) above, the la:er includes cases related to the “implicit speech act of the u:erance” (Quirk et al., 1985, p. 1089), rather than to the proposi.onal content of the matrix clause. In Spanish, similar dis.nc.ons are also found in grammars. The grammar of the RAE-ASALE (2009, p. 3550- 3551) considers two types of condi.onals: condicionales del enunciado and condicionales de la enunciación, which would be equivalent to the direct and indirect types dis.nguished by Quirk et al. (1985). Likewise, Montolío Durán (1999a, p. 3683- 3690) regards the indirect type as a “peripherical” use of the condi.onal construc.on, used to convey politeness, to make metalinguis.c comments or to express relevance. Func.onal-cogni.ve models of condi.onality have also shown dis.nc.ons between content condi.onals, epistemic condi.onals and speech act condi.onals (Sweetser, 1990; Dancygier and Sweetser, 2000, 2005). Content condi.onals correspond to prototypical pa:erns of condi.onality in which the protasis indicates a cause and the apodosis its consequence or effect. Epistemic condi.onals, in turn, are those related to the verbaliza.on of a process of reasoning on the part of the speaker; they can be paraphrased by “If I know [protasis], then I conclude [apodosis]” (Sweetser, 1990, p. 121). An example of an epistemic condi.onal is presented in (5), in which the speaker infers that John went to the party to make Miriam angry. Finally, speech act condi.onals can be paraphrased as “If [protasis], then let us consider that I perform this speech act (i.e., the one represented as the apodosis)” (Sweetser 1990, p. 121), as shown in example (6). (5) If John went to that party, (then) he was trying to infuriate Miriam. (Sweetser, 1990, p. 116) (6) If I may say so, that’s a crazy idea. (Sweetser, 1990, p. 118) In addi.on to the cases of condi.onal construc.ons introduced by if and si already discussed, construc.ons introduced by the same conjunc.ons can also be used as instances of insubordina.on. Evans (2007, p. 367), who coined the term insubordina+on, defines this phenomenon as “the conven+onalized main clause use of what, on prima C. Lastres-López The discourse-pragma@c func@ons of if/si construc,ons - 46 facie grounds, appear to be formally subordinate clauses” (italics in the original). In other words, u:erances that contain formal markers of subordina.on, such as the subordina.ng conjunc.ons if and si, also appear in cases in which the if/si-clause stands alone, without the presence of any accompanying matrix clause, as illustrated in (7) and (8). In these cases, the if/si-clause is reanalysed as a main clause itself. Evans (2007) argues that these construc.ons have conven.onalized their func.ons and, despite appearing as stand-alone clauses, their meanings are fully complete in the discourse situa.ons. For instance, example (7) encodes a request by means of an insubordinate ifconstruc.on and (8) illustrates a wish on the part of the speaker. In general, Van linden and Van de Velde (2014, p. 226) argue that insubordinate construc.ons convey “interpersonal meaning”. In the Spanish literature, these construc.ons are oqen examined under the labels of condicional suspendida (suspended condi.onal) or prótasis suspendida (suspended protasis) (RAE-ASALE, 2009, p. 3547, 3549) or realización independiente con si (independent realiza.on with si) (Montolío Durán, 1999a, p. 3681). 3 (7) If you could give me a couple of 39c stamps please. (Evans, 2007, p. 380) (8) ¡Si acabara la tesis este verano! (Gras, 2011, p. 292) ‘If I could finish the dissertaXon this summer!’ While cases of if/si-condi.onal construc.ons and instances of insubordina.on introduced by the same markers have generally been studied independently, we propose here a categoriza.on that encompasses both types of construc.ons. For such purposes, the theore.cal framework adopted is based on three metafunc.ons dis.nguished in Systemic Func.onal Linguis.cs (Halliday and Ma:hiessen, 2014), allowing us to categorize if/si-construc.ons at the idea.onal, interpersonal and textual levels. Idea.onal if/si-construc.ons include prototypical cases of condi.onal construc.ons, that is, those which display a cause-consequence pa:ern. Interpersonal if/siconstruc.ons, in turn, put the emphasis on the interac.on between the discourse par.cipants. A broad concep.on of interpersonal if/si-construc.ons would encompass the epistemic and speech act subtypes dis.nguished by Sweetser (1990) and by Dancygier and Sweetser (2000, 2005), as well as instances of insubordina.on introduced by if and si, given that insubordinate construc.ons “all share the seman.c property of expressing interpersonal meaning” (Van linden and Van de Velde, 2014, p. 226). Finally, textual if/si-construc.ons are those which explicitly signal cohesion in discourse. As Carter-Thomas (2007, p. 158) argues, condi.onals “serve to instruct or guide the readers in following the development of the text”, fulfilling thus a textual func.on. The adop.on of this model –also applied in Lastres-López (2021) – will thus allow us to explore the interpersonal and textual nature of if/si-construc.ons in spoken academic English and Spanish, in addi.on to their prototypical idea.onal func.on. Prior studies have addressed the relevance of the interpersonal and the textual metafunc.ons for the construc.ons under analysis in this paper. In par.cular, Warchal (2010), who examines wri:en academic discourse, emphasizes how research ar.cles build interpersonal rela.ons between the writer and the reader through condi.onal clauses. In the same vein, with respect to the textual metafunc.on, Biber, Conrad and Cortés (2004, p. 371), 3 For comprehensive analyses of these construc@ons in Spanish, see Schwenter (1996, 1999, 2016a, 2016b), Montolío Durán (1999b), Gras (2011) and Pérez Béjar (2018, 2019, 2022). C. Lastres-López The discourse-pragma@c func@ons of if/si construc,ons - 47 state that condi.onals are among the various construc.ons which can be used as “discourse organizing bundles” in academic discourse. Hence, the interest in studying if/si-construc.ons from a broader perspec.ve –including within the same framework of analysis cases of subordina.on, namely, condi.onal construc.ons, together with instances of insubordina.on – and providing a classifica.on which can encompass the various func.ons that these construc.ons may fulfil in discourse. 3. Corpora and methodology In corpus-based contras.ve linguis.cs, one of the most important concerns to be addressed is the use of comparable corpora. These are corpora which contain original texts in two or more languages and are compiled using the same sampling frame (Johansson, 1998; McEnery and Wilson, 2001; Hunston, 2002). As, to date, there is no contras.ve corpus of spoken discourse that samples English and Spanish in the register under analysis, we have resorted to the spoken academic components of similar reference corpora in both languages. For English, we employ the Bri.sh component of the Interna+onal Corpus of English (ICE-GB) (Nelson et al., 2002), extrac.ng data from the classroom lessons component of the corpus (texts “S1B-001” to “S1B-020”). For Spanish, we use the Spanish component of the Integrated Reference Corpora for Spoken Romance Languages (C-ORAL-ROM) (Cres. and Moneglia, 2005), selec.ng the texts that also contain academic discourse. In par.cular, these are labelled in this corpus as “classroom lessons” (texts “ena:e01” to “ena:e04”) and “conferences” (texts “enatco01” to “enatco04”). Despite the different labelling, their contents are very similar, and no differences have been found with respect to the use of si-construc.ons. Regarding the size of the corpora, the number of texts in the English corpus is 20, totalling 42,210 words. In contrast, the Spanish dataset is composed of 8 texts, which amount to 24,628 words. Given the different size of the corpora analysed, normalized frequencies per ten thousand words will be provided, so that the quan.ta.ve results are comparable. The process of data extrac.on from the corpora was carried out searching for if and si respec.vely in the respec.ve interfaces of the corpora, ICECUP, for ICE-GB, and CONTEXTES, for C-ORAL-ROM. The automa.c extrac.on rendered a total of 206 tokens in English and 69 tokens in Spanish. A manual analysis of the data was then carried out to exclude cases which did not correspond to condi.onal or insubordinate uses of if and si. Once excluded, we annotated the data according to two variables: (i) their gramma.cal status, that is, condi.onal construc.ons or insubordinate construc.ons; and (ii) their metafunc.on, that is, idea.onal, interpersonal or textual. 4. Results and discussion 4.1 Frequency As men.oned in Sec.on 3, the 206 tokens retrieved from the English corpus and the 69 tokens extracted from the Spanish corpus were manually analysed to discard cases that did not correspond to condi.onal or insubordinate construc.ons. Table 1 presents the C. Lastres-López The discourse-pragma@c func@ons of if/si construc,ons - 48 absolute frequency and the normalized frequency per ten thousand words of the tokens obtained from the corpora in both languages. Table 1. Frequency in English and Spanish Type English Spanish N p9w N p9w CondiXonal construcXons 137 32.46 48 19.49 Insubordinate clauses 3 0.71 1 0.41 Complement clauses 11 2.61 12 4.87 ComparaXve clauses (as if, like if, como si) 13 3.08 4 1.62 Concessive clauses (even if) 2 0.47 0 0 RepeXXons, false starts, interrupXons, etc. 40 9.48 4 1.62 Among the cases discarded, both languages presented instances of complement clauses introduced by if and si, as illustrated in examples (9) and (10). (9) So I don’t know if uh many of you came across this but there’s a very nice story in Tarbull’s autobiography <,> (ICE-GB:S1B-005 #173:1:A) (10) O sea que lo que puede interesar a un sociolingüista es si hay una diferencia en la forma de hablar (C-ORAL-ROM, enatco04) ‘That is, what can interest a sociolinguist is if there is a difference in the way of speaking’ Cases of complex subordinators were also found in the samples examined and discarded for further analysis. On the one hand, this includes clauses introduced by as if and like if in English and their Spanish equivalent como si, indica.ng a comparison, as in (11) and (12). On the other hand, instances of concessive clauses –introduced by even if– were only found in the English corpus, probably as a result of the reduced size of the sample analysed in Spanish. An illustra.on of an even if-clause is presented in (13). (11) Uh it was almost as if uhm she had tried to learn a second language and she was always very hal.ng in her use of language (ICE-GB:S1B-003 #106:1:B) (12) Crecía como si fuera un individuo más de la familia (C-ORAL-ROM enatco01) ‘He was growing as if he were another member of the family’ (13) So even if we try to say that the cons.tu.on is that body of rules, we discover that it’s not just rules (ICE-GB:S1B-011 #142:1:A) Finally, both corpora presented cases of typical speech phenomena such as repe..ons, false starts and interrup.ons, as illustrated in (14) and (15); these were also discarded for further analysis. C. Lastres-López The discourse-pragma@c func@ons of if/si construc,ons - 49 (14) I mean <,,> if if if it was present <unclear-words> (ICE-GB:S1B-009 #209:1:D) (15) También podríamos decir si cada día uno escribe sus propios [///] (C-ORAL- ROM, enatco03) ‘We could also say if every day one writes their own’ Quan.ta.vely, as expected, condi.onal construc.ons represent the large majority of the cases in both languages. However, it is worth no.ng a contras.ve difference associated with the frequency of condi.onals in English and Spanish. English condi.onals present a higher normalized frequency (32.46) than their Spanish counterparts (19.49), a finding which indicates that full condi.onal construc.ons seem to be more frequently employed in English spoken academic discourse than in the same register in Spanish. With respect to insubordinate clauses, the results show very low frequencies in both languages. This goes in line with corpus findings from prior studies in other registers, which indicates that condi.onal insubordina.on represents low propor.ons as compared to condi.onal construc.ons (Lastres-López, 2021). Interes.ngly, however, instances of insubordina.on were found in both corpora. Both if/si-condi.onal construc.ons and if/si-insubordinate clauses will be discussed in full in Sec.on 4.2. 4.2. Discourse-pragma7c func7on As already men.oned, both condi.onals and insubordinate construc.ons were manually annotated according to their discourse-pragma.c func.on. For this purpose, as discussed in Sec.on 3, we classified the tokens according to the three metafunc.ons considered in Systemic Func.onal Linguis.cs (Halliday and Ma:hiessen, 2014), dis.nguishing thus three types of if/si-clauses: idea.onal, interpersonal and textual. Table 2 presents the distribu.on of idea.onal, interpersonal and textual if/siconstruc.ons in English and Spanish, including absolute frequencies, percentages with respect to the total number of condi.onals and insubordinate construc.ons in each language, and normalized frequencies per ten thousand words. Table 2. Distribu+on of idea+onal, interpersonal and textual if/si-construc+ons in English and Spanish. Type English Spanish N % p9w N % p9w IdeaXonal 112 80.00 26.53 33 67.35 13.40 Interpersonal 27 19.29 6.40 16 32.65 6.50 Textual 1 0.71 0.24 0 0 0 Total 140 100 - 49 100 - As shown in Table 2 above, idea.onal if/si-construc.ons are the most frequent construc.on in both languages, amoun.ng to 80% of the cases in English and to 67.35% in Spanish. With respect to their normalized frequencies, the idea.onal metafunc.on is twice more frequent in English (26.53) than in Spanish (13.40), showing that, in spoken academic discourse, English resorts more to these construc.ons . Idea.onal condi.onals are revealed as important mechanisms in academic discourse to provide explana.ons, C. Lastres-López The discourse-pragma@c func@ons of if/si construc,ons - 50 since they allow the speaker to establish the cause of an event in the protasis of the condi.onal and express its consequence or effect in the apodosis. Illustra.ons are provided in examples (16) and (17). In (16) the lecturer in a geology class discusses the consequence of dividing the stra.graphic column, whereas in (17) the lecturer in a linguis.cs class men.ons the possible effect of providing a child with a building game. (16) If you were to divide up the stra.graphic column today you would get four main divisions (ICE-GB:S1B-006 #286:1:A) (17) Si se ofrece a un niño piezas de un juego de construcción, es posible que haga lo que hacían los niños del ejemplo de Vigotsky, hacer construcciones (C- ORAL-ROM, enatco03) ‘If a child is offered pieces of a building game, it’s possible that he does what the children from the example of Vigotsky did, make construc.ons’ Interpersonal if/si-construc.ons are second in frequency in both languages. These include 24 condi.onals and 3 instances of insubordina.on in English, and 15 condi.onals plus an insubordinate clause in Spanish. In total, interpersonal if/si-construc.ons represent 19.29% of the total of cases of if in English, whereas the propor.on is higher in Spanish, amoun.ng to 32.65%. Despite these differences with respect to the total number of if/si-tokens, the normalized frequencies show that these construc.ons are similarly frequent in English and Spanish, with frequencies per ten thousand words of 6.40 and 6.50 respec.vely. Therefore, both languages seem to express interpersonality through condi.onals in very similar rates in spoken academic discourse. Interpersonal condi.onals are illustrated in examples (18) and (19). They allow the speaker to engage with the audience, a func.on that Hyland (2005) argues is typical of academic discourse. This engagement is carried out by means of “nego.a.ng meanings and standpoints rather than merely providing informa.on” (Warchal, 2010, p. 141), as would be the case with idea.onal condi.onals. This interpersonal func.on is thus also related to mi.ga.on (Caffi, 2007, p. 65, 67; Chodorowska-Pilch, 2017), since the content is presented as less asser.ve by means of if you like in (18) and si se quiere in (19), respec.vely. (18) So if you like that is the defini.ve statement for the .me being (ICE-GB:S1B- 007 #205:1:A) (19) Una parte de esta ac.vidad que es propiamente cogni.va y si se quiere dirigida a la resolución de problemas (C-ORAL-ROM, entaco03) ‘One part of this ac.vity which is specifically cogni.ve and if you like oriented to the resolu.on of problems’ As discussed above, most of the interpersonal cases belong to condi.onal construc.ons. However, and despite the rela.vely small data sample analysed, instances of insubordina.on are present in both languages, amoun.ng to three tokens in English and only one in Spanish. Examples of if-insubordina.on and si-insubordina.on are shown in (20) and (21) respec.vely. Even though these are construc.ons in which the protasis stands alone, without any accompanying apodosis, they are not considered as cases of incomplete sentences (Evans, 2007), since their meaning is fully complete in the