Distance, evidentiality and counter-argumentation: Concessive future in Spanish

Available online at www.sciencedirect.com

ScienceDirect Journal of Pragmatics 85 (2015) 181--199 www.elsevier.com/locate/pragma

Distance, evidentiality and counter-argumentation: Concessive future in Spanish Susana Rodríguez Rosique * Department of Filología Españ ola, University of Alicante, Campus de San Vicente, Ap. 99, E-03080 Alicante, Spain

Abstract The aim of this paper is to examine the connection between Spanish concessive future and evidentiality. More specifically, it is argued that concessive future does not merely represent a contextual variant of conjectural future (Escandell, 2010) but a new use (Squartini, 2012) which is restricted by the activated status of the proposition (Dryer, 1996). Unlike conjectural future, concessive future does not have an inferential purpose; instead, it plays a role within the (counter) argumentation process. It actually develops a déréalisant function (Ducrot, 1995). Therefore, although conjectural use and concessive use share the deictic value of future, they are the projection of this value over different levels of meaning. More generally, this paper shows that future in Spanish may intersect with several semantic and discourse categories, including -- but not limited to -- evidentiality. © 2015 Elsevier B.V. All rights reserved.

Keywords: Concessive future; Evidentiality; Counter-argumentation; Distance

1. Introduction From Bello ([1847] 1970: 221) to Nueva Gramática de la Real Academia Españ ola (2009: 1767--1768), Spanish traditional grammar has regarded future as a mark of tense; more precisely, it has been defined as a verb morpheme expressing tense. Tense is a particular expression of linguistic time and, since it is grammaticalized as morphological information inside the verb, it constitutes a grammatical category. However, tense is a very special grammatical category indeed, since it inherently fulfills a referential or contextual function through which it becomes a deictic category (Benveniste, 1966; Comrie, 1976, 1985). In fact, tense links the situation denoted by the verb to the now of the speaker, and that happens due to three primary time references: anteriority (1); simultaneity (2); and posteriority (3).1 (1) Fuimospast a Madrid. [We went to Madrid] (2) Estamospres. en Madrid. [We are in Madrid] * Tel.: +34 965 90 34 00x2107. E-mail address: [email protected]. 1 Situations may be directly oriented toward the now of the speaker, as it happens with future tense (Iremosfut a Madrid), or they may have an indirect orientation; that is, situations may be oriented toward an intermediate point which is in turn oriented toward the now of the speaker. This is what happens with conditional
also described as future of the past
(Al día siguiente iríamoscond a Madrid). The difference between absolute and relative tenses becomes clear in the light of these examples (Comrie, 1985). http://dx.doi.org/10.1016/j.pragma.2015.03.017 0378-2166/© 2015 Elsevier B.V. All rights reserved.

182

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

(3) Iremosfut. a Madrid. [We will go to Madrid] Future places the situation denoted by the verb after the now of the speaker and consequently invokes scenes which have not happened yet. As a result, it usually relates to speculation and to different modal values, which somehow depend on context. Thus, when future appears with the first person, it is generally associated with a promise or with the speaker’s commitment to the action expressed by the verb (4); however, its use with the second person may lead us to interpret it as an order, a piece of advice or a prohibition (5); if the subject is a non-volitional third person, future may be related to a clear predictive value (6); and when it denotes general situations and appears with an undetermined subject, it may even trigger a categorical or necessity-related reading (7). (4) Este fin de semana te ayudaréfut. con los deberes. [I will help you with your homework this weekend] (5) Esta tarde limpiarásfut. tu habitacio´n. [You will clean your bedroom this afternoon] (6) Mañana lloveráfut. en Alicante. [It will rain in Alicante tomorrow] (7) Todos moriremosfut. algún día. [Everyone of us will die someday] In spite of the modal values contextually triggered in examples (4) to (7), future still plays a time-related function in all of them: it still places the situation after the ‘‘now’’ of the speaker. Nevertheless, future in Spanish may be used to denote situations which are not located after the point of reference or the moment of speech. When this happens, future is utilized to convey that the proposition reflects a conjecture or a speaker’s calculation (Bello [1847], 1970: 236): ?

(8) - Qué hora es? [What time is it?] -Seránfut. las cuatro. [It must be (befut.) four] In these cases, the temporal form is extracted from its natural context (posteriority), which justifies the fact that some linguists consider it a case of dislocation (Rojo, 1974; Rojo and Veiga, 1999). This use of future has been traditionally related to epistemic modality -- and more specifically, to probability --, since it may be rephrased using a present tense form and an adverb of modality: (9) En estos momentos seránfut. las cuatro. [In this moment, it must be (befut.) four] > Probablemente sonpres. las cuatro. [It is probably four] (Rojo and Veiga, 1999: 2912). Cases such as (10B) are usually included to exemplify a contextual variation of this conjectural or epistemic future: (10) A: Juan es muy simpático. [Juan is very nice] B: Seráfut. muy simpático, pero a mí nunca me saluda. [He may be (befut.) very nice, but he never says hello to me] The behavior of future in (10) has been labeled in European Spanish as concessive, and has been characterized by a number of recurrent features: it is frequently followed by pero conjunction2; it commonly occurs in replies; and it is quite similar to a concessive move -- in fact, it very much resembles the aunque + subjunctive construction in meaning: Aunque 2 Concessive future may also be followed by any other contrastive conjunction or discourse marker which discursively works in a similar way to pero. Gili Gaya actually suggests a structure followed by the discourse marker sin embargo as an example of concessive future: -Fulano es un sabio; -Lo seráfut.; sin embargo, se ha equivocado muchas veces ([1951] 1993: 166) [So-and-so is a wise man; -- He may be (befut.); however, he has made mistakes many times]. Note that sin embargo is used to cancel the conclusion toward which the previous argument is oriented (Martín Zorraquino and Portolés, 1999: 4115). Despite the differences that pero and sin embargo may reveal, both could develop a similar counterargumentation function on some occasions.

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

183

sea muy simpático, a mí nunca me saluda [Even though he is very nice, he never says hello to me] -- (RAE, 1973: 471; Gili Gaya, [1951] 1993: 166; Fernández Ramírez, 1986: 288; Flamenco García, 1999: 3821; RAE, 2009: 1772). Both conjectural -- or epistemic -- future and concessive future have been linked to evidentiality; that is, to the linguistic phenomenon which has analyzed the source of information and the mode of knowledge from Jakobson ([1957] 1971), with different scopes depending on whether the perspective was broad or narrow. It is now widely assumed that the grammatical codification of the aspects included under the label of evidentiality becomes visible in a relatively limited number of languages (Aikhenvald, 2004); however, almost every language has several mechanisms to indicate from where and how knowledge is obtained (Chafe and Nichols, 1986). The aim of this paper is to investigate the relationships between conjectural or epistemic future, concessive future and evidentiality; more precisely, the argument is that concessive future constitutes a new use, no longer defined in inferential terms -- as it happens with conjectural future -- and regulated by the activated status of the proposition where it appears. Conjectural value and concessive value are thus jointly explained due to the deictic character of future, but they actually result from the projection of future’s deictic instruction over different levels of meaning. In a broader sense, the present paper shows that future in Spanish comes across different categories, including -- but not limited to -- evidentiality. Seeking to achieve the aforementioned aim, the next section examines different approaches to the concept of evidentiality and the treatment that conjectural future and concessive future have received from those approaches; this section also offers some data that may prove problematic for the explanations of concessive future proposed so far. Section 3 displays the deictic perspective, which explains the different uses of future in a unified way and serves as a framework to place the concessive future. Section 4 provides a detailed explanation of concessive future as a case of distance toward the utterance and pays attention to the (counter)argumentative consequences deriving from it. Finally, Section 5 summarizes the conclusions drawn from the analysis and explains how a deictic definition of future traverses different semantic and discourse categories. 2. Concessive future in a system of evidentiality From the point of view of evidentiality, or according to its connections with the source of information and the type of evidence, future and conditional shape a micro-system where it is specified that the speaker’s access to information is not direct, but indirect (Squartini, 2001; Escandell, 2010: 22).3 Within this new paradigm, future in Spanish tells us that the source of information is the speaker, and also that the type of evidence or mode of knowledge is an inference, as shown in (11). Concerning conditional in Spanish, it may also convey the same inferential meaning in the past, as can be seen in (12); but it may express that the source of information is external and the type of evidence is a report on some other occasions (13). Table 1 summarizes all these values. Table 1 Future and conditional in a system of evidentiality. Indirect evidence in Spanish Future Conditional

Source: Speaker Type of evidence: Inference Source: Speaker Type of evidence: Inference Source: External Type of evidence: Report

(11) Seránfut. las cuatro. [It must be (befut.) four] (12) Moriríacond. el año pasado. [He must have died (diecond.) last year] (13) El Gobierno seríacond. partidario de recrudecer las medidas para evitar la inflacio´n. [The Government would favor ( favorcond.) reinforcing the measures to prevent inflation] Once these values have become identified, the evidential analysis of future will be closely linked to the scope of the notion of evidentiality. From a wider perspective, evidentiality has been regarded as a functional-conceptual substance 3 Squartini (2001) argues that this system extends to all Romance languages, although he acknowledges that it is unstable and may present variations in each specific language.

184

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

domain that refers to the source of information, the evidence or the justification for the proposition (Boye and Harder, 2009: 14). In Boye and Harder’s view, evidentiality can be understood neither as an exclusively grammatical concept nor as an exclusively lexical coded phenomenon. It should actually be seen as a kind of continuum where grammatical evidentiality does not appear as a static category but as something directly associated with the grammaticalization process; more specifically, it is contemplated as a material that moves progressively toward the status of coded secondariness4 due to its high frequency of use. The work by Hennemann (2013) can be placed within this broad conception of evidentiality. She bases her analysis of conjectural future on the presence of contextual clues to detect a process of inference. Consequently, Hennemann sees as conjectural futures those appearing with causal conjunctions (14), those occurring with subordinating verbs such as calcular [to calculate], concluir [to conclude], vaticinar [to foretell], prever [to foresee], predecir [to predict] (15), or those included in any structure where a reasoning process can be deduced (16). (14) El récord no seráfut. difícil de superar, porque a la máquina le quedan por delante, al menos, tres meses de trabajo hasta abrir un orificio de unos 66 centímetros de diámetro. [The record will not be difficult to overcome, because the machine still needs at least three months of work to open a hole of about 66 inches in diameter] (Hennemann, 2013: 394) (15) Sergio Scariolo, el seleccionador, prevé que el Mundial que empieza mañana en Turquía seráfut. difícil [. . .] [The coach, Sergio Scariolo, foresees that the World Cup which starts tomorrow in Turkey will be difficult] (Hennemann, 2013: 396) (16) ‘‘[. . .] Voy a pasar un tiempo antes en Japo´n, y seráfut. una buena ocasio´n para relajarme [. . .]’’, ha concluido. [‘‘I will spend some time in Japan before, and it will be a good opportunity for me to relax’’, he has concluded] (Hennemann, 2013: 395) Nevertheless, most of the examples included in the work by Hennemann show the future form appearing in situations which are still located after the now of the speaker, thus making difficult to observe the progress toward Boye and Harder’s coded secondariness (2009: 31--38). Concerning concessive future, Hennemann (2013: 339) does not specifically pay attention to it, but she considers it a reportive case -- following Reyes. According to Reyes (1990: 23--24), the reportive use may include any contact relationship between texts where one of them represents the other. Reportive use is therefore understood in a wide sense, and it contains a great variety of phenomena, including contrastive structures with sino [but] and pero [but], causal structures with ya que [since], or even factive constructions. From a diametrically opposed theoretical perspective, evidentiality is exclusively conceived as a grammatical coded phenomenon signaling the source of information. From such perspective, Escandell (2010) offers an analysis of future in Spanish. According to her, future always urges the addressee to assume that the only source of knowledge for the event is a speaker’s internal process. Using the future enables the speaker to convey that he has not a better source for the event, either because it is happening somewhere else or because it is happening in some other time period. For Escandell, future in Spanish is an evidential -- using Aikhenvald’s terms (2004) --; that is, the inferential value is its primary, basic meaning, and it is always present, even in the traditionally considered temporal cases, where, according to Escandell, the function of future is not to place situations in time, but to express that, once again, there is no better source for the event than the speaker. She thus argues that future does not work with expressions that explicitly point to an external source, not even in the most commonly treated temporal cases. However, examples like the following contradict her proposal:5 (17) El paro volveráfut. a subir con fuerza en España en 2012, según la OCDE. [The unemployment rate will strongly rise again in Spain in 2012, according to the OECD] (El País, 28/11/2011)

4

They define coded secondariness as the type of information which cannot be foregrounded under any circumstances. In fact, the common occurrence of future with the phrase según X [according to X] has led Hennemann (2013: 381--383) to consider the possibility of analyzing future as a reportive evidential mechanism in these cases. However, she also acknowledges that future cannot trigger this value on its own, without some kind of contextual reinforcement. Consequently, the role of future would not largely differ from those played by any other temporal form, never related to evidentiality, when occurring with según X, as seen in the following example, with a past form: ‘‘Según los testigos presentes en la sala, Breivik se mostro´ tranquilo durante los 40 minutos que duro´ la comparecencia, y no expreso´ emocio´n alguna’’ [According to the eyewitnesses who were present in the courtroom, Breivik remained calm throughout his forty-minute-long appearance, and did not show any emotion whatsoever] (El Perio´dico, 16/07/2011). 5

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

185

(18) Según el entrenador del Real Madrid José Mourinho, Iker Casillas tendráfut. dificultades para recuperar su puesto titular bajo los palos del equipo. [According to Real Madrid’s coach José Mourinho, Iker Casillas will have difficulty in regaining his post as the first-team goalkeeper] (ESPN Deportes, 31/03/2013) From Escandell’s perspective, concessive future also shows an evidential value, and is considered a contextual variant of conjectural future: (19) A: Juan es muy listo. . . [Juan is very intelligent. . .] B: Seráfut. muy listo, pero no estudia nada. [He may be (befut.) very intelligent, but he does not study at all] (Escandell, 2010: 27) According to her, future urges us to interpret the proposition once again as not verifiable in structures like the one in (19) -- which is in stark contrast with what the addressee has just said. The utilization of future removes the factuality of the utterance, which loses strength in the argumentative process, ultimately leading to a dissociative interpretation of the structure. In a recent paper, Escandell (2014) refines her analysis of future in Spanish as an evidential. She further argues that the meaning of future does not contribute to the propositional content; instead it is situated at the place occupied by high-level explicatures -- i.e. where illocutionary force and speaker’s attitude are located. Concerning the specific type of evidential meaning, Escandell insists on the idea that future invokes an instruction for the addressee to assume that the source of the propositional content is a speaker’s intuitive inference, which is thus incompatible with the reference to any other external source. As for concessive future, she acknowledges that it represents an exception for the evidential value of future, justifying it as an echoic use, though. Her reasoning would be as follows: future conveys that the only source for the event is a speaker’s mental process; it would be false in the case of (19), due to A’s previous assertion; future in B is interpreted in echoic terms -- that is, B attributes the propositional content to A, and identifies a mental process by A as the only source for it. According to Escandell, thus, concessive future is still a contextual variation of the inferential meaning. Squartini’s (2001, 2012) conception of evidentiality and his explanation of conjectural and concessive future are located in between. In this author’s view, evidentiality is a semantic notion which may be grammatically coded in some languages or may be expressed by several mechanisms in others. Concerning the inferential value of future in Spanish or Italian, he does not consider it a primary, basic meaning; so future would not be a grammatical evidential, but an evidential strategy -- in Aikhenvald’s terms (2004: 20). This assumption directly links to the traditional perspective of analyzing the inferential meaning of future as a derived or metaphorical value (Bello ([1847] 1970: 235--242). As for concessive future, Squartini focuses on the case of Italian, although he sometimes refers to Spanish. In his opinion, conjectural future and concessive future are different, and the latter constitutes a new semantic extension. As an argument, Squartini relates concessive use to the factuality of a proposition, linking it to a new parameter of evidentiality: (inter)subjectivity (Nuyts, 2001a,b). (20) A: Fulano è saggio. [Fulano is a wise man] B: Lo sarà; tuttavia, qualche volta si è sbagliato. [He may be (befut.); nevertheless, he has been mistaken at times] (Squartini, 2012: 2119 -- adapted from the Spanish version by Gili Gaya [1951] --) For Squartini, whereas conjectural future has the self as its primary source, the speaker places himself as a secondary source in concessive future, acknowledging information which is intersubjectively shared by other participants. Squartini argues that concessive future is intersubjective both in the sense expressed by Traugott (2010) and as it was defined by Nuyts (2001a,b) or Cornillie (2009). According to Traugott, concessive future evolves from a subjective meaning to a new one which pays more attention to the addressee. According to Nuyts and Cornillie, though, intersubjectivity has to do with the status -- and consequently with the reliability -- of the source. As Nuyts and Cornillie see it, evidence may be due to the speaker; to the speaker and others; or to others. The most reliable evidence is the one shared by the speaker and others. From this perspective, concessive future reflects a concessive move, where the speaker invokes two different sources: he accepts what provides a primary source -- thus sharing responsibility -- and places himself as a secondary source.

186

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

In Squartini’s view, concessive future’s evidential value is related to some other discursive uses, such as (21), which are linked to mirativity -- or to the reception of information by an unprepared mind (DeLancey, 2001). In both cases, he claims that future alludes to an external source which is intersubjectively shared, either through direct perception, report or inference. (21) [Ho dimenticato le chiavi] Sarò scemo! [I forgot my keys. I amfut stupid] (Squartini, 2012: 2125) From a formal perspective, Rivero (2014) has also recently dealt with the evidential value of Spanish future. She considers future morphemes as modals with an evidential base, in which modal strength may vary depending on both the type of source -- realistic or informational, in Kratzer’s terms -- and the anchoring conditions. She regards epistemic future as a strong inferential, and places it at one end of the scale: specifically, epistemic future arises from combining a realistic source with specific anchoring conditions (i.e. it requires a judge who carries out the evaluation, be it the speaker or the main clause’s subject); in this way, the speaker becomes responsible for the information supplied. At the opposite end of the scale, she places concessive future, which is considered a weak mirative: it arises from the combination of an informational source with flexible anchoring conditions, since concessive future only needs a salient proposition; the speaker is not necessarily responsible for the information in these cases, and he may even deny it. Contrary to Escandell, in Rivero’s view, future morphemes are not illocutionary operators but they contribute to propositional content. In short, concessive future poses a challenge when it comes to determining the role of future within a system of evidentiality, as it is followed by the different -- and often contradictory -- explanations that have been offered: (a) Hennemann (2013) points out that concessive future exemplifies a reportive evidential mechanism, but she does not develop this idea -- based on an excessively broad definition of report which goes beyond evidentiality and includes very different phenomena. (b) Escandell (2010, 2014) argues that concessive future is nothing but a contextual variation of the primary evidential value (an instruction for the addressee to assume that the propositional content has a speaker’s intuitive inference as its only source); thus, it is characterized as non-factual and explained as echoic. (c) Squartini (2012) considers that concessive future represents a new semantic extension, the isolation of which stems from its factuality. This new semantic extension refers to an intersubjectively shared external source. (d) Rivero (2014) describes concessive future as a case of mirativity -- placed at one end of a scale where the opposite end would be occupied by epistemic future. The first question to be addressed is whether concessive future represents a contextual variation of conjectural future or it constitutes a new use. If the following examples are observed, a distinction becomes visible between (22) and (23), on the one hand, and (24) and (25), on the other. ?

(22) Pero tú crees que los políticos están haciendo algo para que la gente, para que haya trabajo, para que la gente viva bien? [But, do you think (that) politicians are doing something for people to. . ., for employment to be available, for people to live well?] Hombre, algo estaránfut. haciendo, pero también harán algo para llenarse ellos los bolsillos. [Man, they must be (befut.) doing something, but they will also do something to fill their own pockets] (Azorín, 2002: 550) ?

(23) B: es que estás mala? Parece que tiene mal colorcete. [Are you ill? You look a bit bad] C: [. . .] a lo mejor tendréfut. sueño pero mala no estoy. [Maybe I am (befut.) sleepy, but I am not ill (actually)]. (Briz and Val.Es.Co., 2002: 246) (24) : Porque piensan que así, consintiéndoles, lo mismo están más felices [. . .] [Because they think that, in that way, spoiling them, they could even feel happier] : Sí, a los niños los haránfut. muy felices pero no es eso lo que les interesa. [Yes, they may make (makefut.) children very happy, but that is not what they need] (Azorín, 2002: 374)

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

187

(25) Anto´n: [. . .] Estás en contacto con la naturaleza. Y gozando de la inmensa paz de un hogar como Dios manda. Ves qué orden? Y todos estos souvenirs que parecen comprados en Hong Kong, pues nada más lejos. Todo de El Corte Inglés [ ]. [Anto´n: You are in touch with nature. And enjoying the immense peace of a home ‘‘as God commands’’ (i.e. a real/proper home). Can you see this order? And all these souvenirs, which seem to have been bought in Hong Kong; well, nothing could be more untrue. Everything from El Corte Inglés] Candela: [. . .] La visio´n de todo esto me ataca. Tu casa seráfut. muy decente, pero es repugnante. [Candela: The sight of all this shocks me. Your home may be (befut.) very decent, but it is disgusting] (RAE, CREA, J. J. Alonso Millán: Pasarse de la raya)

?

...

Although they may look similar at first sight, future in (22) and (23) expresses a conjecture that interacts with the speaker’s evaluation of the proposition as probable. This conjectural, epistemic reading is respectively reinforced by P1’s question in (22) ( tú crees que. . .?) [Do you think (that). . .?], and by the adverbial a lo mejor [maybe] which appears at the beginning of C’s own utterance in (23). Note that future forms in (22) and (23) may occur with a subordinating verb expressing calculation without a change in meaning, as can be seen in (26) and (27): ?

(26) Yo creo que algo estaránfut. haciendo, pero también harán algo para llenarse ellos los bolsillos [I think that they must be (befut.) doing something, but they will also do something to fill their own pockets] (27) Yo creo que a lo mejor tendréfut. sueño, pero mala no estoy [I think that maybe I am (befut.) sleepy, but I am not ill (actually)] On the contrary, (24) and (25) show a future form which no longer performs an inferential or an epistemic function. In fact, the propositions where future occurs cannot be embedded into a subordinating verb expressing calculation, as shown in (28) and (29): (28) # Yo creo que a los niños los harán muy felices, pero no es eso lo que les interesa [I think that they may make (makefut.) children very happy, but that is not what they need] (29) # Yo creo que tu casa será muy decente, pero es repugnante. [I think that your home may be (befut.) very decent, but it is disgusting] Whereas future is external to the proposition where it appears in (22) and (23), it remains external to the utterance containing it in (24) and (25), and affects the link between two propositions; expressed differently, future has widened its scope in (24) and (25) (Bybee, Perkins and Pagliuca, 1994: 227). The second question to be addressed is whether concessive future always refers to a non-factual proposition (following Escandell) or to a factual one (following Squartini). In this respect, note that the proposition where future appears in (24) may be interpreted as non-factual, since it recovers information that the first speaker (hT4i) has expressed as hypothetical; on the contrary, the proposition which contains a future form in (25) may be interpreted as factual, since it recovers information that the first speaker (Anto´n) asserts as such. It consequently seems that the distinction between factual and non-factual is not the parameter regulating the utilization of concessive future. In fact, as explained below, the parameter which determines this new use is the activated status of the proposition where future appears (Dryer, 1996). Finally, the third question to be addressed has to do with the ‘‘another voice’’ flavor triggered by concessive future. For Escandell (2014), this nuance is resolved in echoic terms, although her explanation seems to be oriented to stressing that the reference to a speaker’s intuitive inference as the only source remains an unchangeable value of future; there are several cases, however, where future appears to play a number of other functions. With regard to the analysis of concessive future as a reportive evidentiality strategy -- following Reyes (1990) and Hennemann (2013) --, it is important to evaluate its role in (24) and (25) in order to decide whether this concessive future only proves useful to identify the source of information, or, on the contrary, if it expresses the speaker’s attitude as well. From the perspective adopted in the present paper, future does play a role in the (counter)argumentative process invoked by the construction where it appears in these cases -- and accordingly goes beyond the usual function of a reportive evidential strategy. The link between concessive future and mirativity also arises in connection with this issue (Squartini, 2012; Rivero, 2014). It is argued here that future exhibits a common feature in both cases: it operates on the utterance level, although with different effects, as will be shown in Section 5. Contrary to conjectural uses, where the distance conveyed by future is projected over the epistemic level, concessive uses project that distance over the utterance level. Conjectural future and concessive future share a deictic function but they act on different levels of meaning. An explanation of the deictic model on which future is based appears in the next section.

188

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

3. Distance: a powerful template As explained in the introduction, tense is the information about time grammaticalized by the verb. Time has traditionally been represented as a straight line with the present as its zero point or reference point; past would be placed on the left and future on the right (Comrie, 1985: 2). However, a certain imbalance has been assumed from Aristotle onwards between past and present, on the one hand, and future, on the other, which has led to conceive the latter as a branch structure (Dahl, 1985; Jaszczolt, 2009; De Brabanter, Kissine and Sharifzadeh, 2014). In fact, contrary to past and present, future can be neither perceived nor remembered. Likewise, future is undetermined in terms of truth conditions; in other words, it becomes difficult to know when a future event is true, whether at the utterance time or at the moment of verification.6 Furthermore, despite being one of the most typologically marked categories (Dahl, 1985), future can be described as a quite unstable system. Fleischman (1982) shows it for Romance languages: more specifically, Spanish synthetic future competes with the periphrastic form ir a + infinitive and with the present tense followed by time adverbs to indicate posteriority. For all these reasons, and due to the great number of modal values with which it is associated, future has been recently assumed to constitute an essentially modal element, as Giannakidou and Mari (2012) argue for Greek and Italian. According to these two authors, the temporal value of future derives from the modal one: in the absence of any other information, future is an epistemic necessity modal operator, similar to must, which conveys indirect evidence at the moment of speech; the temporal value arises when an adverb or temporal information exist which show that the speaker has direct knowledge of the relevant time -- i.e. of the specific moment when the event would be placed. Nevertheless, the view according to which the temporal value represents the unitary meaning of future has also been vindicated from different perspectives. For instance, from a formal standpoint, Kissine (2008) claims for a unitary nonmodal analysis for future in English which is based on a temporal shift toward the future with respect to the utterance time. After bringing together the different proposals made by Kratzer, Lewis and Stalnaker, Kissine argues that the cases in which will seems to behave exclusively as a modal are better explained as a domain restriction of an epistemic necessity operator under which all the assertions fall. In this new defense of temporal value, even French epistemic future has been explained in temporal terms from the perspectival interpretation of tenses made by De Saussure (2013), in whose opinion future does not change as a tense in epistemic cases; what really changes is the nature of the event: from indicating p to indicating verification of p. Epistemic future therefore represents the verification in the future of an assumption referred to the current state of affairs, and this is achieved through a change of perspective: the state of affairs is allocentrically represented by a third entity located in the future who can actually assert its truth. As a consequence, it is inferred that the speaker cannot assert the truth in the present, so he asserts it in the future; and, furthermore, the proposition is thought to be based on an epistemic attitude which has more strength than a mere possibility (i.e. high probability). In order to reinforce his proposal, De Saussure (2013: 61) suggests that the epistemic value may also arise from periphrastic future, as shown by the following examples taken from English, French and Spanish: [The (30) (31) (32)

bell rings] That’s gonna be the postman Ça va être le facteur Ese va a ser el cartero

Broader scope approaches have arisen in recent years which have as their aim to relate temporality and thought. For instance, from the Default Semantics perspective, and with a strong philosophical component, Jaszczolt (2009) argues that our concept of time is modal, since it depends, on the one hand, on an epistemic detachment from now; and, on the other, on metaphysical time, which appears as an essentially modal construct in contemporary physics. According to her, past, present and future belong to human experience. This psychological or internal time is conceived as a detachment from certainty, or expressed differently, as a fluctuation from the possibilities of future to the certainty of now and from the certainty of now to the fragmented -- and thus uncertain -- image of past. As for the representation of future, Jaszczolt (2009: 140--146) claims that there is no a clear-cut boundary between temporal and modal forms; hence her decision to place them on the same scale, depending on the epistemic commitment that they exhibit (tenseless future > futurative progressive > regular future > epistemic necessity future [must] > epistemic possibility future [might]). In the context of these broader scope approaches, Cognitive Grammar regards tenses as grounding predications (Brisard and Patard, 2011) -- that is, a particular type of deixis which relates propositions to the speech situation. This relationship is conceived as a mainly epistemic one: the speaker shows a special interest in the knowledge status assigned to the event. Within the framework of Cognitive Grammar, tenses are given schematic values which may 6 A higher complexity of future, as opposed to present and past, would thus be expected; however, the data treated by De Brabanter et al. (2014) from the point of view of neurolinguistics turn out to be revealing in this respect: speakers suffering from aphasia do not find more difficulty in conceiving future than in conceiving past.

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

189

materialize in different uses. For instance, English present tense is considered to have a schematic value of epistemic immediacy, the prototypical use of which would be that of temporal coincidence (Langacker, 1999, 2011). Likewise, the schematic value of French imparfait consists in locating a point of view simultaneous to the speaker’s, thus invoking a virtual ground which does not coincide with the speaker’s here and now (Brisard, 2010). Once again, this schematic value has the temporal one as its prototypical use, but it may also explain some other uses of imparfait.7 Neither the proposals which defend the temporal analysis of future nor the approaches which advocate its modal treatment often pay attention to its discursive values, amongst which stands out concessive future. There is something on which consensus seems to exist between all the recent perspectives, though: the attempt to attribute a unitary value to tenses which should be present in all their uses. From the perspective adopted in this paper, future works on a template that is always present: it conveys a deictic instruction specified as ‘‘distance forward’’ and has the speaker as the origo -- his here and now, his beliefs, and his role as a participant in communicative interaction. The link between movement in time and movement in space has been widely assumed in the literature devoted to linguistic time (Traugott, 1978; Fleischman, 1982; Comrie, 1985; Bybee et al., 1994; Haspelmath, 1997). Rather than movement, what can be really generalized in the meaning of future is its distance forward deictic instruction with respect to the speaker. It is this value that appears in all its uses, not only in the temporal one; and it may accordingly be considered its schematic meaning.8 This deictic value is then projected over a subjectivity axle (Fleischman, 1989; Traugott, 1989; Schwenter, 1999: 73 -- footnote 21--), which traverses the different levels of meaning established by Sweetser (1990), as shown in Fig. 1, and makes it possible to explain the different uses of future.9 This schematic definition -- which is based on a distance forward deictic instruction and has the speaker as its reference point -- acquires a perspectival nature which pervades every use of future, and makes it possible for this form to interact with a variety of semantic and discourse categories often regarded as perspectivization mechanisms (Habler, 2010; Hennemann, 2013).10 When the deictic instruction is projected upon the content level, future behaves as a tense and plays a role inside the proposition (Bybee et al., 1994). The Spanish tense system, as that of most Indo-European languages, places future in front of the speaker (Fleischman, 1982; Haspelmath, 1997; Company, 2006; Jaszczolt, 2009).11 At this level, the meaning 7 Brisard’s proposal reformulates that of Coseriu (1976), for whom the main feature of the Romance tense system is that it can be divided into two levels: a level of actuality, which coincides with the time line, has the present as a reference point, and forms a paradigm with past and future; and a level of inactuality, which does not coincide with the time line but runs parallel to it, has the imperfecto/imparfait as its reference point, and forms a paradigm with pluperfect and conditional. From this perspective, future would convey distance forward taking the speaker’s here and now as a point of reference, whereas conditional would convey distance forward taking an alternative, inactual point of reference. This proposal fits well with the one defended in this paper. 8 Langacker (2011) and Brisard (2010) draw a distinction between schematic and prototypical meaning with regard to tenses which seems interesting for the analysis of future. Thus, the schematic meaning of future, which appears in all its uses, is ‘‘distance forward’’. Concerning its prototypical value, it would originally be its temporal meaning: during the early stages of its creation (from the 13th century), the temporal value of synthetic future was about 80%, while the cases in which it exhibited a conjectural, epistemic value accounted for 20%, as Company (2006) proves. In contemporary Spanish, the prototypical use is probably moving toward conjectural epistemic future but this does not mean that the temporal value has disappeared --as shown by its survival to the present day in European Spanish. The schematic meaning, as a distance forward deictic instruction, is present both in the temporal use and in the epistemic, conjectural one, and it also turns out to be productive at the discourse level where concessive future is placed. 9 This distance forward deictic value is also linked to the diachronic origin of Spanish synthetic future. Spanish synthetic future comes from the Latin construction infinitive + habere: the structure, which arises from an initial lexical value of possession (habere), firstly evolves into obligation and predestination; subsequently acquires a value of intention; and finally becomes a tense. The said distance forward deictic value underlies the whole process, as it may be considered a component of ‘‘movement toward a goal,’’ in terms of Bybee et al. (1994: 268). The goal would be the event itself, and the distance may be crossed in a more or less controlled way. The distance forward value thus remains latent in the perfect cycle that Spanish future represents, according to Van der Auwera and Plungian (1998): an initial process of demodalization, until acquiring its temporal role, and a subsequent process of remodalization, which becomes visible in conjectural, epistemic cases. Furthermore, it also explains those cases where future operates at the utterance level and plays a discursive function. 10 The mere conception of space which underlies the deictic instruction becomes a clear example of perspectivization (Cifuentes, 1989; Haspelmath, 1997). 11 Two metaphors serve to conceive time in terms of movement across space: moving time and moving ego (Clark, 1973). According to the former, the observer is stationary and time moves through him; this metaphor becomes especially productive in the creation of temporal expressions (Haspelmath, 1997). And following the latter, time is stationary, and the observer moves through it; this metaphor is responsible for the situation of future in front of us and shows the highest productivity in the tense systems of most Indo-European languages. However, this is not the case in Latin or Ancient Greek (Fleischman, 1982; Company, 2006), and neither does it happen in other languages --including Maori or Aymara (Jaszczolt, 2009). Although Jaszczolt acknowledges that differences may exist depending on the observer’s orientation, similarly to what Levinson (2003) established for the conception of space, this led Jaszczolt to argue that the value of time flow does not play a primary role within our conception of temporality. According to her --as has been already mentioned-- the primary value of temporality consists in showing a detachment from the uncertainty of now. In any case, Jaszczolt’s perspective is not incompatible with the schematic definition of future as a deictic instruction based on distance forward.

190

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

Fig. 1. Projection of the deictic value of future.

of distance forward makes it possible to place the event after the now of the speaker; hence its association with a temporal instruction. As mentioned above, synthetic future has been increasingly complemented with several modal values which may be contextually explained and which are determined to some extent by the syntactic and communicative environment where future occurs. Thus, (33), (34), (36), (37) and (39) have to do with intention, whereas (35) and (38) are closer to prediction or inference. Likewise, future occurs with vague or undetermined time-related expressions (algún día, algún jueves, un día) in some of the examples collected from two oral contemporary Spanish corpora. (33) Ya llamaréfut. después. [Then I will call later] (Briz and Val.Es.Co., 2002: 83) (34) Yo vendréfut. algún día. [I will come some day] (Briz and Val.Es.Co., 2002:162) (35) Pronto saldráfut. humo. [A puff of smoke will come out soon] (Briz and Val.Es.Co., 2002: 173) (36) Pues sí que iréfut. algún jueves. [I will definitely go any Thursday now] (Briz and Val.Es.Co., 2002:182) (37) Luego ya haremosfut. un poco de coloquio y hablaremosfut de otras cosas. [Later we will have a little bit of debate on it and will talk about other things] (Azorín, 2002: 487) (38) Estuve en Helados Alacant hace ya casi un año y el año que viene a lo mejor me llamaránfut. otra vez. [I worked for Helados Alacant almost a year ago now and perhaps they will call me again next year] (Azorín, 2002: 543) (39) Esa es una gran novela, un día te la dejaréfut. [That is a great novel, I will lend it to you one day] (Azorín, 2002: 866) Unlike the examples above, the future forms in examples (40) and (41) -- found in European Spanish press -- locate the event more neutrally, occurring with more precise time-related indications (el pro´ximo jueves, en octubre). (40) La cadena pública estrenaráfut. el pro´ximo lunes el programa ‘‘Entre todos’’. [The public channel will start broadcasting the program ‘‘Entre todos’’ next Monday] (El País, 23/08/2013) (41) La cinta que todos daban por perdida veráfut. la luz en octubre. [The videotape which was thought to be lost by everybody will be shown in October] (El País, 08/08/2013) However, future still places the situation after the now of the speaker both in the examples (33)--(39) and in the pair (40) and (41), which justifies their temporal classification (Dahl, 1985). The ‘‘distance forward’’ deictic instruction invoked by future may be projected over some other levels of meaning, e.g. the epistemic one. In such cases, future no longer acts inside the proposition, but remains external to it

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

191

(Bybee et al., 1994). The deictic value is now used by the speaker to take a distance from the proposition, as shown in the examples below: (42) Hombre, habráfut. gente que necesite pensar en algo, creer en algo para sentirse mejor y tal, yo no. [I mean, there must be (befut.) some people who probably need to think of something, to believe in something in order to feel better and so on; not me] (Azorín, 2002: 548) (43) Tú por tu acento no lo conocerásfut. porque no eres de aquí, de Alicante. [Due to your accent, you must not know (knowfut.) him because you are not from here, from Alicante] (Azorín, 2002: 873) (44) Habrá sidofut. perf. más gastadora tu madre que yo. [Your mother must have been (befut. perf.) more of a spender than me] (Briz and Val.Es.Co., 2002: 135) At an epistemic level, future is linked both to inferential evidentiality and to epistemic modality; however, no unanimous consensus has been reached on how -- or even on whether -- these categories interact (Palmer, 1986; Van der Auwera and Plungian, 1998; Dendale and Tasmowski, 2001; Nuyts, 2001a,b; Squartini, 2008, 2012; Cornillie, 2009). Unlike what happens with past forms, the ‘‘distance forward’’ conveyed by future is translated into a positive distance in epistemic terms; in other words, the speaker evaluates the proposition as close to factuality; hence its interpretation by default as probable -- an intermediate zone between hypotheticality and factuality.12 (45) Hombre, habráfut. gente que necesite pensar en algo. > Probablemente haypres. gente que necesita pensar en algo. [Probably, there are people who need to think of something] The utilization of future indicates that the speaker has not direct evidence for the proposition; i.e. the proposition stems from his own inference. The different syntactic positions where future may occur when it acts at an epistemic level make it possible to draw a continuum through which future ends up behaving as an evidential strategy -- in Aikhenvald’s terms (2004). This continuum also reflects the progression toward the coded secondariness proposed by Boye and Harder (2009: 31--38). More precisely, future may occur in the subordinate clause, being dependent on a main verb that explicitly conveys the notion of conjecture (46); it may occur in a thematic subordinating clause where the main verb has been colloquially omitted, as it happens in (47) -- where one can imply a main verb such as imagino [I imagine] or supongo [I suppose]--; it may work as a main verb followed by a subordinate clause that reinforces the reasoning process (48); or, finally, it may develop an inferential strategy on its own, as is the case in (49)13: (46) Juan supongo que estaráfut. pues hasta las narices de los curas. [Juan, I suppose, must be (befut.) fed up with priests] (Briz and Val.Es.Co., 2002: 92) (47) Esto que lo habrán tira(d)ofut. perf. a la basura y algún crío lo habrá puesto allí. [This. . ., I mean, somebody must have thrown (throwfut. perf.) it in the garbage and a child must have left it there] (Briz and Val.Es.Co., 2002: 225) (48) Seráfut. porque la practica ochenta veces. [It must be (befut.) because he practices it eighty times] (Briz and Val.Es.Co., 2002: 57) (49) Hombre, habráfut. gente que necesite pensar en algo, creer en algo para sentirse mejor y tal, yo no. [I mean, there must be (befut.) some people who probably need to think of something, to believe in something in order to feel better and so on; not me] (Azorín, 2002: 548) 12

According to Van der Auwera and Plungian (1998), it is precisely in inferentiality that epistemic modality and evidentiality overlap. By contrast, Squartini (2008, 2012) or Cornillie (2009) claim that the fact that Italian future may occur with adverbs indicating different degrees of epistemic commitment, such as forse [perhaps] -- weak commitment -- or sicuramente [certainly] -- strong commitment -- is an argument to justify the consideration of evidentiality and epistemic modality as two separate categories. In the case of Spanish, future may occur with adverbs indicating either uncertainty, such as quizás [perhaps], a lo mejor [perhaps], or probability, such as probablemente [probably] and seguramente [surely]. Note that the latter conveys ‘‘great probability, but not certainty’’ (Kovacci, 1999: 755; RAE, 2009: 2351). Concerning the possibility to occur with adverbs that express a stronger epistemic commitment, such as necesariamente [necessarily], an implicative environment is required -- for instance, a causal or a conditional structure. In the absence of adverbs, epistemic future triggers a modal value of probability. 13 The capacity for epistemic future to occur in different subordination environments is one of the arguments that Rivero (2014) adduces to reject it as an illocutionary operator.

192

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

The deictic instruction invoked by future may be projected upon a higher level of meaning; in these cases, distance forward is specified as a distance from the speaker toward the utterance. This entails a new widening of scope: future has changed from being external to the proposition, as in epistemic future, to being external to the utterance; that is precisely what happens in concessive future. The interpretation of distance with respect to the utterance is triggered by information structure. One of the main dichotomies according to which discourse information structure is organized refers to the opposition between given and new information. However, the notion of given information understood in terms of pragmatic information (Dik, 1997), givenness (Clark and Haviland, 1977), or speech context (Coseriu, 1969) -- that is, as shared knowledge, or as all the information that speakers have acquired throughout their lives -- is an excessively wide concept which not always has a linguistic reflection. Consequently, several parameters have been traditionally adduced in order to delimit it. For instance, Dik (1997) draws a distinction between what the participants know before the interaction and what gradually happens during the interaction. Similarly, Prince (1992) opposes what is given or new from the addressee’s point of view to what is given or new from the discourse point of view. Therefore, the contrast ‘‘activated versus non-activated information’’ arises orthogonally to the distinction between given and new information. Activated or salient information is the one that, according to the speaker, is outlined in the addressee’s mind; whereas non-activated information is the one that is not profiled (Chafe, 1976; Lambrecht, 1994; Dryer, 1996). The activated information status has to do with short-term memory (Lambrecht, 1994: 94) and permits to establish a continuum of activation (Dryer, 1996: 481): the focus of attention is the information to which one pays special attention; semi-deactivated entities are those which have been activated but have progressively lost their activation status; in turn, accessible entities are the ones which have not been activated but belong to an activated semantic scheme; and, as opposed to all of them, there is non-activated information. Dryer’s proposal does not only take entities into account but also includes propositions in the description of information structure. It is consequently possible to find activated, semideactivated, accessible, and non-activated propositions. Likewise, information may be either discursive or situationally activated. One of the most attractive aspects of Dryer’s approach lies in the fact that the activation status does not epistemically evaluate the proposition (Dryer, 1996: 483--485). A proposition may be activated after being uttered, and it may be factual or believed too, as it happens in Helsinki is the capital of Finland. However, after saying The Earth is plane, the proposition has become activated despite its falsehood -- it does not form part of our shared knowledge. A proposition may be activated in this way even if it is characterized as uncertain, as shown by the proposition Berta has come in (50): (50) A: Has Berta come? B: I don’t know. The distance forward deictic instruction invoked by future is projected upon the utterance level when the (factual or nonfactual) proposition has been previously activated.14 If this happens in counter-argumentative contexts, one interprets the distance of future toward the utterance in déréalisant terms -- as is explained in the next section. 4. Function of concessive future The concessive use of future differs from conjectural or epistemic future insofar as it is subject to certain contextual constraints (Squartini, 2012). However, these contextual restrictions are not related to the factuality of the proposition, but to its activated status (Dryer, 1996). The future appearing in (24) and (25) -- which are now repeated in (51) and (52) -- has a concessive value but, while the proposition where future occurs in (51) may be interpreted as non-factual, the one in (52) can be seen as factual. Indeed, the proposition which contains future in (51) in T8’s turn has been firstly introduced by as hypothetical; on the contrary, the proposition where future occurs in (52) in Candela’s turn has been firstly introduced by Anto´n as factual information. (51) : Porque piensan que así, consintiéndoles, lo mismo están más felices [. . .] [Because they think that, in that way, spoiling them, they could even feel happier] : Sí, a los niños los haránfut. muy felices pero no es eso lo que les interesa. [Yes, they may make (makefut.) children very happy, but that is not what they need] (Azorín, 2002: 374)

14 Note that whereas epistemic, conjectural future appears frequently in reactive conversational movements (Escandell, 2014), what matters in those cases is the set of speaker’s beliefs, inferences or conjectures to explain something. By contrast, in concessive future, it is the proposition appearing in future that has been previously activated.

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

193

(52) Anto´n: [. . .] Estás en contacto con la naturaleza. Y gozando de la inmensa paz de un hogar como Dios manda. Ves qué orden? Y todos estos souvenirs que parecen comprados en Hong Kong, pues nada más lejos. Todo de El Corte Inglés [. . .]. [Anto´n: You are in touch with nature. And enjoying the immense peace of a home ‘‘as God commands’’ (i.e. a real/proper home). Can you see this order? And all these souvenirs, which seem to have been bought in Hong Kong; well nothing could be more untrue). Everything from El Corte Inglés] Candela: [. . .] La visio´n de todo esto me ataca. Tu casa seráfut. muy decente, pero es repugnante. [Candela: The sight of all this shocks me. Your home may be (befut.) very decent, but it is disgusting] (RAE, CREA, J. J. Alonso Millán: Pasarse de la raya)

?

Both examples share the activated character of the proposition. Therefore, the concessive interpretation of future requires a proposition which has just been activated; and more precisely, which must be the focus of attention. Concessive future is usually followed by pero or any other adversative mark. Adversative structures represent counterargumentation strategies, insofar as they simultaneously invoke two points of view and align with one of them; they are thus polyphonic (Anscombre and Ducrot, 1994). Within such a complex strategy, the member introduced by pero is the strong one, since it imposes its argumentative orientation upon the whole construction; in other words, the pero member owns argumentative sufficiency (Portolés, 1998). Note how ‘‘if something makes a child happy’’ is a very good argument to carry it out in the case of (51), but then the fact that this is not in the child’s interest becomes an even stronger argument not to carry it out: the argument introduced by pero consequently imposes the argumentative orientation upon the whole construction. The structure in (51) represents a case of indirect counter-argumentation: each member provides a line of reasoning which leads to opposed conclusions (Fig. 2). Concerning (52), the condition of a house being very decent is a positive argument for somebody to feel good there; however, the conclusion introduced by pero -- and the one which eventually turns out to be imposed -- is just the opposite: Candela thinks that the house is disgusting. In this case, pero invokes a direct counter-argumentative strategy: the first member introduces an argument oriented toward one conclusion but the conclusion introduced by the pero member is just the opposite (Fig. 3). Something makes a child happy

Something is not in the child’s interest

To carry it out

Not to carry it out Fig. 2. Indirect counter-argumentation strategy invoked by pero.

A house is very decent

To feel good

To feel disgust

Fig. 3. Direct counter-argumentation strategy invoked by pero.

Apart from being polyphonic -- the same as any other counter-argumentation strategy --, structures containing concessive future are diaphonic (Roulet et al., 1991) because the speaker’s utterance includes what the addressee has just said.15 For this reason, they usually appear in interactional contexts, play an interpersonal function (Pérez Saldanya, 2002: 2637) and are intersubjective à la Traugott, in the sense that these structures with concessive future gradually pay increasing attention to the addressee. However, concessive future does not merely report the addressee’s utterance, which makes it difficult to analyze this type of future exclusively as a reportive evidential strategy; it invokes the speaker’s distance from it too. Concessive future has a twofold aim from a discourse perspective. At an exchange level, it may reflect an initial agreement, which fits in with the concessive discourse movement identified by Couper-Kuhlen and Thompson (2000: 383), and attributed by Squartini (2012: 2123) to concessive future in Italian. 15 Quoting Roulet et al. (1991: 70--71): ‘‘In a polyphonic structure [. . .] two voices are combined in the same turn, but the voice of the other is neither the addressee nor a direct link to the addressee, but just a mere object of reference. However, in a diaphonic structure, the speaker does not just react against the other’s words
present or absent
but takes the addressee’s words and reinterprets them in his own speech. Diaphonic structure is thus one of the most preferred ways of negotiating the points of view that characterize any interaction.’’ [The translation is mine].

194

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

A: State something. B: Acknowledge the validity of this statement (conceding reading). C: Go on to claim the validity of a potentially contrary statement. Nevertheless, the addressee’s utterance does not merely represent the weak argument at a turn-of-speech level and within the counter-argumentative strategy,16 its argumentative strength actually becomes even weaker due to the use of future; therefore, concessive future develops a déréalisant function (Ducrot, 1995)17 which reduces the argumentative strength of an already weak member and, as a result, makes the conclusion imposed by the stronger member more prominent (Table 2). Concessive future structure stands close in meaning to aunque + subjunctive construction in Spanish, as shown in (53) and (54), which rephrase (51) and (52), respectively: (53) Aunque los hagansubj. muy felices, no es eso lo que les interesa. [Even if they make them very happy, this is not what they need] (54) Aunque seasubj. muy decente, tu casa es repugnante. [Even though it is very decent, your home is disgusting] In aunque + subjunctive constructions, the weak member in argumentative terms develops a déréalisant function through the subjunctive mood, the non-assertive character of which places an already weak member -- the one introduced by aunque -- at a second level of discourse from the point of view of information structure.18 Both future and subjunctive can weaken the argumentative strength of the member containing them, and they consequently reinforce the argumentative orientation of the stronger member. Whereas the utilization of future by the speaker weakens the argumentative strength showing distance from the utterance, the use of subjunctive allows the speaker to reach his goal placing the information at a second level of discourse -- that is, at the background. Table 3 compares the déréalisant effect of both structures. Table 2 Déréalisant function of concessive future. Counter-argumentation with concessive future Concessive future member

Pero member

Weak member + concessive future = Déréalisé weak member (Higher decrease of argumentative strength)

Strong member imposing the argumentative orientation (A more prominent favored conclusion)

Table 3 Déréalisant weak counter-arguments. Déréalisant counter-argumentation Pero structures Déréalisé counter-argument Weak argument + concessive future

Aunque structures Strong member Pero member

Déréalisé counter-argument Aunque + subjunctive

Strong member Apodosis

Both the activated status of the proposition and the déréalisant function that future plays in the argumentative strategy may explain not only the occurrences of concessive future in interaction but, most importantly, its appearance in monolog contexts too. 16

For a conversational analysis of pero, cf. Garrido Rodríguez (2004: 162--200). Ducrot (1995: 148) uses the term déréalisant for lexical categories that decrease or reverse the argumentative orientation of the member where they appear. For instance, éloigné [distant] in parent éloigné [distant relative] has a déréalisant effect, whereas proche [close] has a réalisant effect in parent proche [close relative]. 18 Aunque may occur with the indicative mood -- concessive constructions
or with the subjunctive mood -- concessive conditional constructions (König, 1986). The subjunctive, as the non-assertion mood, is also related to the activated status of the proposition introduced by aunque: if the information has already been activated, no need exists to activate it again with the assertive mood
i.e. the indicative; so the subjunctive use is imposed. As a non-assertive mood, subjunctive constitutes the weak member of a Gazdar scale (1979) where the indicative acts as the stronger member. As a result of a clausal quantity implicature, subjunctive may trigger a hypothetical reading. If this reading is canceled, the information no longer appears as hypothetical but still finds itself at a second level of discourse, and it is thus translated into an increased irrelevance of the proposition introduced by aunque. For a contrast between pero and aunque, and for an analysis of concessive constructions and concessive conditional constructions, cf. Rodríguez Rosique (2008: 152--212, 343--411). 17

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

195

(55) Los pobrecitos golfos que salen del limbo para llegar al purgatorio después de recorrer unos caminos -- los ibéricos --, llenos de gente como ellos. [. . .] En nuestros caminos los jueces y los suicidas, los maestros y los abogados, hasta los militares, son todos pata negra. Por nuestras carreteras te puedes encontrar hasta a algún ser clo´nico de Alfonso XII. Casi todos proceden de un convento o de una cárcel o de cualquier otra bastardía. [. . .] Ojalá les guste a ustedes esta película de una gente que seráfut. muy imperfecta, pero a mí me gusta mucho. [Those poor scoundrels who leave the limbo to get to the purgatory after traveling some paths -- the Iberian ones -- full of people like them. [. . .] On our roads judges and suicide victims, teachers and lawyers, even soldiers, are all genuine people. On our roads you may find even a clone of Alfonso XII. Almost all of them come from a convent or from a prison or any other kind of improper place. [. . .] I wish you would like this film led by these people who may be (befut.) very imperfect, but whom I do like very much] (RAE, 2015, CREA, J. L. García Sánchez, ‘‘Suspiros de Galiardo y Echanove’’, El Mundo, 17/06/1995) Example (55) shows a proposition appearing in future that has just been activated in the preceding discourse; what is more, it constitutes the focus of attention. In addition to this, the counter-argument of something being very imperfect in relation to the conclusion that the writer likes it is even weaker due to the projection of the distance conveyed by future over the utterance. The déréalisant function developed by concessive future thus contributes to highlight the ultimate conclusion of the text even to a greater extent: the speaker’s adhesion to the characters that he has just described. 5. Conclusions Concessive future consequently represents a new use regulated by the activated status of the proposition where it appears. When the proposition has just been activated, the distance conveyed by future is not projected over the proposition any more -- as it happens in conjectural or epistemic future -- but over the utterance or énonciation level. As a consequence, the member that contains concessive future acquires a déréalisant function, insofar as it weakens the argumentative strength of an already weak member even more. This explanation is consistent with the widening of scope that Bybee et al. (1994) propose for concessive future -- summarized in Table 4. Table 4 Widening of scope in Spanish future. Temporal value Internal to the proposition

Evidential/epistemic value External to the proposition

Concessive value External to the utterance

Beyond the concessive value, the projection of the distance conveyed by future over the level of utterance not only is a very useful mechanism but also may help explain some other values of future forms (Pérez Saldanya, 2002: 2637--2638); in fact, it may explain how future interacts with a number of other semantic or discourse categories.The distance forward invoked by future is also projected upon the utterance level in the following examples: (56) En sus discursos, pretende transmitir que es un hombre más, demo´crata desde la médula, simplemente solidario; tendráfut. cara el tío! [In his speeches, he means to express that he is a man like any other, a full-fledged democrat, simply supportive; such a shameless guy!] (RAE, 2015, CREA, Á. Enríquez Soriano. Estrés. Co´mo aprender en la encrucijada. 1997) (57) La señora inmensa me arrebata su maleta y desaparece la tía sin darme las gracias. Seráfut. guarra. [The huge lady snatches her suitcase from me and disappears without thanking me for it. Such a rude woman] (E. Lindo, Don de gentes. Alfaguara, 2011 ). !

In these cases, future tends to be accompanied by evaluative terms and, occasionally, by exclamation marks. Again, this value is legitimized by the activation status of the information from which the evaluation derives, and this may have situationally happened, or even the speaker himself may have just activated it. In such examples, due to the evaluative clues, the distance toward the utterance is interpreted in terms of surprise toward the information which has just been activated: X surprises me or it is difficult for me to assume that information. This mirative value of information which crashes into an unprepared mind may have become lexicalized; an example can be found in the structure será posible:

196

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

(58) A: Que un cafre se ha cargado unas pinturas rupestres con un spray. [That a savage has destroyed some cave paintings with a spray] B: Será posible! [I can’t believe it!] (Azorín, 2002: 224) !

As mentioned above, both Squartini (2012) and Rivero (2014) relate concessive and mirative examples. According to the former, both cases refer to an intersubjectively shared external source; for the latter, concessive cases would be mirative because contrast forms part of mirativity. However, it deserves to be highlighted that, in the case of concessive future, the contrast is already provided by the adversative structure with pero. The present paper seeks to demonstrate that, although concessive and mirative cases have some things in common, they are not exactly alike. What they both share is the fact that the distance forward invoked by future is projected upon the utterance level and that this projection becomes legitimized by the activated status of the proposition. Relating them to activated information may explain factual as well as non-factual information cases. Likewise, this approach can explain those cases where the information is activated by the speaker himself -- that is, monolog examples. Furthermore, this proposal makes it possible to preserve the distance forward value invoked by future as its schematic meaning, which appears in all its uses: those operating on the propositional level, those at work on the epistemic one, and also the ones which act on an utterance level. The projection that the deictic value of future has upon the utterance level may also explain certain examples traditionally linked to a future of necessity (Fernández Ramírez, 1986), which have proved quite productive in interaction19; i.e. the future of verbs such as comprender, entender, estar conmigo en when they appear in the second person: (59) Yo no es que diga que tu hermano sea una mala persona -- prosiguio´ la Constancia acomodándose enfrente suyo en el mostrador--, pues hay que reconocer que ha repartido mucho trabajo en este barrio y que ha sabido darle otro aire al negocio. Si por el camino se ha quedado con unas cuantas imprentas de por aquí es mejor que haya sido así porque bien sabe Dios que hubieran cerrado de todos modos. Pero reconocerásfut. que es un hombre muy ambicioso y que cuando se propone algo se ciega y lo busca caiga quien caiga [. . .] [It’s not that I am saying (that) your brother is a bad guy -- Constancia continued to say while she settled herself in front of him at the counter--, since it must be acknowledged that he has distributed a lot of work (among people living) in this neighborhood and has managed to give a fresh air to this business. If he has taken over a few local printer shops in the meantime/on the way, well, it is better like that, because God knows that they would have closed down anyway. But you will admit that he is a very ambitious man, and when he wants something, he gets blinded (by the desire) and goes after it no matter what [. . .] (RAE, 2015, CREA, J. Fernández de Castro, La novia del capitán, Madrid, Mondadori, 1987). Once again, the proposition in future has to do with previously activated information; more specifically, it is presented as accessible from the information that has just been activated. Moreover, the speaker distances himself from the utterance, which results in a softening of the speech act illocutionary force (in contrast to what happens if the directive act appears in the imperative). Thanks to these two movements, future contributes to information management; in other words, the speaker urges the addressee to assume the information as necessary -- to take it as uncontroversial and to incorporate it into the shared knowledge. Insofar as the speaker refers to the necessity of assuming the information as an inference that the addressee must logically derive from the information which has just been activated, this persuasive value of the future may be considered intersubjective -- as both Traugott and Nuyts understand it. Concessive, mirative and intersubjective or persuasive cases represent the projection of the deictic value over the utterance that is regulated by the previously activated status of the proposition. Depending on the context, each of them plays a different discourse function: in the concessive, the distance over the utterance has a déréalisant effect upon the counter-argumentation process; in the mirative, the distance over the utterance contributes to the evaluation, which is subsequently considered surprising for an unprepared mind; and finally, in persuasive, intersubjective cases, the distance toward the utterance is interpreted as the necessity to assume the information as uncontroversial. At the same time, these cases share the distance forward deictic meaning of future with the epistemic or conjectural use, though now future no longer works over the proposition, but over the utterance.

19 The definition of future as a unit which codes a speaker’s intuitive inference in all its uses leads Escandell (2014) to assume that cases expressing future of necessity are characteristic of a previous, fossilized system that speakers only learn in formal, academic contexts. However, the productivity of these necessity future cases in interactive contexts poses a problem from this point of view.

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

197

Fig. 4. Projection of the deictic value of future in Spanish.

In sum, the value of European Spanish future which becomes visible in all its uses is a kind of distance-based deictic meaning (Fleischman, 1989). This value may be projected along a subjectivity axle which traverses the different levels of meaning established by Sweetser (1990). In doing so, the deictic value of future is orthogonal to several semantic and discourse categories, which not only include evidentiality but also temporality, epistemicity, counter-argumentation, intersubjectivity (both in Traugott’s and in Nuyts’ terms), or even mirativity, as can be seen in Fig. 4, which completes the scheme presented in Fig. 1. Future in European Spanish may consequently work as an inferential evidential strategy; nevertheless, its deictic value goes even further because it may be projected over different levels, which will subsequently enable it to perform a variety of functions in other categories. Concessive future is just a new step on this road to projection. Acknowledgments I am very grateful to Marta Albelda and María Estellés as well as to the audience at the IPrA’s panel Evidentiality in nonevidential languages (New Delhi, 2013). I would also like to thank the two anonymous reviewers both for the insightful comments and suggestions, and for the references provided. Thanks also go to Víctor Pina for checking my English. This research has been supported by Generalitat Valenciana (Spain), under grant GV/2014/089, and by the Spanish Ministry of Economy and Competitiveness, under grants FFI2013-45693-R and FFI2012-30941. References Aikhenvald, Alexandra Y., 2004. Evidentiality. Oxford University Press, Oxford. Anscombre, Jean-Claude, Ducrot, Oswald, 1994. La Argumentacio´n en la Lengua. Gredos, Madrid. Azorín, Dolores (Coord.), 2002. ALCORE. Alicante Corpus del Español. Universidad de Alicante, Alicante. Bello, Andrés, [1847] 1970. Gramática de la lengua castellana destinada al uso de los americanos. Sopena, Buenos Aires. Benveniste, Émile, 1966. Problèmes de Linguistique Générale. Gallimard, Paris. Boye, Kasper, Harder, Peter, 2009. Evidentiality. Linguistic categories and grammaticalization. Funct. Lang. 16 (1), 9--43. Brisard, Frank, 2010. Aspects of virtuality in the meaning of the French imparfait. Linguistics 48 (2), 487--524. Brisard, Frank, Patard, Adeline, 2011. Introduction: cognitive approaches to tense, aspect and epistemic modality. In: Patard, Adeline, Brisard, Frank (Eds.), Cognitive Approaches to Tense, Aspect and Epistemic Modality. John Benjamins, Amsterdam and Philadelphia, pp. 1--20. Briz, Antonio, Grupo Val.Es.Co, 2002. Corpus de Conversaciones Coloquiales. Arco Libros, Madrid. Bybee, Joan, Perkins, Revere, Pagliuca, William, 1994. The Evolution of Grammar. Tense, Aspect, and Modality in the Languages of the World. The University of Chicago Press, Chicago and London. Chafe, Wallace, 1976. Givenness, contrastiveness, definiteness, subjects, topics, and point of view. In: Li, C. (Ed.), Subject and Topic. Academic Press, New York, pp. 25--55. Chafe, Wallace L., Nichols, Johanna (Eds.), 1986. Evidentiality. The Linguistic Coding of Epistemology. Ablex, Norwood. Cifuentes, José Luis, 1989. Lengua y Espacio. Introduccio´n al Problema de la Deixis en Español. Universidad de Alicante, Alicante. Clark, Herbert H., 1973. Space, time, semantics, and the child. In: Moore, Timothy E. (Ed.), Cognitive Development and the Acquisition of Language. Academic Press, New York, pp. 27--63. Clark, Herbert H., Haviland, Susan E., 1977. Comprehension and the given-new contract. In: Freedle, Roy O. (Ed.), Discourse Production and Comprehension. Erlbaum, Hillsdale, pp. 1--40. Company Concepcio´n, 2006. Tiempos de formacio´n romance II. Los futuros y condicionales. In: Concepcio´n Company (Eds.), Sintaxis Histo´rica de la Lengua Española, vol. 1. Fondo de Cultura Econo´mica, Mexico, pp. 349--418. Comrie, Bernard, 1976. Aspect. An Introduction to the Study of Verbal Aspect and Related Problems. Cambridge University Press, Cambridge. Comrie, Bernard, 1985. Tense. Cambridge University Press, Cambridge.

198

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

Cornillie, Bert, 2009. Evidentiality and epistemic modality. On the close relationship between two different categories. Funct. Lang. 16 (1), 44--62. Coseriu, Eugenio, 1969. Determinacio´n y entorno. Dos problemas de una lingüística del hablar. In: Coseriu, Eugenio (Ed.), Teoría del Lenguaje y Lingüística General: Cinco Estudios. Gredos, Madrid, pp. 282--323. Coseriu, Eugenio, 1976. El Sistema Verbal Románico. Siglo XXI, Mexico. Couper-Kuhlen, Elizabeth, Thompson, Sandra, 2000. Concessive patterns in conversation. In: Couper-Kuhlen, Elizabeth, Kortmann, Bernd (Eds.), Cause, Condition, Concession, Contrast. Mouton de Gruyter, Berlin and New York, pp. 381--410. Dahl, Ӧsten, 1985. Tense and Aspect Systems. Blackwell, Oxford, Basil. De Brabanter, Philippe, Kissine, Mikhail, Sharifzadeh, Saghie, 2014. Future Tense vs. Future Time: An Introduction. Oxford University Press, Oxford, pp. 1--25. De Saussure, Louis, 2013. Perspectival interpretations of tenses. In: Jaszczolt, Kasia M., de Saussure, Louis (Eds.), Time. Language, Cognition and Reality. Oxford University Press, Oxford, pp. 46--72. DeLancey, Scott, 2001. The mirative and evidentiality. J. Pragmat. 33, 369--382. Dendale, Patrick, Tasmowski, Liliane, 2001. Introduction: evidentiality and related notions. J. Pragmat. 33, 339--348. Dik, Simon C., 1997. The Theory of Functional Grammar. Mouton de Gruyter, Berlin and New York. Dryer, Matthew S., 1996. Forms, pragmatic presupposition, and activated propositions. J. Pragmat. 26, 475--523. Ducrot, Oswald, 1995. Les modificateurs déréalisants. J. Pragmat. 24, 145--165. Escandell Vidal, María Victoria, 2010. Futuro y evidencialidad. Anu. Lingüíst. Hispán. 26, 9--34. Escandell Vidal, María Victoria, 2014. Evidential futures: the case of Spanish. In: de Brabanter, Philippe, Kissine, Mikhail, Sharifzadeh, Saghie (Eds.), Future Tense vs. Future Time: An Introduction. Oxford University Press, Oxford, pp. 219--246. Fernández Ramírez, Salvador, 1986. Gramática Española, 4. El verbo y la Oracio´n. Arco Libros, Madrid. Flamenco García, Luis, 1999. Las construcciones concesivas y adversativas. In: Ignacio Bosque, Violeta Demonte (Dirs.), Gramática descriptiva de la lengua española. Espasa-Calpe, Madrid, pp. 3805--3878. Fleischman, Suzanne, 1982. The Future in Thought and Language. Diachronic Evidence From Romance. Cambridge University Press, Cambridge. Fleischman, Suzanne, 1989. Temporal distance: a basic linguistic metaphor. Stud. Lang. 13 (1), 1--50. Garrido Rodríguez, María del Camino, 2004. Conectores Contraargumentativos en la Conversacio´n Coloquial. Universidad de Leo´n, Leo´n. Gazdar, Gerald, 1979. Pragmatics. Implicature, Presuppositions and Logical Form. Academic Press, New York and London. Giannakidou, Anastasia, Mari, Alda, 2012. The future of Greek and Italian: An epistemic analysis. In: Proceedings of Sinn und Bedeutung, 17. pp. 255--270, http://semanticsarchive.net/Archive/Dk3NGEwY/GiannakidouMari.pdf. Gili Gaya, Samuel, [1951] 1993. Curso superior de sintaxis española. Barcelona Indugraf. Habler, Gerda, 2010. Epistemic modality and evidentiality and their definition on a deictic basis. In: Becker, Martin G., Remberger, Eva-Maria (Eds.), Modality and Mood in Romance: Modal Interpretation, Mood Selection, and Mood Alternation. Walter de Gruyter, New York, pp. 95-108. Haspelmath, Martin, 1997. Temporal Adverbials in the World’s Languages. Lincom Europa, München. Hennemann, Anja, 2013. A Context-sensitive and Functional Approach to Evidentiality in Spanish or Why Evidentiality Needs a Superordinate Category. Peter Lang, Frankfurt am Main. Jakobson, Roman, [1957] 1971. Shifters, verbal categories, and the Russian verb. In: Roman Jakobson (Ed.), Selected Writings. Vol. II. The Hague: Mouton, pp. 130--147. Jaszczolt, Kasia M., 2009. Representing Time: An Essay on Temporality as Modality. Oxford University Press, Oxford. Kissine, Mikhail, 2008. Why will is not a modal? Nat. Lang. Semant. 16 (2), 129--155. König, Ekkehard, 1986. Conditionals, concessive conditionals and concessives: Areas of contrast, overlap and neutralization. In: Traugott, Elizabeth C., et al. (Eds.), On Conditionals. Cambridge University Press, Cambridge, pp. 229--246. Kovacci, Ofelia, 1999. El adverbio. In: Ignacio Bosque, Violeta Demonte (Dirs.), Gramática descriptiva de la lengua española. Espasa Calpe, Madrid, pp. 705--786. Lambrecht, Knud, 1994. Information Structure and Sentence Form. Topic, Focus and the Mental Representations of Discourse Referents. Cambridge University Press, Cambridge. Langacker, Ronald W., 1999. Virtual reality. Stud. Linguist. Sci. 29 (2), 77--103. Langacker, Ronald W., 2011. The English present: temporal coincidence vs. epistemic immediacy. In: Patard, Adeline, Brisard, Frank (Eds.), Cognitive Approaches to Tense, Aspect and Epistemic Modality. John Benjamins, Amsterdam/Philadelphia, pp. 45--86. Levinson, Stephen C., 2003. Space in Language and Cognition. Explorations in Cognitive Diversity. Cambridge University Press, Cambridge. Martín Zorraquino, Antonia María and Portolés, José, 1999. Los marcadores del discurso. In: Ignacio Bosque, Violeta Demonte (Dirs.), Gramática descriptiva de la lengua española. Espasa Calpe, Madrid, pp. 4051--4213. Nuyts, Jan, 2001a. Epistemic Modality, Language, and Conceptualization. John Benjamins, Amsterdam and Philadelphia. Nuyts, Jan, 2001b. Subjectivity as an evidential dimension in epistemic modal expressions. J. Pragmat. 33, 383--400. Palmer, Frank R., 1986. Mood and Modality. Cambridge University Press, Cambridge. Pérez Saldanya, Manuel, 2002. Les relacions temporals i aspectuals. In: Joan Solà, Maria-Rosa Lloret, Joan Mascaro´ and Manuel Pérez Saldanya (Dirs.), Gramàtica del català contemporani. Empúries, Barcelona, pp. 2567--2662. Portolés, José, 1998. El concepto de suficiencia argumentativa. Signo y Seña 9, 201--224. Prince, Ellen, 1992. The ZPG letter: Subjects, definiteness, and information status. In: Mann, William C., Thompson, Sandra (Eds.), Discourse Description: Discourse Analyses of a Fundraising Text. John Benjamins, Amsterdam and Philadelphia, pp. 295--325. Real Academia Española [RAE], 1973. Esbozo de una nueva gramática de la lengua española. Espasa-Calpe, Madrid. Real Academia Española [RAE], 2009. Nueva gramática de la lengua española, Morfología. Sintaxis I. Espasa-Calpe, Madrid. Real Academia Española, [RAE], 2015. Banco de datos (CREA) Corpus de referencia del español actual. http://www.rae.es. Reyes, Graciela, 1990. Tiempo, modo, aspecto e intertextualidad. Rev. Españ. Lingüíst. 20, 17--53. Rivero, María Luisa, 2014. Spanish inferential and mirative futures and conditionals: an evidential gradable modal proposal. Lingua 151, 197--215. Rodríguez Rosique, Susana, 2008. Pragmática y Gramática. Condicionales Concesivas en Español. Peter Lang, Frankfurt am Main.

S. Rodríguez Rosique / Journal of Pragmatics 85 (2015) 181--199

199

Rojo, Guillermo, 1974. La temporalidad verbal en español. Verba 1, 68--149. Rojo, Guillermo, Veiga Alexandre, 1999. El tiempo verbal. Los tiempos simples. In: Ignacio Bosque, Violeta Demonte (Dirs.), Gramática descriptiva de la lengua española. Madrid: Espasa-Calpe, 2867-2934. Roulet, Eddy, et al., 1991. L’articulation du Discours en Français Contemporain. Peter Lang, Berne. Schwenter, Scott, 1999. Pragmatics of Conditional Marking. Garland, New York. Squartini, Mario, 2001. The internal structure of evidentiality in Romance. Stud. Lang. 25 (2), 297--334. Squartini, Mario, 2008. Lexical vs. grammatical evidentiality in French and Italian. Linguistics 46 (5), 917--947. Squartini, Mario, 2012. Evidentiality in interaction: the concessive use of the Italian Future between grammar and discourse. J. Pragmat. 44, 2116--2128. Sweetser, Eve, 1990. From Etymology to Pragmatics: Metaphorical and Cultural Aspects of Semantic Structure. Cambridge University Press, Cambridge. Traugott, Elizabeth C., 1978. On the expression of spatio-temporal relations in language. In: Greenberg, Joseph H., Ferguson, Charles A., Moravcsik, Edith A. (Eds.), Universals of Human Language, vol. 3. Stanford University Press, Stanford, pp. 369--400. Traugott, Elizabeth C., 1989. Subjectification in grammaticalisation. In: Stein, Dieter, Wright, Susan (Eds.), Subjectivity and Subjectivisation. Cambridge University Press, Cambridge, pp. 31--54. Traugott, Elizabeth C., 2010. (Inter) subjectivity and (inter)subjectification: a reassessment. In: Davidse, Kristin, Vandelanotte, Lieven, Cuyckens, Hubert (Eds.), Subjectification, Intersubjectification and Grammaticalization. Walter de Gruyter, Berlin and New York, pp. 29--71. Van der Auwera, Johan, Plungian, Vladimir A., 1998. Modality’s semantic map. Linguist. Typol. 2, 7--124. Susana Rodríguez Rosique is an assistant professor of Spanish Linguistics at the University of Alicante. Her research interests focus on the relationships between pragmatics and grammar; specifically, she has worked on the pragmatic values of verb categories, as well as on the connection between grammatical construction and discourse. More generally, she is interested in the recurrence of certain notions, such as modality, scalarity and contrast. She has written the monograph Pragmática y Gramática. Condicionales Concesivas en Españ ol (Peter Lang, 2008), and has co-edited the volumes Spanish Word Formation and Lexical Creation (John Benjamins, 2011) and Verb Classes and Aspect (Forthcoming).