On English turn out and Spanish resultar mirative constructions. A case of ongoing grammaticalization?

1 On English turn out and Spanish resultar mirative constructions A case of ongoing grammaticalization? Mario Serrano-Losada University of Santiago de Compostela This article focuses on the diachronic development of English turn out and Spanish resultar ‘turn out’ mirative constructions. Having undergone processes of semantic generalization over time, both verbs express evidential and, most prominently, mirative nuances in the present-day languages. This study explores the mechanisms that condition the evolution of turn out and resultar from their original meanings as lexical resultative and change-of-state verbs towards their eventual subjectification and grammaticalization as predicates conveying evidential and mirative senses. The present-day mirative constructions take thatand infinitival complement clauses in both languages. The analysis suggested here shows that both verbs exhibit diverging, though closely related, paths and degrees of grammaticalization. Moreover, this study delves into the further development of these predicates as parenthetical expressions. While English parenthetical turns out has already been grammaticalized, Spanish resulta may be on its way to becoming a grammaticalized parenthetical. Keywords: mirativity, evidentiality, grammaticalization, raising predicates, resultar, turn out 1. Introduction The expression of evidential and mirative strategies may take a wide range of forms in the languages of the world. English turn out and its Spanish equivalent resultar are such forms. As raising predicates, these verbs can take both thatand INF-clauses as complements. In addition to their evidential readings (see Cornillie 2007, 2008 for resultar, Heller & Howe Accepted manuscript. Article published as: Mario Serrano-Losada. 2017. On English turn out and Spanish resultar mirative constructions. A case of ongoing grammaticalization? Journal of Historical Linguistics 7(1/2): 160-189. http://doi.org/10.1075/jhl.7.1-2.07ser. 2 2010 for turn out), these constructions express mirativity (see Howe & Heller 2010 for turn out), i.e., they signal information which is newsworthy or unexpected, with overtones of surprise and counterexpectation. The present paper deals with the diachronic development of mirative resultar and turn out verb constructions such as the ones illustrated in (1) and (2): (1) It turns out that elephants have an advanced sense of self, which means in part that they’re smart enough to be capable of really caring about others. (COCA, 2009) (2) En esta ocasión le voy a dar un ejemplo bastante ridículo. Resulta que los lobos han desaparecido oficialmente en Estados Unidos. La cacería descontrolada, los prejuicios y la insaciable voracidad de los explotadores del medio ambiente acabaron con esos nobles animales. ‘This time I’ll give you a pretty ridiculous example. It turns out that wolves have officially disappeared in the United States. Uncontrolled hunting, prejudices and the insatiable voracity of environmental exploiters have put an end to those noble animals.’ (CORPES, 2001) The verbs turn out and resultar in the impersonal constructions in (1) and (2) behave like complement-taking predicates (CTPs), that is, they have an inherently secondary status with respect to the proposition expressed by the complement clause (see Boye & Harder 2007, Van Bogaert 2011). Thus, in spite of their matrix clause status, resulta que and it turns out that introduce secondary information, their main function being that of conveying evidential and, most conspicuously, mirative overtones, as well as that of introducing the propositions under their scope. The study of evidential and mirative CTPs like the ones exemplified in (1) and (2) is especially relevant, since such constructions seem to be evolving towards grammaticalized parenthetical expressions whose functions resemble those of pragmatic markers (see Dehé & Kavalova 2007, Kaltenböck 2013, 2015). This article focuses on resultar and turn out constructions involving thatand INFcomplements. It explores the processes whereby these predicates evolve from their erstwhile status as lexical resultative and change-of-state verbs to their eventual subjectification and grammaticalization as verbs expressing evidential and mirative meanings. Furthermore, it delves into their development as parenthetical expressions. The results show that both constructions exhibit diverging, though analogous, paths and degrees of grammaticalization. 3 While English turns out has already been grammaticalized into a parenthetical, Spanish resulta may be undergoing further grammaticalization into a parenthetical element. I begin by offering an introduction to the expression of evidential and mirative strategies in English and Spanish in Section 2. In Section 3 I outline the present-day usages of turn out and resultar mirative constructions. Section 4 probes into the historical development of turn out, while Section 5 explores the diachronic evolution of resultar. These last two sections focus on qualitative and contrastive aspects, especially as regards the parallelisms and divergences between both developments. Section 6 offers a tentative account of the further grammaticalization of Spanish resultar into a parenthetical. Finally, in Section 7 I put forward some concluding remarks, including some suggestions for further research. 2. Evidentiality and mirativity in English and Spanish Evidentiality and – to a lesser extent – mirativity have received growing scholarly attention in recent times, from both the perspective of typological studies (see, among others, Aikhenvald 2004, 2014, 2015 for evidentiality; DeLancey 2001, 2012, Aikhenvald 2012 for mirativity) and of studies on languages in which these categories are not morphologically encoded (see, for instance, Chafe 1986 for evidentiality in English, Squartini 2001 for evidentiality in Romance, Bermúdez 2005 and Cornillie 2007, for evidentiality in Spanish, Diewald & Smirnova 2010a for evidentiality in German, Hengevald & Olbertz 2012 for mirativity in Andean Spanish, Marín Arrese et al. 2017 for evidentiality in European languages). Despite the wealth of literature devoted to evidentiality and mirativity, their status as linguistic categories is still a matter of considerable debate (see Aikhenvald 2004, 2012, Diewald & Smirnova 2010a, 2010b, DeLancey 2012, among others). This is especially true when it comes to defining these categories, elucidating the way in which they relate to each other and delimiting what kind of elements might be considered evidential or mirative markers. Evidentiality is the linguistic expression of source of information. It is an obligatory grammatical category, expressed morphologically by means of affixes, clitics and/or particles, in about a quarter of the languages of the world (Aikhenvald 2004: 1). Mirativity is a complex category that conveys new information, surprise, and sudden discovery with overtones of counterexpectation and unprepared mind (of the speaker, the addressee or a third person) (see DeLancey 1997, 2001, 2012, Aikhenvald 2012). Grammatical miratives are also well attested in a substantial number of languages, but they are less prevalent than grammatical evidentials in the languages of the world (Aikhenvald 2012: 473). 4 Although mirativity has sometimes been considered a part of evidentiality, there is widespread scholarly consensus that the former constitutes a linguistic category in its own right (see Aikhenvald 2004: 195-215 for an overview of the relationship between both categories, 2012, DeLancey 2012, Hengeveld & Olbertz 2012, Peterson 2013, 2015; contra Hill 2012). However, the boundaries between evidentiality and mirativity are not clear-cut, and both categories seem to be closely interrelated. This is evidenced by the fact that some evidentials can acquire mirative uses. According to Aikhenvald (2015: 260), a non-visual or a non-firsthand evidential may express a mirative meaning when the speaker uses it in first person. Gipper (2014), for instance, provides evidence for the extension from inferential evidentiality to mirativity in Yurakaré, a Bolivian isolate language. In languages like English and Spanish, which do not encode evidentiality or mirativity morphologically, the linguistic realizations of these categories are optional communicative strategies that are most often lexically represented (see Cornillie 2007, 2008, Diewald & Smirnova 2010a, 2010b). Still, evidential and mirative strategies in these languages can be found at different linguistic levels that range from the prosodic through the morphological to the discourse level. Exclamatory intonation, for instance, may convey mirative nuances of surprise. So do a range of discourse-pragmatic expressions, like the English interjection wow! or its Spanish equivalents ¡vaya! and ¡anda!, as well as adverbial expressions such as English surprisingly and unexpectedly and their Spanish counterparts sorpredentemente and inesperadamente. Nevertheless, the expression of these strategies may take place in more grammatical settings in these languages. Unlike certain languages with complex grammaticalized evidential systems, European languages generally feature expressions of indirect evidentiality, while direct evidentiality is left formally unmarked (Diewald & Smirnova 2010: 59). In (3) below, adapted from Bermúdez (2005: 3-4), the Spanish imperfect (3a) is used to express indirect evidence. In fact, salía does not imply any temporal or aspectual reference (the bus might have departed yesterday or may depart today or tomorrow). Rather, it suggests that the speaker is reporting on something which s/he has not witnessed him/herself: s/he might have read that the bus was supposed to leave at 15h after looking it up on the timetable. The use of the imperfect contrasts with the use of the perfect in (3b), where there is no room for indirect evidence, as it indicates that the bus did depart at 15h. (3) a. el bus salía {hoy/ayer/mañana} a las 15h. 5 the bus depart.PST.IMPFV.3SG {today/yesterday/tomorrow} to the 15h ‘the bus is/was supposed to depart at 15h.’ b. el bus salió a las 15h. the bus depart.PST.PFV.3SG to the 15h ‘the bus departed at 15h’ Likewise, mirativity may be grammatically encoded in some varieties of Spanish. In Ecuadorian Highland Spanish the perfect aspect (present perfect) has a secondary function as a mirative marker (Olbertz 2009, Hengeveld & Olbertz 2012). According to Olbertz (2009: 70), the mirative might have been introduced into Andean Spanish through contact with Quechua varieties. In (4), the present perfect signals surprise upon the discovery of new information. Further evidence of this mirative usage is (5), in which the erstwhile present perfect is used to refer to a future event. (4) De albaricoque ha sido. From apricot AUX.PRES.3SG COP.RES ‘It’s from apricot (I see).’ (Hengeveld & Olbertz 2012: 492) 1 (5) El año que viene ha sido bisiesto. The year REL come.PRES.3SG AUX.PRES.3SG COP.RES leap year ‘(I just realize) next year is a leap year.’ (Hengeveld & Olbertz 2012: 493) Considering these examples, it appears that both evidential and mirative strategies may also be expressed resorting to grammatical means even in languages that do not have evidentiality and/or mirativity as obligatory grammatical categories. Although the expression of these categories is far from obligatory in English and Spanish, the use of both turn out and resultar is a further example of the linguistic realizations of evidential and mirative strategies in these languages. 3. Present-day mirative turn out and resultar constructions 1 The translation for example (4) is Hengeveld & Olbertz’ (2012) own. Perhaps a more appropriate translation would be ‘It turns out it is apricot jam’, since ‘I see’ implies direct evidence. 6 As noted above, English turn out and Spanish resultar may convey both mirative and evidential overtones. However, their mirative readings are more salient and tend to overlay the evidential ones. The expressions it turns out that and resulta que in (1) and (2) could be paraphrased using mirative adverbials such as unexpectedly or surprisingly, in the case of (1), or their Spanish equivalents inesperadamente or sorprendentemente, in (2). Such paraphrases reveal that it turns out that and resulta que convey mirative overtones of sudden revelation (to the addressee) and unexpectedness or counterexpectation (both to the speaker and the addressee) and thus of surprise. Moreover, these expressions are used to emphasize the newsworthiness of the propositions within their scope. Conversely, the omission of the turn out and resultar matrix clauses in (1) and (2) would exclude their mirative interpretations. In (1′) and (2′), the mirative meanings of these expressions are evaluated by means of a simple entailment test: (1′) It turns out that elephants have an advanced sense of self, # not that this is a newsworthy, unexpected or surprising discovery. (2′) Resulta que los lobos han desaparecido oficialmente en Estados Unidos, # no es que esto sea un descubrimiento destacable, inesperado ni sorprendente. ‘It turns out that wolves have officially disappeared in the US, # not that this is a newsworthy, unexpected or surprising discovery.’ The negations in (1′) and (2′) prove odd, if not contradictory. The uses of these verbs convey newsworthiness and surprise and, in turn, overtones of unexpectedness and counterexpectation. Thus, the mirative meanings conveyed by the turn out and resultar constructions can be regarded as rather conventionalized and not as mere conversational implicatures. Concomitant to their use as mirative strategies, these verbs also tend to convey nonfirsthand evidential nuances. The expressions it turns out that and resulta que in (1) and (2) could also be paraphrased using hearsay evidential expressions such as I’ve been told or its Spanish counterpart me han dicho. The evidential scope of these verbs can be tested as well: (1″) It turns out that elephants have an advanced sense of self, # but elephants might not have an advanced sense of self. 7 (2″) Resulta que los lobos han desaparecido oficialmente en Estados Unidos, # pero podría haber lobos en Estados Unidos ‘It turns out that wolves have officially disappeared in the United States, # but there could be wolves in the United States.’ The infelicitous result of examples (1″) and (2″) corroborate that the speakers rely on external sources of information to make their claims. Compare these examples with the felicitous combination of the inferential use of seem in Wolves seem to have officially disappeared in the United States, but there could be wolves in the United States. As Cornillie (2007: 49) argues, the speaker is not involved in the conclusion expressed by resultar, a claim that can also be extended to turn out. The speakers in (1″) and (2″) are reporting on someone else’s findings (most likely scientists), and do not evaluate the proposition. The constructions it turns out and resulta que are thus used to convey hearsay evidentiality. However, the array of evidential meanings expressed by these verbs may also cover inference, deduction or reasoning from direct evidence. In fact, this is not an unusual development, as resultatives or other forms with completive meanings may acquire an additional inferential overtone, or more generally, overtones of non-firsthand information (Aikhenvald 2015). Sections 3.1 and 3.2 provide an overview of the main mirative construction types with these verbs. Although there might be some geographic differences regarding the use of resultar and turn out mirative constructions, the exploration of diatopic variation (e.g., European Spanish vs Puerto Rican Spanish) is beyond the scope of this article. Nevertheless, resultar and turn out mirative expressions are well attested on both sides of the Atlantic. 3.1 Mirative turn out constructions Mirative strategies in English have not received much scholarly attention (see, for instance, Gentens et al. 2016 for no wonder miratives). In Present-day English, mirative uses of turn out may occur in three different constructions: raised subject constructions (6), impersonal constructions (7) and parenthetical constructions (8-9). In the raised subject construction turn out takes a to-INF clause. The verb turn out in this construction functions in a similar way to a semi-auxiliary, modifying the proposition contained in the INF-clause. This infinitive is generally copular be, although other infinitives may occur in the construction occasionally (Serrano-Losada, forth.). In (6), the former 2016 8 presidential candidate Hillary Clinton is commenting on her 2015 email controversy, in which she used her private email server for her official email communications while she held the office of United States Secretary of State: (6) So, look, I did what was, as I said, allowed. I said it wasn’t the best choice. And it turned out to be a mistake, in retrospect. (COCA, 2015) In example (6) Clinton’s expectations are challenged by the outcome: she made a poor decision that was ultimately revealed to be a mistake. Thus, the unexpected result portrayed in (6) entails overtones of sudden realization and is intended to express surprise on the speaker’s part and to astound her addressee. In turn, raised subject turn out constructions tend to express evidential overtones of inference from direct evidence. In (6) the speaker seems to have deduced that what she did was a mistake after reflecting upon (and reevaluating) the experience (in retrospect). Contrast the use of the raised turn out construction with an instance such as and it was a mistake, where the deductive process would not be acknowledged. Thus, a piece of information (her mistake) is presented as being logically inferred from another piece of information (namely her decision to use her personal email to conduct government business and the consequences this brought about). This can thus be interpreted as inference based on results or assumption. According to Aikhenvald (2004: 146), the fact that the meanings of deduction and inference are linked to the results of something that has already been achieved explains why resultatives tend to develop inferential overtones. Moreover, the evidential and mirative use of raising turn out in (6) can be related to the concept of “deferred realization” (Aikhenvald 2015: 262), whereby a “post-factum inference” is made based on something that the speaker experienced before but that s/he could only interpret and realize in retrospect. These revelations tend to involve a sudden realization whereby the speaker is enabled to interpret what has occurred. Such unexpected realizations reveal the speaker’s “unprepared mind” and also involve the speaker’s distancing her/himself from the actual events. As illustrated in (7), raising turn out can also occur in impersonal construction with it as subject and a that-clause. As argued in Huddleston & Pullum (2002: 960), in such constructions the subject is semantically empty and the content clause is the only argument of the matrix clause. A different analysis is put forward by Quirk et al. (1985: 1183), who 9 consider the that-clauses in examples like (7) cases of obligatory subject extraposition. Huddleston & Pullum (2002: 960-962), however, analyze such constructions as impersonal and argue for the inability of the extraposed subject interpretation, suggesting, inter alia, that these alleged extraposed subjects cannot occupy preverbal position. In this construction, the turn out matrix clause has a focalizing function in discourse with forward scope: it is generally used to introduce new information. Kaltenböck (2015: 124-126) refers to such turn out clauses as “presentational matrix clauses”. In his view, these are “weakly stressed”, “filler matrix predicates” that assert the existence of the second clause. Thus, although syntactically the turn out clause is a matrix clause, semantically and discursively this matrix clause can be considered secondary to its complement clause (Boye & Harder 2007). (7) We just spent an hour in the car together. It turns out that his best friend is one of my best friends (COCA, 2011) In (7) the speaker uses the turn out impersonal construction to express sudden revelation to the addressee and unexpectedness and surprise given the fact that after having talked to someone for an hour, s/he has discovered that they have a very good friend in common. As regards evidentiality, the construction seems to entail hearsay, as the speaker has learned this from talking to somebody. According to Heller & Howe (2010, slide 18), the meaning of turn out impersonal clauses is evidential in the sense that it indicates an external source of evaluation. In spite of the contextual information provided by the speaker, the impersonal turn out construction introduces a reportative reading that would not be as readily available if the turn out construction were omitted. Unlike the raised subject construction, which usually conveys either inference or reportative evidentiality, impersonal turn out constructions tend to convey reportative evidential meaning. Mirative turn out can also appear in parenthetical expressions. These may be either bare (8) or adverbial (9) parentheticals: (8) She looked at her husband with an astonished glance. “Well, the CEO, turns out, is an absolute prick.” (COCA, 2008) (9) I had scheduled […] an interview with the CEO of UnitedHealthcare, the largest health insurance company in the United States and my health insurance company, as 16 Although the raised subject construction and the impersonal construction seem to have emerged around the same time, the former is more robust in the data. The fact that first attestations for raised subject constructions such as (20) and impersonal constructions like (19) are found in letters and in epistolary literature, genres which one might regard as two of the closest to the spoken register in the corpus, point to the idea that the emergence of mirative turn out, might have taken place long before it made it into the written record (Serrano-Losada, forth.). A further development regarding mirative turn out constructions is the emergence of parenthetical uses. Throughout the history of English, raising verbs have often developed functions similar to those of comment clauses and have even been grammaticalized into parenthetical expressions or pragmatic markers (see Boye & Harder 2007, Brinton 2008, Kaltenböck 2013, 2015, López-Couso & Méndez-Naya 2014, 2015). The CLMET3.0 data (Figure 2) reveal that parenthetical instances of the verb are first attested in its adverbial parenthetical form as it turns/turned out (21) during the nineteenth century, although they do not start growing in frequency until the first half of the twentieth century, an increase that is traceable in the American English data: adverbial parentheticals rise from 0.72 tokens per mil. words during the first decade of the 1900s to 3.14 in the 1950s in COHA. This construction is similar to adverbial parentheticals like as it seems, which, as claimed by López-Couso & Méndez-Naya (2014: 209), clearly predate bare parentheticals (in this case, parenthetical it seems). (21) although perhaps, it would, as it turned out, have been much better for me, personally, if I had gone there again, under all the disadvantages which I had to anticipate, […] (CLMET3.0, 1820–22) Bare turn out parenthetical constructions seem to have developed from impersonal turn out constructions (see Torres Cacoullos & Walker 2009, Kaltenböck 2015). They are first attested in twentieth century American English (22), although it is not until the 1990s that turn out parentheticals without anticipatory it are documented (23). (22) She is of German birth, it turns out, despite her name, which isn’t her original one. (COHA, 1927) 17 (23) a. And my mom, turns out, was a budding starlet in the early ‘50s, […] (COCA, 1996) b. I thought I’d looked everywhere. Everywhere but up, turns out. (COCA, 2009) Data from COCA show that parenthetical turn out amounted to 0.05 tokens per mil. words during the period 1990–1994. Its presence increased to 0.98 tokens per mil. words by the 2010–2015 period. These bare parentheticals are most common in the magazine and spoken components of the corpus. Hence the corpus data evidence that parenthetical turns out is has become entrenched in American English. This is also supported by several factors, including anticipatory it deletion, the parenthetical’s ability to appear in non-initial positions and the fact that the tense of the parenthetical is independent of the verb in the host clause, as exemplified in (23) above (see Heller & Howe 2010, Howe & Heller 2010, Kaltenböck 2015). Figure 2. Semantic change of turn out (from Serrano-Losada, forth.) Figure 2 schematizes the plausible semantic changes undergone by turn out on its way to becoming an evidential and mirative predicate during the Late Modern English period. It could thus be argued that a subjectification process triggered semantic innovations which resulted in the encoding of evidential and mirative nuances and in further syntactic changes. However, this explanation is not without difficulties. As examples (18-20) above reveal, all the different turn out senses in Table 1 as well as its copular and raising uses are first attested in the 1740s, within a very short period of time. Thus, the chronological evidence leaves little room to consider a gradual grammaticalization process that comprises different stages leading to the emergence of raising turn out and its evidential and mirative senses. Conversely, the data suggest that this might have been a rather cataclysmic change (Hendrik De Smet, p.c.) in the sense of Petré (2012), who argues for the “abrupt” copularization of Old English becuman and weaxan during the twelfth and thirteen centuries as an outcome of analogical modeling and not as the result of gradual grammaticalization. ‘to result, ‘to become ultimately ‘to be ultimately to eventuate’ (and be found to be)’ found to be’ resultativity evidentiality mirativity change-of-state 18 Petré (2012: 28) refers to the abruptness with which these two Old English copulas emerged and became entrenched as sudden categorial incursion after De Smet (2009), who defined categorial incursion as a “non-gradual” analogy-driven mechanism whereby a construction becomes a member of a new category, which, however, already exists as an established category. According to De Smet (2009: 1748), such a change could be considered an “analogical extension” of a construction into the realm of another. In fact, Petré (2012: 61) claims that a certain degree of similarity is necessary for constructional attraction to occur. In the case of turn out, a process of (pseudo)copularization triggered the possibility of evaluative complements (17), which in turn allowed for incipient evidential interpretations, which opened the door to analogical reasoning. Thus, the verb’s abrupt incursion into the paradigm of raising verbs was brought about by means of both concrete and structural analogical modeling (see Fischer 2015) after preexisting established members of this category such as seem, appear (see Gisborne & Holmes 2007), happen or prove – as well as after other members of this class that were grammaticalized during the same period, including promise and threaten (see Traugott 1997) – all of which share, most prominently, their status as raising predicates that can be used to express either evidential meanings, in the case of seem and appear, or mirative ones, in the case of happen and prove. 5. A diachronic overview of resultar Unlike mirative turn out, the emergence of evidential and mirative resultar can be accounted for by means of a more traditional grammaticalization process in which three different stages can be posited. The dataset for this section consists of 596 instances of the resultar + INF string and 3137 of the resultar + que-clause string drawn from CORDE. 5.1 Stage I Spanish resultar is a cultismo, a learned word which only penetrates the Spanish lexicon during the fifteenth century, most likely via Italian. None of this verb’s Latin meanings encompass its resultative Spanish sense. Thus, Latin RĔSULTO, meaning ‘to jump, to bounce’, i.e., a physical displacement from one point to another, comes to acquire resultative and change-of-state senses by means of a metaphoric semantic change: abstract (as well as physical) change is conceived in terms of movement. First attested in Enrique de Villena’s Los doze trabajos de Hércules (CORDE, 1417), resultar was incorporated into Castilian 19 quite abruptly during the fifteenth century (CORDE yields 340 hits in 106 documents for the period 1417–1500). The verb appears in a variety of constructions from its onset. It fits the x resulta de y ‘x results from y’ schema: resultar is a lexical resultative verb that takes a preposed prepositional de-complement, as in (24). Its early occurrences do not express conventionalized evidential or mirative senses, and its evaluative readings will only appear later on. (24) por tanto el juyzio dela rrazon non puede ser thus the judgment of the reason not can PRES.3SG be.INF cierto […] delo qual rresulta que en los sueños non true […] of the which result. PRES.3SG that in the dreams not podemos pecar njn tanpoco meresçer can.PRES.1SG sin.INF nor neither achieve.INF ‘Thus judgment of reason cannot be true […] from which results that we cannot either sin nor achieve in our dreams’ (CORDE, 1445) 2 The most frequent resultar sequences are those preceded by de do(nde), de que, de lo cual, de él and other similar consecutive sequences that express result. This verb is generally followed by an INF-clause or a que-clause as subject. The resultar que-clause construction can be represented as (25a). In this construction, the que-clause functions as the extraposed subject of resultar, as shown in (25b): (25) a. [de X]C resultar [que-clause]SBJ b. de lo qual resulta que los spiritus malignos non of which result. PRES.3SG that the spirits evil not pueden tomar cuerpos can.PRES.3SG possess.INF bodies 2 In order to reflect the different syntactic patterns of earlier Spanish resultar in the translations, I have opted for less idiomatic renderings of the examples. The intransitive verb result is used in the translations of earlier examples meaning “to become, turn out (in a specified manner)” (OED, s.v. result, v., 1. c), in contrast to evidential/mirative turn out. 20 ‘From which results that evil spirits cannot possess bodies’ (CORDE c. 1445) From its earliest attestations, resultar can also co-occur with extraposed INF-clauses in subject function, as illustrated in (26): (26) a. [de X]C resultar [INF-clause]SBJ b. De do rresulta ser prouado nuestro of where result. PRES.3SG be.INF prove. PST.PTCP our yntento & proposito. intention and purpose ‘From which results to be proven our intention and purpose.’ (CORDE, 14771485) Both resultar + nominal INF-clauses and que-clauses are synonymous at the time. However, que-clauses are much more numerous and are attested earlier on. Just in the 1425–1474 period, the corpus yields 42 instances of resultar + que-clause and one instance of resultar + INF. Moreover, while que-clauses are chiefly Romance, nominal INF-clauses have a strong Latin feel to them. Thus, it seems plausible to assume that such INF-clauses emerged due to the influence of Latin syntactic models. Fifteenth century Castilian underwent a process of intensive Re-Latinization (see Pountain 1998, Pons 2007, 2008). The influence of Latin models brought about the introduction of syntactic innovations such as the imitative structures of Latin Accusativus cum Infinitivo (ACI), exemplified in (27): (27) Vistas estas tres maneras de escrivir, podemos dezir Seen these three ways of writing can.PRES.1SG say.INF el estilo de aquestas coplas ser sátiro e comedio the style of these couplets be.INF satirical and comical ‘Given these three ways of writing, we can claim the style of these couplets to be satirical and comical’ (Mena, Coronación ca. 1439, in Pons 2007: 274) In fifteenth-century imitative ACI constructions the INF-clause functions as the direct object 21 of the main verb and has a subject different from that of the main clause (Pons 2007, 2008). In contrast, the INF-clause in the resultar construction functions as the subject of the main verb, resultar, and has its own subject embedded within, as illustrated in (28) below: (28) de ser bajos los tributos resulta ser los indios ociosos, pues se entiende que no trabajan más de cuanto les es necesario para la paga de los dichos tributos y para su continuo y pobre sostenimiento […] ‘Out of having low tributes the natives result (to be) idle, since it is known that they do not work more than it is necessary to pay those tributes and for their continuous and poor sustainment’ (CORDE, 1569–1570) Example (28) has two INF-clauses, one embedded in the complement (de ser bajos los tributos) and one in subject function (ser los indios ociosos). In both, the infinitives (ser) have their own subjects (los tributos and los indios, respectively). Although these sequences are not instances of ACI proper, their emergence was crucial for the rise of the imitative ACI in fifteenth century Castilian (Pons 2007: 277), and in fact essential for the subsequent development of raising resultar. Nominal INF-clauses in this construction do not present any restrictions regarding the infinitive, which explains the diversity of infinitives found from the fifteenth to the eighteenth century, with up to 30 different attested infinitives in the construction. From the early eighteenth century onwards resultar the resultative constructions in (25a) and (26a) start to recede. 5.2 Stage II Towards the mid seventeenth century, resultar undergoes a process of semantic generalization whereby its resultative meaning is blurred in favor of evaluative subjectivized readings. Incipient (inferential) evidential readings are first attested at this stage. Although instances corroborating this change are first documented in the mid seventeenth century, such readings do not become conventionalized until the mid-eighteenth century. In the case of resultar + INF-clause, this semantic change leads to ambiguous contexts in which the sequence allows for a double interpretation whereby the infinitive might be considered as head of the infinitive clause with subject function (resultar + ser-clause) or 22 interpreted together with resultar in a raised construction (resultar ser + C). Such ambiguous contexts facilitate the construction’s reanalysis, after which several actualizations took place. The interposition of evaluative adjectives between the infinitive and the subject (29) points towards the actualization of this reanalysis. Moreover, the erstwhile embedded subject of the infinitive agrees with resultar (30), whereas earlier examples of the construction like (28) do not exhibit such agreement (see Cornillie 2008). (29) Esparta tuvo rigor en no admitir a su República forastero; de que resultó ser tan corta su población, que […] ‘Esparta was strict in refusing foreigners in their Republic; from which resulted that their population was so little that […]’ (CORDE, 1626) (30) como estas fincas vienen a recaer en la religión por fin del religioso, resultan ser tantas las fincas de una y otra especie que poseen, […] ‘Since these estates end up in the hands of the clergy, the estates that they possess turn out to be so many […]’ (CORDE, 1747) During the eighteenth century the resultar INF-clause construction and the grammaticalized raised subject construction coexist. However, there is a gradual substitution over time (see Figure 4). The demise of the first clears the way towards the generalization of the new construction. The resultar + que-clause construction follows a parallel development, brought about by the semantic change indicated above. In (31), the resultar que-clause construction does no longer require a de-complement (although the por-phrase seems to occupy that slot). Moreover, the construction can be interpreted as impersonal, since the que-clause does no longer seem to act as the subject of resultar. The verb’s resultative meaning is somewhat obscured and, in turn, it conveys an emergent inferential nuance. (31) Por la idea que va de este resumen, resulta que el diezmo señalado por arrendamiento de las tierras es muy favorable al colono ‘From this summary, it turns out that the tithe fixed for the leasing of these lands is very positive for the settler’ (CORDE 1768) 23 For instance, in (31) the writer infers that the amount received for leasing the lands is favorable to the settlers based on the evidence provided in the summary. Although the evidential reading is confirmed in the context (Por la idea que va de este resumen), this example points to the incipient conventionalization of evidential overtones. As Aikhenvald (2015: 268) claims, the connection between resultative meaning and non-firsthand evidentiality is a widespread tendency across the languages of the world. 5.3 Stage III The third stage, which coincides with the late eighteenth century and early nineteenth century, evidences the conventionalization of hearsay evidential and mirative extensions, and the entrenchment of both the raised subject construction and the impersonal construction. A further actualization affecting resultar + INF occurs during this period. The last step towards the consolidation of the raised subject construction is subject raising itself (32), which seems to have taken place during the late eighteenth century. During this stage, there is a rise in frequency of ser, which is consolidated in the construction to the detriment of other possible infinitives in the nineteenth century (see Figure 3). (32) ambas á dos resultaron ser superiores á la de Villafeliche […] ‘Both of them turned out to be superior to Villafeliche’ (CORDE, 1791) Figure 3. Infinitives in the resultar + INF string in CORDE (frequencies per mil. words) 0 1 2 3 4 5 6 15th c. 16th c. 17th c. 18th c. 19th c. 20th c. ser other infinitives 24 The impersonal construction is also consolidated during this stage. In (33) the impersonal resultar construction expresses both non-firsthand evidentiality and mirativity. (33) Aun quando […] el sexto [abuelo hubiese sido] el primero que fuese bien educado en el [catolicismo], sin oir hablar ya de cosas de judios, resulta que desde el quinto abuelo pueden ser todos los ascendientes de los que viven ahora unos christianos tan puros, y asegurados en la fé, como los que jamás tubieron un ascendiente infecto. ‘Even if the sixth grandfather had been educated in Catholicism, without having ever heard about Judaism, it turns out that, from the fifth grandfather on, all of the ascendants of those who live now can be as pure and as rooted in the faith as those Christians who never had an infected ascendant.’ (CORDE, 1797) Example (33) is taken from a text in which the narrator explains the workings of the limpieza de sangre (literally ‘purity of blood’) tests enforced by the Spanish Inquisition to detect Jewish or Muslim ancestry. The resultar construction in the example expresses hearsay evidence, as the narrator is reporting on these appalling tests. In fact, resulta que in (33) could be replaced with other hearsay expressions like aparentemente ‘apparently’ or supuestamente ‘reportedly’. In turn, it conveys mirative nuances of newsworthiness and sudden revelation that are intended for the audience. 5.4 Summary Throughout its history, resultar underwent a process of semantic generalization whereby subjectivized meanings were progressively developed, resulting in the conventionalization of evidential and mirative nuances. These semantic shifts entailed a series of syntactic changes which brought about the weakening of the verb’s argument structure and its eventual reanalysis as a raising evidential and mirative predicate. 25 Figure 4. Resultative (+INF-clause) and evidential/mirative (-INF-clause) resultar + INF in CORDE (frequencies per mil. words). The different constructions developed at different stages still coexist during the nineteenth century. The effects of layering (see Hopper 1991, Hopper & Traugott 2003) become thus apparent by analyzing resultar in a single work. The examples in (34), all taken from Concepción Arenal’s La cuestión social (CORDE, 1880), demonstrate the effects of syntactic persistence in the case of the resultar + que-clause sequence. While (34a) exemplifies the survival of the early de X resulta Y construction, (34b-34c) illustrates the grammaticalized impersonal resultar construction: (34) a. De lo dicho resulta que la propiedad no es un hecho arbitrario, […] ‘From the aforementioned results that property is not an arbitrary fact’ b. Resulta que un hombre, en virtud de su personalidad, tiene derecho á ser propietario en general, pero no á serlo de una cosa particular. ‘It turns out that a man, in virtue of his personality, has the right to be an owner in general, but not of something in particular.’ c. Resulta, que á un pobre que no tiene hacienda ni industria ni comercio, […], para eximirse del servicio militar paga lo mismo que un magnate 0 0.5 1 1.5 2 2.5 3 3.5 4 +inf-clause ±inf-clause -inf-clause 32 M. W. Dixon, 1-51. Oxford: Oxford University Press. Aikhenvald, Alexandra Y. 2015. Evidentiality and its interrelations with other categories. Linguistic Typology 19:2.239-77. Bermúdez, Fernando. 2005. Evidencialidad: La codificación lingüística del punto de vista. Stockholm: Stockholm University dissertation. Boye, Kasper & Peter Harder. 2007. Complement-taking predicates: Usage and linguistic structure. Studies in Language 31:3.569-606. Brinton, Laurel. 2008. The comment clause in English. Cambridge: Cambridge University Press. Brovetto, Claudia. 2002. Spanish clauses without complementizer. Current issues in Romance languages: Selected papers from the 29th Linguistic Symposium on Romance Languages (LSRL), Ann Arbor, 8–11 April 1999 ed. by Teresa Satterfield, Christina Tortora & Diana Cresti, 33-46. Amsterdam: John Benjamins. Chafe, Wallace. 1986. Evidentiality in English conversation and academic writing. Evidentiality: The linguistic coding of epistemology ed. by Wallace Chafe & Johanna Nichols, 261-272. Norwood, NJ: Ablex. Cornillie, Bert. 2007. Evidentiality and epistemic modality in Spanish (semi)auxiliaries: a cognitive-functional approach. Berlin: Mouton de Gruyter. Cornillie, Bert. 2008. On the grammaticalization and (inter)subjectivity of evidential (semi)auxiliaries in Spanish. Theoretical and empirical issues in grammaticalization ed. by Elena Seoane and María José López-Couso, 55-76. Amsterdam: John Benjamins. De Smet, Hendrik. 2009. Analyzing reanalysis. Lingua 119:11.1728-1755. De Smet, Hendrik. 2010. Grammatical interference: Subject marker for and the phrasal verb particles out and forth. Gradience, gradualness and grammaticalization ed. by Elizabeth C. Traugott & Graeme Trousdale, 75-104. Amsterdam: John Benjamins. Dehé, Nicole & Yordanka Kavalova, eds. 2007. Parentheticals. Amsterdam: John Benjamins. DeLancey, Scott. 1997. Mirativity: The grammatical marking of unexpected information. Linguistic Typology 1.33-52. 33 DeLancey, Scott. 2001. The mirative and evidentiality. Journal of Pragmatics 33:3.369382. DeLancey, Scott. 2012. Still mirative after all these years. Linguistic Typology 16:3.529-564. Diewald, Gabriele & Elena Smirnova, eds. 2010a. Evidentiality in German: linguistic realization and regularities in grammaticalization. Berlin: Mouton de Gruyter. Diewald, Gabriele & Elena Smirnova, eds. 2010b. Linguistic realization of evidentiality in European languages. Berlin: Mouton de Gruyter. Fischer, Olga. 2015. ‘The influence of the grammatical system and analogy in processes of language change: The case of the auxiliation of have to once again’. Studies in linguistic variation and change: from Old to Middle English ed. by Fabienne Toupin & Brian Lowrey, 120-150. Newcastle: Cambridge Scholars. Gentens, Caroline, Ditte Kimps, Kristin Davidse, Gilles Jacobs, An Van Linden & Lieselotte Brems. 2016. Mirativity and Rhetorical Structure. Outside the Clause: Form and Function of Extra-Clausal Constituents, ed. by Gunther Kaltenböck, Evelien Keizer & Arne Lohmann, 125-156. Amsterdam: John Benjamins. Gipper, Sonja. 2014. From inferential to mirative: An interaction-based account of an emerging semantic extension. Usage-based approaches to language change ed. by Evie Coussé & Ferdinand von Mengden, 83-116. Amsterdam: John Benjamins. Gisborne, Nikolas & Jasper Holmes. 2007. A history of English evidential verbs of appearance. English Language and Linguistics 11:1.1-29. González Fernández, M. Jesús & Ricardo Maldonado. 1998. La perfectividad como fuente de contraexpectativas: resulta que X finalmente Y. Memorias del IV Encuentro de Lingüística en el Noroeste, 61-82. Hermosillo: Universidad Autónoma de Sonora. Heller, Brooke & Chad Howe. 2010. Raising parentheticals as discourse particles. New Reflections on Grammaticalization 4. 16-19 July 2008. University of Leuven. Hengeveld, Kees & Hella Olbertz. 2012. Didn't you know? Mirativity does exist! Linguistic Typology 16:3.487-503. Hill, Nathan W. 2012. ‘Mirativity’ Does Not Exist: ḥdug in ‘Lhasa’ Tibetan and Other Suspects. Linguistic Typology 16:3.389-433. Hopper, Paul J. & Elizabeth C. Traugott. 2003. Grammaticalization. 2nd ed. 34 Cambridge: Cambridge University Press. Hopper, Paul J. 1991. On some principles of grammaticization. Approaches to grammaticalization (vol. I) ed. by Elizabeth C. Traugott and Bernd Heine, 17-36. Amsterdam: John Benjamins. Howe, Chad & Brooke Heller. 2010. Turns out they weren’t much of a stretch: Variable patterns of structural persistence. Paper presented at 39th New Ways of Analyzing Variation, 4-6 November. University of Texas at San Antonio. Huddleston, Rodney D., & Geoffrey K. Pullum. 2002. The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press. Kaltenböck, Gunther. 2013. The development of comment clauses. The verb phrase in English ed. by Bas Aarts, Joanne Close, Geoffrey Leech & Sean Wallis, 286-317. Cambridge: Cambridge University Press. Kaltenböck, Gunther. 2015. Processibility. Corpus pragmatics: A handbook ed. by Karin Aijmer & Christoph Rühlemann, 117-142. Cambridge: Cambridge University Press. Kotwica, Dorota. 2015. Evidential al parecer: Between the physical and the cognitive meaning in Spanish scientific prose of the 18th, 19th and early 20th centuries. Journal of Pragmatics 85.155-167. López-Couso, María José & Belén Méndez-Naya. 2014. On the origin of clausal parenthetical constructions: Evidential/epistemic parentheticals with seem and impersonal think. Diachronic corpus pragmatics ed. by Irma Taavitsainen, Andreas H. Jucker & Jukka Tuominen, 189-212. Amsterdam: John Benjamins. López-Couso, María José & Belén Méndez-Naya. 2015. Evidential/epistemic markers of the type verb + complementizer: Some parallels from English and Romance. New directions in grammaticalization research ed. by Andrew D. M. Smith, Graeme Trousdale & Richard Waltereit, 93-120. Amsterdam: John Benjamins. Marín Arrese, Juana I., Gerda Haßler & Marta Carretero. 2017. Evidentiality Revisited: Cognitive grammar, functional and discourse-pragmatic perspectives. Amsterdam: John Benjamins. Olbertz, Hella. 2009. Mirativity and exclamatives in functional discourse grammar: Evidence from Spanish. Web Papers in Functional Discourse Grammar 82.6682. 35 Peterson, Tyler. 2013. Rethinking mirativity: the expression and implication of surprise. Unpublished paper. Available at http://semarch.linguistics.fas.nyu.edu/Archive/2FkYTg4O/Rethinking_Mirativity.pdf Peterson, Tyler. 2015. Grammatical evidentiality and the unprepared mind. Review of Cognitive Linguistics 13:2.314-3252. Petré, Peter. 2012. General productivity: How become waxed and wax became a copula. Cognitive Linguistics 23:1.27-65. Pons Rodríguez, Lola. 2007. El infinitivo no concertado latino en el castellano del siglo XV: propiedades formales. Actes du XXIV Congrès International de Linguistique et de Philologie Romanes (Aberystwyth, 1-6 août 2004) ed. by David A. Trotter, 273-286. Tübingen: Max Niemeyer Verlag. Pons Rodríguez, Lola. 2008. Las construcciones imitativas del Accusativus cum infinitivo: modelos latinos y consecuencias romances. Revista de Historia de la Lengua Española 3.117-148. Pountain, Christopher J. 1998. Learnèd syntax and the Romance languages: the ‘accusative and infinitive’ construction with declarative verbs in Castilian. Transactions of the Philological Society 96:2.159-201. Quirk, Randolph, Sidney Greenbaum, Geoffrey Leech & Jan Svartvik. 1985. A Comprehensive Grammar of the English Language. London: Longman. Serrano-Losada, Mario. Forthcoming. Raising turn out in Late Modern English: The rise of a mirative predicate. Review of Cognitive Linguistics. Squartini, Mario. 2001. The internal structure of evidentiality in Romance. Studies in Language 25:2.297-334. Torrego, Esther. 1983. More Effects of Successive Cyclic Movement. Linguistic Inquiry 14:3.561-565. Torres Cacoullos, Rena & James A. Walker. 2009. On the persistence of grammar in discourse formulas: A variationist study of that. Linguistics 47:1.1-43. Traugott, Elizabeth C. 1989. On the rise of epistemic meanings in English: An example of subjectification in semantic change. Language 65:1.31-55. Traugott, Elizabeth C. 1997. Subjectification and the development of epistemic meaning: The case of promise and threaten. Modality in Germanic languages: Historical and comparative perspectives ed. by Toril Swan & Olaf J. Westvik, 36 185-210. Berlin: Mouton de Gruyter. Traugott, Elizabeth C. 2003. Constructions in grammaticalization. The handbook of historical linguistics ed. by Brian D. Joseph and Richard D. Janda, 624-647. Oxford: Blackwell. Tyler, Andrea & Vyvyan Evans. 2003. The semantics of English prepositions: Spatial scenes, embodied meaning and cognition. Cambridge: Cambridge University Press. Van Bogaert, Julie. 2011. I think and other complement-taking mental predicates: a case of and for constructional grammaticalization. Linguistics 49:2.295-332. Author’s address Mario Serrano-Losada Department of English and German Faculty of Philology University of Santiago de Compostela 15782 Santiago de Compostela Spain [email protected]s