Raising turn out in late Modern English: The rise of a mirative predicate

1 Raising turn out in Late Modern English: The rise of a mirative predicate Mario Serrano-Losada University of Santiago de Compostela Abstract: Present-day English turn out is used in several constructions with mirative and evidential overtones. Among these, the raised subject construction and the impersonal construction stand out. The current paper provides a diachronic account of the changes that led to the emergence of these constructions. It examines the semantic and syntactic configuration of turn out and the mechanisms whereby this verb developed evidential and mirative readings during the Late Modern English period. The record shows that turn out developed abruptly from a full lexical (predominantly resultative and change-of-state) control verb into a raising verb in the course of the eighteenth century. This change was triggered by a process of semantic generalization and subjectification whereby the verb acquired evidential and, most notably, mirative nuances. Analogy seems to have played an important role as well, since the evidence suggests that other constructions somewhat similar in form (syntactic structural usage) and meaning facilitated this process. The bulk of the data examined in this study were drawn from the Oxford English Dictionary and the Corpus of Late Modern English Texts, version 3.0. Keywords: turn out, Late Modern English, subjectification, mirativity, evidentiality 1. Introduction Languages tend to convey different mirative meanings—i.e., surprise, sudden discovery, unexpectedness—in different ways, and these do not always have to be grammatically  I would like to thank Teresa Fanego and Belén Méndez-Naya for their careful reading of my manuscript, as well as Hendrik De Smet, Olga Fischer and Sune Gregersen for their valuable feedback and insightful suggestions on earlier versions of the article. All remaining shortcomings are, of course, my own. For generous financial support, I thank the Spanish Ministry for Education, Culture and Sport (grant FPU13/02618) and the Spanish Ministry of Economy and Competitiveness (National Program for Excellence in Scientific and Technical Research, grant FFI2014-52188-P). Accepted manuscript. Article published as: Mario Serrano-Losada. 2017. Raising turn out in late Modern English: The rise of a mirative predicate. Review of Cognitive Linguistics 15(2): 411-437. http://doi.org/10.1075/rcl.15.2.05ser. 2 encoded (Aikhenvald 2012: 474). In the case of Present-day English (PDE), such means range from prosody (e.g. exclamatory intonation) to discourse-pragmatic expressions (e.g. the interjection wow and the adverbs surprisingly and unexpectedly). However, mirativity can also be conveyed through more grammaticalized means. Such is the case of the raising predicate turn out, exemplified in (1) and (2): (1) The tea came to $1.51. […] After the turkey club […] I sipped the, by then, cooled green tea, which turned out to be Earl Grey. I didn't let it get to me though. […] I said Green, she heard Grey. I convinced myself the sounds were similar. An honest mix-up. Green. Earl. What's the difference? (COCA:2011:FIC:AmerScholar) (2) 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:SPOK:CBS:Sun_Morning) Whether used as a raised subject construction (with a to-infinitive complement clause), as in (1), or as an impersonal construction (with a that-complement clause) as in (2), turn out conveys mirative overtones of sudden revelation to the addressee as well as unexpectedness or counterexpectation in relation to both the speaker and the addressee, and thus surprise. We could in fact paraphrase examples (1) and (2) as (1′) and (2′) below: (1′) Unexpectedly/surprisingly/contrary to what I expected, the tea was not Green tea. It was Earl Grey. (2′) Unexpectedly/surprisingly/contrary to what one may expect, elephants have an advanced sense of self. 3 Concomitant to its mirative nuances, turn out also expresses evidentiality 1 , by either inference (1″) or hearsay (2″): (1″) I sipped the cooled green tea, which turned out to be Earl Grey. (‘I infer it from sensory evidence and/or from general knowledge’). (2″) It turns out that elephants have an advanced sense of self (‘I’ve been told so’). In (1″), evidentiality is expressed by means of inference, as the speaker identifies the tea as Earl Grey and not as the expected green tea, based on his/her previous experience with and knowledge of tea types, and on sensory evidence; in (2″), on the basis of hearsay, as the speaker reports on someone else’s findings—presumably those of a scientist. Nevertheless, in both (1) and (2) the mirative readings seem to be more salient and to assert themselves more than evidential ones. In this study I aim to account for the origin and development of turn out constructions such as those found in (1) and (2). Thus, I will explore the semantic and syntactic changes undergone by the verb, focusing on the Late Modern English (LModE) period. As will be noted, over the course of the eighteenth century turn out underwent abrupt semantic changes, whereby it acquired evidential and mirative readings, shifting, at the same time, from a resultative (change-of-state) lexical verb to a raising catenative verb (see Quirk et al. 1985: 146-147; Huddleston & Pullum 2002: 1194-1198). This raising predicate gave way to two divergent though related constructions: the raised subject turn out construction (1) and the impersonal turn out construction (2). 1 An anonymous reviewer suggests that turn out in (1) could also express direct (i.e. visual) evidentiality, since the evidence for the tea being Earl Grey could well be a tea bag with the name ‘Earl Grey’ on it. However, if one pays close attention to the context, the speaker actually comes to the realization that the tea is Earl Grey and not green tea after having sipped it, thus smelling and tasting it. The type of evidence provided by the speaker in (1), then, is inferential. Even though the speaker’s evidence relies on sensorial traces of an event, it ultimately implies an inference or deduction on the part of the speaker (see Aikhenvald 2004: 159). 4 Given its diachronic scope, this study takes a corpus-based approach. The historical data used are drawn largely from the Oxford English Dictionary (OED), the Corpus of Late Modern English Texts, version 3.0 (CLMET3.0) and the Old Bailey Corpus (OBC), although other historical sources were also consulted. I have referred to the Corpus of Contemporary American English (COCA) to document present-day usage. The paper is organized as follows: Section 2 provides a brief account of the concepts of evidentiality and mirativity. Section 3 offers an overview of turn out constructions in PDE. Section 4 is the historical study proper, while Section 5 presents a summary and wraps up the discussion of the diachronic changes examined. Finally, in Section 6, I set out some concluding remarks and recommendations for future research. 2. Evidentiality and mirativity Evidentiality and mirativity have attracted a great deal of attention in recent years. Nevertheless, there is still considerable debate on how to define these linguistic categories, how they relate to each other, and indeed which elements can be considered as evidentials and miratives. Hence it is important here to set out some basic notions regarding how evidentiality and mirativity are understood in the current study. Evidentiality—roughly defined as the linguistic encoding of the source of information or knowledge (Chafe & Nichols 1986; Willett 1988; Aikhenvald 2004)—and mirativity—a category that conveys new or unexpected information with overtones of surprise (DeLancey 1997, 2001, 2012; Lazard 1999; Aikhenvald 2012)—are both grammatical categories in many of the world’s languages. According to Aikhenvald (2004: 1), evidentiality is a grammatically encoded category in about a quarter of the languages of the world: in these, it is expressed morphologically by means of affixes, 5 clitics or particles, among others (see Faller’s 2002 account of Quechua evidentials). Grammatical miratives, though less prevalent, are also well attested in a considerable number of languages (Aikhenvald 2012: 473). Even though mirativity has often been accounted for in close relation to evidentiality, even seen as part of it, there is general scholarly agreement that enough evidence exists to consider them independent, though related, linguistic categories (Aikhenvald 2004: Ch. 6 for an overview of the relationship between the two categories, 2012; DeLancey 2012; contra Hill 2012). In general, typologists have argued that languages such as those of the Germanic or Romance branches, which do not express evidentiality and/or mirativity grammatically—i.e., those languages in which these categories are not morphologically encoded as markers—, do not have evidentiality and/or mirativity (Aikhenvald 2004, 2012; DeLancey 2012). Instead, such languages have evidential and mirative communicative strategies that are most often lexically represented. However, this fact does not mean that evidentiality (or mirativity) lacks grammatical status in languages like English or Spanish (Cornillie 2007: 7). Indeed, imposing such clear boundaries between grammar and the lexicon is, at the very least, counterintuitive from a cognitive-functional perspective, and it is for this reason that Diewald and Smirnova (2010b: 1) refer to evidentiality as a “semantic-functional domain”, a broad label that could also be extended to mirativity and which would cover all linguistic expressions of these categories. Evidentiality and mirativity have garnered increasing interest, not only in the form of typological studies such as the ones mentioned above, but also from the perspective of languages in which these categories are not ‘morphologically’ encoded (see, among others, Chafe 1986 for evidentiality in English; Squartini 2001, 2004 for evidentiality in Romance; Cornillie 2007 for evidentiality in Spanish; Diewald & Smirnova 2010a for 6 evidentiality in German, 2010b for evidentiality in the European languages; Olbertz 2009 for mirativity in Andean Spanish; López-Couso & Méndez-Naya 2015 for a comparative perspective on Romance and English evidentials; Gentens et al. 2016 for mirativity in English). Evidential and mirative strategies are found at different linguistic levels, from the prosodic through the morphological to the discourse level. As regards lexis, for instance, adverbial expressions such as English in the end or Spanish finalmente seem to have acquired mirative counterexpectative readings (see González & Maldonado 1998): (3) In any case, she told me that she was looking for a new song, so I wrote the song for her. They didn't use it in the end. (COCA:2011:MAG:HarpersBazaar) (4) Después de hacer varios intentos de viajar a Chile, finalmente pudieron conseguir una oportunidad de trabajo en España ‘After trying to travel to Chile several times, in the end they were able to find a work opportunity in Spain’ (CDE:19-N:Col:Semana:834) In (3) and (4) the resultative meaning encoded in the bolded expressions gives way to a change of state which contravenes the speaker’s expectations, thus conveying unexpectedness and even surprise. The expression of mirativity and evidentiality also takes place in more grammatical contexts. Evidential readings, for instance, have been attributed to Spanish future tense morphology (Rivero 2014). In example (5), one of the girl’s parents remarks that their daughter is not looking well. The future tense used in the other parent’s reply, tendrá (lit. ‘will have’), is dubbed a ‘fake’ future tense, since it has an inferential reading based on inference from direct evidence and does not express future tense. (5) - La niña tiene muy mala cara hoy. 7 the girl have:PRES.3.SG very bad face today 2 ‘Our little girl doesn’t look good today.’ - Tendrá fiebre. have:FUT.3.SG fever ‘She may/might/must have a temperature.’ (Rivero 2014: 197-198) In Ecuadorian Highland Spanish the resultative or experiential perfect aspect (the present perfect) is used to express mirativity (Olbertz 2009, 2012; Hengeveld & Olbertz 2012). It is generally accepted that the mirative was introduced into Andean Spanish through contact with Quechua (Olbertz 2009: 70). In (6), taken from Olbertz (2009: 76), the speaker is recounting an episode in which she suddenly stumbles and reaches for what she thinks is a tree, while walking through the jungle: (6) Me agarré de un árbol, que no REFL.1SG hold:PAST.PF.1SG PREP a tree REL NEG había sido. AUX.PAST.IMPF.3SG COP.RES ‘I held on to a tree, which turned out not to be one.’ The mirative present perfect may even be used for future reference in this variety of Spanish, as in (7): (7) El año que viene ha sido bisiesto. the year REL come:PRES.3SG AUX.PRES.3SG COP.RES leap year 2 Abbreviations for the glosses in examples (5) to (7) are as follows: 3 = 3rd person, AUX = auxiliary, COP = copula, FUT = future, IMPF = imperfective, NEG = negation, PST = past, PF = perfective, PREP = preposition, PRES = present, REFL = reflexive, REL = relative, RES = resultative, SG = singular. 8 ‘(I just realize) next year is a leap year.’ (Hengeveld & Olbertz 2012: 493) These examples show how both evidentiality and mirativity can be grammatically conveyed in languages which have not traditionally been considered to possess these categories. As will be argued in this paper, raising turn out is also used in specific constructions to express mirative and evidential nuances. 3. A synchronic overview of mirative turn out constructions Despite their widespread extension in PDE, turn out mirative constructions have received very little scholarly attention. DeLancey (2012: 533) acknowledges (it) turns out (that) as a “conventionalized” construction that expresses mirativity. Given his focus in that study on grammatical mirativity in Kham (Tibeto-Burman) and Hare (Athabaskan), he does not go into further detail regarding the English construction. In an unpublished paper on raising verb parentheticals, Howe and Heller (2010) also refer to the mirative properties of turn out (“counterexpectation or surprise”, slide 32). Kaltenböck (2015: 124-131) analyzes turn out matrix clause constructions like (2) above—which he names ‘presentational matrix clauses’—as constructions that are undergoing grammaticalization as potential pragmatic markers. However, he does not consider their mirative nuances. Mirative turn out constructions in PDE can be ascribed to three basic types: raised subject constructions, impersonal constructions, and parentheticals. Examples of the last type here can occur either as bare parentheticals (turns out) or as adverbial parentheticals (as it turns out), as in (8): (8) He didn't need the money. As it turned out, special ops talents were wellcompensated these days (COCA:2012:FIC:Bk:LuckyInLove). 9 However, due to space limitations, and given the fact that such parentheticals do not become entrenched until the twentieth century (Serrano-Losada, forthcoming), these constructions are not discussed in depth in the present study. Whether used with a to-infinitive or a that-complement clause, mirative turn out typically introduces a proposition that is argumentatively opposed to an earlier proposition or that challenges or contradicts the speaker’s and/or the addressee’s beliefs or expectations. Thus, turn out encodes unexpectedness and counterexpectation, and hence surprise, both to the speaker and the addressee, and overtones of sudden revelation, especially for the addressee. In what follows, both types of PDE mirative constructions are explored individually. 3.1 Raised subject construction In this type of mirative construction, raising turn out appears in catenative constructions in which it functions in a similar way to a (semi)auxiliary, adding evidential and mirative overtones to the proposition contained in the infinitive clause. (9) I headed into the shower but it turned out to be a window. (COCA:2012:FIC:TheatreForum) (10) Though she was trying hard to keep it together now that her intruder had turned out to be a furry bandit, she was clearly shaken up. (COCA:2012:FIC:Bk:RunFromFear) In both (9) and (10) the speakers’ expectations are challenged by the final outcome. In (9) the speaker intends to walk into the shower, but the shower is ultimately revealed to be a window. In (10) the narrator explains how a woman struggles to keep calm after believing she has a trespasser in her home. Eventually, the woman learns that the intruder is a 16 prototypical reading of location in the sense of being away from a central locus in an outward direction 8 . Such is the case with (19a), where turnde (ut of) has the sense of ‘going away’. Note that turnde ut in (19a) is not yet a phrasal verb. Related to this locative sense is the meaning of exclusion or removal found in (19b): (19) (a) he turnde ut of þe burh into wilderne (HC, c1225) ‘he went out of the city into the wilderness’ (b) Baronius goes on to tell us, that Sozomen affirms, Liberius was turned out of his Church, for not consenting to the Faith at Ariminum. (EEBOCorp1.0, 1695) However, as a verb particle out would also develop a meaning of completion related to reaching an end-point in a process, as in (20): (20) Our expedition up Vesuvius turned out very well (OED, 1830) Some of the most frequent earlier senses of the verb registered in EEBOCorp1.0 and CLMET3.0 include ‘to expel’ (OED s.v. turn, v., turn out, 3.), ‘to dismiss or eject from office or employment’ (OED s.v. turn, v., turn out, 5.) and, by extension, ‘to go away, to depart’ (OED s.v. turn, v., turn out, 15.). According to the OED these meanings came into use in the Early Modern English period. EEBOCorp1.0 yields approximately 4000 hits for phrasal turn out, an overwhelming majority of which correspond to these three senses. However, as (19a) indicates, they were already incipient in Middle English (albeit in a physical sense). Example (19b) reflects the ‘being expelled’ sense. In turn, example (20) has a resultative meaning triggered by the ‘reaching an end point’ sense of out. 4.2 On the path towards mirative turn out: focus on Late Modern English 8 See Tyler and Evans (2003: 200-209) for an analysis of out in terms of cognitive semantics. 17 As exemplified by (20), turn out underwent a further semantic shift over the course of the eighteenth century. The verb developed new meanings expressing evaluative and subjective nuances, bringing about the evidential and mirative senses of the verb. (21) a. ‘to come about in the end; to result, eventuate.’ (OED s.v. turn, v., turn out, 20.a.) b. ‘with complement: to come to be, become ultimately’ (and so be found or known to be). (OED s.v. turn, v., turn out, 20.b.) c. ‘to be ultimately found or known, to prove to be (without implication of becoming).’ (OED s.v. turn, v., turn out, 20.c.) The three OED senses listed under (21) are best accounted for as the result of subjectification, a process which Traugott (2003: 126) defines as: “the mechanism whereby meanings come over time to encode or externalise the SP[eaker]/W[riter]’s perspectives and attitudes as constrained by the communicative world of the speech event, rather than by the so-called ‘real-world’ characteristics of the event or situation referred to.” Subjectification operates by semantically encoding senses which were previously contextual or pragmatic in nature and which reflect a speaker-oriented perspective. The development of both evidential and mirative overtones can be considered as an outcome of a process of subjective perspectivation. In this regard turn out seems to have undergone a process of subjectification whereby change-of-state senses acquired inferential meaning extensions and, eventually, mirative overtones. Broadly speaking, turn out seems to have developed along the lines of the processes adduced for other raising verbs like its Spanish equivalent resultar (Cornillie 2007, 2008), English promise and threaten (Traugott 1997), German drohen ‘threaten’ (Heine & Miyashita 2008), or seem and appear (Gisborne & Holmes 2007) and their 18 Spanish counterpart parecer (Cornillie 2007, 2008). In one way or another, all of these authors posit subjectification as the mechanism at work behind the development of these verbs. Interestingly, while raising seem and appear can be traced back to the Middle English period (Gisborne & Holmes 2007), raising turn out—like coeval promise and threaten—developed much later, as it is not attested in the record until the eighteenth century. In the remainder of this subsection, I explore the different senses enumerated in (21) and the consequences of their emergence. 4.2.1 Sense I The first of these senses (21a) is exemplified in (22) and (23), in which turn out means ‘to come about in the end or issue; to result, eventuate’. First recorded in the 1730s, turn out in this sense is almost synonymous with to turn up: (22) I have been ill and fatigu'd, so can't express myself as I would. 'Tis a base Prosecution, and I believe will turn out so. (OBC, 1739) (23) Who knows what may turn out? (OED, 1786) This meaning reflects a prototypically resultative Aktionsart in which something results in something else. In this sense, the verb may have adverbial evaluative complements (e.g. well, fine). However, it has not yet undergone subjectification. It does not convey evidential nuances, nor does it express mirativity. Moreover, the verb in (22) and (23) is a lexical control verb. 4.2.2 Sense II The next sense, (21b) ‘to come to be, become ultimately (and so be found or known to be)’, is also recorded first in the eighteenth century. While (22) and (23) emphasized time, 19 examples (24) to (26) are figurative in nature. Unlike the former sense, this one expresses change of state, not merely result: (24) The fine gentleman formed upon reading the former [books] will almost always turn out a pedant, and he who forms himself upon the latter [the stage], a coxcomb. (CLMET3.0, 1749) (25) For some time they enjoyed their triumph; but it turned out at last to their unspeakable mortification. (CLMET3.0, 1753) (26) The day has turned out better then I expected it. (OED, 1769) In his Historical Syntax, Visser (1963, §235) includes turn out in cases such as (24)-(26) among the verbs which exhibit copular behavior. He claims that “copular turn out” is a “copula of modality” that has been in use since the middle of the eighteenth century. Indeed, this inchoative use of turn out exhibits a behavior very similar to that of copulas. At this point turn out shows signs of having undergone subjectification, a process most likely induced by the semantic generalization of the verb and which in turn explains its copular status. In their diachronic analysis of raising verbs of appearance, Gisborne and Holmes (2007: 6) claim that verbs like seem “acquire propositional arguments as their meanings undergo semantic bleaching, and this in turn leads to the development of evidential senses.” Their explanation fits well with the present case. In (24)-(26), the speakers convey inference from direct evidence: they have some form of external evidence on the basis of which they infer the propositions expressed. In (25) and (26), for instance, the inferences are drawn from the speakers’ own experiences, that is, they are linked to the results of something already experienced. According to Aikhenvald (2004: 274), verbs which refer to existence may give rise to inferred evidentials. In the case of turn out, inferential evidential overtones are acquired in a process whereby a speaker 20 reinterprets an objective, externally observed change in terms of his own internal perceptual or cognitive evaluation. Thus the speaker’s perspective is expressed in the proposition by means of inferential knowledge. This subjectification pathway, which is in line with Traugott’s (1989) paths of semantic change, is not uncommon (e.g. Lauwers & Duée 2011). 4.2.3 Sense III A third sense arises during the eighteenth century which is linked to the emergence of the raising verb. Turn out begins to appear with an infinitive complement clause, with the sense of ‘to be ultimately found or known, to prove to be (without implication of becoming)’ (21c). (27) After all, I hope the best: but if this should turn out to be a plot, I fear nothing but a miracle can save me. (CLMET3.0, 1740) (28) And though this incident has not turned out to answer all I wished from it, yet is it not necessary, if I would acquaint thee with my whole circulation (CLMET3.0, 1748) Examples (27) and (28) illustrate the emergence of raising turn out. These instances still do not suggest full raising, in that the subjects in both cases continue to hold some thematic relation to turn out. In fact, the data bears witness to the layered development of the raising construction. As Traugott (1997: 188) puts it, far from being a parametric phenomenon, raising represents a continuum from non-control to full control. According to Boye (2010: 95), raising verbs can be seen as the outcome of the reanalysis of the semantic structural properties of control verbs or of control verb constructions. In the case of turn out, reanalysis could have been semantically 21 instantiated, brought about by a process of semantic generalization and subjectification whereby the verb first acquired copular characteristics (21b) and was later reinterpreted as a raising verb (21c). Unfortunately the record does not provide ambiguous instances or bridging contexts to exemplify how this hypothetical reanalysis might have taken place. At this point the resultative and inchoative senses become blurred and give way to a further subjectivized sense. There is no need to have an actual change: turn out is now being used to convey that something has been discovered or proven to be a certain way. The mirative meaning is thus encoded in the construction. As Aikhenvald (2004: 203) claims, evidential strategies may develop mirative extensions. Through a further process of subjectification, the verb comes to express surprise as a result of the sudden or unexpected realization or discovery as regards the proposition. (29) I was robbed in a field between Mary-le-bone turnpike and Kilburn Wells. I had just passed through a gap, when I saw a man who turned out afterwards to have a crape over his face; when I first saw him, from his face appearing black, I took him for a negro. (OBC, 1775) (30) I will call Mrs. Brumidgham, who will tell you positively that they are not the same, and have all been altered; if this turns out to be the case, surely it must make a great impression upon your minds, and it is not a very extraordinary thing that Mr. Snelling is not here (OBC, 1787) Alongside its inferential evidential nuances, turn out in examples (29) and (30) expresses mirative overtones of counterexpectation. The earliest examples in the data seem to indicate that counterexpectation is the most primitive mirative nuance expressed. Other mirative extensions seem to have developed from this earlier counterexpectative value. 22 As to evidentiality, turn out with infinitive complements usually expresses inferential evidentiality. However, in certain contexts it is also used as a reporting or recounting strategy to express hearsay. The literature has observed a tendency for resultative and inchoative senses to develop evidential (Aikhenvald 2004: Ch. 9) and mirative readings (see González & Maldonado 1998). In a recent paper, Gipper (2014) found evidence for the extension from inferential evidentiality to mirativity in Yurakaré, a Bolivian isolate language. According to Heine and Kuteva (2002: 213), some languages have encoded the mirative in the copular system as a distinction that is expressed in the verbal system by means of finite constructions built on copulas. Interestingly, unlike other raising verbs such as seem or happen, which can combine with a wide range of infinitives, turn out basically occurs alongside copular be in the great majority of cases. This could be due to its copular status. A further innovation of LModE mirative turn out constructions involves the emergence of the impersonal turn out construction. Just as turn out was reanalyzed as a raising verb, it also began to appear in non-raised contexts. (31) But after all this mystery, it does not turn out that there is any thing great in this match, but the greatness of the secret. (CLMET3.0, 1743) Example (31), the first impersonal turn out construction attested in CLMET3.0, belongs to a letter written by Walpole. This early instance of the construction features negation in the matrix clause. This is a considerable difference with regard to PDE impersonal turn out constructions, in which negative particles are usually embedded in the that-clause 9 . 9 COCA yields 19 examples with negation in the main clause, all of which, with one exception, have the structure to not turn out that way, e.g. It did not turn out that way. Khan changed titles but kept access to his labs (2004:NEWS:WashPost). 23 Examples (32) and (33) represent more canonical turn out matrix clauses. However, there is still tense agreement and the complementizer that is expressed, features that will become less prominent as the construction grammaticalizes into a parenthetical in PDE (see example (17) above). (32) Lightowler was taken over the way to the Green Man; then I talked a deal to him. Some time after he was sent back to the office, and it turned out that he had signed this confession then. (OBC, 1792) (33) I went in search of some of my property, and it turned out that some of the prosecutor’s property was on the same premises (OBC, 1838) Unlike raised subject constructions, impersonal turn out constructions convey hearsay evidentiality in most cases. This may be explained by the impersonal nature of such constructions. The semanticization of hearsay evidentiality can also be considered an outcome of subjectification. Hearsay or reportative evidentiality involves information that the speaker has acquired from a third party and which is therefore not a matter of his own conclusion or experience. Just as the speaker in Sense II conceptualized change in his own subjective terms, which gave rise to inference, he now chooses to signal hearsay using the same evidential expression: turn out. The data reveal that impersonal constructions are commonly employed in reporting as a means of focalizing and highlighting certain statements. In (32), the speaker uses the turn out construction to introduce information about an event that he did not witness (the signing of the confession). In (33), the speaker learns from hearsay that some of the belongings in the premises were the prosecutor’s. Mirative extensions are also present with impersonal turn out. In both (32) and (33), the speakers introduce and highlight an unexpected element. 24 (34) I said you had got a pair of white stockings, and that the other had garters on. I thought it had been so by the appearance in the dark, but it turned out to be darkcoloured stockings (OBC, 1772) The construction exemplified in (34) is, apparently, a raised subject construction. However, there is a lack of concordance between the singular subject (it) and the plural complement (dark-coloured stockings). Examples such as this are not uncommon, and could be regarded as blends of the raised subject construction and the impersonal construction. 4.3 Mirative turn out in CLMET3.0 This section provides an overview of mirative turn out constructions from their earliest attestations in CLMET3.0. Specifically, it focuses on to-infinitive and that-complement clauses. The corpus, compiled by De Smet et al. (2013), covers the LModE period, comprising texts that span the years 1710-1920. The corpus is divided in 3 sub-periods. Table 1 provides a summary of the total number of words per genre and sub-period. Genre 1710-1780 1780-1850 1850-1920 Total Narrative non-fiction 4,642,670 4,830,718 6,311,301 15,784,689 Narrative fiction 1,863,855 1,940,245 958,410 4,762,510 Drama 407,885 347,493 607,401 1,362,779 Letters 1,016,745 714,343 479,724 2,210,812 Treatise 1,114,521 1,692,992 1,782,124 4,589,637 Other10 1,434,755 1,759,796 2,481,247 5,675,798 10,480,431 11,285,587 12,620,207 34,386,225 10 The category ‘Other’ includes mainly (auto)biographies, travel literature, historical and fictional treatises and other unclassified texts (37 texts in total). 25 Table 1. Total number of words per genre and subperiod All in all, I retrieved a total of 1173 tokens of turn out, 48 (1.40 per million words) of which were followed by a that-clause and 210 by a to-infinitive clause. All instances were selected manually, removing from the concordance those examples which did not fit the patterns. From the 210 instances for turn out followed by an infinitive, only 191 (5.5 per million words) corresponded to the mirative construction. Figure 1. Mirative turn out with that and to-infinitive clauses in CLMET3.0 (per million words) Figure 1 represents the diachronic distribution of both turn out constructions in CLMET3.0. Their steady growth reflects gradual entrenchment of both constructions in the language. However, infinitival constructions are better represented in the corpus data. The fact that that-complement constructions are less frequent in CLMET3.0 may be explained by the fact that these were more readily found in spoken language, as is still the case in PDE 11 . 11 In COCA, the search string it turns out that yields 7.15 tokens (per million words) in the spoken component, against 4.86 tokens in magazines, 2.43 in academic writing, 1.98 in newspapers and 1.59 in fiction. 0 2 4 6 8 10 12 to-infinitive that-clause 32 However, in view of the evidence, the emergence of evidential and mirative turn out during the eighteenth century might in fact be the result of a more abrupt change in the sense of Petré (2012), who explains the sudden copularization of Old English becuman and weaxan in the twelfth and thirteen centuries as a consequence of analogical modeling and not as the result of a gradual grammaticalization process. According to Petré (2012: 28), the processes whereby these new copulas emerge and become entrenched, are instances of sudden categorial incursion, which De Smet (2009: 1748) defines as a “non-gradual”, analogy-driven mechanism in which a construction is recategorized as a member of a pre-existing category, that is, an “analogical extension of one construction into the domain of another”. To posit such an abrupt, analogy-driven change for turn out makes sense given that the (pseudo)copularization of the verb triggered the appearance of evaluative complements, which in turn allowed for incipient evidential interpretations, bringing about the possibility of analogical reasoning. It is thus possible that the abrupt emergence of raising turn out was facilitated through analogical modeling following preexisting members of the evidential (seem, appear) and mirative (happen, prove) verb paradigm, but also other emerging raising predicates that grammaticalized evidential and mirative nuances around the same period, including promise and threaten, which acquired raising properties and developed into evidential predicates during the LModE period. 6. Concluding remarks The present paper has dealt with the emergence of evidential and mirative turn out constructions in LModE. It is intended to contribute to the already large body of work on evidentiality and, to a lesser extent, mirativity in those languages that have traditionally 33 been described as lacking evidential or mirative systems. The English language does not possess evidentiality and mirativity as grammatical categories in the sense that they are not morphologically encoded; that is, these categories cannot be compartmentalized in a paradigm of obligatory morphological affixes like tense or number. However, they do cover grammatical functions in a broader sense, as they are expressed by means of specific constructions in which they fulfill very precise functions. Raising turn out and its evidential and mirative readings are an eighteenth-century innovation. A semantic drift in turn out triggered the emergence of the subjective readings which gave rise to the evidential and mirative senses. Likewise, these semantic changes allowed syntactic innovations to take place, which facilitated the emergence of turn out as a copula and as a raising verb. However, the data suggest that an abrupt, analogy-driven change must have taken place, given the dates in which the different senses and constructions are first documented. Nevertheless, further research is needed in order to arrive at a more complete characterization of turn out and the encoding of evidential and mirative strategies, in particular regarding the verb’s complementation patterns, which would allow for a better understanding of the different changes at work. Moreover, the role of analogy remains to be more fully explored. Further analysis of the developmental pathways of similar mirative constructions with verbs such as prove, happen and end up would also shed light on the nature of raising and its relation to evidentiality and mirativity. Data sources 34 CLMET3.0 = The Corpus of Late Modern English Texts, version 3.0. Hendrik De Smet, Hans-Jürgen Diller & Jukka Tyrkkö. 2013. Available at: https://perswww.kuleuven.be/~u0044428/clmet3_0.htm. COCA = The Corpus of Contemporary American English. Mark Davies. 2008-. Available at: http://corpus.byu.edu/coca/. EEBOCorp1.0 = Early English Books Online Corpus 1.0. Peter Petré. 2013. Available at: https://lirias.kuleuven.be/handle/123456789/416330. HC = The Helsinki Corpus of English Texts. Matti Rissanen, Merja Kytö, Leena KahlasTarkka, Matti Kilpiö, Saara Nevanlinna, Irma Taavitsainen, Terttu Nevalainen and Helena Raumolin-Brunberg. 1991. OBC = The Old Bailey. Corpus Spoken English in the 18th and 19th centuries. Magnus Huber, Magnus Nissel, Patrick Maiwald, & Bianca Widlitzki. 2012. Available at: www.uni-giessen.de/oldbaileycorpus. OED = Oxford English Dictionary Online. Oxford University Press. Available online at: http://www.oed.com/. References Aikhenvald, A. Y. (2004). Evidentiality. Oxford: Oxford University Press. Aikhenvald, A. Y. (2012). The essence of mirativity. Linguistic Typology 16(3). 435-85. Boye, K. (2010). Raising verbs and auxiliaries in a functional theory of grammatical status. In K. Boye & E. Engberg-Pedersen (Eds.), Language usage and language structure (pp. 73-104). Berlin/New York: Mouton de Gruyter. Boye, K., & Harder, P. (2007). Complement-taking predicates: Usage and linguistic structure. Studies in Language 31(3), 569-606. 35 Brinton, L. (2008). The Comment Clause in English. Cambridge: Cambridge University Press. Chafe, W. 1986. Evidentiality in English conversation and academic writing. In W. Chafe & J. Nichols (Eds.), 261-272. Chafe, W., & Nichols, J. (Eds.). (1986). Evidentiality: The linguistic coding of epistemology. Norwood, NJ: Ablex. Cornillie, B. (2007). Evidentiality and epistemic modality in Spanish (semi)auxiliaries: a cognitive-functional approach. Berlin/New York: Mouton de Gruyter. Cornillie, B. (2008). On the grammaticalization and (inter)subjectivity of evidential (semi)auxiliaries in Spanish. In E. Seoane & M. J. López-Couso (Eds.), Theoretical and empirical issues in grammaticalization (pp. 55-76). Amsterdam/Philadelphia: John Benjamins. De Smet, H. (2009). Analyzing reanalysis. Lingua 119(11), 1728-1755. De Smet, H. (2010). Grammatical interference: Subject marker for and the phrasal verb particles out and forth. In E. C. Traugott & G. Trousdale (Eds.), Gradience, gradualness and grammaticalization (pp. 75-104). Amsterdam/Philadelphia: John Benjamins. DeLancey, S. (1997). Mirativity: The grammatical marking of unexpected information. Linguistic Typology 1(1), 33-52. DeLancey, S. (2001). The mirative and evidentiality. Journal of Pragmatics 33(3), 369382. DeLancey, S. (2012). Still mirative after all these years. Linguistic Typology 16(3), 529564. 36 Diewald, G., & Smirnova, E. (Eds). (2010a). Evidentiality in German: linguistic realization and regularities in grammaticalization. Berlin/New York: Mouton de Gruyter. Diewald, G., & Smirnova, E. (Eds). (2010b). Linguistic realization of evidentiality in European languages. Berlin/New York: Mouton de Gruyter. Faller, M. (2002). Semantics and pragmatics of evidentials in Cuzco Quechua. Standford, CA: Stanford University dissertation. Gentens, C., D. Kimps, K. Davidse, G. Jacobs, A. Van Linden & L. Brems. (2016). Mirativity and Rhetorical Structure. In G. Kaltenböck, E. Keizer & A. Lohmann (Eds.), Outside the Clause: Form and Function of Extra-Clausal Constituents (pp. 125-56). Amsterdam/Philadelphia: John Benjamins. Gipper, S. (2014). From inferential to mirative: An interaction-based account of an emerging semantic extension. In E. Coussé & F. von Mengden (Eds.), Usagebased approaches to language change (pp. 83-116). Amsterdam/Philadelphia: John Benjamins. Gisborne, N., & Holmes, J. (2007). A history of English evidential verbs of appearance. English Language and Linguistics 11(1), 1-29. González Fernández, M. J., & Maldonado, R. (1998). La perfectividad como fuente de contraexpectativas: Resulta que ‘x’ finalmente ‘y’. In Memorias del IV encuentro de internacional de lingüística en el Noroeste (Vol. 2) (pp. 61-82). Hermosillo: Universidad Autónoma de Sonora. Heine, B., & Kuteva, T. (2002). World Lexicon of Grammaticalization. Cambridge: Cambridge University Press. 37 Heine, B., & Miyashita, H. (2008). Accounting for a functional category: German drohen ‘to threaten’. Language Sciences 30(1), 53-101. Hengeveld, K., & Olbertz, H. (2012). Didn't you know? Mirativity does exist! Linguistic Typology 16(3), 487-503. Hill, N. (2012). “Mirativity” does not exist: ḥdug in “Lhasa” Tibetan and other suspects. Linguistic Typology 16(3), 389-433. Howe, C., & Heller, B. (2010). Turns out they weren’t much of a stretch: Variable patterns of structural persistence. Paper presented at 39th New Ways of Analyzing Variation (NWAV 39), University of Texas at San Antonio, 4-6 November. Huddleston, R., & Pullum, G. K. (2002). The Cambridge Grammar of the English Language. Cambridge: Cambridge University Press. Kaltenböck, G. (2013). The development of comment clauses. In B. Aarts, J. Close, G. Leech & S. Wallis (Eds.), The Verb Phrase in English (pp. 286-317). Cambridge: Cambridge University Press. Kaltenböck, G. (2015). Processibility. In K. Aijmer & C. Rühlemann (Eds.), Corpus Pragmatics (pp. 117-142). Cambridge: Cambridge University Press. Lauwers, P., & Duée, C. (2011). From aspect to evidentiality: The subjectification path of the French semi-copula se faire and its Spanish cognate hacerse. Journal of Pragmatics 43(4), 1042-1060. Lazard, G. (1999). Mirativity, evidentiality, mediativity, or other? Linguistic Typology 3(1), 91-109. López-Couso, M. J., & Méndez-Naya, B. (2014). Epistemic parentheticals with seem: Late Modern English in focus. In M. Hundt (Ed.). The syntax of Late Modern English (pp. 291-308). Cambridge: Cambridge University Press. 38 López-Couso, M. J., & Méndez-Naya, B. (2015). Evidential/epistemic markers of the type verb + complementizer: Some parallels from English and Romance. In A. D. M. Smith, G. Trousdale & R. Waltereit (Eds.), New Directions in Grammaticalization Research (pp. 93-120). Amsterdam/Philadelphia: John Benjamins. Olbertz, H. (2009). Mirativity and exclamatives in functional discourse grammar: Evidence from Spanish. In E. Keizer & G. Wanders (Eds.), The London papers I [Special issue]. Web Papers in Functional Discourse Grammar 82, 66-82. http://home.hum.uva.nl/fdg/working_papers/WP-FDG-82_Olbertz.pdf (January 30, 2016) Olbertz, H. (2012). The place of exclamatives and miratives in grammar: a functional discourse grammar view. Linguística 8(1), 76-98. Petré, P. (2012). General productivity: How become waxed and wax became a copula. Cognitive Linguistics 23(1), 27-65. Quirk, R., Greenbaum, S., Leech, G., & Swartik, J. (1985). A comprehensive grammar of the English language. London: Longmans. Rivero, M. L. (2014). Spanish inferential and mirative futures and conditionals: An evidential gradable modal proposal. Lingua 151(B), 197-215. Serrano-Losada, M. (forthcoming). On English turn out and Spanish resultar mirative constructions: A case of ongoing grammaticalization? Squartini, M. (2001). The internal structure of evidentiality. Studies in Language 25(2), 297-334. Squartini, M. (2004). Disentangling evidentiality and epistemic modality in Romance. Lingua 114(7), 873-895. 39 Traugott, E. C. (1989). On the rise of epistemic meanings in English: An example of subjectification in semantic change. Language 65(1), 31-55. Traugott, E. C. (1997). Subjectification and the development of epistemic meaning: The case of promise and threaten. In T. Swan & O. J. Westvik (Eds.), Modality in Germanic languages: Historical and comparative perspectives (pp. 185-210). Berlin/New York: Mouton de Gruyter. Traugott, E. C. (2003). From subjectification to intersubjectification. In R. Hickey (Ed.), Motives for Language Change (pp. 124-139). Cambridge: Cambridge University Press. Tyler, A., & Evans, V. (2003). The semantics of English prepositions: Spatial scenes, embodied meaning and cognition. Cambridge: Cambridge University Press. Van Bogaert, J. (2011). I think and other complement-taking mental predicates: A case of and for constructional grammaticalization. Linguistics 49(2), 295-332. Visser, F. Th. (1963). An historical syntax of the English language: Syntactical units with one verb (Vol. 1). Leiden: Brill. Willett, T. (1988). A cross-linguistic survey of the grammaticization of evidentiality. Studies in Language 12(1), 51-97. Author’s address Mario Serrano-Losada Department of English and German Faculty of Philology University of Santiago de Compostela 40 15782 Santiago de Compostela Spain [email protected]s About the author [About the author]