Not always a stage. A typical patterns in Spanish copular clauses

an International Journal of Theoretical Linguistics Ángel L. Jiménez-Fernández Editor-in-Chief Universidad de Sevilla Pilar Barbosa Universidade do Minho Victoria Camacho-Taboada Universidad de Sevilla Michelle Sheehan University of Cambridge Volume 5 Issue 2, December 2013 ISSN: 1989-8525 Editorial Team Editors Ángel L. Jiménez-Fernández, Editor-in-Chief, Universidad de Sevilla, Spain Pilar Barbosa, Universidade do Minho, Portugal Victoria Camacho-Taboada, Universidad de Sevilla, Spain Michelle Sheehan, University of Cambridge, UK Layout editor Pedro J. Carrillo-Gomez, Chief Layout Editor, Universidad de Sevilla, Spain Victoria Camacho-Taboada, Universidad de Sevilla, Spain Alicia González Seco, Universidad de Sevilla, Spain Estefanía Guerrero Robles, Universidad de Sevilla, Spain Copy editors Megan Devlin, University of Ulster, UK Rebecca Woods, University of York, UK Proofreaders Victoria Camacho-Taboada, Universidad de Sevilla, Spain Pedro J. Carrillo-Gomez, Universidad de Sevilla, Spain Principal Contact Ángel L. Jiménez-Fernández, Phd Phone: (+34) 954 551 546 Fax: (+34) 954 551 516 Email: [email protected] Support Contact Pedro J. Carrillo-Gomez Phone (+34) 620 581 485 pe[email protected] Mailing Address Facultad de Filología Universidad de Sevilla C/ Palos de la Fra. s/n 41004 Sevilla (Spain) Volume 5 Issue 2, December 2013 ISSN: 1989-8525 Scienti�c Committee Caroline Heycock, University of Edinburgh, UK Anders Holmberg, University of Newcastle, UK Selçuk Issever, Ankara University, Turkey Mary Kato, State University of Campinhas, Brazil Michael Kenstowicz, MIT, USA Itziar Laka, University of the Basque Country, Spain Alazne Landa, University of the Basque Country, Spain Mark Jary, University of Roehampton, UK Maria-Rosa Lloret, Universitat de Barcelona, Spain, Spain Manuel Leonetti, Universidad de Alcalá, Spain Ángel J. Gallego, Centre de Lingüística Teòrica & Universitat Autònoma de Barcelona, Spain Mireia Llinas, Universitat Autònoma de Barcelona, Spain Giuseppe Longobardi, University of Trieste, Italy Ana Maria Martins, University of Lisbon, Portugal Jürgen M. Meisel, University of Hamburg, Germany and University of Calgary, Canada Amaya Mendikoetxea, Autonomous University of Madrid, Spain Shigeru Miyagawa, MIT, USA Andrew Nevins, Harvard University, USA Jairo Nunes, University of São Paulo, Brazil Ana Ojea, University of Oviedo, Spain Francisco Ordoñez, Stony Brook University, USA Javier Ormazabal, University of the Basque Country, Spain Sumru Ozsoy, Boğaziçi University, Turkey Orin Percus, University of Nantes, France Josep Quer, Institució Catalana de Recerca i Estudis Avançats & University of Pompeu i Fabra, Spain Andrew Radford, University of Essex, UK Eduardo Raposo, University of California at Santa Barbara, USA Gemma Rigau, Universitat Autònoma de Barcelona, Spain Luiggi Rizzi, University of Siena, Italy Ian Roberts, University of Cambridge, UK Juan Romero, University of Extremadura, Spain Alain Rouveret, University Paris 7, France Vieri Samek-Lodovici, University College London, UK Uli Sauerland, Centre for General Linguistics, Typology and Universals Research (ZAS), Germany Vassilis Spyropoulos, University of Athens, Greece Esther Torrego, University of Massachusetts, USA Ana Maria Brito, University of Porto, Portugal Ignacio Bosque, Universidad Complutense de Madrid, Spain Zeljko Boskovic, University of Connecticut, USA Jonathan Bobaljik, University of Connecticut, USA Elena Benedicto, Purdue University, USA Adriana Belletti, University of Siena, Italy Artemis Alexiadou, University of Stuttgart, Germany Gorka Elordieta, University of the Basque Country, Spain Utpal Lahiri, University of the Basque Country, Spain Gabriela Matos, University of Lisbon, Portugal Ines Duarte, University of Lisbon, Portugal Maria Joao dos Reis de Freitas, University of Lisbon, Portugal David Adger, University of London, UK Teresa Cabre, Autonomous University of Barcelona, Spain Hector Campos, Georgetown University, USA Anna Cardinaletti, University of Venice, Italy Juan Uriagereka, University of Maryland, USA Myriam Uribe-Etxebarria, University of the Basque Country, Spain Susanne Wurmbrand, University of Connecticut, USA Francisco Garrudo, University of Seville, Spain Maria del Pilar Garcia-Mayo, University of the Basque Country, Spain Teresa Fanego, University of Santiago de Compostela, Spain María Victoria Escandell, UNED, Spain David Embick, University of Pennsylvania, USA Violeta Demonte, Autonomous University of Madrid, Spain Hamida Demirdache, University of Nantes, France Paola Crisma, University of Trieste, Italy Joao Costa, New University of Lisbon, Portugal Guglielmo Cinque, University of Venice, Italy Cedric Boeckx, Institució Catalana de Recerca i Estudis Avançats & Universitat Autònoma de Barcelona, Spain Olga Fernández-Soriano, Autonomous University of Madrid, Spain Luis Eguren, Autonomous University of Madrid, Spain Víctor Acedo, Universitat Autònoma de Barcelona, Spain Volume 5 Issue 2, December 2013 ISSN: 1989-8525 an International Journal of Theoretical Linguistics Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori Consejo Nacional de Investigaciones Cientí�cas y Técnicas. CONICET- Argentina [email protected] Volume 5.2, 2013 1-37 ISSN: 1989-8525 doi:tba Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori Consejo Nacional de Investigaciones Cientí�cas y Técnicas CONICET- Argentina [email protected] Abstract The present paper focuses on cases that challenge the classical de�nition of the Spanish copula estar as stage-level, temporary or unstable predication by patterning with individual-level predicates (associated, by de�nition, with ser, rather than with estar). More importantly, data also indicates that the choice for estar over ser is not semantically nor syntactically trivial, even in those contexts where similar aspectual implications are involved (i.e., even in the delivery of IL predicates). Accordingly, we aim to show that (i) the distinctive semantic properties of estar occurrences follow from its conceptual construal as a location (in either concrete or abstract space); and that (ii) a di�erent implementation of an implied comparison approach to the IL/SL distinction (cf. Franco & Steinmetz 1986) could succeed in capturing the two kind of predications rendered by estar in way that is more in tune with its primary (locative) semantic properties. Moreover, we will claim that a view on semantic content sensitive to cognitive operations available for locative predicates (e.g., perspectival location) may correctly account for di�erent facets of meaning classically ascribed to this copula (e.g., contrastiveness, subjectivity) as well as for the semantic and syntactic patterns restraining the selection of both the DP subject and the adjectival predicate. Keywords: IL-Predication, Locative Semantics, Abstract Location, Stage Level Predicate, Copular Sentences. 1 Introduction In previous work (Mangialavori Rasia 2011, 2013a, 2013b, 2013c) we have shown that the distribution of estar, as well as numerous semantic and syntactic de�ning properties of this copula, can be predicted and explained on the basis of its locative semantic content, as long as an abstract reading of the spatial relation conveyed the copula is consented. On this account, and against the largely defended split analysis on estar supporting a di�erentiated analysis of attributive and locative clauses, we provided empirical data —accompanied by the corresponding formal appreciations— disclosing a structural analogy between the these con- Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 1 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori structions both at semantic and syntactic levels1which, moreover, explains the selectional restrictions that do not follow from estar’s aspectual properties (i.e., incompatibility with Ns). In addition to the signi�cant homomorphism between semantic and syntactic structure2, we entertained the proposal that this simpli�cation can also be supported with regard to conceptual structure. Thus, by regarding states as abstract places (in the sense of Jackendo�1983, 1990) and following the claim that the mind does not manufacture abstract concepts but rather adapts machinery already available (cf. Jackendo�1983: 189); we propose that both locative and attribute constructions yielded by estar stem from a single conceptual (locative) structure, thus making them fall nicely inside a general theory grounded on estar’s spatial properties. In this article, we will set out to explore new data sets which remain elusive to mainstream descriptions and generalizations about the semantic (aspectual) entailments of this copula, at least as far as our knowledge goes. In particular, we will address occurrences as those presented in (1) which —as evidence to be introduced 1Which we argue to be delivered by a same verb (a copula); and, more importantly, showing that the structural analogy can be extended to the alternative heads combined with it (by regarding PAs and PPs/AdvPs as structurally alike, following the claim raised by Hale & Keyser (2002), Mateu (2002) i.a.). 2Within the body of work alluded, which exploits the locative content of estar, some of the contributions we made converge in the postulation of a �ner structure in which the copula is not seen as a primitive, simple verb. Instead, it is seen as a complex predicate resulting from the (prelexical) combination of a P projection [place, schematically represented by at] selected by a stative verbal head (a pure copula [BE]). In this structure, the complement of the locative copula can be alternatively realized either as a PP or as an AP, according to the degree of abstraction of the place conveyed (i.e., the situation in which the subject is located), along similar lines to those to be presented here with regard to (27)-(28). Thus, we assume estar’s locative content to be syntactically (though prelexically) assembled. On this account, we leave clear that we part ways with Gallego and Uriagereka’s (2009) de�nition (estar=ser+P), as we believe it is a primitive BE head (a pure copula), combined with a locative P, that gives rise to estar and not ser. Among other things, because ser features semantic [speci�cally, aspectual] implications which are not found in estar. Among other consequences, this allows us to consider the two main guises in which estar clauses come —traditionally regarded as the manifestation of two (sometimes semantically, sometimes syntactically) di�erent verbs: the copula yielding attributive constructions (e.g.Juan está feliz) and the intransitive verb delivering locative predications (e.g.Juan está en su casa)— in the light of one common lexical-syntactic structure (same verb [copula] and structurally similar predicates. Besides, from this perspective, patterns and restrictions shown also by attributive constructions are seen to follow naturally from the locative content of the verb, which remains anyhow consistent through in the di�erent constructions, thus providing also a signi�cant simpli�cation. Third, as these �ndings comprehend locative constructions, which have been traditionally excluded from the analysis of estar as a copula, important facts emerge accounting for the distribution of estar in the two complementary alternations in which it is engaged (ser/estar and estar/haber), touching on both grammatical and cognitively-rooted conditions (which eventually become grammatically relevant). Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 2 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori next will show— are not necessarily captured by parameters often involved in the analysis of estar, such as stage-level [SL] predicate (cf. also Fábregas 2012:22), briefness, perfectiveness, telicity, inchoativity, resultativity, changing state, nonstability, transiency, etc. (1) a. El guiso está fenomenal ‘The stew is amazing’ b. Esta muchacha está muy linda ‘This girl is very pretty’ c. Ese programa está muy interesante ‘That show is very interesting’ Yet, and on the other hand, we will also focus on the fact that, even when featuring similar aspectual implications, these constructions are not equivalent, nor similar, to those delivered by ser. Rather, the choice of estar is related to a peculiar semantic �avor. For instance, the viewpoint of the speaker seems peculiarly engaged: indeed, the property conveyed through estar is traditionally described as subjective (cf. Fernández Leborans 1999:2928 and references therein, Gili Gaya 1943§46). Moreover, its use is restricted to certain adjectives in a fashion that does not follow from its aspectual de�nition as SL, temporally bounded or transitory predication. Furthermore, di�erent truth conditions and implications are involved. Against this background, we will put forward the idea that —apart from the di�erences regarding the temporal spread of the states denoted— the cases in (1) may be also derived from the original spatial meaning of estar, which remains, according to our proposal, somehow constant across its di�erent occurrences, allowing it to deliver topographic as well as abstract situations (i.e., states of a�airs), on the basis of a similar semantic construal and involving similar conceptual functions (in the sense of Jackendo�2002). Furthering the proposal developed in Mangialavori (2011, 2013a, 2013c3, we will suggest that estar constructions convey the location, at di�erent degrees of abstraction, of an entity; and that the contrast between occurrences like (1) and the prototypical semantic (aspectual) implications of estar can be related to the kind of comparison implied, following basic cognitive tenets. Additionally, this view makes the right predictions as to the restrictions imposed on both the DP subject and the AP predicate (not triggered by ser), thus suggesting an interesting relation between facts touching on cognitive levels and morphosyntactically-instantiated patterns. 3Of course, we are not unaware that the semantic entailment of estar clauses has been related to its locative content in other studies, such as Marín (2010) and Gallego & Uriagereka (2009); and that an integrative analysis of estar has been suggested since Demonte (1979), Suñer (1990:99), Brucart (2010) among many others. Still, the approach pursued here di�ers in several respects (cf. n. 3). Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 3 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori The paper is structured as follows: after this brief introduction, in section 2 and 3 empirical tests show how data under consideration challenge mainstream semantic (more speci�cally, aspectual) tenets associated with Spanish copulas (with special focus on the IL/SL approach), as well as classical generalizations about their mutual implication. Section 4 introduces an alternative analysis of the IL/SL distinction based on implied comparisons, whereas in section 5 we put forward a di�erent implementation of this proposal: in particular, one which is more in tune with of estar’s locative meaning and following cognitive tenets associated with locative predicates. In the meanwhile, section 6 addresses the restrictions observed on DPs and APs combined with estar. The general proposal is furthered (and summarized) in section 7 by distinguishing two di�erent scenarios accounting for ILP and SLP readings. Conclusions are o�ered in section 8. Let us start with a brief review of mainstream descriptions of estar’s semantic entailments. 2 Parameters classically associated with estar Even when we are not unaware of the fact that we can hardly do justice to the complexity and subtlety of the debate within the limits of this presentation, in very general and simpli�ed terms, we could say that, given minimal pairs like (2), the combination with ser (2a) is traditionally seen to imply that a given property (Radha’s blondness) extends through an unspeci�ed time spread, probably running along the subject’s lifetime. By contrast, the use of estar in (2b) renders the entailment that this property will hold for a relatively short time span; or, at least, for a shorter extent than the one covered in (2a). In few words —and abstracting away from �ner-grained distinctions— the general wisdom concerning Spanish copulas is, anyhow, laid on the idea succinctly summarized in (3). (2) a. Radha es rubia ‘Radha is blond’ b. Radha está rubia ‘Radha is blond [now]’ (3) Ser describes a state that is permanent or more prolonged in time or stable than the one expressed by estar, which is mostly expected to be limited, transitory or to change at some point. Departing from this basic insight, a huge range of proposals and parameters have been postulated with a view to capturing and/or predicting the semantic underpinnings and distribution of Spanish copulas, along with (in most cases) their aspectual implications. Among them, one of the notions most recurrently engaged in the analysis of Spanish copulas is the division between individual-level [ILP] and stage-level predicates [SLP], respectively—following Carlson’s (1977), in�uential work—especially as Spanish has been argued to lexicalize this partition through its pair of altern- Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 4 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori ating copulas (cf. Bosque 1993 i.a.). In fact, ILPs and SLPs have been associated with ser and estar respectively in a great many works (Bosque 1993, Demonte 1999, Leonetti 1994, Fernandez Leborans 1999, Arche 2006 i.a.) following the claim —roughly simpli�ed here— that states delivered by ser apply to individuals, whereas estar denotes states applying to stages.4 Although not free of controversy, this distinction nevertheless conforms to the standard temporal implications associated to each copula (3). Indeed, predicates denoting properties attributed to an individual or a class of individuals are eventually interpreted as more lasting or holding for a rather extended period of time —if not through the whole lifespan of the subject (more on Lifetime e�ect later—; whereas predicates applying to a stage are de�ned by denoting temporary (Kratzer 1995) properties or traits holding within a phase or period of the lifespan of the individual (i.e., transitory situations/states in which the entity can be found). Therefore, according to the essential tenets of this criteria, and also taking into account the successive readjustments postulated in order to grasp the semantic underpinning of Spanish copulas more accurately (cf. Escandell & Leonetti 2002, Arche 2006, Marín 2010), sentences like (2) are usually argued to denote (a)an individual property which, is in some sense, possessed by the individual (cf. Arche 2006:16); or, rather, (b) a relatively stable property which de�nes the individual by belonging to a class; the one outlined by people whose hair is of a pale yellow[ish] nature (cf. Leonetti 1994:199).5 By contrast, as the blondness expressed by (2b) is seen as applying to temporally delimited stage, it is usually characterized as an unstable (cf. Escandell & Leonetti 2002) or non-permanent quality (Camacho 2010), likely to change or resulting from a change. In other words, the application of the ILP/SLP distinction on Spanish copulas leads to the claim that when ser is involved the property denoted by the AP “is predicated of the individual as such”. By this, it assumed that the speaker claims that the subject is a yellow-haired person (if the terms exists), whereas through estar the speaker “predicates the properties of the subject on a particular occasion, linked to external reasons (Arche 2006:2; emphasis added). In fact, such a link is relevant in that it would explain the widely known fact that an utterance like Radha está rubia can be argued to imply the idea of a contingency, transiency or contrast with a usual situation (i.e., that the girl had her hair bleached and that she is, say, a brunette). To be fair, estar has been extensively related to the restriction to a speci�c situation (since Carlson 1977:128) or con- 4As temporally bounded manifestations of the individual, according to Carlson’s (1977: 115). Regarding the applicability of this de�nition on Spanish data (cf. Leonetti 1994:184). 5Note that even if we de�ne ILPs as classi�catory instead of as permanent properties, it is also acknowledged that the nature of the concepts we conceive of as classi�catory is such that the ascription of an individual to a class tends to produce a lasting result (cf. Escandell & Leonetti 2002:160), and, thus, the de�nition eventually matches mainstream descriptions of ser. Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 5 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori (21) a. *? El guiso estuvo delicioso en pocos segundos ‘The stew was delicious a few seconds’ b. El guiso estuvo caliente en pocos segundos ‘The stew was hot in a few seconds’ Further aspectual parameters heavily used in estar’s characterization, such as inchoativity (Camacho 2010, Zagona 2009 i.a.) do not seem appropriate either, since this verb delivers a number of constructions which can hardly be considered inchoative. On these grounds, the occurrences under consideration here would add up to such (challenging) sets of data, since cases like (20b) clearly do not convey the idea that the stew became delicious at a given point in time.18 In practical terms, these instances of estar do not seem to match its description as transitory (Pustet 2003, Maienborn 2005), changeable or situation-dependant (Escandell & Leonetti 2002), unstable (Givón 1984), or modi�cable (Gili Gaya 1943) states. On similar grounds, these cases do not seem to pattern with the de�nition of estar as contrasting topic situations associated with the same subject referent. Such a claim —raised following Clements’ 1998 postulation of the [Nexus] feature— is taken up among others by Maienborn 2005 in the study of estar. According to this proposal, estar represents the discourse-dependent variant of ser, whose use is licensed through some type of topic situation [*s in (23)] contrasting with other topic situations where the predicate did (or will) not apply to the (same) subject referent, as ((23)b) sketchily illustrates. In turn, the lack of topic situation contrast (23a) would accommodate ser occurrences (Maienborn 2005:172). (22) a. Mónica es impaciente Qs* [z | ⌧(s*) ⇢⌧(z), Q(z)] ‘Monica is an impatient person.’ b. Mónica está impaciente Qs* [z | ⌧(z) ⇢⌧(z), Q(s*)] ‘Monica is feeling impatient’. (Maienborn 2005:169) With that said, we cannot help noting that, in our examples, the property denoted by the predicate does not necessarily "contrast with previous or later topic situations in which the predicate does not apply to the subject referent", according to Maienborn’s (2005:172) de�nition of estar. In fact, if we applied the di�erentiation put forward by Husband (2010), then it could be argued that these cases indeed convey properties instead of states. In practical terms, Este guiso está delicioso is not necessarily read as a topic situation contrasting with other topic situations in which the state delicioso [tasty] did not apply to the same referent (the erties: besides the end-point test, it also shows no subinterval property among other telicity-related characteristics. Cf. n. 15. 18On this account, these cases may challenge also Camacho’s (2010:15) claim that a progressive aspect projection is selected by estar, involving a beginning boundary, encoded by an [INCH] feature. Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 12 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori very same stew). For that purpose, if a contrast were at stake, it would be in any case a contrast between the property ascribed to a token in contrast to a kind (the average stew), rather than a contrast between a speci�c topic situation/state of the (same) token-stew and a di�erent one where delicioso does not hold of it —which could be the case in an environment triggering a SL reading, like El guiso está delicioso hoy, mañana estará ácido, where there is indeed a contrast between two topic situations of the same token. Hence, under regular conditions, El guiso está delicioso corresponds to a case in which a token-stew is being described individually, the AP predicate conveys a property rather than a state, and it would imply in any case a comparison held against a (contrasting) individual, rather than against a (contrasting) stage or situation applying to the same referent. Apart from this, if we still were to follow through on Maienborn’s analysis, then it may also be noted that in our cases the topic time would fall completely within the K-state time, and thus the situation would (once more) match ser’s de�nition rather than estar’s.19 On the other hand, given the arguments (and empirical data) o�ered above, the characterization as a speci�c state anchored (in Maienborn’s terms) on a given context/situation becomes troublesome for other reasons. Actually, restriction to a particular topic situation20 (23b) is not free of controversy either as these expressions are usually uttered the �rst time that the speaker comes to perceive the entity and, therefore, a Nexus with a prior state is not likely to be established (cf. Escandell & Leonetti 2002, Roby 2009). In other words, here the speaker has no mental point of reference to compare against how Radha looked or the stew tasted 19According to Maienborn’s model, the selection of ser determines the imperfectivity of the clause, as the topic time falls within the K-state (de�ned as abstract objects for the exempli�cation of a property P for a holder x at at time t, Maienborn 2005:168) time, and, therefore, Monica in (23a) is described as impatient independent of any speci�c context or time gap. By contrast, the perfectivity of typical estar cases like (23b) is (allegedly) derived from the fact that the K-state time falls completely within the topic time, and thus Monica is described as impatient in (23b) within a speci�c topic time which contrasts with previous or later ones; a situation which, clearly does not apply to those occurrences under question. Moreover, even if we were sympathetic to the idea that topic situation contrasts [s*] could predict any given use of estar with an AP, it may also be noticed that the temporal dimension against which s* is manifested in constructions like Este guiso está fenomenal encompasses the (referent of) subject entirely (i.e., the entire span of the stew being described) and not just parts of it; thus raising a question as to which might be the topic situations against which the contrast is held. As we will see later, quirky occurrences of estar lead us to consider an alternative scenario, whereby the referent of the subject is considered individually, and the property described with estar as a quality by which this token is highlighted and contrasted with the other members of the class. 20Even if we took another course as to the concept of a focus involved in the ILP/SLP distinction —cf. Raposo and Uriagereka’s (1995) claim that IL and SL predicates are di�erentiated by virtue of the constituent that scopes out to a functional position: which can be the individual denoted by the DP subject (IL) or the event (SL)—; then the instances under consideration still pattern with ILPs rather than SLPs. Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 13 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori at the utterance time, this being [allegedly] the �rst time he/she comes to perceive the object. We will take up the anticipated state hypothesis next, on the basis of examples like (26). For the moment, let us emphasize that, in point of fact, Radha está muy bonita does not bear the largely known implication regularly associated with estar that makes utterances like Estás bonita so o�ensive; in short, there are no reasons to assume that the speaker is implying that the addressee was not stunning at one time, as it is the case in the classical example just raised. Summing the contents of this section, so far we presented evidence indicating that the property expressed through estar in these cases (i) is not associated to a particular moment or circumstance, but rather to the individual itself, and (ii) it does not seem to imply a change or result (and it is not linked to external reasons if we follow Arche’s de�nition); moreover, (iii) is not assumed to hold for a slice in the timespan determined by the existence of the individual, (iv) shows Lifetime e�ects and (v) it conveys a property rather than a state, and (vi) it does not necessarily contrast with previous or later topic situations in which the predicate does not apply to the (same) subject referent. Taken together, facts suggest that estar can deliver predicates with aspectual implications usually associated to ser in classical descriptions. On the other hand, even if both copulas may succeed in delivering IL predications, we still have to consider that the copula choice is not trivial. In fact, ILPs delivered by ser like Radha es fenomenal or El guiso es delicioso are not —by any means— synonymous to those rendered by estar and these di�erences are sig- ni�cant at diverse levels. Yet, before we turn to this issue, a formal introduction of our proposal, which would account for this contrast between ser and estar IL- predication, is in order. 3Ser not always implies estar Before we proceed to lay out our proposal on the semantic properties of estar enabling these peculiar occurrences, there is a further traditional claim we want to bring under consideration. In particular, the examples proposed above represent an interesting challenge to the classical assumption that ser predication implies estar, but not the other way round. More speci�cally, the idea behind this generalization is that the state denoted by ser refers to a stretch of time that can be decomposed into several delimited, successive, time intervals; and that each one of those intervals can be described by estar as well. In Lujan’s terms, all ser-predicates have to have been estar-predicates once (1981:177). Accordingly, a statement like Radha es bonita would imply that Radha está bonita applied (theoretically) at least once; whereas the state predicated by estar bonita may apply even without ser bonita ever applying, as (24) sketchily intends to summarize. Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 14 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori (23) ser bonita ==> estar bonita estar bonita =/=> ser bonita However, as those estar occurrences under discussion yield individual properties spanning over the subject’s whole lifespan (or, at least, for an inde�nite temporal gap) instead of applying exclusively to limited or brief time spreads only —i.e., intervals shorter than the ones covered by ser—, then the relation in (24) does not necessarily hold either. In fact, a case like (25) precisely illustrates the idea that the property denoted by ser does not necessarily implies that estar applied, not even once. (24) a. Esta almohada es blanda, pero [para mí] {no está/nunca estuvo} blanda. ‘This pillow is soft, but {it’s not/it has never been} soft [to me]’ b. Radha será guapa, pero no {está/*es} guapa. ‘Radha may be good-looking, but she’s not good-looking today’ On the other hand, even when two copular predications like those articulated ser blando and estar blando can be argued to refer to permanent, stable properties not applying only to a speci�c situation or stage in the lifespan of the entity, they are not synonymic and one does not imply the other. On this account, another fact we should consider is that the fact that (25) does not imply a contradiction is not due to the length of the time frame covered; actually, it has to do with the fact that estar implies the ascription of the property from the viewpoint of the speaker. So, what this example says is that the pillow may be individually regarded (in general) as holding a certain property (e.g., blanda); but that this same pillow is not deemed individually as such by the speaker (and, once more, the property rendered by estar is not implied to hold for a smaller time gap or expected to change). Thus, the fact that both clauses in (25) may be de�ned as ILPs —or, at least, have similar aspectual implications— does not imply (nor allow) a semantic correlation or synonymy in view of a number of semantic properties that clearly exceed temporal entailments, starting with the subjective �avor implied by estar. In order to address this issue more clearly, we should �rst introduce the theoretical framework. 4 An alternative approach Indeed, we showed a particular interest on the IL/SL distinction as main axis for the review outlined above. This is due to the fact that these concepts have been formulated in a way that seems semantically promising for estar and the facts presented so far; and, in particular, for a cognitively-sensitive take on its semantics. Speci�cally, we will take pains to show that a di�erent implementation of an IL/SL model of analysis based on implied comparisons might succeed in handling data Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 15 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori falling out of the scope in previous studies. Moreover, the approach we are about to introduce may prove fruitful in revealing semantic underpinnings shared by the di�erent uses of estar. Among the di�erent options explored in the literature, Franco & Steinmetz (1983, 1986) make use of the IL/SL distinction to claim that the di�erence between the semantic implications of ser and estar is grounded on implied comparisons. However, this proposal is quite di�erent to Clement’s Nexus and Maienborn’s discourse-anchored models. According to Franco & Steinmetz’ proposal, ser implies or signals a comparison between the entity being described and other of a same (or similar) kind. The idea is formalized in terms of an X/Y relation, whereby the entity (X) is compared against other(s) (Y) providing the standard by which the quality/trait is attributed to X. On the other hand, estar is argued to imply an X/X relation; that is, the entity [X] is compared against itself. On this account, a sentence like Radha es bella is argued to assert that Radha is pretty in the sense that her beauty is greater than, namely, that of some putative average woman. Thus, the quality attributed to Radha is relative to (the degree of that same trait in) other entities and relative to the set of women providing the standard for that quality. In turn, as estar expresses an X/X kind of comparison, an utterance like Radha está bella asserts that Radha’s (present) beauty is “greater than it is known or thought usually to be” (Franco & Steinmetz 1986: 377). In more Carlsonian terms, the quality is relative to other stages/situations in which Radha was (or will be) found. As Roby (2009) also suggests, Franco & Steinmetz’s interpretation of estar can be related to Falk’s (1979) concepts of individual norm vs. general norm in order to appreciate certain coincidences with classical descriptions of ser, some of them succinctly mentioned above. Thus, if ser is understood as drawing a comparison in reference to the general norm of a class or set of (comparable) entities, whereas estar is de�ned by drawing a comparison in reference to an individual norm (Franco & Steinmetz 1986: 285), somehow, the classical depiction of ser as a classi�catory predication (recently vindicated by Escandell & Leonetti 2002 i.a.) seems to be accounted for. In other words, this postulation also captures the idea that when a predicate appears in combination with ser the subject is categorized as belonging to the class denoted by the predicate; only that, in this case, the proposal highlights the fact that this is achieved through the comparison with other entities. By contrast, estar is used to compare the subject’s current state against its own usual state of a�airs (i.e., that individual’s norm). Anyhow, the occurrences under discussion in this paper seem to challenge this approach also, since —in view of allthe facts brought into consideration above— they do not seem to entail an X/X comparison whatsoever. In other words, it is clear by this point that constructions like El guiso está celestial do not imply a comparison against prior stages or the usual condition of the same X (token), but Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 16 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori rather a comparison between a token (deemed individually) and others (thus, Y) setting the standard by which the trait ‘celestial’ is ascribed to X (the stew). In other words, IL predicates delivered by estar stem from an X/Y, instead of an X/X relation. Yet, estar is used instead of ser, even when it is the latter that is associated with X/Y comparisons in the model quoted. Franco & Steinmetz (1986: 381) attempt to accommodate an example similar to ours —Este acero está duro [This steel is hard] in their version— as a “quite natural” extension of the same (X/X) kind of comparison. According to their tweak, these cases would be due to the fact the comparison is laid between the speaker’s perceptual expectancy prior to utterance time (and also prior to any sensory experience involving it) and the speaker’s sensory perception of the entity. Thus, the speaker would not be comparing X (este acero) with what he knows to be the usual state of X; instead, they claim, the speaker would be comparing the current state of X against an anticipated state of X. In few words, in order to deal with atypical occurrences, estar is analyzed as implying two di�erent comparisons: that between X’s current state vs. X’s own usual/prior states, or, alternatively, one between the actual state of X and the speaker’s expectations about X (i.e., X’s anticipated state). However, as Roby (2009: 26) also notices —and also as we indicated above with regard to Maienborn’s proposal (p.9)— the anticipated state hypothesis [ASH] as a variable in the choice of estar over ser runs into problems in, at least, three respects. On the one hand, because the selection of estar comprise examples that can perfectly occur even if the speaker experiences or describes an object/entity without any prior knowledge or expectation about it. What is more, the “discovery interpretation of estar” (Maienborn 2005:160) —which can be lined up with Querido’s (1976: 354) assertion that estar is the appropriate choice of copula for describing a �rst sensorial experience— fails to predict the copula choice in cases in which estar is employed to describe a usual state or condition. Secondly, the prediction of estar’s distribution based on the ASH is also �awed as discovery interpretations can also be reported through ser. For instance, one could very well claim Los limones son ácidos after tasting lemons for the very �rst time, just as one could easily say Las amebas son diminutas [Amoebas are minuscule] after discovering them in the microscope (and, in fact, this case would make a much better candidate for comparison against anticipated states). (25) a. Estos limones están ácidos ‘These lemons are sour’ b. El cielo está celeste [en un día despejado] ‘The sky is blue’ [on a clear day] Abstracting away from the fact that it would be very hard to prove that an expression like Radha está fenomenal does imply a contrast against an expec- Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 17 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori ted/anticipated state, ASH as a variable in the choice of estar over ser breaks down against utterances like (26) for other reasons. First, because these sentences denote exactly what one could expect from their subjects (i.e., lemons and the sky, respectively). On the other hand, and following from this, because it can also be argued that what sentences like (26a) actually convey is not that lemons are deemed sour in relation to what one might expect from lemons, but rather that lemons are found sour in relation to something else; most likely, in comparison to other lemons. Accordingly, these cases can be accommodated in our schema as long as they can be argued to imply a comparison between the acidity of some given lemons and either the average acidity of lemons in general (kind) or the acidity of other lemons (as tokens; e.g., other lemons tasted before). In other words, the shortcoming could be easily solved by acknowledging the possibility that these cases may instantiate an X/Y kind of comparison, which can be held (alternatively) between a token and the class or between two (lemon) tokens. Thus, sentences like (26a) would denote a property, which is established not in relation to expectations but to the average degree of that property in the class represented by Y (thus, setting the standard by which property P is ascribed to X, in Franco & Steinmetz’s terms); or, alternatively, against other (comparable) lemon-tokens which may not be deemed as sour according to the speaker. This situation contrasts sharply with (26b), where the predicate indeed denotes a state (the light-blueness of the sky) held comparison against the referent’s average condition, its anticipated state, or against a di�erent situation (e.g., how the sky looks like on a cloudy day). On this account, a legitimate X/X implied comparison is therefore instantiated, as the property is ascribed in relation to other stages/states of this same (token) sky. As a matter of fact, (26b) allows the explicitation of the particular stage in which the property holds of the sky (e.g., El cielo está celeste hoy); whereas a similar speci�cation is not consented by (26a) —indeed, Estos limones están ácidos hoy does not make an acceptable construction, at least under normal circumstances.21 In our view, this represents another indicator of the fact that the implied comparison is held individually between tokens (or between a token and a class) rather than between stages or contrasting situations of the same token (anticipated states included).22 Against this background, we will set out to show that an (abstract) spatial representation of these notions could be very revealing, apart from being more in tune with the semantic content of this copula. In particular, an abstract reading 21Even if it might be deemed possible in a context in which, for instance, we are repeatedly tasting the same lemons (in di�erent days); yet, the comparison would still be of an X/Y kind. 22On this account, pairings like (26) could be argued to show that the anticipated state hypothesis does not handle quirky uses of estar like (26a), and it does not necessarily capture regular cases (26b) either. We want to thank an anonymous reviewer for the indication allowing us to make a stronger point here. Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 18 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori of estar’s locative meaning would not only give us the possibility to account for the occurrences in question without abandoning the X/X-X/Y implied comparison model; but it would also involve the additional advantage of capturing the di�erences between the two types of estar clauses on the basis of a same core semantic content. In addition, and what is more interesting, such an approach will allow us to explain the semantic features distinguishing them from ser clauses even in those cases where the aspectual implications are not that di�erent —i.e., in delivering IL predicates. 5 Space and abstraction As we already argued in the introduction —in reference to previous works (cf. Mangialavori 2013a)— an abstract reading of estar’s spatial content carries a number of advantages in the analysis of this copula, especially concerning its semantic implications and grammatical properties. In particular, this hypothesis shows that similar grammatical patterns emerge across apparently unrelated semantic constructions rendered by this copula (i.e., between spatial/existential clauses and those conveying states of a�airs). In consequence, relevant insights about its distribution in the two complementary alternations involving this verb (estar/ser and estar/haber) arise. In the present case, abstract location will allow us to introduce the hypothesis that the semantic underpinnings of quirky uses of estar become predictable if a di�erent view on the implied comparison model (presented above) is pursued. In particular, we will take pains to show that a localist conceptualization of the comparison model put forward by Franco & Steinmetz would nicely �t the phenomenon represented by estar clauses; and, indeed, it may account for relevant facets of meaning making these constructions di�erent to those articulated through ser, even when conveying stable, long-term properties. More interestingly, this view could support a connection between cognitively-based facts and grammatical patterns in Spanish. In particular, we will depart from two postulations. On the one hand, the widely known assumption that states can be conceived as abstract or metaphorical locations (since Fillmore, 1968; Gruber, 1965 and Jackendo�, 1983 i.a.); on the other hand, the fact that locative predicates serve as structural templates for other (nontopographical) expressions (cf. Miller & Johnson-Laird, 1976:375 i.a.). This means that in exploring the organization of concepts such as IL/SL predication in the case of estar we do not have to start de novo. Rather, we can constrain the possible hypotheses by adapting, to a certain extent, the conceptual structure of spatial relations to our new purposes. In other words, we want to entertain the hypothesis that, as estar can express the concrete topographical sense of objects located in space in its regular (or original) use, it seems logical to consider that it may also be employed to de- Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 19 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori liver a more abstract, often psychological (Talmy 2000:291), meaning. Thus, states (therefore, the occurrences of estar under consideration here) could be easily accommodated within the proposed (locative) semantics of this copula, as abstract situations in which an entity can be found. All in all, the proposal of a same conceptual construal (to be in a certain situation/position) connecting estar’s di�erent guises �ts the widely known claim that the same conceptual functions we use when dealing with physical space can also be applied to our conception of abstract space (Gruber 1965 and Jackendo�1990 i.a.23). Therefore, and against the general wisdom in Spanish grammar (cf. Fernández Leborans 1999:2424), the concept of temporary situation, originally posited only for —allegedly non-copular24 — constructions like (27), could be argued to apply to cases like (28) just a well, following Jackendo�(1983:194) (Cf. Mangialavori Rasia 2013a). (26) El guiso está en la heladera ‘The stew is in the fridge’ (27) El guiso está caliente ‘The stew is hot’ Two further endorsements for this view —or, rather, for the proposed simpli�cation that comes from the idea that estar can be used to render non-spatial predicates, such as states of a�airs, just like many other spatial verbs, on the basis of a same semantic structure— come from the (psychological) claim that the mind does not manufacture abstract concepts out of thin air, but it rather adapts machinery already available (cf. Jackendo�1983: 189); and also from the observation that spatial organization is of central importance in human cognition (since Lyons 1977: 718). On this account, cases like (28) —and, therefore, also quirky estar occurrences like El guiso está celestial— can also be argued to conform to the semantic content of this copula as ‘sense of objects in space’, by virtue of an abstract representation of the predicative relation rendered in (27). In fact, locative semantics25 may account for great part of the facts under consideration here. 23We refer to the ‘Thematic Relations Hypothesis’ (Jackendo�1983, 1990 inspired on Gruber 1965), according to which in any semantic �eld of states and events, the principal functions are a subset of those used for location and motion (Jackendo�1983:188). 24Considering, as posed in detail by Mangialavori Rasia (2013), that locative content has traditionally been claimed to be absent in copulative occurrences of estar, as opposed to the (alleged) intransitive locative verb. 25From our perspective, it is important to emphasize that cognitive operations, like those deriving atypical uses of estar, are enabled by its locative meaning, which we believe to be (following Mateu 2002) a function of both conceptual content and syntactically-encoded semantic construal (cf. n. 3); and, indeed, such options are not available for a non-spatial copula like ser. So, even if conceptual content has been de�ned as grammatically opaque (cf. Mateu 2002 i.a.), data o�ers strong reasons to believe that conceptual structure may not be homomorphically related to syntax, but might match it on crucial respects. In the case of estar, the convergence of structural and conceptual spatial semantic features may not only set a constant among estar occurrences, but Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 20 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori In particular, we want to focus on the fact that information passed on by conceptual domains —on the basis of both the locative meaning of estar— could account for the delivery of both IL and SL predicates, and, in particular, for the special semantic features driving the choice for estar in the delivery of ILPs. We refer to the fact that location, from the view point of cognition and conceptual construal, is relative. More speci�cally, the spatial disposition of a (focal) object in a scene is largely characterized in terms of a further object, which is also selected within the scene, and whose location and sometimes also geometric properties are already known and so can function as a reference object (Talmy 2000: 182). Thus, human perception of space and location stipulate that the site of an object —which, in our view, also applies to states such as those conveyed by estar in (28)— is indicated in terms of distance from or in relation to the location of a further, secondary object. Summing up, location of an object in a scene is conceived in relation to another’s. Going back to the analogy between (27) and (28) above, this suggests that stative expressions also de�ne a one-dimensional "pseudospace" (Jackendo�1983: 134) in which entities are located. Thus, in its abstract use, estar predication is about things that are located not only in concrete, topographical space, but in abstract situations or states26 as well. To appreciate the force of the parallelism, consider the inference patterns of corresponding spatial and stative expressions. In the spatial construction, the function be at —corresponding to the locative copula— maps a thing into a spatial frame and asserts that the thing occupies a speci�c point within this space. When shifted into the stative domain, the function be at maps a thing into a state and asserts that the thing occupies a determined site within this [pseudo-]spatial frame. Along these lines, estar could be argued to denote the situation of entities in concrete or abstract space; nonetheless, in being articulated by a locative verb, this situation will be essentially characterized in terms of a further object. Accordingly, the reference frame set up by the (implied relation with) a secondary object would explain the peculiar semantic �avor of these clauses —mostly related to subjectivity, contrastiveness and relativity of the property ascribed— and, therefore, the choice for this copula instead of ser. In particular, this relativity of location can be easily lined up with the implied comparison model we are advocating for, as X (the subject) is always described in relation to something else (be it against other instances of itself [X/X] or other individuals [X/Y]). In practical terms, it predicts the contrastiveness described in also operate in di�erentiating them from other copular constructions (i.e., ser-IL predications) and predicting patterns not following from the IL/SL distinction, which makes it all fall nicely inside a general theory explaining its semantic qualities. 26Let us clarify that we use the word state here in the sense of the state [of a�airs] rendered by estar clauses like (28), to di�erentiate them somehow from topographical constructions such as (27); as we a are well aware that (27) is also a stative predication. Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 21 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori (33) a. La The mujer wife ?({de of Pepe/esa}) Pepe/that está is buena pretty ‘Pepe’s wife is attractive’ / ‘That woman is attractive’ b. La The casa house *?({de of la the esquina/del corner/of-the Intendente/esa}) Mayor/that está is espectacular amazing ‘I �nd {the house on the corner/the Mayor’s house/ that house} amazing’ By the same token, it may also be noted that the fact that determiners such as un deliver contexts allowing a choice between ser and estar (35a) can be largely related to its ambiguity. In broad terms, the combination with estar leads to the interpretation of un as a quanti�er signaling one entity within a set or group —though not as an inde�nite/unspeci�c determiner, in the fashion allowed by ser—, thus forcing a disambiguation sketchily illustrated in (35c). This might be more visible in the glosses, as the two interpretations correspond to two di�erent determiners (a/one) in English. (34) a. {El/este/aquél} actor {es/está} fenomenal ‘The/This/That actor is great’ b. {Cierto/Un} actor {es/?está} fenomenal ‘Certain/An actor is great’ c. Un[o] [de los] actor[es] {es/está} fenomenal ‘One of the actors is great’ 7 Two kind of locations To �nish our proposal, we will present the localist conceptualization of the implied comparison model accommodating SL and IL readings. From our perspective, the existence of two kinds of stative predications delivered by estar (SLP/ILP) is due to the convergence of two phenomena: (i)the fact that location is established in relation to a secondary object setting a parameter of contrast; and (ii)the fact that this secondary object can be instantiated either by a di�erent stage/state went through by the same individual (i.e., a location relative to a prior/former location of the same entity) or by another individual. Taking up Franco & Steinmetz’ model of implied comparison, what we mean is that the relativity of location makes it possible to locate an object in space either in a X/X or in a X/Y relation, as we sketchily illustrate in (36). (35) a. Position occupied by X [individual1]...(relative to)...Position occupied by Y [tokenindividual2/kind] b. Position occupied by X [stage1].........(relative to)...Position occupied by X [stage2] Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 28 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori Let us now see how this approach holds up against the data discussed in this paper. On the one hand, the SLP reading is easily explained by an X/X relation. Thus, the ‘location’ or ‘situation’ conveyed by estar in its regular use (37) is understood as contrastive, transitory or temporally bounded given the implied comparison with respect to (i)other situations in which the same entity (X) was/will be found or (ii) the usual position occupied by X (which must be di�erent to the current one). In sum, the situation would match the proposal laid out by Franco & Steinmetz in the sense that it implies a comparison against the entity and other instances of itself. (36) Radha está fenomenal [antes {no lo estaba tanto/estaba pasada de peso}] ‘Radha is splendid’ [{she wasn’t that way/she was overweight} before] By contrast, as the occurrences now under discussion pattern with ILPs rather than with SLPs, then an X/Y kind of comparison should be expected. Interestingly enough, the locative perspective allows a cognitively-based explanation for an X/Y relation in estar. We can start by noting that, as soon as you work with individuals, rather than with stages, you can give them a place (i.e., locate/classify them) either (a) within a group or (b) in relation to the position marked by others (individually). According to this, atypical estar constructions could be argued to imply the comparison of X against either (a) one or more entities (Y) providing the standard by which the quality denoted by the AP is attributed to the subject of the clause (X) —i.e., a group—, or (b) against a single (comparable) entity (Y) occupying a di�erent position from the speaker’s standpoint or perspective, thus X’s site indicated in terms of distance from or in relation to this Y (as posed in section 5 above) —i.e., a position marked by others.39 In our view, this is what determines the main di�erence with respect to ser, as both copulas succeed in delivering IL predicates though they are not trivially interchangeable, and the clauses yielded by them are not synonymic. Indeed, both copulas may eventually draw a comparison in reference to the norm of members of a class or set possessing that quality (X/Y relation). However, as we saw above in section 2, ser is essentially characterized by expressing a property that is somehow ‘possessed’ by the subject (cf. Arche 2006 i.a.). In turn, estar also de�nes the entity as an individual, though it does not imply possession, as it does so by locating the entity within a reference frame (set by the secondary object Y). This frame may 39By means of illustration, note that these cases can be paraphrased with other originally spatial predicates that conform to a similar conceptual construal (38) (roughly put: �nd/locate X at Z). (i) Yo la encuentro fenomenal [en relación a las otras mujeres] ‘I �nd her stunning’ [in relation to other women] Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 29 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori either (a) conform the norm according to which fenomenal is imputed to Radha (38a), or (b) o�er the reference in relation to which Radha is located within a given set —that of women deemed fenomenal by the speaker— in (38b), in agreement with the two options for location of individuals o�ered above. (37) a. Radha está fenomenal [!pertenece al grupo de mujeres consideradas excepcionalmente atractivas/norma: ‘mujer fenomenal’] ‘Radha is a very attractive woman’ [!she belongs to the set of women regarded as exceptionally attractive/’attractive woman’ norm] b. Radha está fenomenal [{María no tanto/Carla está pasada de peso}] ‘Radha is a very attractive woman’ [{María is not that attractive/Carla is overwight}] In this respect, we want to emphasize that the ‘comparison’ is clearly not an extension of the X/X relation (as posited by Franco & Steinmetz), since in neither case the entity is compared against itself. Indeed, the evidence gathered above clearly indicated that these predicates are not about stages, but about individual properties, and that the individual is located as a whole (thus triggering, among other things, LifeTime e�ects), and, as we already claimed, a X/Y implied comparison is expected on conceptual grounds.40 Therefore, by readjusting Franco & Steinmetz’ proposal, we would be able to account for the fact that in regular estar occurrences like (37) or (39a) Radha is described as fenomenal; the implication being be that Radha’s present appearance (i.e., her situation) is held in relation to the situation/position in which she usually is found (e.g., her usual looks) or the situation in which she was in a previous temporal gap, perhaps because she dropped some pounds or had a makeover. By contrast, in quirky cases like (38a,b) the implication would be that her beauty is prominent compared to other women —either by comparison with one or more women positioned by the speaker as not-good looking (38b), or with one or more entities providing the standard fenomenal (38a). This alternative could be summarized in pairings like (39), which intend to make explicit the di�erent entailments associated with regular estar occurrences (represented by (39a) and the alternative use described in this paper (39b). 40Apart from this, our proposal has the advantage of avoiding a strict dependency on (though there might be sensitivity to) discursive levels (cf. Maienborn’s approach, which by claiming two semantically null copulas fail to explain many facts observed). Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 30 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori (38) a. Radha está fenomenal [antes estaba bastante �oja] ‘Radha is splendid [she was pretty plain before]’ b. Radha está fenomenal [su hermana no es ni de cerca tan bonita] ‘Radha is splendid [her sister is not nearly as good-looking]’ All in all, the implication of an implied comparison (be it of the X/X or Y/X kind) can be explained on the basis of the conceptual tenets introduced in section 5, which also account for �ner-grained conceptual entailments. For instance, given this relativity, it remains implicit that a secondary object is needed as a reference by virtue of its known spatial (situational) disposition. Accordingly, the object is located as follows: the primary object (i.e., the subject referent) features unknown spatial properties (situation) to be determined; therefore, the secondary object acts as a reference entity, having known properties that can characterize the primary object’s unknowns. Indeed, this matches certain cognitive implications (in the Talmian sense) that can be noticed in estar clauses, as the primary object is seen as more recently located on the scene (e.g., in awareness) and appears as an object of greater concern or relevance (more salient, once perceived), whilst the secondary object is located earlier on the scene or in memory41 and can be characterized as an object of lesser concern/relevance (more backgrounded, once primary object is perceived). 7.1 Further features captured by Perspectival Location Apart from this, the approach o�ered here also captures an already largely described �avor (cf. Gili Gaya 1943 [2001] §46, Escandell & Leonetti 2002, F. Leborans 1999), though yet unaccounted for in connection with estar’s spatial content. The general intuition we refer to is that estar implies a classi�cation depending on the speaker’s opinion; whereas ser describes the subject as possessing a certain quality regardless of the speaker’s standing. In our account, the relativity of location is directly associated with the fact that the property is ascribed from the viewpoint of the speaker —in fact, this component is what predicted estar’s compatibility with certain APs, as seen in section 6— and this �ts the standard conceptualization of locating an object in space. Indeed, the involvement of the viewpoint matches the conceptual terms involved in the setup of location, as long as the concept of perspectival location (cf. Talmy 2000: 68) is considered. Thus, the ‘subjectivity’ in estar is due to the fact that the conceptual system establishes a perspective point from which the entity (X) is cognitively regarded (situated), be it either in relation to other stages or other individuals. This also predicts the compatibility patterns for AP predicates laid out above. 41Thus, matching the (debatable, though largely known) claim that estar compares the object either with another similar, already known, object or with the speaker’s expectation, as we mentioned with regard to (26). Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 31 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori Finally, this proposal also �ts other implications signaling a di�erence between ser and estar IL predications. For instance, there is a general consensus as to the fact that estar predication implies (i) the existence of the (token) referent, (ii) a judgment about a speci�c item identi�ed by the speaker and (iii) the direct appreciation of this item by the speaker, as (41) indicates (cf. F. Leborans 1999:2428). By contrast, ser seems to convey a generic statement implying no direct experience on the speaker’s end, nor the existence of such entity. In fact, the sentences in (40) can be uttered even if the speaker never met Radha nor tasted the pizza because, by using ser, there is no opinion or evaluation necessarily implied. In our view, and in sheer contrast with the case of estar, not being a perspectival location set by the speaker, no viewpoint is involved. (39) a. La pizza es buena [yo no la probé] ‘Pizza is good42 [I haven’t tasted it]’ b. Radha es buena [yo no la conozco] ‘Radha is (a) good (person) [I haven’t met her]’ (40) a. La pizza está buena [*yo no la probé] ‘The pizza is (very) good [*I haven’t tasted it]’ b. Radha está buena [*yo no la conozco] ‘Radha is stunning [*I haven’t met her]’ In fact, the perspectival point from which the location/situation is established must be somehow speci�ed with estar, whereas this requisite does not hold for ser.43 Moreover, and even if we will not proceed any further on this issue, note that ser allows a discrepancy that becomes problematic with estar (42), as it already involves a location set from the speaker’s standing, and therefore, a di�erent position is not liable. (41) a. Andar sin comer y sin dormir no es bueno (pero yo creo que sí es bueno) ‘To go without food and without sleep is not good (though I think it is good)’ 42In the sense, namely, that pizza is good for your health; though we would nowadays disagree. 43For example, if we say Andar sin comer y sin dormir no es bueno [To go without food and without sleep is not good]’ it can be understood as a general belief not involving a speci�c viewpoint. By contrast, in the case of Andar sin comer y sin dormir no está bueno [To go without food and without sleep is not a ride] a dative complement explicitating in sentential syntax what is conceptually implied is expected (e.g. para ella). The fact that this can be syntactically elided is not an indication of viewpoint not being relevant; in fact what happens is that it can be retrieved somehow (contextually or deictically) or, otherwise, it is assumed that this is held form the speaker’s standing; whereas none of this necessarily happens with ser. Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 32 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori b. Andar sin comer y sin dormir no está bueno (*pero yo creo que sí está bueno) ‘To go without food and without sleep is not a ride (though I think it is a ride)’ 8 Summing up Although estar has traditionally been related to SL predication, in this paper we present evidence indicating that this copula also delivers predicates with IL properties. By suggesting a di�erent implementation of Franco & Steinmetz’s analysis of SL/IL predication, we relate this to the entailment of two di�erent kinds (X/X vs. X/Y) of [implied] comparisons, accommodating SL and IL predicates respectively. Still, even if the delivery of ILPs by estar can be lined up with the semantic properties usually ascribed to ser (e.g., LifeTime e�ects, absence of temporal boundedness, etc.), the copula choice is not semantically trivial. In particular, we assume that when estar is involved we are dealing with a locative predication no matter whether it describes a topographical or a more abstract sense of objects in space, or whether there is temporal boundedness in the situation conveyed or not. If the latter does not occur, instead of proposing a mutation/coercion from SL to IL predicate, we put forward an alternative approach which follows directly from its spatial semantics. On this account, instances like (1) become predictable and consistent with the semantic underpinnings of estar by virtue of a cognitively-determined fact (i.e., location is relative, set in relation to a secondary object. Thus, quirky estar constructions can be argued to denote the location of an entity (X) as well, the only di�erence being that, in this case, the position it is not set in relation to other points formerly or successively occupied by the same entity (X/X comparison). Instead, the entity (X) is located as a whole, individually, in relation to other (comparable) individuals (Y) setting the standard by which a given site is ascribed to X; hence, a X/Y kind of implied comparison —which happens to be the one associated to IL predicates— arises. Whereas the implied comparison model accounts for the contrastive �avor largely described in estar in a way that also �t cognitive entailments related to locative predicates, the fact that this situation/position may be construed from a perspectival location would explain both the involvement of the speaker’s perspective —thus capturing the judgmental/subjective �avor described since Gili Gaya (1943)— and the contrastiveness traditionally pointed out in semantic studies. Moreover, it captures not only �ner-grained semantic entailments (novelty, prominence) and truth conditions, but also grammatically-relevant patterns drawn by restrictions on both the DP subject and the AP predicate. To our understanding, these considerations are interesting not only in view of grammatical facts determined by cognitive operations available in a natural language featuring a locative copula —thus, also applicable to languages with a similar Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 33 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori features like Portuguese—, but as indicators of the interaction between semantic information encoded by extralinguistic (conceptual) systems and grammar. 9 References Arche, Maria. 2006. Individuals in time. Amsterdam: John Benjamins. Bosque, I. 1993. Sobre las diferencias entre los adjetivos relacionales y los cali�cativos. Revista Argentina de Lingüística 9:9–48. Brucart, J. M. 2010. “La alternancia ser/estar y las construcciones atributivas de localización”. In A. Avellaneda (ed.), Actas del V Encuentro de Gramática Generativa, Neuquén: UNCo, 115-152. Camacho, José. 2010. ’Ser’ and ’Estar’: Individual/Stage level predicates or aspect?. In Hualde, J. I. & Olarrea, A. (eds.) Blackwell Handbook of Hispanic Linguistics, Amsterdam: Blackwell. Camacho, José. 2012. “Ser and Estar: The Individual/Stage level Distinction”, en: Hualde, J. I., Olarrea, A. & O’Rourke, E. (eds): The Handbook of Hispanic Linguistics, vol. 22, Chichester: John Wiley & Sons. Carlson, Gregg. 1977. Reference to Kinds in English. New York: Garland. Clements, J. Clancy.1988. “The semantics and pragmatics of the Spanish <Copula+Adjective> construction”, Linguistics, 26, 779-822. Davidson, Donald. 1969. “The Individuation of Events”. In Rescher, N. (ed.). Essays in Honor of Carl G. Hempel, 216-34. Dordrecht: Reidel. Demonte, Violeta. 1999. “Clases de adjetivos: La posición del adjetivo en el SN”. In I. Bosque and V. Demonte (eds.) Gramática Descriptiva de la Lengua Española, 129–215. Madrid: Espasa. Di Tullio, Angela. 2005. Manual de gramática del español. Buenos Aires: La isla de la Luna. Dixon, Robert. 1977. Where have all the adjectives gone? Studies in Language 1,19–80. Dowty, David. 1979. Word Meaning and Montague Grammar. Dordrecht: Reidel. Escandell, Victoria & Manuel Leonetti. 2002. “Coercion and the stage/individual distinction” en: J. Gutiérrez Rexach (ed.) From Words to Discourse, 159-179. New York/Amsterdam: Elsevier. Fernández Leborans, M. Jesús. 1995. “Las construcciones con el verbo estar: aspectos sintácticos y semánticos”, Verba 22, 253-284. Fábregas, Antonio. 2012. A guide to IL and SL in Spanish. Borealis 2012(1/2), 1-71. Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 34 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori Fernández Leborans, M. Jesús 1999. "La predicación: Las oraciones copulativas", Bosque, I. & Demonte, V. (eds.) Vol. 2, 2357-2460. Madrid: Espasa. Franco, Fabiola & Steinmetz, Donald. 1983. Scr y estar+adjetivo cali�cativo en español. Hispania 66, 176-184. Franco, Fabiola & Steinmetz, Donald. 1986. Taming ser and estar with predicate adjectives. Hispania 69: 377-386. Gallego, Angel. & Uriagereka, Juan. 2009: "Estar=Ser+P", en: XIX CGG, April 1-3 2009, Euskal Herriko. Available at http://goo.gl/xseBX [12/25/2013] Gili Gaya, Samuel. 1943. Curso superior de sintaxis española. México: Minerva. Reprinted in 2001. Barcelona: Vox. Gruber, Je�rey. 1976. Lexical Structures in Syntax and Semantics. Amsterdam: North-Holland. Hale, Ken and Samuel J. Keyser. 2002. Prolegomenon to a Theory of Argument Structure. Cambridge: MIT Press Husband, Matthew. 2010. On the compositional nature of stativity. PhD dissertation, Michigan State University. Jackendo�, Ray. 1983. Semantics and Cognition, Cambridge, MA: MIT Press. Jackendo�, Ray. 1990. Semantic Structures, Cambridge, MA: MIT Press. Jackendo�, Ray. 2002. Foundations of language. Oxford: Oxford University Press. Jiménez-Fernández, Ángel. 2012. “What information structure tells us about Individual/Stage Level Predicates”, Borealis 2012(1/1), 1-32. Kearns, Kate. 2007. “Telic Sense of Deadjectival Verbs” Lingua 117, 27-66 Kratzer, Angelika. 1995. “Stage-level and individual-level predicates”, en: Carlson, G. N. & Pelletier, F. J. (eds.), The Generic Book, 125-175. Chicago, London: The University of Chicago Press. Kratzer, Angelika. 2000. “Building Statives”, en: Conathan, L. J. (ed.) Proceedings of the 26th Annual Meeting of the Berkeley Linguistics Society. Berkeley: BLS, 385-399. Leonetti, Manuel. 1994. “Ser y estar: estado de la cuestión”, Pliegos de la Ínsula Barataria 1, 182-205. Leonetti, Manuel. 2008. “De�niteness e�ects and the role of the coda in existential constructions”. In Klinge, Alex & Henrik Hoeg-Müller (eds.) On nominal determination. Amsterdam: John Benjamins, 131–162. Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 35 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori Leonetti, Manuel & Ma. Victoria Escandell 1995. “Construcciones existenciales y oraciones de relativo” in Ru�no, Giovanni (ed.) Atti del XXI Congresso Internazionale di Linguistica e Filologia Romanza, Tübingen: Max Niemeyer Verlag, 261-272. Luján, María. 1981. “The Spanish copulas as aspectual indicators”, en: Lingua 54, 165-210. Lyons, John. 1977. Semantics. Cambridge: Cambridge University Press. Maienborn, Claudia. 2005. “On the limits of the Davidsonian approach: The case of copula sentences”. In Theoretical Linguistics, 31(3), 275-316 Mangialavori Rasia, M. Eugenia. 2011. Una semántica léxica uni�cada para estar. Propuesta de análisis integrador de sus diversas ocurrencias. Handout presented at I Simposio Nacional de la Lengua Española, Montevideo, September, 2011. Mangialavori Rasia, M. Eugenia. 2012. Ni perfectividad ni telicidad ni coerción. In C. Muñoz et al. (eds.) Actas del V CIL, Buenos Aires: FFyL. Mangialavori Rasia, M. Eugenia. 2013a. “Conciliating states and locations. Towards a More Comprehensive and In-depth Account of the Spanish Copula Estar”, Studies in Hispanic & Lusophone Linguistics 6:1, 37-57. Mangialavori Rasia, M. Eugenia. 2013b. The expression of (abstract) location /situation: the case of estar. Capturing new challenging data. Presented at Permanent Colloquium on Theoretical Linguistics, CSIC, Madrid, July 2013. Mangialavori Rasia, M. Eugenia. 2013c. Lexical Syntax of the AP and the architecture of Spanish copular constructions. Handout presented at EGG 6, General Roca, August, 2013. Marín, Rafael. 2010. “Spanish adjectives within bounds”, In: Cabredo Hofherr, Patricia and Ora Matushansky (eds.), Adjectives: Formal Analyses in Syntax and Semantics, 307–332. Amsterdam: John Benjamins. Mateu, Jaume. 2002. Argument Structure. Relational Construal at the SyntaxSemantics Interface. PhD Dissertation. Barcelona: Bellaterra. Miller, George & Philip Johnson-Laird. 1976. Language and Perception. Cambridge, MA: Belknap Press. Moro, Andrea. 1997. The Raising of predicates. Cambridge: Cambridge University Press. Musan, Renate. 1997 “Tense, Predicates, and Lifetime E�ects “, Natural Language Semantics 5-3, 271-301. Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 36 Not always a stage. A typical patterns in Spanish copular clauses Maria Eugenia Mangialavori Raposo, Eduardo and Juan Uriagereka. 1995. “Two Types of Small Clauses”. In A. Cardinaletti and M. T. Guasti (eds.), Small Clauses, New York: Academic Press. 179-206. Roby, David. 2009. Aspect and the Categorization of States: The case of ser and estar in Spanish. Amsterdam: John Benjamins. Schmitt, Cristina. 1992. “Ser and estar: a matter of aspect”, en: Broderick, K. (ed.), Proceedings of NELS 22, 411-26. Amherst, MA: GLSA. Suñer, Avelina.1990. La predicación secundaria en español, PhD thesis, Universidad Autónoma de Barcelona. Available at <http://www.tdx.cat/handle/10803/4895> [01/01/2013] Svenonius, Peter. 1996. “Predication and Functional Heads” in Camacho, José and Lina Choueri (eds), Proceedings of the Fourteenth West Coast Conference on Formal Linguistics, California: CSLI, pp. 493-507. Talmy, Leonard. 2000. Toward a Cognitive Semantics. Volume I: Concept Structuring Systems. Cambridge, MA: MIT Press. Zagona, Karen. 2009. “Ser and Estar: Phrase structure and aspect”, en Nishida, C. & Russi, C. (eds.), Selected proceedings of Chronos 8. Amsterdam: Rodopi. Reception date: 08/23/2013 Revision date: 09/16/2013 Acceptation date: 11/24/2013 Maria Eugenia Mangialavori Consejo Nacional de Investigaciones Cientí�cas y Técnicas (Argentina) [email protected] Iberia: IJTL |Volume 5.2, 2013, 1-37 ISSN: 1989-8525 doi: tba http://revistas.ojs.es/index.php/iberia/ 37