When focus goes wild : An empirical study of two syntactic positions for information focus

My goal in the present paper is to carry out an analysis of the syntactic and discourse properties of Information Focus (IF) in Southern Peninsular Spanish (SPS) and Standard Spanish (SS) varieties. Generally, it has been argued that IF tends to occur last in a sentence since new information is placed in final position, following the End-Focus Principle as well as the Nuclear Stress Principle (Zubizarreta 1998). Focus fronting has been hence reserved for those cases in which a clear contrast between two alternatives is established (Contrastive Focus and Mirative Focus; cf. Cruschina 2012). The starting hypothesis here is that IF can appear as a fronted element in a sentence and that SPS speakers show a higher degree of acceptability and grammaticality towards such constructions, as opposed to SS speakers. This points toward a certain degree of microparametric variation, which will be tested by means of a survey run among both SPS and SS speakers.


Introduction
This paper explores the discourse-syntax properties of a specific type of focus construction in Spanish, namely Information Focus Fronting.It has been generally asserted that Spanish lacks this type of discourse-based movement on the basis that the focus found in the Left Periphery LP) of a sentence always conveys some sort of contrast, thereby instantiating what is termed as Contrastive Focus.Cruschina (2012) explicitly groups Italian and Spanish together in that both can have Focus Fronting only if it bears a contrastive flavour, following Rizzi (1997) for Italian, and Zubizarreta (1998Zubizarreta ( , 1999) ) and López (2009) for Spanish.This is illustrated in (1).
Pedro buy-past.3sgapples c. #Manzanas compró Pedro.'Pedro bought apples.' As indicated by #, sentence (1c) is regarded as non-felicitous.It is not acceptable as an answer to (1a).The reason provided in the literature is that in Spanish Information Focus must always be post-verbal, and when focus is preposed it expresses contrast.
I show that the unacceptability of examples such as (1c) is subject to microparametric variation (sensu Roberts 2012 andBiberauer et al. 2010).Some varieties of Spanish (Southern Peninsular Spanish, including Andalusian and Extremaduran Spanish) seem to obviate the postverbal condition on Information Focus.I discuss the syntax of different types of focus, namely Contrastive/Corrective Focus (CF), Mirative Focus (MF) and Information Focus (IF), and propose that Information Focus has a designated position in the left periphery in SPS, which is missing in Standard Spanish (SS).
Descriptive surveys such as the one carried out by RAE-ASALE does not take into account the different types of focus and includes all kinds of fronting within a single group regardless of the distinct salient properties at the syntactic, interpretive and phonological levels.In this group phenomena are mixed up such as Contrastive Focus, Mirative Focus (Cruschina 2012;Jiménez-Fernández 2015), Resumptive Preposing (Cinque 1990, Leonetti & Escandell 2009), and Quantifier fronting/Negative Preposing (Barbosa 2001;Bosque 1980).Researchers have agreed that in all these cases contrast is present: (2) a.
¡Por Dios, dos botellas se han bebido!(MF, adopted from Cruschina 2012) Studies in generative grammar have emerged that distinguish some of these IS phenomena (Bianchi 2012, Cruschina 2012, Haegeman 2012).In Spanish, some works have identified focus preposing other than purely contrastive (Uriagereka 1988, Quer 2002, Gallego 2007, Leonetti & Escandell 2009), yet all these analyses agree Spanish focus fronting is never associated with purely new information.Fronting is a device used to express either contrast on a specific constituent or on the sentence polarity (Verum Focus).
In this work, I show that interpretative and syntactic properties can be used to establish a more accurate typology of focus, concentrating on the peripheral position that IF can fill in some varieties of Spanish (Andalusian, Extremaduran). 1 An experiment is conducted among speakers in which they have to judge the grammaticality of fronted focus constructions.
In this survey, informants were faced with examples in which IF occurred in post-verbal position and in the LP.The (A) sentences below provide the context where the informants have to rightly place the sentences that follow: (3) A: El chocolate que había puesto en la nevera ya no está.be the most natural option, namely the elliptical clause.However, I have tried to avoid the word-by-word repetition of the question in the answer, since this would add to the clumsy flavour that answers echoing the material in the question has for speakers.In addition, I assume that (at least some) pre-verbal subjects are placed in the CP-domain, sitting in an A'-position (Alexiadou and Anagnostopolou 1998;Barbosa 1995;Frascarelli 2007; among others).However, I depart from these authors in that I claim that this is not a topic position, rather it is a focus position.
The informants have been divided into 2 different groups.The first group includes areas from northern Spain and Madrid (this is what I will call Standard Spanish), whereas the second group comprises the south (corresponding with Southern Peninsular Spanish).My initial hypothesis is that SPS speakers accept focus fronting when this focus conveys purely new information, whereas SS don not allow naturally for this type of focus fronting.This confirms the microparametric distinction argued for in this paper.
The article is organized as follows.In section 2 I discuss the different types of fronted foci that can be found in the LP.In section 3 I present the data and the test run among Spanish speakers.Section 4 discusses the results of the test.Section 5 summarises my main findings.

The focused constituent: different types of Focus
The label Focus is often applied to phrases performing two discourse functions, namely (a) the introduction of new information (IF) and (b) the introduction of a contrast (CF) (Kiss 1998).A number of scholars have argued in favour of a clear-cut distinction between these discourse categories, based on syntactic, phonetic, phonological and discourse factors (cf.Kiss 1998, Zubizarreta 1998, Nespor & Guasti 2002, Donati & Nespor 2003); Cross-linguistic evidence supports the necessity of a distinction between different types of Focus, which is syntactically encoded (cf.Molnár, 2006, Bentley 2007, Cruschina 2012, Bianchi & Bocci 2012, Bianchi 2013), and differently interpreted at the interfaces (Frascarelli & Ramaglia 2013).
Within Generative Grammar two main approaches to the notion of focus ca be found.In the Alternative Semantics approach (Rooth 1992, Beaver & Clark 2008), Focus generates a set of alternatives: given a question like "what does John want?", a set of propositions varying in the position of the focused direct object ({John wants coffee, John wants tea, ...}) constitutes congruent answers to it.
On the other hand, the Structured Meaning approach (Krifka 2006) instead partitions the proposition into a Focus (e.g., 'coffee' in the example above) and a background (the denotation of the rest of the clause, i.e. the property of being something that John wants).
When the different types of focus are addressed, we are confronted with different semantic operations, which are reflected in the syntactic derivation.Hence I concentrate on both interpretive and syntactic properties which are used to distinguish the type of foci.

Information Focus
According to the Structured Meaning approach (Krifka 2006), question-answer congruence requires that the Focus in the answer should correspond to the interrogative phrase of the question, and that the backgrounds should be identical.The element in the answer satisfying the information request in the question constitutes new information.This is illustrated in ( 12): (5) A: ¿A quién viste en la playa?In this dialogue the information provided by the object a Marta in the answer fully satisfies the information request in the question, and hence it stands as the Information Focus (IF).This focus is usually placed in final position in Spanish, as claimed by Zubizarreta (1998), Gutiérrez Bravo (2008) and López (2009).

Mirative Focus
Following Cruschina (2012), Bianchi et al. (2014, to appear)  Âmbar (1999: 41) terms a similar kind of fronting in Portuguese as evaluative construction, involving the fronting of an emphatic and evaluative element to a dedicated functional projection (i.e., the Evaluative Phrase) in the left periphery.Due to its unexpectedness, MF has some exclamative flavour Mirative is not dependent on a question-answer context.Contrast is established with an element that is part of the shared knowledge of the participants and can be semantically characterized as a "proposal to negotiate a shared evaluation" (Bianchi 2012).The set of alternatives is therefore very large.In example (6), the DP dos botellas stands as one alternative among many other alternatives

Corrective/Contrastive Focus
When Focus marks a constituent that is a direct rejection of an alternative, either spoken by the speaker himself ('not A, but B') or by the hearer, the Focus is 'Corrective'.Correction thus implies removal of information (cf.among others Gussenhoven 2007), as seen in ( 7), and can also apply to given material, as illustrated in ( 8): Correction implies a Focus-Background partition and the set of alternatives is very restricted (limited by the semantic properties of the rejected item).However, this type of Contrast may not be associated with a corrective import, as is shown in (9) from Frascarelli & Jiménez-Fernández (2013) and (10) from Kratzer (2004) Pure Contrast implies a Focus-Background partition (which can be created by the sentence itself) and the set of alternatives is restricted.In Romance languages, CF has a specific position in the LP.This is illustrated from Spanish in (10) from Hernanz (2011: 251, her (3a)): (11) LAS ACELGAS detestan los niños ( y no la pasta).
the chards detest-pres.3sgthe children (and not the pasta) 'Children detest chard, not pasta' As stated earlier, it is generally assumed that the focus position in the LP is for CF.Recall that MF can also occupy a focus position in the LP (see Cruschina 2012, Bianchi et al. 2014, to appear, and Jiménez-Fernández 2015 for crucial differences between CF and MF).However, IF has been claimed to always occupy a low position.Ortega-Santos (2005) and Jiménez-Fernández & İşsever (2012) suggest this position is the specifier of vP; cartographic analyses such as the one in Belletti (2004) argue for a designated Focus position in the low periphery.I leave this question aside since it does not affect the work presented here To summarise, Focus is encoded as a syntactic feature [+Focus] which characterises all types of focus, and conveys purely new information.The distinction of types is realised by different combinations of features (a general line pursued by Jiménez-Fernández 2015):

Methodology and data
Recall that my working hypothesis is that in Southern Peninsula Spanish IF can also occur in a high position at the LP.To prove the validity of this hypothesis, a test has been created for a systematic interface analysis of focus fronting (FF), for Mirative, Corrective/Contrastive Foci, but most especially for IF both in situ and fronted in two varieties of Spanish: Southern Peninsular Spanish (SPS, Andalusian and Extremaduran) and Standard Spanish (SS, Asturian, Catalan Sp., Basque Sp., and Madridian).
It is really complicated to identify what can be called Standard Spanish, as argued in RAE-ASALE (2009/2011), but given the unifying character of Spanish in general, it is taken to be the variety spoken by educated people in all dialects (though I am not taking into account American varieties).The divide between SS and SPS should thus be seen as geographically and linguistically motivated (see Alvar 2004, for the defining properties of SPS).
In the survey, all types of focus to be examined here have been taken into consideration.FF with the specific discourse functions of CF and MF have been systematically compared with FF with an IF function on purpose.In-situ IF has also been tested to make sure that informants were giving the right pragmatic interpretation to the relevant sentences.
Informants (27 SPS and 33 SS native speakers) were given a set of sentences preceded by a context inducing a specific focus reading of the preposed or in-situ constituent (judgments could be expressed as */??/OK).This was presented in writing.In particular, a sentence introducing the context was provided for speakers to react.This created the environment to induce the appropriate focus interpretation of both preposed and non-fronted foci.Speakers were confronted with sentences in a random order to avoid any sort of predetermined pattern in their answers.
In the examples of the experiment, sentences A constitute the stimulus context and focused constituents in sentences B are in underlined.For presentation in this work, I have grouped sentences depending on the type of focus and the type of FF involved.Also the object/subject asymmetry has been taken into account since movement of subjects does not necessarily have an impact on the linear ordering of sentence constituents.Examples (B) contain post-verbal focus, whereas those in (B') comprise preverbal focus.For the stimulus sentence I just provide the English translation, but for the sentences to be judged I offer both the gloss and the translation.Le gusta nada.

Feature-based typology of foci
her like-pres.3sgat.all 'She is eating pasta and it does surprise me since she doesn't like it at all.' For the sake of clarity, I have divided the data into five groups depending on the type of focus involved in the relevant sentence.In the first two groups either the subject or the object has the discourse function of IF.For examples in ( 12) and ( 14) a transitive verb is used and the reply in B shows the use of postverbal subject with a clear IF function, whereas that in B' illustrates the use of preverbal subject performing the role of IF as well.Sentences in (13) include the intransitive (unaccusative) verb ir 'go', but the discourse category of the only argument of this verb is also that of IF.
The third group contains IF on the object.Sentences in (15) convey new information via a non-animate object, whereas sentences in ( 16) and ( 17) involve IF developed by a human DP object, thereby requiring personal a.The reply in B shows an in-situ IF, hence postverbal object, while the reply in B' displays the possible use of a fronted IF.
The fourth set of sentences represent CF, making clear its contrastive flavour by inserting the alternative with explicitly negated.Again the reply in B involves in-situ CF, whereas that in B' contains a fronted CF.Finally, the fifth group illustrates MF, both in situ (reply B) and fronted (reply B').The mirative interpretation is induced by explicitly inserting the unexpected reaction against the assertion where MF has been used.

Results and discussion
In this section I present the results of the experiment carried out among SPS and SS speakers.Each different discourse category and its distinct variants will be followed by a brief discussion of the figures obtained.Figures include the number of speakers who gave a positive answer about the grammaticality of the relevant construction, alongside the percentage represented by this group of participants.

IF on subject (SPS) Position of Focus
Preverbal Speakers were faced with sentences in which the subject of a transitive verb functioned as IF.In an SVO language such as Spanish, the preverbal position for a subject does not necessarily indicate that it is dislocated in the CP-system.It may as well be place in spec-TP.As a consequence, with no other device than the context inducing the focus reading, speakers avoid the preverbal position and clearly prefer the postverbal position (for both SS with 82%, and SPS with 78% of OK answers), which has been described in the literature as unambiguously reserved for IF-subjects (Zubizarreta 1999, López 2009, Ortega-Santos 2006).The figures obtained in the survey confirm the validity of this analysis.However, the preverbal position for IF is not considered totally ungrammatical for both SS and SPS speakers.
preference to leave the element conveying the speaker's unexpectedness in situ, as put forth in Table 5.However, the high position is still an option in both SS and SPS, with 67% and 70% of positive answer respectively.Economy may be somehow the reason for this preference.

Conclusions
In this paper, I have shown that in some varieties of Spanish (SPS) a specific position for IF is available at the LP of the sentence.This position has always been argued to denote some sort of contrast.However, the data presented in this work support the view that the focus category detected in the CP may simply convey new information in SPS, thereby carry the feature [+Foc].This position, however, is not activated in Standard Spanish, which confirms that microvariation in SPS (with respect to SS) affects the Information Structure/Syntax interface.This microparametric variation has been supported by experimental evidence in the form of a test run among SPS and SS informants.
heard you met Fred yesterday.B: No, I met Bill.(8) A: I know John is going on vacation with Fred and Bill.B: No, he's only leaving with Bill.
and Jiménez-Fernández (2015), I consider Mirative Focus (MF) as not purely informative.It provides new information and additionally, based on the speaker's knowledge of the hearer's expectations, signals that such information will be unexpected: