Date: Tue, 31 Jan 2006 19:50:45 +0000 From: Kim Schulte Subject: Variación sintáctica en español: Un reto para las teorías de la sintaxis
EDITORS: Knauer, Gabriele; Bellosta von Colbe, Valeriano TITLE: Variación sintáctica en español SUBTITLE: Un reto para las teorías de la sintaxis SERIES: Linguistische Arbeiten 494 PUBLISHER: Max Niemeyer YEAR: 2005
Kim Schulte, Department of Hispanic Studies, School of Modern Languages, University of Exeter, UK
This book is an edited collection consisting of an introductory chapter by the editors and contributions by twelve different authors, most of which are based on conference papers presented at the 13th meeting of the German Association of Hispanists that took place at the University of Leipzig in March 2001. The book is written entirely in Spanish; the English translation of the title is ''Syntactic variation in Spanish: A challenge for syntactic theories.''
The stated aim of the book is to bring together a wide range of papers on syntactic variation with different theoretical and methodological approaches and backgrounds, in order to narrow the gap between theoretical explanation and empirical description. The contributions can be subdivided into two distinct groups. The introductory chapter and the following three papers investigate theoretical issues such as the compatibility of theoretical frameworks with corpus-based studies, the notion of grammaticality, and the comparison of different methods of analysis of variation present in corpora. The remaining papers discuss specific instances of syntactic variation in Spanish, resorting to a variety of different explanatory and methodological approaches.
In what follows, I will briefly summarize of each contribution to this volume, providing English translations of the titles.
1. Gabriele Knauer and Valeriano Bellosta von Colbe: ''Syntactic variation as a theoretical challenge: an introduction” The editors begin their introductory chapter by contrasting two approaches to linguistic analysis, the ‘anomalist’ and the ‘analogist’ approach. To the ‘anomalists’, languages is essentially irregular despite some identifiable regular patterns; they view the abstract system of language as the result of the sum of utterances in a linguistic community, which in turn makes the use of corpora an important prerequisite for a correct linguistic analysis. The ‘analogists’, on the other hand, view language as essentially regular and are concerned with the competence of the individual and how the speaker generates language; their preferred methodological approach is introspection by individual speakers.
Following this, the concept of linguistic variation is discussed in some detail. The authors observe that most theoretical frameworks are reluctant to accept the existence of syntactic variation, whilst it is generally accepted that variation exists at other levels of linguistic description, as shown by the presence of relatively uncontroversial concepts such as allophony, allomorphy, synonymy and polysemy. A crucial problem is the so-called 'theoretical paradox of variation', caused by a clash between fundamentally different analyses of absolute and partial synonymy.
The authors then discuss why theoretical frameworks have been reluctant to accept or include syntactic variation in their analyses. An exception to this trend is the neurocognitive approach of 'competing motivations' (e.g. DuBois 1985); otherwise, it is almost exclusively the field of sociolinguistics that has concerned itself with the phenomenon over the past decades, leading to an increasing divide between theoretical linguistics and sociolinguistics.
The final section of the introduction provides a brief overview and summary of the papers in the volume.
2. Guido Mensching: ''Syntactic variation, corpus linguistics and generative grammar: theories, methods and problems'' This paper discusses the role and usefulness of the study of syntactic variation in a generative framework. Mensching argues that the internal language of the individual (I-language) must be the object of (generative) linguistic analysis. Corpora, representing the external language of a speech community (E-language), are thus of very limited use, as the only insight they provide is the presence of different structures in the grammars of different members of a speech community. Only if it can be shown that there is variation in the E- language of individual speakers is variation of interest to the generativist syntactician, who then has various options of analysis. Such variation might be due to 'actuation phenomena' outside his I- language, triggered by socio-linguistic factors etc., or to the availability and choice of different lexical options.
As the object of generativist analysis is the I-language, and the appropriate and preferred tool to identify grammatical structures is introspection, the frequency with which a particular structure occurs is of little importance: infrequent structures may be marked, but what counts is only whether or not the speaker considers them to be grammatical. However, as the tool of introspection is not available for past stages of a language, Mensching defends the use of corpora and their statistical analysis for diachronic syntactic studies: a decrease in the frequency of a structure over time, eventually leading to its complete disappearance, indicates that this structure has disappeared from the speakers' I-language.
3. Josse De Kock: ''Corpus, frequency and grammaticality: thirty competing constructions in three corpora'' In this paper, the author picks fifteen variable structures that can take different syntactic forms but are semantically equivalent. By comparing the frequency with which these alternants occur in a corpus of informative prose texts, he establishes the degree to which each structure is present in the linguistic system, i.e. the 'degree of grammaticality' of each alternant. Based on their relative frequency, structures can thus be positioned along a scale of grammaticality; the author also observes a correspondence pattern between low relative frequency of a structure and the degree of speakers' doubts regarding its grammaticality.
A statistical analysis of the occurrence of the same variant structures in corpora of spoken Madrid Spanish and spoken Buenos Aires Spanish reveals that their relative frequencies differ from those in the informative prose texts. The author analyzes this as evidence that structures do not have a single, fixed 'degree of grammaticality', but that it depends on factors such as medium and place. This means that speakers' grammaticality judgements will always depend not only on the environment in which they have acquired the language, but also on the environment in which they are using it.
4. Nicole Delbecque: ''Corpus analysis as a tool for cognitive grammar: towards an interpretation of the alternation between SV and VS order.'' This paper deals with different theoretical approaches to a specific issue, the factors determining the position of the subject in Spanish. To a certain extent, this contribution is an academic autobiography of the author, who describes how and why her own approach has changed over time.
Her initial analysis focuses primarily on the corpus-based quantification of language-internal factors such as constituent length or the thematic role of the argument, with the aim of setting up probabilistic rules that link these factors with the choice of subject position. This type of primarily formal analysis is, however, not fully satisfactory because the statistical findings tend to become an end in themselves, with little real explanatory value.
The next stage of development is a move towards a more functionalist approach, based on the semantic features of the verb, on its relation to the subject, and on ideas developed in text linguistics and discourse analysis, such as thematic progression. Though a number of patterns can be identified, the author nevertheless considers this approach unsatisfactory, as it does not provide a global, uniform explanation. She therefore moves on to a probabilistic cognitive analysis, in which markedness plays a crucial part. She identifies preverbal subjects as tending to be the point of departure for a 'flow of energy', whilst postverbal subjects are typically associated with a more stative perception of the event.
The paper concludes by emphasizing the advantages of combining the findings of quantitative analysis, the identification of frequent lexical and syntactic patterns, and the application of cognitive parameters, to arrive at a more holistic understanding of syntactic variation.
5. Alicia González de Sarralde: ''On the relation between subject position and narrative structures'' This paper also deals with the variability of the subject position in Spanish, choosing a cognitive-functional approach. The author's corpus, consisting of 29 different speakers' descriptions of the exact same sequence of events, allows for an onomasiological and semasiological analysis. The onomasiological analysis reveals the different strategies by which the same state of affairs, with the same subject referent and the same degree of verbal agentivity, can be expressed; the semasiological analysis provides insights into what it is that is expressed by the choice between preverbal and postverbal subject position. The analysis reveals that subject postposition is not uniquely associated with a single function, but that four central functions can be identified as typically associated with postverbal subjects.
6. Valeriano Bellosta von Colbe: ''Syntactic variation in 'Role and Reference Grammar': object position in ditransitive clauses.'' This paper investigates the motivations underlying the varying position of the direct and indirect object in ditransitive clauses, within a 'Role and Reference Grammar' framework. Combining corpus analysis and the grammaticality judgement of individuals, the author identifies a number of competing semantic, syntactic and pragmatic factors that are instrumental in determining the constituent order of such clauses. In an analysis based on the principles of Optimality Theory, the author comes to the conclusion that the competition between these factors is resolved in different ways, depending on the context and communicative situation at the moment of utterance.
7. Pedro Martín Butragueño: ''The prosodic construction of the focal structure in Spanish'' Applying an autosegmental analysis of intonation to recordings of Mexican Spanish speakers, this paper examines the highly complex relations between prosodic, informative (pragmatic), and syntactic focus. The author identifies a minimum of three different intonational patterns that are used to mark the prosodic focus of a sentence, and he also distinguishes two different types of informative focus: neutral and contrastive.
Arguing that focus cannot be accounted for in purely syntactic terms, and refining a proposed analysis by Zubizarreta (1998, 1999), he convincingly shows that there is little syntactic predictability if intonation and prosody are left out of the equation. Instead, it is a combination of a constituent’s syntactic position and its prosodic realization that allows the speaker to mark the focus of a sentence.
8. Amparo Morales: ''Language acquisition in Puerto Rican children: on the null subject hypothesis'' This paper deals with the null-subject parameter. Whilst a binary distinction between languages that do and don’t allow deletion of pronominal subjects is usually taken for granted, this paper shows that the situation can, in fact, be more complex.
Whilst standard Spanish is a typical case of a null-subject language, Caribbean Spanish is far more permissive of variation between subject pronoun deletion and retention. Considering that other syntactic patterns which are normally characteristic of non-subject-deleting languages are also optionally present in Caribbean Spanish, it would appear that this variety lies somewhere between the two prototypes.
Analysing the use of pronominal subjects in the speech of children during language acquisition, the author shows that statistically, there are some subject pronouns that are more prone to deletion than others. On the one hand, anaphorically used pronouns resist deletion to a far greater extent than deictically used ones, which may be attributed to a later acquisition of the concept of anaphora.
A clear distinction can be made in terms of grammatical person. The 1st and 2nd person singular pronouns tend to be retained to a far greater extent than other pronouns, which the author interprets as a pragmatic manifestation of children’s self-centred perception of the world, in which the explicit emphasis of the contrast between themselves and the other interlocutor is highlighted by frequent explicit pronominal reference. This, the author suggests, may be supported by the possibility of using subject pronouns as topicalization markers in adult Caribbean Spanish.
9. Ulrich Detges: ''The grammaticalization of prepositional accusatives in Ibero-Romance: a pragmatic hypothesis'' This paper investigates the historical motivation for the present-day variation between direct objects with and without the 'prepositional' marker ''a''. Generally speaking, prepositional marking of objects is often pragmatically motivated, increasing the relevance of the respective object. In medieval Ibero-Romance, ''a'' functions as a focalizing particle for the rheme of a sentence, whilst it can have at least three different functions in the theme of a sentence: contrastive focus marker, non-contrastive focus marker, or non-focal thematic linking element.
In modern Spanish, the marker 'a' has developed an important discourse pragmatic function, being used by speakers as a strategy of forcefully taking or retaining their turn in conversation. This usage has its origin in the original focalizing function of ''a'' (i.e. contrastive focus between the speaker and the other interlocutor(s)), but the focalizing function has gradually been bleached away in favour of a purely discourse pragmatic use.
Expanding his analysis regarding strategies of turn-taking, the author identifies a link between three superficially unrelated phenomena: left- dislocation of direct objects, anaphoric reference to such objects by means of a 'redundant' clitic, and the use of the 'prepositional' object marker itself.
10. Eugeen Roegiest: ''Pronominal variation in Spanish: the dative pronoun between syntax and semantics'' This paper investigates the phenomenon of 'leísmo', i.e. a loss of opposition between direct and indirect object pronouns when referring to human referents, with the originally indirect object pronoun ''le'' also being used for direct objects. Based on a quantitative analysis of a corpus of modern Spanish authors, it is shown that there is considerable variation in the degree to which individual authors/speakers retain the opposition for masculine referents, whilst the frequency of the dative pronoun for feminine direct objects is relatively low but stable among authors.
It is then examined which syntactic contexts favour the use of ''le'' for direct objects. For constructions which can be either bivalent or trivalent, there is a strong tendency to retain the indirect object pronoun for the constituent that is the indirect object of the trivalent construction, but the direct object of the bivalent one. A similar phenomenon is identified for factitive infinitival constructions, in which the strong semantic cohesion between the two predicates leads to their syntactic fusion, effectively making the construction trivalent. From a semantic point of view, the more active or agentive the referent is perceived to be, the more likely it is that the dative pronoun is chosen. This can be explained by the fact that indirect objects are intermediate between prototypical agents and prototypical patients in terms of their agentivity.
The second environment examined, bivalent constructions that can have either a direct or an indirect object (verbs of perception or expression of emotion), can be understood to be influenced by the same agentivity parameter, with the direct object pronoun being the preferred option when the focus is on the act of perception rather than on the perceived action and its agent.
11. Rena Torres Cacoullos: ''Syntactic variation in diachronic perspective: the intensifying dative'' This paper examines the non-argumental use of the indirect object pronoun for pragmatic purposes in Mexico and New Mexico, e.g. the ''le'' in ''ándale'', where the pronoun has no referent, but instead functions as an intensifier of the action expressed by the verb. This usage can be contrasted with the 'ethical dative', which does make reference to a specific (evaluating) individual.
One important factor in the development of this pronominal usage is a diachronic process in which ''le''-constructions have increasingly lost their transitive value. This is reflected in the gradual decline of 'leísmo' in Mexican Spanish between the 16th and the 19th centuries, for which a diachronic frequency analysis is provided. The second factor is the loss of the pronoun's referentiality, which is reflected in the co- occurrence of an indirect object and ''le'' (originally itself an indirect object pronoun) in the same clause.
12. Irania Malaver: ''A comparative analysis of adjectival expressions of age in the speech of Seville and Caracas'' This paper examines the choice and variation of the copular verb linking a (human referent) noun with an adjectival expression of age, comparing the patterns found in Caracas and Seville.
In the Spanish of Caracas, the choice implies a semantico-pragmatic distinction, with the copula ''ser'' being used to assign a person to an age group, i.e. classifying that person, whilst the copula ''estar'' is used for a more subjective characterization or judgment regarding the person's behaviour or appearance. However, this opposition is only present in the discourse foreground; in the discourse background it is neutralized, leading to an indistinct use of either of the two copulas in such contexts.
In the Spanish of Seville, on the other hand, adjectival expressions of age are almost exclusively linked by ''ser'', both in fore- and backgrounded contexts; the foreground distinction, made by means of copula choice in Caracas, is achieved by other (e.g. lexical) means in the speech of Seville.
13. Dexy Galué: '''Me acuerdo que...': pronominal verbs and the 'queísmo' phenomenon'' This paper examines the variation between the presence and absence of the preposition ''de'' before object clauses. The absence of the preposition where it is normatively required is referred to as 'queísmo'.
The analysis is based on a sociolinguistic corpus of Caracas speech. In a probabilistic approach, the author examines the degree to which four different variables influence this variation: (1) the syntactic structure of the main clause, (2) linguistic material separating main verb and subordinate clause, (3) presence/absence of phonological sequences similar to ''de'' preceding the subordinate clause, and (4) whether the referent of the main clause subject is the speaker, and whether he identifies with, or disassociates himself from, the content of the subordinate clause. The statistical analysis reveals that only (1) and (3), as well as the speakers' socio-economic background, are relevant variables. In particular, the nature of the main verb is identified to be relevant; reflexive verbs, which normatively require the presence of ''de'', are subject to 'queísmo' particularly frequently. This is viewed as a possible analogical extension of the non-reflexive subordination pattern without 'de', and it is analyzed as a syntactic reorganization of the construction: where ''de'' is omitted, the conjunction ''que'' is, according to the author, reanalyzed as taking the place of the preposition separating the two clauses.
This volume clearly achieves its main aim of bringing together a wide variety of approaches and methodologies used in the analysis of syntactic variation. It has a good balance between primarily theoretical papers and more specific case studies.
The main insight that can be gleaned from this book is the fact that syntactic variation cannot be explained or accounted for if syntax is viewed on its own; other variables such as pragmatic and sociolinguistic factors, but also formal elements such as prosodic structure, have to be factored into the equation to understand the mechanisms at work. In particular, the many different variables identified as relevant in this volume show how complex and multi- faceted the interaction between syntax and other levels of linguistic analysis is. In this respect, the contribution by Guido Mensching is perhaps of particular value, as it shows why formal syntactic approaches such as generativism/minimalism are of little value for the analysis of syntactic variation: By excluding most factors that can determine the choice between syntactic structures from the domain of the analysis, such models have little explanatory value. WHY a speaker makes a particular choice, and what it is that triggers this choice of one syntactic structure over another in a particular utterance, is a crucial question that, as shown by this volume, requires an analysis that goes beyond the purely syntactic domain.
Many papers in this volume are characterized by the extensive use of corpora and probabilistic analyses of which contexts are more likely than others to trigger the use of one syntactic variant rather than another. One great strength of this methodological approach is that it can identify patterns of syntactic variation even where the speaker has a genuine choice between competing structures; it can identify tendencies even where the choice in any individual utterance is unpredictable. In a diachronic perspective, such tendencies or statistical preferences are, of course, a likely starting point for a subsequent entrenchment of the more likely or 'preferred' structure in a particular context. Synchronically, it casts some doubt on the view that grammaticality is a binary property, as shown in the contribution by Josse De Kock, who argues that a syntactic structure's degree of grammaticality can be variable, even for a single speaker.
Though some of the more data-based papers in this volume could have done with a more extensive theoretical discussion of their findings, and in some cases of existing work dealing with the respective phenomenon, the standard of the contributions is generally high, and they provide genuinely new insights into the range of factors underlying the phenomenon of syntactic variation.
DuBois, John, 1985. 'Competing Motivations', in ''Iconicity in Syntax: Proceedings of a Symposium on Iconicity in Syntax, Stanford, June 24- 26, 1983'' ed. by J. Haiman, 343-365. Amsterdam/Philadelphia: Benjamins.
Zubizarreta, María Luisa, 1998. ''Prosody, Focus and Word Order.'' Cambridge, MA: MIT Press.
Zubizarreta, Maria Luisa, 1999. 'Las funciones informativas: tema y foco', in ''Gramática descriptiva de la lengua española, Vol.3: Entre la oración y el discurso. Morfología'', ed. by I. Bosque and V. Demonte, 4215-4244. Madrid: Espasa-Calpe.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Kim Schulte is lecturer at the University of Exeter, UK, where he teaches Spanish, Portuguese and Romance linguistics. His research interests include pragmatic causation in syntactic change in a comparative Romance perspective, the evolution and emergence of non-finite structures, and contact-induced language change.