Language Documentation: Practice and Values edited by Lenore A. Grenoble and N. Louanna Furbee (John Benjamins Publishing Company) Language documentation, also often called documentary linguistics, is a relatively new subfield in linguistics which has emerged in part as a response to the pressing need for collecting, describing, and archiving material on the increasing number of endangered languages.
Language Documentation details the most recent developments in this rapidly developing field with papers written by linguists primarily based in academic institutions in North America, although many conduct their fieldwork elsewhere.
The articles in Language Documentation position papers and case studies focus on some of the most critical issues in the field. These include (1) the nature of contributions to linguistic theory and method provided by documentary linguistics, including the content appropriate for documentation; (2) the impact and demands of technology in documentation; (3) matters of practice in collaborations among linguists and communities, and in the necessary training of students and community members to conduct documentation activities; and (4) the ethical issues involved in documentary linguistics.
The papers in Language Documentation stem from a collaboration of several years sponsored by the Linguistic Society of America (LSA) and funded by the National Science Foundation. At the request of the Society, scholars involved in the documentation and archiving of endangered languages engaged in discussions among themselves on issues arising in this new field in the profession. Among their activities, they planned a conference on these topics, which was held during the LSA Linguistic Institute at MIT/Harvard in 2005. Originating in that conference, this book presents a statement about the content and conduct of language documentation at a turning point in its development, when it only recently has become a recognized area in linguistics. Edited by Lenore A. Grenoble, University of Chicago and N. Louanna Furbee, University of Missouri, Columbia, archivist for the Linguistic Society of America, the volume is organized around position papers and case studies that identify and illustrate existing possibilities and inadequacies, as well as desirable directions for the growth of the enterprise. The editors believe that Language Documentation should be seen both as a characterization of challenges afforded by language documentation at this point in time and as a set of informed suggestions for directions to be pursued.
The group of specialists engaged in these considerations represented the stakeholders in the emerging activity of language documentation of endangered languages, such as major archiving projects and electronic repositories for endangered languages, funding agencies, standards-setting initiatives, training programs, indigenous communities, and the LSA Committee on Endangered Languages and Their Preservation. Initially numbering 21 people, the group of participants became known as the LSA Conversation on Endangered Languages and Their Archiving, or simply the conversation group. A significant number of people joined the original conversation group and added their perspectives and expertise. The following describes the goals and contents of Language Documentation.
The conversation group organized a conference event, "Extending the LSA Conversation on Archiving Endangered Languages," which formed a final session of the conference on July 11, 2005, and which included about half of the original conversationalists, plus several other conference participants.
The interest groups collaborated throughout the next six months and then reported back at the 2006 LSA Annual Meeting, January 5-8, 2006, in Albuquerque. They reported to the Committee on Endangered Languages and Their Preservation, the Committee on Computing, and the Executive Committee. Suggestions ranged from resolutions such as one supporting the International Year of Languages, and another urging acceptance of documentary studies of endangered languages as appropriate topics for dissertations, to quite meaty and sometimes controversial works on issues such as (1) the state of the field for endangered language documentation; (2) linguistics, the public sector, and the documentary team, and (3) enhancing creative interactions among scholars, avocationalists, and students.
The "Conference on Language Documentation: Theory, Practice, and Values" examined issues that the process of language documentation raises for linguists, heritage speakers, and their respective communities. It attempted to set directions for aspects of the documentary enterprise (e.g., collaborations, appropriate training of field linguists, ethical issues, the documentary linguists' role in language revitalization efforts), as well as to present case histories of attempts at documentation and approaches to documentation and training.
Many of the papers in Language Documentation originated as posters or talks, reworked to fit the themes of the volume, focusing on key issues that emerged from the original topics that the conversation group identified the role of technology in documentation; the push among linguists, activists, and community members to jointly define language documentation and revitalization projects; and, above all, the ethical and moral issues underlying all of this work.
The contributors to Language Documentation range from seasoned linguists to undergraduate students, as well as to dedicated activists and community members, who all share a sense of commitment and enthusiasm for the hard work of language documentation. Although they present many perspectives, their works all exhibit a preoccupation with the ethical practice of language documentation. As those persons labor to save languages that are endangered, or at least save a persistent and useable record of them, they are more concerned with the impact of the manner of their work than many of their predecessors have been. This preoccupation makes their suggestions especially interesting since many offer truly original ways of incorporating and accommodating the interests of the communities who speak or once spoke these languages.
The contributors bring a wealth of experience working with different languages and communities to the discussion, and expertise in all aspects of the documentation process. Useful reading for anyone contemplating, embarking on or engaged in a language documentation project. Marianne Mithun, University of California, Santa Barbara
This is an indispensable volume. A terrific collection of rich, readable, thought-provoking, and very practical chapters. Jane Hill, University of Arizona
The drive to document languages is a new pressing imperative, but a dense thicket of issues intellectual, practical, social, ethical threaten to frustrate their attempts to fulfill it. This book points out the hazards, and charts a path through them, combining focused position papers with the revealing experiences of dozens of practitioners. Nicholas Ostler, Foundation for Endangered Languages
An exciting, wide-ranging exploration. It highlights the roles of technological advances and of ethical considerations in moving fieldwork from a solo enterprise to a multipurpose enterprise undertaken by and for diverse stakeholders. Nancy Dorian, Bryn Mawr College
Language Documentation will again extend the LSA Conversations on endangered languages and their archiving, continuing the creative engagement of the participants who helped to create it all the conversationalists, all the conference participants, and all those who prepared post-conference papers to situate and orient the sections.
A Romance Perspective on Language Knowledge and Use: Selected Papers from the 31st Linguistic Symposium on Romance Languages (LSRL), Chicago, 19-22 April 2001 edited by Rafael Nunez-Cedeno, Luis Lopez, and Richard Cameron (Amsterdam Studies in the Theory and History of Linguistic Science. Series IV, Current Issues in Linguistic Theory, V. 238: John Benjamins Publishing)
The essays summarized below offers a snapshot of the present volume. It contains a collection of twenty-one articles presented at the 31st Linguistic Symposium on Romance Language which was hosted by the University of Illinois-Chicago on April 19—22 of 2001. The content has been organized thematically to cover a variety of theoretical issues ranging from phonology, morphology, and syntax to their contextual use in Romance linguistics as seen through pragmatics and sociolinguistics. We are thus capturing the essence of the sixty-one topics treated in the conference which are summarized next, ordered alphabetically by sections.
In the section on phonology and morphology, Julie Auger examines the morpho-phonological status of pronominal clitics in Picard French. She questions whether these elements are independent syntactic elements or morphological affixes, thus echoing a previous hypothesis in which she suggested that subject clitics in Picard have been reanalyzed as affixal agreement markers. While her evidence for agreement marking was strong, the affixal analysis in that essay was supported by fewer arguments. In this article, Auger provides additional evidence showing that the affixal analysis was indeed the correct one. A comparison of the behavior of vowel epenthesis at word boundaries, inside clitic + verb clusters, and inside words, confirms that clitic + verb clusters behave more like morphological words than like syntactic phrases.
While looking at the oftentimes researched topic of syllable-final /s/ in Spanish dialectology, Teresa Brown and Torres Cacoullos compare /s/ reduction in both initial and final word and syllable positions in data from northern Mexico. Contrary to the prevailing assumption that Is reduction diffuses from syllable final to syllable initial position, they find higher aspiration/deletion rates in syllable initial position (as in /e.se/) than in syllable final position (/es.te/) and, in contrast to well-studied Caribbean dialects, higher reduction rates in prepausal and prevocalic than in preconsonantal contexts. Variable rule analyses reveal that preceding phonological environment is more important syllable initially, while following environment
has a greater effect syllable finally. At the same time, lexical factors are involved in both syllable positions. The authors conclude that syllable initial aspiration will occur in those Spanish dialects in which syllable final /s/ reduction is favored more before a vowel than before a consonant and occurs at relatively low rates.
Moving from synchronic analyses of Romance languages to a diachronically focused research, Fernando Martinez-Gil studies various phonological mechanisms involved in the emergence of the so-called intrusive consonants in Old Spanish and Old French. The addition of consonants involves a type of epenthesis that breaks up heterosyllabic consonant-liquid clusters of raising sonority, a process commonly described in terms of syllable phonotactics as a repair strategy that targets bad syllable contacts. His Optimality theoretical account, based on the framework of Correspondence Theory, shows that a formal analysis of this type of consonantal epenthesis enjoys several important advantages over standard serial accounts of the process.
Also based on Optimality Theory is Mario Saltarelli's essay on raddoppiamento. He re-examines this phenomenon, which has been considered an independent phono-syntactic rule, and proposes instead that it is an effect of the interaction of general quantity restrictions regulating the duration of vowels and consonants in word peripheral and medial contexts. Saltarelli shows that visibility of lexical stress contrasts at phonetic interface requires durational enhancement of the syllable lengthening of the vowel medially and of the consonant peripherally (radoppiamento). The lengthening asymmetry is favored by the location of geminate consonant contrast to word medial position.
Next we turn to the section of this collection in which we have grouped research that may be identified as Pragmatic or Sociolinguistic in method and con-tent. Pragmatics and Sociolinguistics are generally regarded as two different fields of inquiry. However, they interface at many different points. Both fields share similar objects and methods. Researchers from both fields are interested in structural and functional elements of language. By and large, data comes from language in use and, as such, researchers may share interests in issues such as acts, actions, ideologies, or significant categories of social experience such as gender, ethnicity, and class. Both fields of study may borrow techniques from one another, such as the use of statistics to resolve issues of syntactic analysis or the use of Speech Act categories to say something about cultural difference. Likewise, both fields may build upon or within neighboring theoretical frameworks in pursuit of saying something insightful about the relation of context to form or of form to the construction of identity. Therefore, when planning the conference parasession, we elected to include Pragmatics and Sociolinguistics under one heading so as to emphasize their interconnectedness and to provide for cross-fertilization.
Having said this, we may now look briefly at various representatives of Pragmatics and Sociolinguistics. Beginning with the work of Claire Beyssade, Jean‑Marie Marandin and Annie Rialland, we find a clear example of research in pragmatics which builds on neighboring theoretical frameworks. Specifically, they argue that "the lack of a conceptual framework within which pragmatic notions are given precise definitions has resulted in much confusion and contradiction in the literature". They propose to overcome this situation by providing clear, operational, definitions of focus, ground, given and discourse topic within an Illocutionary Semantics framework.
The work of Bonnie Fonseca-Greber and Linda Waugh is an illustration of cross-fertilization between sociolinguistics and pragmatics. Basing their work on tape-recorded conversations in everyday spoken Swiss French as well as on conversations, classroom discussions, and service encounters in Parisian French, the authors identify a complicated set of interrelated changes in progress which involve morphology, semantics, reference, and the syntax of pro-drop languages. These changes include the morphologization of first and second person clitics as inflectional prefixes with third person clitics lagging slightly behind. The in-definite on has replaced first person plural nouns as second person to has come to replace indefinite on in the expression of indefinite or generic reference. Also, third person plural elles is being replaced by ils. As a result of the changes (and others which they also identify), two key points emerge. First, French is acquiring a sharp difference between spoken and written language. Second, unlike many other diachronic studies of pro-drop languages in which change appears to go from null to non-null subject status, their work indicates a change towards null subject status.
The work of Javier Gutierrez-Rexach and Scott Schwenter turns to a topic more clearly aligned with pragmatics, the relation between form and scalar inferences. They note that although most Spanish negative polarity items (NPIs) belong to the class of n-words (nada, nadie, ninguno, etc.), there is a small class characterized by the property of having a full propositional form. The members of this class are mostly parenthetical expressions headed by the negative particle ni (ni lo pienses, ni sonarlo, etc.). Their contribution to the content of the sentence they attach to is to strengthen the associated denial. In this paper, they analyze another propositional NPI, que digamos, which exhibits clearly differential behavior with respect to the other propositional NPIs mentioned above. The essential pragmatic property of this NPI is its scalar nature, but, unlike other well-known NPIs (e.g. en absoluto), it does not signal an "extreme" point on a pragmatic scale. Rather, it invokes a pragmatic scale of expectations and indicates that the proposition it marks is less expected, or more unexpected, than some alternative proposition(s) located on the same scale. The resulting interpretation of the proposition marked by que digamos is an attenuated denial.
Michela Ippolito's paper, like that of Gutierrez-Rexach and Schwenter, investigates syntactic forms, the interpretation of which necessarily involves implicature. Like the work of Beyssade, Marandin and Rialland, discussion of these forms also entails reference to semantics. She discusses the modal uses of the Italian Imperfect both in main and embedded clauses. In particular, she offers an account of peculiar properties of Imperfect Conditionals, which distinguish them from both indicative and subjunctive conditionals. She provides an account of the modal uses of the Imperfect in main clauses based on the claim that past tense may not be interpreted inside the proposition where it superficially occurs. Cross-linguistic evidence is considered. Turkish conditionals offer syntactic and morphological support for the theory presented in this paper.
As with the work of Fonseca-Greber and Waugh, in the article by Heloisa Maria Moreira Lima Salles and Maria Marta Pereira Scherre, we find the use of quantitative sociolinguistic methods in pursuit of a syntactic analysis. Specifically focusing on indirect objects variably headed by the prepositions a or para, the researchers find a clear effect of two pragmatic constraints: +l– (potentially) lightness of the verb and +l– referentiality of the DO nominal. Their statistical findings provide a basis for subsequent analysis of argument structure in general which leads to the discovery of a curious parallel with English dative constructions that do not permit dative alternation.
The next article, by Rosina Marquez Reiter, shows the interaction of a classic topic in Pragmatics, speech act theory, with a sociolinguistic theme of dialect difference. This article examines the results of a contrastive empirical study of conventionally indirect requests in two varieties of Spanish: Uruguayan and Peninsular Spanish. The results reveal both pragmatic similarities and differences in the realization of conventionally indirect requests in these two language. varieties. Most of the pragmatic similarities were found at the level of the linguistic mapping of utterances, with both Uruguayan and Peninsular Spanish speakers showing a negative correlation between (in)directness and social distance. The less familiar the interlocutors are with each other, the more likely it is for their requests to be realized indirectly. On the other hand, differences were found in terms of the tentativeness conveyed by the requests. More specifically, Uruguayan Spanish requests were more tentative than those in Peninsular Spanish. This tentativeness was achieved by a more frequent and more varied use of external modifications of the down-grading type. It is argued that differences in the tentativeness conveyed by those requests might form the basis of generalized perceptions by which Spaniards are seen as more a direct than Uruguayans.
Dialect differences give way to a study of dialect formation in a context of language contact in the work of Ariana Mrak. Focusing on the Spanish spoken by the Mexican-American community in Houston, Texas, the researcher evokes arguments of simplification in attempting to determine if third person accusative clitics are undergoing a change in progress. If these accusative pronouns are progressively being lost from use, then a series of hypothesized consequences may.follaw. These include reiteration of the NP, use of the pronoun eso, expansion of dative clitics into accusative slots, duplication, reassignment of gender and/or number, and finally, omission of the clitic. Results indicate that of these various possibilities only three appear in the data. The most frequent, aside from use of the accusative clitics, is repetition of the NP. In turn, speakers may also express clitics that do not agree in number or gender and, finally, speakers may also substitute, though infrequently, dative clitics for the accusatives.
Though also based in an extensive corpus of spoken language, the work of Francisco Ocampo on the expression of topic in the spoken Spanish of La Plata, Argentina, goes to a key topic in functional syntax: topic and word order. Evoking work on information status and the structure-building framework of the psycholinguist Morton Ann Gernsbacher, Ocampo provides a succinct analysis of the interaction of topic shift or continuity with information status, word order, and stress (primary, secondary, and tertiary). Yet, the correlations between these various elements prove to be strong statistical tendencies, not absolutes. Such a finding, of course, is very much in keeping with quantitative sociolinguistics which finds systematicity in statistical patterns of use.
The final article from Pragmatics and Sociolinguistics is the contribution of Ricardo Otheguy and Naomi Lapidus. Their work utilizes and explores such methods and topics as quantitative analysis, issues of reference, change in progress, and simplification in contexts of language contact. Briefly, they note that the notion of simplification has often been proposed as an explanation for contact-induced change, a topic we see in Mrak's work. In a modest but significant extension of the explanatory value of simplification, they suggest that contact-induced changes should additionally be understood in terms of the notion of adaptation. In turn, they apply the notion of adaptation to the study of that most common of all con-tact phenomena, namely cross-language loan lexical insertions. In this case these insertions are English nominal lexical insertions in the Spanish spoken in New York City. The use of adaptive reasoning permits tests of specific predictions regarding simplificatory changes involving these nouns. In particular, they focus on the nearly complete lack of applicability of the Spanish arbitrary gender system to this portion of their informants' vocabulary. The data come from 33 sociolinguistic interviews conducted in New York City with speakers from a variety of age groups and of Latin American national origins. These speakers also differ with respect to whether they were born in New York or, if born in Latin America, in the length of time they have lived in the City. The new facts which are revealed by these predictions may be seen as useful expansions of our empirical knowledge regarding Spanish in the United States, and as support for the idea that contact-induced change is adaptive.
The final section includes a selection of papers on syntax, semantics and their interfaces, as well as a paper on the acquisition of syntax. Each of them presents a significant contribution to current debates in linguistic theory.
Tonia Bleam enters an old controversy from a Minimalist perspective: the structure of verbs that take two complements. Following intuitions that hark back to Juan Uriagereka's dissertation, Bleam presents new arguments that a double-object type of structure obtains in Spanish when the indirect object is doubled by a dative clitic. Differences between English and Spanish double object constructions are explained by independent features of these languages.
Alicia Cipria's paper centers on the formal semantics of "sequence of tense", which in the Spanish literature has been discussed mostly within the context of subjunctive complements. Perception verbs take indicative complements, yet the tense combinations (of main and complement) have certain restrictions, whose import is not as simple as previously claimed. The interaction of aspectlaktionsart and pragmatics with the lexical requirement of the main verb plays a role in the resultant temporal interpretations. Past tense complements are particularly interesting due to the different interpretations they receive when embedded under perception verbs.
Advancing the theoretical underpinnings of Optimality Theory, Lisa Davidson and Geraldine Legendre compare acquisition of the Catalan and French verbs, de-parting from the observation that while children acquiring French exhibit an "optional infinitive" stage (see the earlier work of Amy Pierce and Ken Wexler) Catalan learners do not (as reported by J. Grinstead). An examination of 3 French and 3 Catalan learners revealed further differences: (1) Catalan children use 3rd singular present forms as defaults, not NRFs. (2) Catalan children develop AGR first, whereas French children exhibit TNS first. (3) While Catalan children show a linear increase in the acquisition of both TNS and AGR, French children exhibit a stage where there is competition for the realization of functional morphemes. Further-more, both Catalan and French children show a gradual decrease in NRFsldefaults, which cannot be accounted for by traditional OT analyses. Such variation can be captured by an Optimality Theoretic account in which "floating" constraints prohibiting functional structure are balanced against constraints requiring faithfulness to inputs with TNS and AGR.
Viviane Deprez's article explores the syntax/semantics interface. She investigates the interpretations of bare nouns in several Creole languages. In earlier work, she argued that a Plural Parameter distinguishes two broad sets of languages ac-cording to whether or not their nominal projections necessarily contain a Num projection with a counter. In the present article she argues that this parameter, coupled with the observation that languages may have overt or covert determiners, provides an adequate account of the range of variation found among the bare nouns of Creole languages.
Jon Franco and Alazne Landa's contribution discusses definite null objects, which have been documented in two Spanish contact varieties: Andean and Basque Spanish. After arguing that Liliana Sanchez' earlier analysis for Andean Spanish cannot be imported into Basque Spanish, Jon Franco and Alazne Landa propose an analysis that suggests that Basque Spanish is a topic-oriented language, similar in some respects to Chinese, as analyzed by James Huang.
Kate Paesani's article proposes an analysis of auxiliary choice (avoir or titre) in pronominal verb constructions in the French passe surcompose. She hypothesizes that three key factors determine the Spell-Out form of the auxiliary compound in all syntactic contexts: (1) the presence or absence of an abstract DlP° head in the syntax; (2) participial AGR; and (3) the have for be, parameter. A distinct advantage of this analysis is that dialectal forms of the passe surcompose auxiliary compound are accounted for by a minor parametric variation of a type already motivated for Romance: Richard Kayne's have for be parameter.
Ever since Heles Contreras suggested that pre-verbal subjects in null-subject languages are in an A'-position, a rich literature has emerged with arguments pro and against this proposal. Margarita Super's Minimalist article argues that, at least in Spanish, the evidence indicates that preverbal subjects can be found in an A-position, specifically Spec,T. She uses distributional, interpretational and binding sources of evidence to make her argument. She further argues that Spanish does not resemble Greek, as described by Artemis Alexiadou and Elena Anagnostopoulou, in this respect.
Maria Luisa Zubizarreta's contribution focuses on wh-in-situ in French. Cedric Boeckx has recently identified three properties of wh-in-situ in French (for some speakers): exhaustivity, intervention effects and locality. Zubizarreta discusses the first two and adds a fourth: lack of pair-list readings. She argues that the intervention effects cannot be considered a syntactic, minimality type, phenomenon. Instead, they lie in the syntax/semantics interface. Additionally, she relates intervention with exhaustivity and the absence of pair-list readings. Her analysis hinges on her approach to contrastive focus, based on a `A-not-A' operator.
The Atoms of Language: The Mind's Hidden Rules of Grammar by Mark C. Baker (Basic Books) A major scientific breakthrough into the common elements of all languages, which give us a deeper insight than ever before into how the mind works.
Whether all human languages are fundamentally the same or different has been a subject of debate for ages. This problem has deep philosophical implications: If languages are all the same, it implies a fundamental commonality-and thus the mutual intelligibility-of human thought.
We are now on the verge of answering this question. Using a twenty-year-old theory proposed by the world's greatest living linguist, Noam Chomsky, researchers have found that the similarities among languages are more profound than the differences. Languages whose grammars seem completely incompatible may in fact be structurally almost identical, except for a difference in one simple rule. The discovery of these rules and how they may vary promises to yield a linguistic equivalent of the Periodic Table of the Elements: a single framework by which we can understand the fundamental structure of all human language. This is a landmark breakthrough, both within linguistics, which will thereby become a full-fledged science for the first time, and in our understanding of the human mind.
Perspectives on Negation and Polarity Items by Jack Hoeksema, Hotze Rullmann, and Victor Sanchez-Valencia (Linguistik Aktuell / Linguistics Today, No 40: John Benjamins) Excerpt: The study of negation and negative polarity items has seen a tremendous upswing in recent years. There have been more conferences on this topic in the 1990's than in the nine preceding decades of the 20th century. The present volume contains papers from one such conference, entitled `Perspectives on Negation' and held at the University of Groningen in August 1996. In these proceedings, one can find contributions on various aspects of the syntax and semantics of negation, including the old and notorious problem of scope, the vexing issues surrounding polarity items, the interaction of negation with presuppositions and implicatures, and the acquisition of polarity items.
Most of the papers in this volume are directly or indirectly devoted to matters of polarity sensitivity, from a variety of analytical perspectives. Questions to be addressed are, among others,
The first paper, by Jay Atlas, concerns itself with the first question, more precisely with the difference between anti‑additive and anti‑multiplicative quantifiers. Anti‑additive quantifiers are expressions such as nobody or nothing, whereas not everybody or not everything are anti‑multiplicative. Both classes of negative expressions differ from negation itself in that they each do not obey one of the De Morgan laws for negation (albeit a different one for each class). In a sense, then, both classes of quantifiers are weaker than negation, but neither is necessarily weaker than the other, just different. However, as pointed out in the work of Zwarts and others, it appears that anti‑additive quantifiers are stronger triggers for polarity items than anti‑multiplicative quantifiers. Atlas' paper tries to make sense of this observation, by noting that antiadditive quantifiers are truly more negative than anti‑multiplicative quantifiers when the matter is considered intuitionistically, rather than from the perspective of classical logic.
Anastasia Giannakidou's paper looks at items (such as the Greek indefinites of the kanenas‑paradigm) which impose weaker requirements on their licensing context than standard polarity items such as ever. She proposes that the general notion of nonveridicality, which includes, but is not restricted to the regular contexts of ever, is relevant for the semantic licensing of these items. Giannakidou's work has already been fruitful in that sensitivity to nonveridicality in other languages than Greek has become an object of investigation, and has turned up a number of cases that pattern much like kanenas.
The paper by Jack Hoeksema and Hotze Rullmann is concerned with issues of sensitivity and triggering (the first and second of the above questions) and studies two closely related adverbial expressions in Dutch, ook maar and zelfs maar. The former has figured quite prominently in the literature on Dutch polarity items, but its near equivalent zelfs maar has been much less well‑studied. From an extensive corpus investigation, it appears that these expressions differ along a number of dimensions: (a) scopal preferences, (b) choice of focus expressions to be modified, (c) triggering. For instance, triggering by non‑antiadditive expressions is significantly more common in the case of zelfs maar. In addition, it is shown that the division of labor between the two expressions is of recent origin, and has become more pronounced in recent years.
Larry Horn's paper is concerned with a wide variety of naturally occurring examples of indirect licensing, whereby the licensing element appears to be a pragmatically‑established negative implicature. His conclusion is that a strictly syntactic or strictly semantic account of polarity licensing will fail for these cases. The paper includes a discussion of squat and similar expressions of scatological origin, which have recently undergone part of the Jespersen Cycle, rendering them negative in force even in the absence of any overt negation: They told him squat = They didn't tell him squat.
Utpal Lahiri's paper discusses the licensing of scalar indefinite NPIs in Hindi. Ever since the work of Fauconnier and Ladusaw, scalarity has been viewed as an essential property of many polarity items. Lahiri extends his analysis of these NPIs (indefinites with the added scalar particle bhu `also, even') to cases of triggering inside definite noun phrases and correlative clause constructions.
Chris Kennedy's paper is concerned with the difference between "positive" adjectives such as long and their negative counterparts like short. It discusses ways in which the monotonicity properties of these adjectives can be formally captured without losing sight of other relevant properties, e.g. the impossibility of comparative subdeletion with antonymic adjectives, cf. #He is taller than she is short.
Issues of antonymy also crop up in Henny Klein's paper, which discusses polarity sensitive adverbs of degree. She shows that these adverbs exhibit multiple semantic and even connotational dependencies. For instance, some adverbs are sensitive both to the presence of negation and the presence of negatively evaluated adjectives or verbs. Both types of "negative" elements are related in ways which have only just begun to be explored. Klein argues that Israel's (1996) pragmatic theory of polarity licensing should be extended to account for evaluative (in her terms, "emotional") aspects of use.
The paper by Johan van der Auwera deals with the second question ("What types of expressions may be sensitive to polarity?"), as it concerns the interaction of negation with modality from a typological perspective. Many languages have modals which are reserved for negative statements. The various lexicalization patterns for combinations of modals and negation in 29 languages of Europe and India are compared in this paper and some general tendencies are identified. It turns out that there are multiple patterns of polarity sensitivity, both with weak (can, may) and strong modals (must),
The paper by the late Gabriel Falkenberg is likewise related to the second question, and discusses the German negative polarity verbs. He shows that these verbs do not form a random sample from lexicon, but fall into a number of semantic classes. The semantic classes involved carry over to related languages like Dutch and English. It is to be hoped that the groundwork laid in this paper will one day lead to a more general typological study of the factors which cause polarity sensitivity in this area. At the moment, only indefinite pronouns (Haspelmath 1997) and to a lesser extent modal auxiliaries (De Haan 1997; Auwera, this volume) have been studied in detail.
Anita Mittwoch's paper concerns the proper analysis of negative sentences involving negation in combination with the temporal connective until. The long‑standing debate on the issue whether until is sometimes a polarity item within the scope of negation, or a non‑polarity item scoping over negation, but sensitive to the aspectual effects of negation, is resolved here in favor of a double‑jointed analysis: until can have both these roles. The paper discusses not just English data, but compares them with data from other languages.
There are two papers in this collection that strictly deal with negation, as opposed to negative polarity phenomena. One, the paper by Kenneth Drozd, is concerned with the use of no and not in early language acquisition. Drozd provides arguments that presentential negation, which in later stages disappears, is used to express metalinguistic negation in the sense of Horn. If this is correct, this finding will have some consequences for the acquisition of polarity items as well. In particular, the expectation would be that negative polarity items are illicit in the scope of presentential negation.
The other paper, by Lucia Tovena, deals with the notorious problem of Neg-raising. The idea that Neg-raising should be treated as a transformation is nowadays no longer held. Tovena treats the phenomenon as a lexicalization of the "closed-world assumption." In Artificial Intelligence, the notion of negation as failure to find a proof is a natural one when dealing with finite data. It should be mentioned that Neg-raising has great indirect importance for the study of polarity items. Items which require a local (=clause-mate) licensor may lift this requirement in the case of Neg-raising.
The paper by Gertjan Postma, finally, deals with a number of issues. In giving an in‑depth analysis of various taboo terms used as polarity items in Dutch, he deals with the internal syntactic structure of these items (in terms of small clauses), licensing conditions and various other issues surrounding the grammar of polarity items. More than any of the other papers, this one brings out the rich complexity of the field of polarity items, and the theoretical challenges which they offer.
Phonetics and Phonology Interactive: The De Gruyter Interactive Introduction to Phonetics and Phonology by Jurgen Handke (Mouton: Walter De Gruyter) This self-guided course is designed for students who are attempting to master the broad areas of current phonetics and phonology. It lacks the detail that most people would want so basic texts are still needed. It works great as an introduction at undergraduate and high school level. (Platforms: Windows 95 up, Macintosh 8.1 or higher; 32 RAM, 30 MB hard disk space CD-ROM drive, Soundcard, SVGA graphics board or screen resolution 800 x 600).
Frequency and the Emergence of Linguistic Structure edited by Joan Bybee and Paul Hopper (Typological Studies in Language: John Benjamins) A mainstay of functional linguistics has been the claim that linguistic elements and patterns that are frequently used in discourse become conventionalized as grammar. This book addresses the two issues that are basic to this claim: first, the question of what types of elements are frequently used in discourse and second, the question of how frequency of use affects cognitive representations. Reporting on evidence from natural conversation, diachronic change, variability, child language acquisition and psycholinguistic experimentation the original articles in this book support two major principles. First, the content of people's interactions consists of a preponderance of subjective, evaluative statements, dominated by the use of pronouns, copulas and intransitive clauses. Second, the frequency with which certain items and strings of items are used has a profound influence on the way language is broken up into chunks in memory storage, the way such chunks are related to other stored material and the ease with which they are accessed to produce new utterances. Contributions by. Joan Bybee and Paul Hopper; Sandra A. Thompson and Paul I. Hopper; Joanne Scheibman; Naomi Hallan; Betty S. Phillips; Janet B. Pierrehumbert; Stefan Frisch, Nathan R. Large, Bushra Zawaydeh and David B. Pisoni; Mary L. Hare, Michael Ford and William D. Marslen-Wilson; Greville Corbett, Andrew Hippisley, Dunstan Brown and Paul Marriott; Daniel Jurafsky, Alan Bell, Michelle Gregory and William D. Raymond; Nathan Bush; Catie Berkenfield; Manfred G. Krug; Joan Bybee; K. Aaron Smith; Joyce Tang Boyland; Shana Poplack, Gertraud Fenk-Oczlon; Brian Macwhinney; Osten Dahl.
A legacy of the structural tradition in linguistics is the widespread acceptance of the premise that language structure is independent of language use. This premise is codified in a variety of theoretical distinctions, such as Saussure’s langue and parole and Chomsky’s competence and performance. A further premise of this legacy is that the study of structure is a higher calling than the study of usage and is a potentially more promising avenue for uncovering the basic cognitive mechanisms that make human language possible.
In contrast, outside linguistics it is widely held that cognitive representations are highly affected by experience. In humans and non-humans detailed tracking of probabilities leads to behavior that promotes survival. Even within linguistics, certain usage-based effects permeate the general lore that practitioners and theoreticians accept: unmarked members of categories are more frequent than marked members; irregular morphological formations with high frequency are less likely to regularize; regular patterns have a wider range of applicability; and high frequency phrases undergo special reduction. Many of these effects had been catalogued and described by George K. Zipf in a pioneering work from the 1930s, The Psycho-Biology of Language. Zipf is known these days chiefly for his "law that the length of a word is inversely proportional to its frequency and his explanation through the "principle of least effort. While this aspect of Zipf's work is often criticized he anticipated many of the themes of more recent investigations of the relationship between frequency and structure, such as the fusion of pronouns with auxiliaries in forms like we're, you'll, etc. and their significance for the genesis of inflection. Zipf coined the term "dynamic philology for the quantitative study of language change and its relevance for linguistic structure.
Zipf's work in linguistics was taken up only sporadically in the discipline as linguists focused their attention on the theoretical questions of how to define the structural units of language exclusively through local combinatorial possibilities. However, by the 1980s, a number of linguists had begun to think of linguistic structure (grammar) as a response to discourse needs, and to consider seriously the hypothesis that grammar comes about through the repeated adaptation of forms to live discourse. The parallel question of how experience with language as reflected in frequency could affect cognitive representations and categorization and thus the internalized grammar of language users also began to occupy researchers at this time in both linguistics and psychology. A related development, symptomatic of the increasing impatience with studies of individual "competence and growing suspicion regarding the reliability of intuitions as a source of data, was the rise in the 1990s of the new field of corpus linguistics. Starting from trends that had begun with "computational linguistics going back as far as the 1950s, corpus linguistics has been made possible by the exponential increase in data storage and high-speed processing. While the corpus is the prime tool for frequency studies in general, with many linguists it also serves as a heuristic for new facts about linguistic structure. One especially important claim coming out of corpus studies is that the dividing line between grammar and lexicon, which has been virtually a dogma in linguistics, cannot be sustained. Time and again the operation of linguistic rules has been found to be limited by lexical constraints, sometimes to the point where a construction is valid only for one or two specific words.
Increasingly, then, in many quarters structure has come to be seen not as a holistic autonomous system but as something more fluid and shifting. An influential concept here has been that of emergence, understood as an ongoing process of structuration. Structuration in recent sociology refers to "the conditions that govern the continuity and dissolution of structures or types of structures. Emergence in this sense is distinct from ontogenesis, which refers to the origins and development-the history-of an existent organism or of a system. By contrast, emergent structures are unstable and are manifested stochastically. The fixing of linguistic groups of all kinds as recognizably structural units (word and phrase units) is an ongoing process; it is the result at any point in time of the "constant resystematization of language. From this perspective, mental representations are seen as provisional and temporary states of affairs that are sensitive, and constantly adapting themselves, to usage. "Grammar itself and associated theoretical postulates like "syntax and "phonology have no autonomous existence beyond local storage and real-time processing. The notion of language as a monolithic system has had to give way to that of a language as a massive collection of heterogeneous constructions, each with affinities to different contexts and in constant structural adaptation to usage.
The notion of emergence constitutes a break with standard ideas about grammar that envisage it as a fixed synchronic system. It relativizes structure to speakers' actual experience with language, and sees structure as an on-going response to the pressure of discourse rather than as a pre-existent matrix. It follows that accounts of grammatical (and phonological) structure must take note of how frequency and repetition affect and, ultimately, bring about form in language. Now work on the notion that frequency of exposure and use is an important factor in the establishment and maintenance of linguistic structure has begun to branch out in many directions. One of the goals of this book is to represent some of the findings of this research.
The papers in Frequency and the Emergence of Linguistic Structure build on these two strands of research into language use-the heuristic of frequency and the metalinguistic principle of emergence-to illustrate certain general principles that are robustly documented by empirical investigations of various sorts: distribution in natural conversation, diachronic change, variability, child language acquisition, and experimentation. Two major principles are addressed here:
The distribution and frequency of the units of language are governed by the content of people's interactions, which consist of a preponderance of subjective, evaluative statements, dominated by the use of pronouns, copulas and intransitive clauses. The frequency with which certain items and strings of items are used has a profound influence on the way language is broken up into chunks in memory storage, the way such chunks are related to other stored material and the ease with which they are accessed.Each of the chapters of this volume treats several issues related to these two principles, so that organizing them thematically has been difficult. The organization we have settled is: Patterns of Use. These are papers that deal with patterns of occurrence of morphosyntactic structures in natural conversation; Word-level frequency effects, that is, the papers that deal with the direct and indirect effects of frequency of use on change and structure at the word level; Phrases and constructions, which contains papers that demonstrate that many of the same principles found at the word level also operate in multi-word sequences; and General. In this category are placed papers that reference and model multiple phenomena and therefore do not fit easily into the first three categories.
Derivation and Explanation in the Minimalist Program edited by Samuel David
Epstein, T. Daniel Seely (Generative Syntax, 6: Blackwell)
To sum up, we believe that we all "must become philosophical and must enter upon
a thorough criticism of the theory's own foundations" (Whitehead). This
willingness to first carefully examine "the objects of our own creative fancy"
(Pierce) and then to criticize them, and be "discontent with higher level
stipulation" (Reuland), underlies the highly characteristic effort to increase
explanatory depth by minimizing theoretical apparatus (Nash, as noted above,
and, of course, Chomsky's Minimalist Program). Minimization-throughderivation,
as discussed here, asks "Can we explain the laws by finding 'a way of thinking
such that the laws are evident'?" (Feynman). Can we grow or generate the
macroscopic ("sentential/syntactic") regularities from the bottom up? (J.
Epstein). These are the very kinds of questions, all regarding deduction and
explanation, which we think it is never wrong to ask (and always
counterproductive to denounce), that each paper in this volume seeks to
explore, in one way or another, and, in some cases, with productive disagreement
regarding the role of derivation in explanation.
contains a collection of articles by leading researchers, each of which
investigates aspects of the fundamental questions raised above. All papers are
concerned with explanation, consonant with Reuland's summary of the shift from
GB to Minimalism within which "Contentment with higher level stipulations has
been replaced by pursuing the question of why they would hold;" and all assume
some form of minimization. The majority develop a derivational approach, Brody's
contribution being an important exception. As for the more specific proposals
raised above, a few (all too brief) comments on the papers are in order.
research goal, as specified in section 2 above, is to deduce as much as possible
from the arguably necessary (but cf. Brody) syntactic structure-building
machinery; i.e. the universal and minimal rules and their mode of recursive
application. What is to be avoided, among other things, on this view is defining
relations on already-built-up trees, such definitions being, in our view,
nonexplanatory. Kayne and Zwart, for example, don't define binding-theoretic
domains on trees but rather attempt to account for constraints on the binding
of Y by X, under the assumption that X and Y are firstmerged as sisters (the
simplest rule-created relation) and then X undergoes independently constrained
movement. Thus, Kayne, "adopting the derivational perspective of Chomsky's
recent work, [explores] the idea that antecedentpronoun relations should be
rethought in movement terms, including what we think of as Condition B and
Condition C effects." This represents a significant shift from defining
binding-theoretic domains on trees to seeking an explanation of them in terms of
independently motivated, and seemingly ineliminable, apparatus (movement theory)
and a highly simplified version of certain referential relations (posited to
hold under sisterhood, as created by Merge when X and Y are concatenated by a
single rule application). In a related paper, Zwart seeks to account for both
local and nonlocal binding relations under a similarly derivational approach.
forms of minimization were reviewed above, including the minimization of levels
(at most the interface levels are posited), minimization of lexical items (the
interpretability condition: no features other than those interpreted at the
interface), and the minimization of syntactic objects (the inclusiveness
condition: no new features are introduced by Chl). Collins provides a classic
form of representational minimization, attempting to eliminate phrasal category
labels entirely, and argues that "as representations are simplified (e.g.,
labels are eliminated), it becomes necessary to articulate the precise nature of
syntactic derivations (e.g., minimize search, the Locus Principle, etc.)."
respect to Inclusiveness, Hornstein and Uriagereka argue that "[The] 'loss of
information' scenario [i.e. where information is available at one point in the
derivation but not at another] fits poorly into models which make use of
enriched representational codings (including traces, indices, and similar
elements) because whatever information may have been lost in the course of the
derivation can be [arguably incorrectly] reconstructed through some abstract
[and sometimes Inclusiveness-violating] coding." And Zwart, as part of his
development of a derivational approach to binding phenomena, argues that "with
each new operation Merge, the element merging to the structure acquires
additional features which are interpreted at LF, but no trace or chain is
created, and no information can be gathered from these entities."
Kitahara, "under the framework of Epstein et al. 1998, ... articulates how ...
mechanisms of scrambling interact with the interpretive procedures applying in
the course of a derivation." And McCloskey "argues that the form of the
complementizer in Irish A'-binding constructions depends not on the mechanism of
Specifier Head agreement (as has been widely assumed), but rather is sensitive
to the mode of introduction of the specifier material (whether it is introduced
by Merge or by Move)." Thus, mere inspection of the representational output is
"develop[s] an approach to the derivation that generates the material at the
root of the tree first, and adds new material to the bottom of the existing
structure," and explores "the surprising prediction that under certain
circumstances, A' movement should share with A-movement the property of being
unable to skip an intervening A-position."
seeks to account "for cross-linguistic variation involving subject-tosubject
raising over an Experiencer," arguing that "the analysis must be couched in a
framework of derivational economy that imposes strict derivationality...." Thus,
the Minimal Link Condition arguably must be a movement constraint, not an
and Gutmann "argue that an optimal derivational system, at least from a
computational point of view, is a system that generates only objects that are
well-formed and satisfy conditions imposed by the interface systems." Further,
"we demand that the computational system be crash-proof. That is, that no
filters are imposed on the end products of derivations...."
Seely, also pursuing minimization, argue that Spell Out applies inside each
transformational rule application, and that "this conception of Spell Out allows
for the elimination of: levels, special categorial phases, lookahead, lookback,
and simultaneous rule application (as well as "principles" like the EPP)."
important alternative view, Brody argues that "no pure derivational theory of
narrow syntax exists." And further, it is argued that "the derivational
explanation of the asymmetry of the notion of c-command ... like other
explanations of c-command (and also like other derivational explanations) is not
this introduction, each author's abstract is presented, in alphabetical order.
Michael Brody: "On the Status of Representations and Derivations" (Chapter one):
the point here that current mixed theories of syntax that involve both
derivations and representations are redundant and in principle less restrictive
than their pure representational or pure derivational counterparts. Next I show
that no pure derivational theory of narrow syntax exists. To be minimally
adequate, derivational theories must be mixed, hence the arguments against mixed
theories apply to these too. In addition to this, I argue that everything else
being equal, and with no additional stipulations added, current derivational
theories with the rule of move are less restrictive than representational
theories with the concept of chain. In the third section of the chapter I
consider the derivational explanation of the asymmetry of the notion of
c-command and conclude that this explanation, like other explanations of
c-command (and also like other derivational explanations) is not successful. I
suggest instead that we should eliminate c-command from the grammar and replace
it by simpler interacting notions.
Collins: "Eliminating Labels" In this chapter, I argue that it is possible to
eliminate the labels of phrasal categories. I consider four areas where labels
have been used in syntactic theory: X'-Theory, Selection, the Minimal Link
Condition, and the PF Interface. In each case, I show that it is possible to
capture the relevant generalizations without the use of phrasal category labels.
This chapter illustrates the relationship between derivational and
representational approaches to syntax. As representations are simplified (e.g.,
labels are eliminated), it becomes necessary to articulate the precise nature of
syntactic derivations (e.g., minimize search, the Locus Principle, etc.).
David Epstein and T. Daniel Seely: "Rule Applications as Cycles in a Level-free
Syntax" (Chapter three) In a number of important recent papers, Chomsky
postulates multiple, cyclic Spell Out, and seeks to explain (not stipulate) this
central, derivational departure from the long-dominant Y-model. Since features
are valued iteratively, i.e. derivationally, and since once valued, Spell Out
cannot properly operate, Chomsky suggests that it follows that Spell Out must
itself also apply iteratively. We reveal potential problems confronting this
important explanation of cyclic Spell Out. We argue that the explanation can be
maintained, while overcoming the problems noted, by "driving Spell Out inside
each transformational rule application." We argue further that this conception
of Spell Out allows for the elimination of: levels, special categorial phases,
lookahead, lookback, and simultaneous rule application (as well as "principles"
like the EPP). Although far from conclusive, this eliminative approach
identifies (and pursues to some degree) what we believe to be the null
hypothesis regarding the architecture of UG.
John Frampton and Sam Gutmann: "Crash-proof Syntax" (Chapter four):
Minimalist Program is guided by the idea that the syntactic system of the
language faculty is designed to optimally meet design specifications imposed by
the interface systems. In this chapter we argue that an optimal derivational
system, at least from a computational point of view, is a system that generates
only objects that are well formed and satisfy conditions imposed by the
interface systems. We explore the feasibility of constructing such a system, one
in which there is no notion of "crashing derivations." We also argue that both
the interface conditions and the derivational system which is designed to
satisfy them are real, and in this sense we expect to find a pervasive
redundancy in linguistic explanation. We demand that the computational system be
crash-proof, that is, that no filters are imposed on the end products of
derivations, and that no global filters (e.g. "comparison of derivations")
assign status to derivations as a whole. By exploring the properties of a
crash-proof system, we hope to clarify the boundary between the derivational
system and the interface conditions, and to make progress in understanding both.
Norbert Hornstein and Juan Uriagereka: "Reprojections" (Chapter five):
chapter explores the possibility that projected labels may change in the course
of the derivation (hence the name reprojection). Label changes do not alter
phrasal geometry, and in that sense are not transformational processes;
however, representational dependencies are altered via reprojection in such a
way that, for instance, head-specifier relations are redefined. The main
consequence of this is that it provides a natural syntax for binary quantifiers,
in much the same way that transitive verbs take their arguments within their
(extended) projection. Interestingly, it is these quantifiers which induce
islands for various LF dependencies across them. In our terms this follows from
the fact that, after reprojection, a putative LF dependency would have to be
across a noncomplement branch. Unary quantifiers need not reproject in order to
meet their syntactic/semantic demands, and hence do not induce LF islands of
this sort. The chapter extends these results to other, less obvious domains.
Richard S. Kayne: "Pronouns and their Antecedents" (Chapter six):
the derivational perspective of Chomsky's recent work, I explore the idea that
antecedent-pronoun relations should be rethought in movement terms, including
what we think of as Condition B and Condition C effects. This allows me to
capture the compelling similarity between Condition C and the prohibition
against downward movement, and to dispense with the notion of "accidental
coreference " in Lasnik's sense. All instances of "intended coreference/binding"
involve a movement relation. In the spirit of the Lebeaux/ Chomsky proposals
concerning Condition A, I argue that neither Condition B nor Condition C is a
primitive of UG. The effects of both follow, in a derivational perspective, from
basic properties of pronouns and basic properties of movement. What underlies
Condition B effects underlies the existence of reflexives. The movement approach
explored, which leads to some use of sideward movement, has consequences for
circularity effects, for reconstruction effects, and for backwards
Kitahara: "Scrambling, Case, and Interpretability" (Chapter seven): Saito (1992)
examines a number of binding phenomena involving scrambling and proposes that
the binding theory applies not solely at LF, but at some preLF-level as well.
In this paper, I elaborate Saito's proposal under the framework of Epstein et
al. (1998). I first specify the mechanisms of scrambling in Japanese, adopting
certain aspects of the probe-goal system (Chomsky 2000, 2001). I then articulate
how these mechanisms of scrambling interact with the interpretive procedures
applying in the course of a derivation.
McCloskey: "Resumption, Successive “Cyclicity, and the Locality of Operations"
(Chapter eight): This chapter is concerned with the mechanisms by which "long"
A'-binding relations are constructed, and in particular with the role of
intermediate positions in constructing those relations. The main empirical
issue it grapples with is the way in which in Irish the form of C varies
whether or not a clause hosts an instance of A'-binding;
what kind of A'-binding (binding of a trace or binding of a resumptive
pronoun) it hosts.
The paper shows that long A'-binding can involve "mixed chains," some of whose
links are created by movement, and some of which are created by merge of a
binding operator in the specifier of CP.
The paper shows that long A'-binding can involve "mixed chains," some of whose
links are created by movement, and some of which are created by merge of a
binding operator in the specifier of CP.
argues that the form of the complementizer in Irish A'-binding constructions
depends not on the mechanism of specifier-head agreement (as has been widely
assumed), but rather is sensitive to the mode of introduction of the specifier
material (whether it is introduced by Merge or by Move). The core characteristic
of the system that emerges is its locally blind character. Choices are made at
intermediate positions with no "foresight" about what the consequences for
semantic interpretation will be. Locality is more highly valued than ease of
Richards: "Very Local A' Movement in a Root first Derivation (Chapter nine):
Richards (1999), following Phillips (1996, to appear), developed an approach to
the derivation that generates the material at the root of the tree first, and
adds new material to the bottom of the existing structure. The approach made the
surprising prediction that under certain circumstances, A' movement should share
with A-movement the property of being unable to skip an intervening A-position.
In this paper I try to show that the prediction is true in a range of cases,
under circumstances largely predicted by the theory.
Torrego: "Arguments for a Derivational Approach to Syntactic Relations Based on
Clitics" (Chapter ten): A well-known complication concerning the data studied
for subject-to-subject raising over an Experiencer is the grammatical source of
cross-linguistic variation. Concentrating on Romance in the area of clitics, I
show how certain aspects of the particulars of dative Case influence this
variation for the languages considered. The analysis must be couched in a
framework of derivational economy that imposes strict derivationality of the
sort argued by Epstein and Seely (1999). Relevant comparison between languages
of the French type and of the Spanish type attributes a range of grammatical
differences between the two groups to a general property of "doubling" clitics,
and the range of clitics that license pro.
Zwart: "Issues Relating to a Derivational Theory of Binding" (Chapter eleven):
The derivational approach to syntactic relations (DASR, Epstein et al. 1998),
adopted here, implies (in its strongest form) that syntactic relations between a
and /3 exist iff a and /3 are merged as sisters at some point in the derivation.
In the spirit of DASR, Kayne (this volume) proposes that coreference results
from the circumstance that the antecedent and the pronoun are merged as sisters,
after which the antecedent moves up to the appropriate Case and theta-positions.
This paper adopts the general outlook, but argues that Kayne's proposal applies
to local anaphor binding only, not to nonlocal coreference/ bound variable
anaphora. The approach assumes that a reflexive pronoun is the marked Spell Out
of an abstract category (bundle of features) PRONOUN, which acquires the feature
[+coreferential (with a)] iff it is merged with a in a sisterhood configuration.
Since the antecedent moves away from the PRONOUN via A-movement, the fact that
the locality conditions on binding and A-movement coincide is accounted for.
insert content here