LINGUIST List 26.1401
Fri Mar 13 2015
Review: Discourse Analysis; Syntax; Typology: van Gijn, Hammond, Matić, van Putten (2014)
Editor for this issue: Sara Couture <saralinguistlist.org>
Date: 16-Oct-2014
From: Amy Cruickshanks <a.cruickshanks
student.unimelb.edu.au>
Subject: Information Structure and Reference Tracking in Complex Sentences
E-mail this message to a friend Discuss this message Book announced at
http://linguistlist.org/issues/25/25-2071.html
EDITOR: Rik van Gijn
EDITOR: Jeremy Hammond
EDITOR: Dejan Matić
EDITOR: Saskia van Putten
TITLE: Information Structure and Reference Tracking in Complex Sentences
SERIES TITLE: Typological Studies in Language 105
PUBLISHER: John Benjamins
YEAR: 2014
REVIEWER: Amy Cruickshanks, University of Melbourne
Review's Editor: Helen Aristar-Dry
SUMMARY
This edited volume, “Information Structure and Reference Tracking in Complex Sentences”, stems from the 2011 Workshop on Information Structure and Subordination: South America and Beyond (Nijmegen) which was organised jointly by The Traces of Contact Project (Radboud University Nijmegen) and the Syntax, Typology and Information Structure Group (Max Planck Institute for Psycholinguistics, Nijmegen). Comprising thirteen individual contributions discussing aspects of information structure and reference tracking in a variety of unrelated languages, the book is divided into two parts, each containing six chapters: 1) Information structure in complex sentences, and 2) Reference tracking in complex sentences. Preceding part one is an introductory overview of the topic area.
The introductory chapter, “Information structure and reference tracking in complex sentences” (pp.1-41), written by Dejan Matić, Rik van Gijn and Robert D. Van Valin, Jr. provides an overview of the concepts the works in this volume are concerned with. “Information structure” (IS) is defined as “common ground management” (p.2) and the authors use this notion as a springboard in describing traditional IS concepts: “topic” highlights the part of common ground which will be added to by a new proposition, and “focus” shows the section of a proposition not yet part of common ground. “Reference tracking” (RT) is described as “the capability of the interlocutors to unequivocally determine the referent(s) of a linguistic expression” (p.2). In relation to complex sentences and the relationship between their (clausal) units, a distinction is made between coordinate and subordinate structures. Subordination is further categorised into i) daughter-subordination, where the subordinate clause functions as a core argument of the verb in the main clause, and ii) ad-subordination, where the subordinate clause is not directly embedded in the main clause and functions as an adjunct modifier.
The interaction between IS and complex sentences is investigated. The authors distinguish between two types of IS interactions: i) ‘external IS’ in which an entire clause in the complex sentence functions as topic or focus, and ii) ‘internal IS’ in which a single constituent within one of the clauses in the complex sentence has an IS role. Each of these interactions is examined and the likelihood of their occurrence with daughter-subordinate and ad-subordinate clauses discussed. In relation to RT, the authors propose a typology based on four parameters: i) the way a referent is marked, ii) the discourse scope of referential understanding of an RT element, iii) the location of the RT marker, and iv) the syntactic, semantic and pragmatic characteristics of controllers and pivots. The authors briefly discuss the effects of IS on RT, claiming that IS often plays a role in switch reference systems.
In the first chapter in part one, “Subordinate clauses and exclusive focus in Makhuwa” (pp.45-69), Jenneke van del Wal discusses the two alternate verb forms in Makhuwa (Bantu) known as conjoint and disjoint, showing that the conjoint form is employed when a noun or simple (i.e. non-clausal) adverb following the verb has “exclusive focus” (p.46). Van del Wal demonstrates this pattern holds regardless of whether the element is an argument of the verb or an adjunct. She then provides evidence that this exclusive focus interpretation also extends to instances where the element following the conjoint form is an adverbial clause rather than a simple phrase or word. No such exclusive focus interpretation exists for complement clauses, however, which can follow conjoint and disjoint forms with no apparent difference in meaning. The author proposes that this difference between the two types of subordinate clauses is due to adverbial clauses not having an internal IS whilst complement clauses do have internal IS. This means that they, unlike adverbial clauses, cannot function as a single informational unit in the main clause.
Saskia van Putten, in “Left dislocation and subordination in Avatime (Kwa)” (71-98), discusses the phenomenon of elements found in a non-canonical clause initial position in Avatime (Kwa, Niger-Congo). There are two functions of left-dislocation in this language: i) to indicate set-membership and ii) to introduce a new referent into the discourse. Left dislocation appears to occur in both main and subordinate clauses; corpus data as well as elicitation reveals that in subordinate clauses, it is most frequently found in complement clauses, however it also occurs in adverbial clauses, specifically in purpose clauses, reason clauses, temporal clauses and conditional clauses. Although there is some question as to whether all these clauses are truly subordinate in Avatime, the author states that in some instances it is clear they are. The claim that left dislocation is not possible within presupposed clauses is discussed and van Putten demonstrates that in Avatime this also appears to be the case – left dislocation only occurs in clauses which are asserted.
In the next chapter entitled “Chechen extraposition as an information ordering strategy” (pp.99-126), Erwin R. Komen investigates the relationship between extraposed relative clauses and narrow focus. The author claims that in Chechen extraposed relative clauses can either be ‘non-restrictive/appositive’ (i.e. the head of the relative clause is identifiable as it stands) or ‘restrictive’ (i.e. the head of the relative clause is unidentifiable as it stands), and that both clause types can occur with either argument or adjunct heads. He shows that whilst heads of non-restrictive extraposed relative clauses can appear in any position in the clause and can carry a topic or focus status, heads of restrictive extraposed relative clauses only ever occur in the narrow focus position immediately preceding the verb, and that this focus is often overtly contrasted in the discourse.
The fourth chapter in part one, “Questions and syntactic islands in Tundra Yukaghir” (pp.127-161) by Dejan Matić, discusses the relevance of information structure for ‘questioned islands’ - syntactic islands with question words. Focusing on relative clause and adjunct clause questioned islands, Matić argues that no element in these clauses can carry focus-marking; focus-marking is a feature that appears at the level of the matrix clause only. While relative and adjunct clauses are often thought to be unfocusable (p.150) and unavailable for inquiry (p. 151), the author demonstrates that in Tundra Yukaghir, questioned islands as a whole can act as focus, with the entire syntactic island behaving like a question word in simple questions. There are no apparent restrictions in terms of what grammatical function the questioned island has in the matrix clause, nor are there any restrictions on the grammatical role of the question word within the island. The reason why these questioned islands can be focused, Matić proposes, is due to the size of the question focus. He states that in question islands it is not purely the identity of the question word being questioned, but the identity of the entire island (p. 156).
In the next contribution, “Constituent order and information structure in Karitiana” (pp. 163-191), Luciana R. Storto discusses characteristics of focus in Karitiana, demonstrating that this language has a focus position at the left-edge of a clause. Subject noun phrases (NPs), object NPs and prepositional phrases (PPs) can all be focussed; where the focussed constituent is an object NP, the verb additionally carries a focus marker. The form of this marker is determined by the type of clause the verb heads: ‘a-’ if the clause is declarative, and ‘ti-’ elsewhere. Storto goes into some detail as to the origins of these two morphemes claiming that ‘a-’ derived from the language’s passive prefix while ‘ti-’ potentially began as an “object subordinate clause nominalizer” (p. 188). In a short excerpt comprising four explanatory sentences and a five-sentence excerpt from a narrative, the author shows objects in subordinate clauses can have the role of focus. She also proposes the language encodes when the subject is a new (shifted) topic by repeating the subject constituent at the end of a sentence.
The final chapter in part one is Patxi Laskurain Ibarluzea’s “Mood selection in the complement of negation matrices in Spanish” (pp. 193-227). The author presents a novel analysis of doubt and negation complements, proposing that the driving factor behind their having subjunctive mood is information structure rather than the complement’s presupposition or nonassertion. Ibarluzea asserts that it is “the relation between negation and active referents” (p.203) which determines the use of the subjunctive in these complement clauses, and that the subjunctive mood indicates both the speaker and hearer have the complement’s proposition in mind (p.204). He claims this is on par with the use of the subjunctive in comment complements. Ibarluzea further addresses the issue of both subjunctive and indicative moods being used with complements of negative doubt predicates such as ‘no dudar’ (meaning ‘not to doubt’) claiming the subjunctive is used when the speaker believes the listener has in mind the complement’s propositional content whilst the indicative is used when this is not believed to be the case.
In the first chapter in part two, “Topic management and clause combination in the Papuan language Usan” (pp. 231-261), Ger Reesink discusses how topical referents are tracked in Usan discourse. Usan is a language that has a switch reference system and contains clause chaining structures, i.e. groups of verbs in which all but the final verb is devoid of tense-aspect-mood markers. Reesink states that a topic is introduced into the discourse with a “presentational construction” (p. 234), and with the topical noun accompanied by ‘ger’ ‘indefinite article’. In all additional appearances, the topical noun is marked by ‘eng’ (translated as ‘that’), however once a topic has been firmly established in the mind of the listener, it is often expressed through the switch-reference verbal morphology alone. The author argues that the clauses in the clause chains are coordinated and that subjecthood and topichood plays a part in which switch-reference marker is selected, demonstrating that speakers can orient their switch-reference choices around a particular referent when there are multiple topics in the discourse (p. 242). Reesink also discusses subordination in Usan, proposing that subordinate clauses are nominalisations. He points out that both overt topical nouns and subordinate clauses are accompanied by the determiner ‘eng’, which signifies givenness.
Jeremy Hammond, in “Switch-reference antecedence and subordination in Whitesands” (pp. 263-290), describes the switch reference system in this language, showing that it differs from canonical switch-reference systems in that it is the initial clause in a chain which hosts the main predicate with all subsequent linked predicates indicating co- or disjoint-reference. The main predicate is always fully inflected for person and it acts as the controller of tense and illocutionary force across the chain. Co-reference is indicated by the echo reference (ER) prefix, ‘m-’, occupying the person/tense slot on the verb, whereas disjoint reference is indicated by a fully inflected verb. Hammond claims that unlike in Usan (Reesink pp. 231-261), switch-reference in Whitesands is not controlled by the most salient participant in the discourse, but rather it is used to show the continuation of a particular referent across clauses; the ER always indicates the subject of the clause is co-referential with the preceding clause’s subject. The author also demonstrates that the ER system does not carry across to relative clauses or complement clauses.
In the next chapter, “Repeated dependent clauses in Yurakaré” (pp. 291-308), Rik van Gijn discusses two constructions in Yurakaré: ‘tail head linkage’ (THL) and ‘repeated citation constructions’ (RCC). Van Gijn claims there are four THL types in Yurakaré ranging from verbatim repetition of the final clause in one chain as the initial clause in the subsequent chain to what he calls ‘association’, where the initial clause in a chain shows no repetition of an earlier predicate but builds on the events in the preceding clause. Within clause chains, a switch reference system is at play with switch reference markers differentiating a) the co- or disjoint referentiality of subjects, and b) (where the subjects are co-referential), whether the clausal proposition is what the author classes as “realis” or “irrealis”. In RCCs, van Gijn shows that a cited predicate is repeated with the second instantiation of the verb carrying one of the two subject switch-reference markers, depending on the mood of the main clause. He claims that THL and RCC constructions are similar not only in structure, but also in that they both work in asserting an event within the discourse whilst simultaneously backgrounding it.
The next contribution, “Clause chaining, switch reference and nominalisations in Aguaruna (Jivaroan)” (pp. 309-340), by Simon E. Overall investigates reference tracking in this Amazonian language. Similar to the others languages discussed in the second part of the book thus far, Aguaruna displays clause chaining structures and a switch reference system that Overall argues marks syntactic information and also contributes to discourse structuring by encoding clauses as either backgrounded or foregrounded (p. 331). The author demonstrates this language also contains THL structures, though the head verb of the second chain is never a verbatim repetition of the tail predicate but rather a pro-verb such as ‘nunit’ and ‘dutikat’ translated as ‘do thus’ (p. 331). Non-restrictive relative clauses are reanalysed as being an extension of the clause-chaining system, with their nominaliser markers being likened to a switch reference indicator.
In “The multiple coreference systems in the Ese Ejja subordinate clauses”(pp. 341-371), Marine Vuillermet investigates the switch reference system in Ese Ejja (Takanan). Rather than having a canonical SS/DS switch reference system, this language has two ‘tripartite’ switch-reference systems. Vuillermet states the first system is found in reason, conditional and ‘before’ clauses. It is driven by the coreferentiality of a participant with the subject in the matrix clause (MC), but also takes into account the semantic role of the MC subject. The second system is found only in temporal clauses and is argued to be dependent clause (DC) driven. In this system, the grammatical function of the argument in the DC which is coreferential with the MC subject determines the switch reference form used.
The final chapter in this book is Ana Vilacy Galucio’s “Argument marking and reference tracking in Mekens” (pp. 373-396). Galucio shows that in simple clauses, argument agreement on the verb follows an absolutive pattern, however in constructions with multiple clauses one finds nominative alignment in the language’s two RT systems. The first system is found with transitive verbs. The author argues that when the subjects of clauses are coreferential, the subject in one of the clauses is omitted, whereas when the subjects of the two clauses are not-coreferential, both are overtly marked. The second RT system is found with intransitive and auxiliary verbs; it tracks referents through subject (non)agreement markers on the verb. Galucio claims the reference tracking systems in Mekens show characteristics associated with neither a switch reference system nor a logophoric system.
EVALUATION
“Information Structure and Reference Tracking in Complex Sentences” is both an informative and interesting book. On the whole, the chapters are detailed and coherent, providing insight into some of the similarities and differences in IS and RT strategies and the ways in which information management processes and complex sentences interact. The contributions complement each other and their cohesion is increased by the introductory overview chapter, which neatly shows how the findings in each chapter fit into the overall typologies given.
The fact that the majority of works in this volume are on lesser studied languages adds to its value. Some of the investigations highlight areas in which we still have much to learn. In particular, the discussions in the second part of the book show that reference tracking strategies don’t always fit well into the categories of switch-reference and logophoric systems; whilst languages may display some characteristics associated with one or both of these systems, they may differ significantly in other ways.
Although the interaction between IS and RT is touched on briefly in a few chapters (e.g. in the introductory chapter as well as in Hammond’s and Reesink’s individual contributions), more discussion and focus on this topic would have added an extra depth to the subject area overall. As this volume is the first major work dealing specifically with IS and RT in complex sentences, it is understandable that not all topics could be covered; however more work looking specifically at the effects of IS on RT in complex sentences would be a welcome addition to the field.
The book has been edited well, though a very few typographical errors and inconsistencies were noted, two in language data examples: on page 324, in example (16) b. DS should be SS, and on pages 376 and 378, examples (4) and (7) are identical, except in the former the subject pronoun is written as a clitic and in the latter as a free pronoun. Additionally, there are some inconsistencies related to terminology throughout the book. Very notable is that in the introductory chapter, there is a detailed discussion about subordination, IS and RT with new terms presented and defined. Yet these terms and definitions are not used or referred to in any other contribution. For example, in the first chapter, we are introduced to the terms “ad-subordinate” and “daughter subordinate”, but throughout the remaining chapters, the traditional terms “adverbial clause” and “complement clause” are typically used. Having said that, the terms used in each chapter are transparent and well explained where necessary, and each contribution can easily be read (and fully understood) as a “stand-alone” paper.
Aimed at those with some understanding of syntactic theory, “Information Structure and Reference Tracking in Complex Sentences” is ideal for linguists and (young) research scholars interested in information structure, reference tracking and discourse analysis.
ABOUT THE REVIEWER
Amy Cruickshanks is a PhD candidate at the University of Melbourne. She is currently researching argument realisation in Pawnee (Caddoan). Her interests include language description and the documentation and preservation of indigenous languages.
Page Updated: 13-Mar-2015