Review of The Syntax of Topic, Focus, and Contrast
Ad Neeleman and Reiko Vermeulen open this edited volume with a chapter entitled “The Syntactic Expression of Information Structure”, which establishes one crucial premise: that the syntax-information structure and syntax-semantics interface account proposed in this volume is at odds with the approach taken by researchers such as Luigi Rizzi and colleagues working in the cartographic program. Proponents of the cartographic program argue that information pertinent to discourse is directly encoded in dedicated functional syntactic projections with accompanying (lexical) features. The initial justification for such an approach is based in large part on observations by Abels (2012), who shows that the relative order of topics and foci in Italian (which formed the initial basis for both Rizzi’s (1997) extension of the left periphery and the cartographic program) may be independently derived as an effect of Relativized Minimality (e.g. Starke 2001). Therefore, locality of movement asymmetries between so-called ‘topical’ and ‘focal’ constituents allows us to greatly simplify the number of left-peripheral functional projections; however, the authors note that despite such simplification, we may not do away with some sort of hierarchy of discourse/semantic notions altogether. Their goal is to establish via data from English, Russian, Dutch, Japanese, and Korean that word order restrictions operative in the mapping of information structure to the syntax do not require a pre-determined hierarchy of functional projections nor associated features. They narrow their attention to three particular primitives of discourse/information structure in this volume: topic, focus, and contrast. The notions [topic/focus] combines with [±contrast] to net Aboutness topics [topic, -contrast], (new) Information focus [focus, -contrast], Contrastive topics [topic, +contrast], and Contrastive focus [focus, +contrast].
The remainder of the introduction clarifies the difference between the latter two notions, showing that although both may exhibit movement and take scope over other elements (unlike their non-contrastive counterparts), they differ in that “focus is a notion relevant to propositions, whereas topic is (a) notion relevant to utterances” (p. 21, from Tomioka 2010). For a contrastive topic then, the relevant alternative is an utterance (one of which is claimed to be false); for a contrastive focus, the relevant alternative is a proposition (one of which is claimed to be false). In what remains of the opening chapter, they examine A’-scrambling of contrastive topics and contrastive foci in Dutch. They show that when these constituents remain in situ, they may appear in any order because they do not mark scope, which is proposed to be the domain of contrast (DoC) of these constituents; under A’-scrambling, however, both are quantificational in that they mark scope, and thus a DoC. A’-scrambling then creates a scope-freezing effect, by which a constituent c-commanded by an A’-moved element may not take scope over it. Relative ordering effects of contrastive topics and foci fall out from discourse-interpretive facts: a proposition (focus) may fall within an utterance (topic), but not vice versa. This is what prevents A’-scrambling of a contrastive focus over a contrastive topic. Additionally, it captures the relative ordering of topic and focus, as well as the oft-noted observation that negative quantifiers may not function as topics. The chapter concludes with an overview of the chapters that follow.
In Chapter Two, “Towards a Unified Encoding of Contrast and Scope”, Ad Neeleman and Hans van de Koot attempt to reconcile the notion of Domain of Contrast (DoC), essentially the theory of contrastive scope introduced in Chapter 1, with the theory of quantifier scope. They develop a proposal for scope and scope extension, which shares much with Reinhart’s (1983, 1995, 2006) view of Quantifier Raising as scope extension, thus seeking to explain asymmetries between scope marking via overt movement and scope marking via covert movement (QR). They propose a Condition on Scope Shift (CSS), by which no node may inherit two scope indices, and show via examples in Dutch, English, and German that quantifier scope and contrastive scope both obey the CSS, which strongly suggests the feasibility of uniting the two theories. Therefore, in the case of both quantifiers and contrastive elements, if a constituent (overtly) moves into a scope-marking position over another, the latter, which is c-commanded by the former, may only escape the scope of the former via A’-movement, and not via covert, QR-type movement. This lends further support to the idea that topic and focus ordering is independently motivated: if a contrastive focus constituent moves into a configuration in which it c-commands a contrastive topic, the contrastive topic may not escape the DoC by covert means, thus yielding an uninterpretable configuration for the information structure.
Chapter Three, “Word order variation and information structure in Japanese and Korean”,
by Reiko Vermeulen, examines discourse-related syntactic variation in Korean and Japanese, two languages which, despite similarities, display important differences captured by the three information structure primitives introduced in Chapter 1. The author shows that both languages have the same rule operative for [topic], requiring these constituents to appear in clause-initial position, in order to transparently mark a topic-comment structure in the syntax. Both languages also have the same rule for [contrast], which requires A’-scrambling, which as in Dutch, marks scope/DoC. The two languages diverge with respect to how they handle Contrastive Topics, which, according to the author, creates a conflict between two rules that cannot be satisfied simultaneously. In this case, the author proposes that Japanese opts for the [topic] rule, thus only allowing contrastive topics in clause-initial position; Korean however, opts for the [contrast] rule, which explains why contrastive topics in this language may appear in a variety of clausal positions. This represents a departure from much of the literature on topics in Korean and Japanese. Nevertheless, this characterization is shown to be adequate, therefore predicting the same ordering restrictions on contrastive topics and foci as those in Dutch. Finally, the author examines the ambiguity of contrastive ‘wa’ phrases in Japanese and contrastive ‘nun’ phrases in Korean, showing that these may be either topics or foci in a given sentence. Curiously, ‘wa’ and ‘nun’ phrases that are interpreted (functionally) as contrastive foci do not display the syntactic behavior of contrastive topics -- facts which lend further support to the thesis advocated in this volume regarding the status of the primitives [topic], [focus], and [contrast].
Chapter Four, entitled “Encoding focus and contrast in Russian” by Elena Titov, further motivates the primitives [focus] and [contrast], each of which is claimed to be responsible for a particular syntactic effect. In Russian, contrastive focus constituents as well as new information focus (i.e. narrow-focus) constituents appear in clause-final position, but only contrastive focus constituents are allowed to undergo A’-movement further leftward. Support for this notion is provided for by the reconstruction of clause-initial contrastive focus elements to a clause-final position. Therefore, the author proposes that while [focus] is the relevant information structural feature for clause-final constituents, it is [contrast] that licenses movement to the left periphery. Notably, the target of A’-movement is not proposed to be a fixed syntactic position, but rather a strategy for disambiguating contrastive focus from, e.g., information focus.
In Chapter Five, “Against FP Analyses of Clefts”, Matthew Reeve argues against an FP (FocusP) analysis of cleft constructions in English and Russian. He first shows that É. Kiss’s (1998) account of English clefts does not capture the fact that they bear distinct semantic and syntactic similarities to specificational copular sentences (e.g. ‘The one that Mary hit was JOHN.’). He then examines Russian cleft constructions, which ostensibly bear more similarities to the construction in Hungarian; however based on the undesirable modifications needed for an F-based account for the Russian construction, he rejects this possibility. As an alternative unified account, the author proposes that despite syntactic differences, clefts in Russian and English may be analyzed as specificational copular sentences that encode equative semantics. He then provides a number of empirical arguments against the FP analysis: evidence suggesting that subject pronouns in clefts are not expletives, similarities between the presuppositional content of clefts and specificational sentences, the properties of ‘wide-focus clefts’ in Russian, adjacency conditions on the pronoun and focus constituent in narrow-focus clefts in Russian, the possibility of focus movement within clefts, the fact that English clefts behave like extraposed clauses, and obligatory versus optional contrastivity facts. He concludes that a specificational account of clefts better captures the above phenomena than an FP account, thus obviating the need for the functional projection and its associated feature(s) in order to encode information structure.
In Chapter Six, entitled “Focus movement can be destressing, but it need not be”, Kriszta Szendrői examines focus movement within the noun phrase and whether such movement serves to mark the domain of contrast (DoC) as it does at the clause level. She first examines adjective reordering in English, which has been argued in the literature to involve focus movement within the DP. She presents data that lead her to determine that adjective reordering allows for reconstruction and is a scope-marking (DoC) operation; however, since the interpretative effect of this movement does not establish focus on the fronted adjective and does not appear to target a pre-determined position, it does not constitute movement to FocP. She next examines polydefinite constructions in Greek, which have also been argued to involve DP-internal focus movement. On the basis that polydefinite constructions do not display the hallmarks of A’-movement and do not mark DoC, she proposes that this construction marks givenness. She then examines the theoretical possibility that givenness-marking and DoC-marking are two sides of the same coin, as proposed in Wagner’s (2007, 2012) theory, based on Schwarzschild (1999). Despite the attractiveness of this proposal, she presents data suggesting that these two systems (focus/DoC-marking and givenness-marking) need to be distinguished as separate operations, in the same way that stress strengthening should be distinguished from destressing.
In “Types of Focus and their Interaction with Negation” (Chapter Seven), Ad Neeleman and Reiko Vermeulen address what they label “an empirical challenge to the decompositional approach to contrastive focus, which arises from the interaction between focus and negation”. The empirical challenge at stake in this chapter lies in focus-sensitive particles such as ‘only’, ‘even’, and ‘also’, some of which result in unexpected meanings (at the level of focus semantics) when they interact with negation. Were it the case that contrastive foci and foci accompanied by focus-sensitive particles uniformly move to FocP, or uniformly undergo LF-raising, a number of phenomena would go unexplained, such as the fact that ‘even’ and ‘also’ may appear under the scope of negation, but may not take scope over it, or that ‘not Y but X’ cannot LF-raise at all. The authors propose that focus-sensitive particles that trigger negative alternatives components (i.e. a negative component of meaning) such as ‘only’ and ‘some’, which also net surprising meanings under negation (e.g. ‘John did not invite only Pia’, which actually implies that John invited someone in addition to Pia) are most appropriately treated as positive polarity items (PPI). Since PPIs cannot be interpreted in the scope of negation they trigger a non-default mapping of negation.
In Chapter Eight (“Concluding remarks”), Ad Neeleman and Reiko Vermeulen, the editors of the volume review the data presented in the preceding chapters and their application to the primitives [topic], [focus], and [contrast]. They also provide an explanation for why they do not consider the notions ‘old’, ‘given’, or ‘discourse-anaphoric’: primarily, such ‘familiar topics’, which often refer to the discourse topic at hand, do not involve selection from a set of alternatives. Additionally, not all given material can be topical, given that topics cannot contain foci. Furthermore, they provide examples from Japanese and Dutch that discourse-anaphoric material frequently escapes generalizations that hold for contrastive and aboutness topics, such as clausal distribution (initial in Japanese) and the fact that discourse-anaphoric topics do not resist focus movement across them as other topics do. They suggest then that such familiar, discourse-anaphoric topics should be analyzed independently from the topic/focus system. In the second section of the conclusion, they discuss the conceptual-intentional interface as the point at which an utterance may be divided into topic-comment or focus-background, and at which domain of contrast may be more transparently mapped to interpretation. Although languages differ with respect to the mapping rules employed, there are many patterns that are predicted to remain constant. Although the contributions in the volume do not involve the phonology interface, they do not rule out the PF considerations that may come to bear as well. They end the chapter with a final treatment of the cartographical approach showing that the postulation of a ContrastP in addition to FocusP and TopicP to capture featural decomposition cannot account for the data presented in this volume and makes inaccurate predictions, not only when such constituents appear in situ, but also when they appear in the left periphery.
While all the contributions make notable advances, Chapter 2 stands out in making less accessible scope readings more transparent -- even unacceptable ones; however, these stand in contrast to more problematic and (admittedly) subtler scope judgments, which at times are so subtle that the differences are practically indistinguishable. Chapter 7 is also remarkable for its detailed and accessible analysis of the scope of negation; it seems however that the importance of [contrast] and its role in providing information about (sets of) alternatives gets lost in the argumentation. If [contrast] marks scope (presumably overtly in the syntax), it is unclear how this feature influences non-default mappings of negation with discourse-sensitive particles such as ‘even’ and ‘only’. Chapter 8 is notable in bringing to light a problematic issue for researchers working with Romance varieties, in which CLLD constituents may be topical or discourse-anaphoric. While the authors propose a solution noted for Italian, it is not clear that this is sufficient for Western Iberian Romance varieties, in which (contrastive) foci may not precede a CLLD topic, discourse-anaphoric or otherwise.
The volume’s unified approach to address issues related to topic, focus, and contrast at the interface of syntax with semantics as well as the discourse/information structure module is refreshing for a number of reasons: first, all of the contributions speak about similar information structure concepts with similar terminology and technology -- a welcome departure from the multiplicity of approaches and definitions found in the literature; second, the contributions share the common goal of a simplified left periphery. The former is desirable from a practical perspective if generative syntacticians wish to speak about interface phenomena from a cross-linguistic perspective. The latter is desirable from the perspective of explanatory adequacy, assuming that one of our chief goals remains explaining how one arrives at a linguistic competence that is largely unconscious. Clearly, one of the goals of syntactic research is describing the orders attested in natural language, but at a certain point we have to question the predictive power afforded us by a proliferation of supposedly lexical features that have more to do with discourse functions than the lexicon proper. Therefore, this volume appears to share the goal of a number of other recent proposals (e.g. López (2009), Kempchinsky (forthcoming)), which aim to simplify the acquisitional and processing load related to interfaces.
Abels, Klaus. 2012. The Italian left periphery: a view from locality. Linguistic Inquiry 43(2). 229-254.
É. Kiss, Katalin. 1998. Identificational focus versus information focus. Language 74. 245-273.
Kempchinsky, Paula. forthcoming. CLLD as a window on the left periphery. Selected proceedings of the 15th Hispanic Linguistics Symposium. Somerville, MA: Cascadilla Proceedings Project.
López, Luis. 2009. A derivational syntax for information structure. Oxford: Oxford University Press.
Reinhart, Tanya. 1983. Anaphora and semantic interpretation. London: Croom Helm.
Reinhart, Tanya. 1995. Interface strategies. Ms., Utrecht University.
Reinhart, Tanya. 2006. Interface strategies. Cambridge, MA: MIT Press.
Rizzi, Luigi. 1997. On the fine structure of the left periphery. In Belletti, Adriana & Luigi Rizzi (eds.), Elements of Grammar, 281-337. Dordrecht: Kluwer.
Schwarzschild, Roger. 1999. Givenness, avoid F and other constraints on the placement of accent. Natural Language Semantics 7. 141-177.
Starke, Michal. 2001. More dissolves into merge: a theory of locality. Genève: Université de Genève dissertation.
Tomioka, Satoshi. 2010. Contrastive topics operate on speech acts. In Zimmerman, Malte and Caroline Féry (eds.), Information structure: theoretical, typological, and experimental perspectives, 115-138. Oxford: Oxford University Press.
Wagner, Michael. 2007. Givenness and locality. In Abdurrahman, Muhammad, Anisa Schardl, and Martin Walkow (eds.), Proceedings of the North East Linguistic Society 38, 415-428. Amherst, MA: University of Massachusetts GLSA.
Wagner, Michael. 2012. Focus and givenness: a unified approach. In Kučerová, Ivona and Ad Neeleman, Contrasts and positions in information structure, 102-147. Cambridge (UK): Cambridge University Press.
| ABOUT THE REVIEWER:
Timothy Gupton is Assistant Professor of syntax in the Department of Romance Languages at University of Georgia. His primary research interest is theoretical syntax, in particular the syntax of subjects, clitics and left-peripheral elements and their interaction with information structure. I additionally employ a variety of experimental methods based primarily on second language acquisition research in order to elicit quantitative psycholinguistic grammar judgment data. My current research interests include the prosody of contrast and CLLD, the L2 acquisition of discourse-configurational word order in Spanish, and subject expression and reference in Caribbean Spanish.