The Circle is a weekly forum dedicated to discussion of the descriptive, experimental, and theoretical study of syntax and semantics, featuring presentations of ongoing research by members of the Berkeley Linguistics Department and other departments, as well as discussion of previously published works.


Fridays, 3:00-4:30pm


1303 Dwinelle Hall


Nico Baier

Katie Sardinha

previous semesters

fall 2013
spring 2013
fall 2012
spring 2012
fall 2011
spring 2011
fall 2010
spring 2010
fall 2009
spring 2009
fall 2008
spring 2008
fall 2007
spring 2007
fall 2006
spring 2006
fall 2005


University of California, Berkeley
Department of Linguistics


9 may
Judith Tonhauser (OSU/Stanford):

Null arguments in Paraguayan Guaraní

In Paraguayan Guaraní (Tupí-Guaraní), cross-reference markers, independent pronouns, noun phrases and combinations thereof may realize the arguments of clausal predicates. The language also has null arguments, i.e. arguments that are not realized by a cross-reference marker, an independent pronoun or a noun phrase. In this talk, I argue that null arguments receive anaphoric interpretations and that the interpretation of null arguments partially determines their distribution: null arguments are only licensed in contexts with a maximally salient, familiar antecedent. Two additional factors affect the distribution of null arguments. First, an argument may not be null but must be realized by an independent pronoun or a noun phrase when it is information-structurally prominent. Another factor is a morphosyntactic one: an argument may only be null if the cross-referencing system of the language does not require it to be realized with a cross-reference marker. Thus, null arguments in Paraguayan Guaraní are similar in their interpretation but not in their distribution to null arguments of languages like Mandarin and Japanese. I conclude the talk with a discussion of the distinctions between null arguments, existentially interpreted implicit arguments and elided arguments.

2 may
Formal Approaches to Slavic Linguistics 23

25 april
Maziar Toosarvandani (UCSC):

Agreement in Zazaki and the nature of nominal concord (Joint work with Coppe van Urk, MIT)

Does a single grammatical mechanism underlie both verbal agreement and nominal concord? We identify an unusual pattern of nominal concord in Zazaki (Northwestern Iranian: Turkey) that reveals a deep similarity between agreement in the nominal and verbal domains. It arises in part, we argue, because nominal concord in Zazaki is subject to the same restriction on verbal agreement that blocks phi-agreement with DPs bearing an inherent case (Rezac 2008, Preminger 2011). To derive this pattern, we extend the logic of Béjar and Rezac’s (2009) cyclic Agree to the nominal domain, in which a probe first looks downward before then looking upward.

18 april
Anna Marie Trester (Georgetown): Graduate Student Career Workshop

11 april
Michael Diercks (Pomona College):

What Kinds of Agreement are Possible? Lubukusu Complementizer Agreement and (Non)Anaphoric Feature Valuation

Lubukusu (Bantu, Kenya) displays a form of complementizer agreement (CA) where a complementizer agrees in gender, number, and person features with the main-clause subject, a distinct pattern from the more familiar West Germanic cases of CA with an embedded subject. As can be seen in (1), the embedding complementizer in Lubukusu agrees in noun class with the subject of the matrix clause (cardinal numbers in examples represent NOUN CLASS, not PERSON):

(1) Alfredi a-a-bolela baba-andu a-li ba-kha-bile
1Alfred 1S-PST-tell 2-person 1-that 2S-FUT-conquer
‘Alfred told the people that they will win.’

A wide variety of syntactic diagnostics are employed to demonstrate that the complementizer agrees with the subject of the most local clause (including causatives, passivization, multiple embeddings, NP intervention effects, complex NPs, and split-antecedence relations). I propose that the agreeing complementizer is the result of anaphoric agreement, a feature valuation process that is identical to that underlying the derivation of self-reflexives. The talk includes discussion of the impact of these conclusions on theories of the directionality of probing of Agree and theories of anaphora. The talk therefore is very rich in empirical documentation and argumentation, but also includes discussion of the theoretical impacts of the conclusions.

4 april
CLS Practice Talk: Nico Baier (UCB)

Multiple Copy Spell-out in Seereer: Implications for Intermediate Movement

In this paper I examine novel data from long distance wh-dependencies in Seereer, a North Atlantic language of Senegal. In addition to providing striking evidence that long distance wh-movement proceeds successive cyclically through the edge of each clause, I argue that the Seereer data show that movement to all CP phase edges is motivated by agreement with features on C. Seereer long distance wh-questions are characterized by (a) the presence of an obligatory pronoun at the edge of each embedded clause and (b) the presence of special morphology on each verb along the path of the extraction. I show that this verbal morphology spells out a valued OP-probe on C, which triggers the movement of a wh-phrase to its Spec. I argue that the embedded pronouns are overtly spelled out copies of the moved wh-phrase. I further argue that overt copies surface at the edge of the CP precisely because they enter into a feature valuation relationship. Specifically, I propose that valuation of an OP-probe defines a copy as the head of an Ā-bar chain. Thus, the application of successive cyclic movement does not result in one long chain, but instead in a series of shorter overlapping chains. This principle, when combined with the independently motivated principle of spelling out the heads of chains, results in the spelling out of intermediate copies in Spec-CP in long distance movement chains.

28 march
Spring Break

21 march
Kristen Greer (UC Davis):

A unified analysis of many and few

Most quantifying determiners find their semantic arguments in the overt syntax, with the NP argument of the determiner providing the restrictor and the VP the scope. I propose that the semantic arguments of many and few are determined pragmatically: the restrictor represents what is presupposed in the discourse, and the scope what is asserted by the quantified sentence. Each of these arguments is complex: the restrictor is the intersection of either the N' or the V' and the restricted universe, and the scope is the intersection of the N' and the V'. To retain the 'mapping hypothesis', whereby semantic arguments are determined by (some level of) syntactic structure, I propose that many/few are always associated with a covert partitive PP complement. This PP dominates a definite NP containing two variables that stand for the presuppositions of the discourse (one in the position of the head N and the other in the position of a V embedded in a restrictive relative clause), as in (1).

(1) Many/few of the Ns who V

Because this PP occupies their structural complement, many/few must combine with the N they are associated with in overt syntax as adjuncts, with the result that these expressions are adjectives and not determiners. Many/few therefore operate over what is essentially a 'bare plural' structure, and this allows me to import parts of Diesing's (1992) analysis of bare plurals to explain the range of interpretations that are available when many/few are paired with individual- or stage-level predicates.

In addition to justifying a pragmatic treatment of the arguments of many/few and outlining their resultant syntactic structure, I describe how this account (a) explains the 'Compset' anaphora pattern after few-quantified sentences (i.e., 'Few students ate their ice cream. They threw it around the room instead.') and (b) can be extended to account for the adverbs often and seldom. Overall, one of the most important insights to emerge from the present account is that the syntactico-seman-pragmatic structure of many and few can be specified in an entirely unambiguous way.

14 march
Eva Csipak (Universität Göttingen):

Decisions and factive subjunctives

In this talk I discuss subjunctive data from German.

(1) Da wäre Pizza im Kühlschrank.
There is.SUBJ pizza in the fridge.

A speaker can use (1) in a context where she knows that there is, in fact, pizza in the fridge – in those cases, (1) has the flavour of an offer. I show that using (1) presupposes the existence of a decision problem, and that the subjunctive in (1) signals that the speaker is contributing to the decision problem in a particular way: knowing that there is pizza in the fridge is relevant for a decision maker who has certain preferences. The speaker signals that she is ignorant about her interlocutor’s preferences. I use a decision theoretic model to show the effect of an utterance like (1) on the decision problem of the agent, and I discuss how this effect is carried over into relevance conditional utterances, and the difference between (2) and (3).

(2) Wenn du Hunger hast, da wäre Pizza im Kühlschrank.
If you are hungry, there is.SUBJ pizza in the fridge.

(3) Wenn du Hunger hast, da ist Pizza im Kühlschrank.
If you are hungry, there is.IND pizza in the fridge.

7 march
Eric Prendergast (UCB):

Macedonian pronominal clitics: clitic doubling pronouns or differential object agreement?

This talk will present and contrast two perspectives on the syntax of Macedonian pronominal object clitics: 1) that they are the clitic counterparts to full accusative and dative pronouns; 2) that they are object agreement markers, part of the complex of verbal morphology that encodes the phi-features of nominal referents in order to render their Case interpretable at the conceptual-intentional interface. I will outline the structure of the Macedonian pronominal system and illustrate the ways in which pronominal object clitics behave both like pronouns and like object agreement morphology. In particular, I will highlight these clitics’ simultaneous roles as semantico-referential and pragmatic operators. Evidence from the former functions suggest that pronominal clitics arrive at their surface position through A-movement in order to license Case; evidence from the latter suggest that pronominal clitics are either differential object marking morphology or independent constituents that arrive at their position through A’-movement and acquire their phi-features through an Agree relation with appropriate direct or indirect objects to the verb. I will sketch a syntactic analysis of Macedonian pronominal clitics that bridges these contradictory perspectives by relating the licensing of object clitics to an Agree relation with the CP. I will also discuss the challenges that this data poses to anaylses of clitic doubling that have been developed in the Minimalist framework of theoretical syntax, particularly for Romance languages.

28 february
Roberto Zariquiey (PUCP):

Transitivity agreement and transitivity harmony in Panoan and elsewhere: towards a typology

In this talk, I explore one of the most typologically interesting features of Panoan transitivity systems: the encoding of the lexical transitivity of verbs at multiple sites within the clause by means of varied processes of transitivity agreement and transitivity harmony (Valenzuela 2011). A detailed discussion of transitivity agreement and transitivity harmony is provided, arguing that, contrary to the common practice, these two categories must be distinguished. While transitivity agreement seems to be rare, transitivity harmony is attested in different languages of the word. Examples from those languages are also offered in this talk, in order to preliminarily propose some parameters of cross-linguistic variation in association with this phenomenon.

21 february
Nico Baier (UCB) Seereer Infinitives

Katie Sardinha (UCB) Internal causation and event construal: A case study of q'ʷəlisa in Kwak'wala

14 february
Jenny Lederer (SFSU):

Conceptual structures in immigration discourse: How we talk, think, and reason about ‘anchor babies’

In this talk, I offer a detailed account of the pejorative nature of the term ‘anchor baby’, an increasingly common phrase used to frame the children of undocumented immigrants within the United States. Using cognitive linguistic methodology within the Critical Discourse Analysis paradigm, I explore language from both sides of the immigration debate to show, step by step, how a seemingly simple compound can subconsciously introduce or affirm deep value-laden judgments about both authorized and unauthorized immigrants. My explanation rests on the theory of Conceptual Blending in which multiple domains of knowledge including those of seafaring, immigration, and family structure blend together giving rise to a new (and offensive) concept of procreation as a tool to gain legal status, resources, and permanent ties to the U.S. I show that hidden meaning and reasoning, instinctively labeled as offensive by many, but often characterized in vague terms, can in fact be systematically described given the right conceptual framework. I propose the theory of Conceptual Blending can be utilized as a uniquely effective tool to dissect associative processes in the study of semantics and pragmatics.

7 february
40th Meeting of the Berkeley Linguistics Society

31 january
Eva Havu (Uni. Helsinki):

Detached constructions in French: dislocation vs. secondary predication