Skip to content Skip to navigation

Syntax and Morphology Circle

The Stanford Syntax and Morphology Circle (SMircle) is an informal forum for the presentation and discussion of new research in syntax and morphology, their interconnections, as well as their connections with semantics and phonology. Everyone is welcome!

Where and When

In Winter Quarter 2018, the group will typically meet on Fridays 1:30-2:50 pm in the Greenberg Room in Margaret Jacks Hall (460-126). Please see below for details and exceptional meeting times.

Winter 2018 Schedule

  • January 12, 2018

Bonnie Krejci and Beth Levin (Stanford University)

Talking about the weather: Two construals of precipitation events in English

Weather expressions such as It is raining have proven challenging for researchers; languages show considerable variation in how they encode such events (Eriksen et al., 2012). In Romance languages in particular, there has been controversy over whether verbs denoting weather events are unergative or unaccusative (Benincà & Cinque, 1992; Bleotu, 2013; Meulleman & Stockman, 2013; Ruwet, 1991). We show that verbs denoting precipitation events in English (rain, snow, hail) pose the same challenge, and we offer an analysis that explains their apparent hybrid nature. We argue that the unergative/unaccusative behaviors of these English verbs arise from the availability of two distinct event structures, which in turn reflect the availability of two different construals (in the sense of Levin & Rappaport Hovav (2005)) of precipitation happenings. English precipitation events may be construed as substance emission events (1) or directed motion events (2), leading to their variable unergative/unaccusative behaviors.

(1)  Substance emission event
a.   The well gushed (oil).
b.   It rained (a light rain/sulfuric acid).

(2)  Directed motion event
a.   An apple fell on the ground.
b.   A light rain rained on my head.

Our analysis of English precipitation events helps resolve the controversy over the status of weather verbs in Romance languages: when precipitation verbs show unaccusative behavior, they show the hallmarks of a directed motion event structure, and when they show unergative behavior, they pattern as activities (Benincà & Cinque, 1992). More broadly, precipitation verbs further support the association of activities with unergative behavior and of directed motion (or scalar change in general) with unaccusative behavior.
  • January 26, 2018

Erik Zyman and Nick Kalivoda (UCSC)

XP- and X⁰-movement in the Latin Verb: Evidence from Mirroring and Anti-Mirroring

A great deal of crosslinguistic evidence supports the Mirror Principle (Baker 1985): if a Root and some functional heads (F⁰s) above it are packaged into a morphological word, then the higher such an F⁰ is syntactically, the farther from the Root it appears in the word. In Latin, however, verbs seem to both obey and disobey the Mirror Principle (in different inflectional forms). In Passive Subjunctive forms such as (1), Mood occurs much closer to the Root than does Voice (Calabrese 1985 via Cinque 1999:197), seemingly violating the Mirror Principle.

'(that) he/she/it may be praised'

The Mirror Principle being a mainstay of morphosyntactic theory, apparent counterexamples to it deserve scrutiny. We propose an analysis of Latin verb-building on which Latin does in fact obey the Mirror Principle, but this is obscured in forms such as (1) because Latin verb forms are derived by a combination of head movement and phrasal (specifically, vP) movement (cf. Bailey 2010, Gianollo 2016, Danckaert to appear). The analysis has the following positive consequences:
1) The apparent counterexample to the Mirror Principle is eliminated. 
2) Uniformity: All Latin finite verb forms, whether apparently "mirroring" or "anti-mirroring," have the same syntactic derivation.

3) Generality: Though the puzzle arose among the finite forms, the analysis extends naturally to nonfinite and even nonverbal forms.
4) Syntactic fit: The analysis makes correct predictions about adverb placement and verb movement.

The analysis also supports the hypothesis that (morpho)phonological "words" need not be syntactic constituents (Julien 2002, Kayne 2017).

  • February 2, 2018

Eva Portelance (Stanford University)

Probabilistic MInimalist Grammar Induction

In the interest of bridging the gap between natural language processing and syntactic theory, the following project explores the possibility of automating the process of learning a minimalist grammar for a language given a corpus of sentences and our prior knowledge of such formalisms. I will present the current sketch of a generative model for minimalist grammars and preliminary results of grammar induction, and will also describe future applications for this research to both the fields of syntax and natural language processing.

  • February 16, 2018

Virginia Dawson (UC Berkeley)

The emergence of case matching in discontiguous DPs (joint work with Emily Clem)

Many languages show case concord among various DP-internal elements. Some of these languages, such as Warlpiri, additionally allow discontinuous DPs and preserve concord in these split structures. In this talk, we examine novel data from two unrelated languages, Tiwa (Tibeto-Burman; India) and Amahuaca (Panoan; Peru), which show a distinct, previously unanalyzed pattern: case concord is possible in discontinuous DPs but not in continuous DPs. We argue that this pattern of case concord only under discontiguity arises as a result of multiple DP layers and a restricted mechanism of concord we call "D-concord''. We demonstrate that this style of analysis correctly predicts instances of case stacking in Tiwa and the interactions of case concord and differential case marking found in both languages. From these data, we conclude that while the patterns in Warlpiri and the patterns in Tiwa and Amahuaca may both pretheoretically be identified as concord, they arise due to two distinct mechanisms. Thus, case concord only under discontiguity represents a new type of concord altogether. 

  • February 23, 2018

Maziar Toosarvandani (UCSC)

Variation and uniformity in constraints on clitic combinations (joint work with Steven Foley)

Languages that have clitic pronouns frequently prohibit certain combinations of these clitics (e.g., the Person-Case Constraint). Why do these constraints restrict just clitic pronouns, not arguments more generally? And, why are only some combinations of clitics prohibited and not others? We identify two patterns in the clitic combinations that are allowed across languages and across phi-domains (across person and gender). These patterns arise, we propose, from how clitics are licensed syntactically; certain asymmetries point, in particular, to the universal role played by a cyclic version of Agree in clitic licensing. The attested variation across languages in how they constrain clitic combinations can then be derived entirely from variation in their lexicons.

  • March 2, 2018

Tessa Scott (UC Berkeley)

Syntactic deletion within resumptive pronouns in Swahili

Swahili has a pattern of resumption in relative clauses which is unique but predicted in several theories of resumption. There are two types of resumptive pronouns (RPs) in Swahili: those derived through movement and those derived through base generation and binding. The two types are morphologically distinct: the movement RP lacks person features but the bound RP obligatorily matches the head in person features. Data for movement RPs come from parasitic gap constructions and data for bound pronouns come from islands. I analyze this pattern, using common theoretical frameworks (Minimalism and DM), by building on van Urk (2017)'s theory of copied pronouns in which movement copies are subject to a sophisticated deletion algorithm that is sensitive to phonological requirements of certain positions. In Swahili, strict disyllabic word minimality requires monosyllabic prepositions to spell out their complement. This requirement is at odds with another part of the algorithm that tries to delete lower copies, resulting in partial deletion. I argue that this type of algorithm favors economy over syntactic phase-hood and deletes as much as is possible, given the inventory of vocabulary items. The data present new empirical evidence that not only can one language have two types of resumptive pronouns, but they can be morphologically distinct. The data also shed light on the nature and derivation of agreement mismatches in movement chains. 

  • March 16, 2018

Anastasiia Ionova (Leiden University & UCSC)

When and why clitics move: On the interaction of second position cliticization and ellipsis

The phenomenon of second position cliticization remains a mystery in many of its aspects, one of which is the timing of clitic movement. In this talk, I focus on Serbo-Croatian and argue that clitics move to the second position at PF. A new argument for this approach comes from the interaction of second position cliticization and ellipsis: a study has shown that ellipsis bleeds clitic movement to the second position, as in (1).

(1)        Mi        ih         nismo  videli,   a          oni       su        rekli     da        (*ih)      jesu     <videli    (ih)>.

            we       them    aren't   saw     but       they     are       said     that      them     are       saw     them              

            “We didn’t see them, but they said that they did.”

The bleeding effect can be easily accounted for in terms of the relative timing of ellipsis and second position cliticization: ellipsis precedes clitic movement. Two questions arise: (i) what exactly blocks clitic movement to the second position and (ii) why there is no violation of the second position requirement if the clitic ih in (1) stays inside the VP? I answer both questions by suggesting that putative clitics inside the ellipsis site never receive their clitic status.

I argue that the choice between a clitic and an accented form of a pronoun or an auxiliary is made no earlier than at the stage of Vocabulary insertion, and the clitic form comes with the requirement to appear in the second position. If ellipsis is the non-insertion of vocabulary items, a pronoun inside the ellipsis side would never become a clitic and would have no motivation to move to the second position.

In sum, there are no second position clitics in syntax, and clitic movement occurs after Vocabulary Insertion, i.e. at PF.


Spring 2018 Schedule

  • April 6, 2018

Amy Rose Deal (UC Berkeley)

  • April 13, 2018

Annie Zaenen (Stanford University)

  • April 27, 2018

Peter Jenks (UC Berkeley)

  • May 11, 2018

David Basilico (University of Alabama at Birmingham)

  • May 25, 2018

Zuzanna Fuchs (Harvard University)


Autumn 2017 Schedule

  • October 13, 2017

Reuben Cohn-Gordon (Stanford University)

Amharic exhibits an unusual phenomenon, where what seems to be an object marker can appear on intransitive verbs. I discuss the semantic effects of these markers, and then draw some conjectures as to their syntactic analysis and diachrony, based on Ruth Kramer's analysis of Amharic object markers as doubled clitics.

  • October 20, 2017

Boris Harizanov and Line Mikkelsen (Stanford University, UC Berkeley)

Resumption and Cyclic Chain Reduction in Danish VP Left Dislocation 

  • November 3, 2017

Bonnie Krejci (Stanford University)

Animacy and agreement with conjoined nominals in Russian

In certain clauses in Russian, when the subject apparently consists of two conjoined noun phrases, the intransitive verb may agree with the first noun phrase, resulting in First Conjunct Agreement (FCA), as in (1).

(1) V  polnoč'     pojavilas'            luna              i       odna         zvezda.
     at  midnight  appeared.FSG  moon.FSG  and  one.FSG  star.FSG
    "At midnight the moon and one star appeared."
FCA in Russian has a limited distribution: (a) it may only occur when the subject appears postverbally, and (b) it may only occur when subjects are inanimate/nonagentive. I present a novel analysis of FCA that accounts for these constraints, in which Russian FCA is the result of the conjunction of two verb phrases, rather than two noun phrases. I follow Glushan (2013) in arguing that semantic properties of the subject, such as animacy, have structural correlates that play a role in the availability of FCA, and I present the results of a judgment study supporting that conclusion. Finally, I show that conjunct-sensitive concord in the nominal domain is also sensitive to the animacy of the conjuncts, and discuss the potential connections between FCA and conjunct-sensitive concord.
  • December 1, 2017

Scott Borgeson (Stanford University)

Verb-Second in Estonian

Most clauses in Estonian follow the verb-second pattern shown below.

1. Eile             taht-s-in               ma        jaluta-da. 
yesterday want-PAST-1SG 1SG.NOM walk-INF

2. Ma              taht-s-in               eile        jaluta-da.
1SG.NOM want-PAST-1SG  yesterday walk-INF

Traditionally, such systems have been explained by positing that 1) the finite verb moves to some phrasal head at the left edge of the clause, and 2) that a single constituent moves to the specifier position of that head (den Besten 1983, Vikner 1995, etc.). Most apparent exceptions to this pattern that are found in the literature (Lindstrom 2001 and 2007, Erelt 2003), can easily be explained with a few simple stipulations— that is, without abandoning the above framework.

In this talk, I document and examine a new construction, which is universally accepted by some speakers, only accepted in certain environments by others, and unacceptable in any environment by a third group. These are clauses which resemble the verb-second pattern, but in which multiple phrasal constituents appear before the fronted verb.

3. Ma              seda            taht-s-in             eile         teh-a. 
1SG.NOM this.PART want-PAST-1SG yesterday do-INF

I argue that the traditional verb-second framework is sufficient to capture such sentences if we posit that more than one constituent can move to the specifier position mentioned above. This account has much in common with multiple-specifier accounts of multiple-WH-movement (see Rudin 1988 and Richards 1998), including the prediction that the ordering of the constituents involved must be the same before and after movement, which I will show is indeed borne out. I will also flesh out an alternative account involving multiple phrase heads with a single specifier each, and will demonstrate that it cannot easily capture the data we observe.
  • December 8, 2017

Andrew Hedding (UC Santa Cruz)

Anti-agreement in Somali

In Somali, subject clitic pronouns can optionally double an R-expression that is in its base-generated position, but are obligatory if the subject is a topic or a pronoun, and prohibited if the subject is in focus. Additionally, under various types of A'-extraction of a subject (e.g. focus, wh-movement, relative clause construction) the verb displays a partial agreement paradigm, and the subject appears in non-nominative case. In this talk, I argue that previous treatments of Somali syntax cannot account for this pattern, and I discuss the issues that arise when applying other theories of anti-agreement to the Somali facts. Additionally, I sketch a potential analysis that highlights the relationship between clitic doubling and resumption in the language. 

Mailing List

Want to receive email updates about all of our activities?

Join our mailing list!


The workshop coordinator is currently Vera Gribanova.

Upcoming Events

Currently, no future events are scheduled.

See all events