Issue of Monday, November 9th, 2026

Syntax Square 11/10 - János Egressy (UCLA)

Speaker: János Egressy (UCLA)
Title: Williams Cycle and Successive Cyclicity unified
Time: Tuesday, November 10, 1-2 PM
Location: 32-D461

Under standard Phase Theory, the locality constraint deriving successive cyclicity is the PIC (Chomsky 2001), which requires movement to proceed via the edge of phases. For example, a CP phase can only be escaped via Spec,CP. The Williams Cycle (Williams 2003; Poole 2023) generalizes selective opacity, a different kind of locality: It is a clause’s relative size that makes it opaque to probes that are not sufficiently high in the next lowest clause. For example, probes on C but not on T can search into CP because C>T in the functional sequence. Most accounts of selective opacity (Keine 2020) or the Williams Cycle effects (Meadows 2023) handle successive cyclicity and Williams Cycle effects as independent phenomena, which need to be derived from separate constraining mechanisms. That is, while PIC may necessitate successive cyclicity out of a CP because it’s a phase, and independently, the same CP may exhibit Williams Cycle effects because it’s too large. 
 
This talk provides a unified account that derives successive cyclicity and Williams Cycle effects. Specifically, I introduce a novel constraint, which dictates movement steps must proceed along well-formed fseq-fragments, i.e. must be turned into steps that are clause-internal in the relevant sense. Cross-clausal movement can proceed via two strategies. My account predicts that with the first strategy, successive cyclic sub-extraction, Williams Cycle effects are obviated because movement from the embedded clause’s edge is free to the matrix clause (e.g. in Mongolian (Fong, 2019) and P’urhepecha clauses (Zyman, 2017)). If sub-extraction cannot proceed successive cyclically from a clause (e.g. German CPs), sub-extraction can still proceed via an alternative strategy: The clause moves to an equal-sized matrix projection. Since the clause’s movement is delayed till the matrix clause reaches the size of the embedded clause, clauses requiring extraposition for transparency will exhibit Williams Cycle effects.
 
Personal website: https://sites.google.com/view/egressyjanos/
 

Phonology Circle 11/9 - Amy Li (MIT)

Speaker: Amy Li (MIT)
Title: Inventory crowding and within-context variability of speech sounds
Time: Monday, November 9, 5pm-6:30pm

Location: 32-D831
Abstract: Many authors have speculated that more crowded inventories tolerate less variability (Martinet, 1952, p. 4; Flege, 1989, and references in pp. 127-8; Cortés Kandler, 2026, and references therein; Zhang & Chodroff, 2026, and references therein). For example, speakers of a language that contrasts /i, a, u/ may vary the exact realization of each of their vowels more than speakers of a language with /i, e, a, o, u/. However, no mechanism explaining the possible effect has been suggested, and empirical evidence is mixed. In response, I propose the “probabilistic dispersion theory of contrast” (augmenting the “dispersion theory of contrast”, Flemming, 2002), which optimizes inventories of speech sounds to balance the probability of misperception with the difficulty associated with reducing variability. The central problem for crowded inventories is high probability of confusion. This model claims that tolerance for confusion is upper bounded in all languages and predicts that languages that permit maximal confusion exhibit the hypothesized relationship. Three new case studies comparing within-context vowel variability across localized differences in inventory crowding find some evidence for more crowded inventories having less variability, supporting the proposed theory.

LingLunch 11/12 - Ahmed Alhuwayshil (University of Edinburgh/MIT)

Speaker: Ahmed Alhuwayshil (University of Edinburgh/MIT)
Title: Redundant Scalar Implicatures
Time: Thursday, November 12th, 12:30pm - 2pm
Location: 32-D461

Abstract: In the long-standing debate around the processing of scalar implicatures (SIs), experimental work has focused on whether scalar implicatures are ‘default’ or ‘contextual’. Defaultism covers possibilities whereby SIs are automatic enrichments of lexical meaning or achieved by the activation of a grammatical exhaustivity operator (Exh). In the experimental literature, contextualism is usually associated with a Relevance Theoretic (RT) approach (Sperber & Wilson 1986) in which SIs arise only if they are germane to current discourse needs. My work focuses on an understudied prediction which distinguishes default from RT accounts. Default accounts predict that SIs arise for listeners even when their content is informationally redundant whereas RT accounts balance an enrichment’s positive cognitive effects against the listener’s processing effort. Using a priming paradigm across three mouse-tracking experiments and one button-press reaction-time experiment, I provide evidence suggesting that SIs are calculated even when they are informationally redundant. I argue that these findings challenge the claims of the RT account and support formal semantic approaches in which exhaustification is triggered by a grammatical exhaustivity operator (Exh) as proposed by Chierchia (2004), Fox (2007) and Chierchia et al. 2012).

Katochoritis @ TLC at the University of Vienna

Our fourth-year student Ioannis Katochoritis gave a talk at the Theoretical Linguistics Colloquium (TLC) at the University of Vienna on November 6, 2026, entitled “It’s just a phase, it’ll pass: The key to unlocking and moving on”. Here is the abstract:

Rackowski and Richards (2005) propose that phases can be ‘unlocked’ to become transparent for extraction if they first Agree with a higher probe, which may then attract a goal from within the phase’s domain. Provided that phases count by default as potential movable interveners due to their featural make-up (Abels 2003), then prior Agree with the phase allows the probe to skip the phase boundary, as per Richards’ (1998) Principle of Minimal Compliance in (1). Thus, in many languages, extraction out of CPs or DPs requires the matrix verb to (φ-)agree with the containing phase.

(1) Once a probe P Agrees with a goal G, P can ignore G for the rest of the derivation.

Although the research program of Phase Unlocking has been prolific (see a.o. van Urk & Richards 2015; Branan 2018; Halpert 2019; Hedding & Yuan 2023; Ershova 2024; Lee & Yip 2024), it still raises important questions: (i) Are unlocking and successive cyclicity both required to escape a phase or not? (ii) How is unlocking compatible with the Phase Impenetrability Condition (Chomsky 2000) and spell-out? (iii)  How does the Principle of Minimal Compliance allow to ignore an intervening locality domain? (iv) Is unlocking an exceptional primitive operation or part of a broader strategy to undo opaqueness?    

To address these issues, I propose the condition in (2), for which I will provide novel empirical evidence from Austronesian Malagasy and Tagalog:

(2) Unlocking of phase XP requires remerge of XP to the edge of the probing head H, before an XP-embedded goal YP can subextract to a(n outer) SpecHP.

Thus, Agree is not sufficient for unlocking; it is only movement of the containing phase into the projection of the probing head that makes that phase transparent for extraction, yielding a multiple-specifier configuration. I suggest that (2) can ultimately provide an answer to all four open questions above.

In the end, (2) will also support the special status of phases as absolute cyclic locality domains, arguing against Halpert and Zeiljstra (2026), who subsume all phase-like locality effects to Relativized Minimality (Rizzi 1990).