Research

In my postdoc project at Michael Wagner’s prosody lab, we are working on the effect of intonation on the interpretation of ambiguous Focus-particles such at English at least. The project is motivated by the intuition that a sentence like (1) can either convey uncertainty regarding the number of cats that peed on the floor, or a – probably odd – preference for some rather than no cats having peed on the floor, depending on the pitch accent on ‘some’. The goal is to link such prosodic effects, in addition to syntactic factors, to a theory of how and what kind of alternatives Focus evokes. We plan to investigate this issue experimentally using both production and comprehension tasks.

(1) At least SOME of the cats peed on the floor.

One of my previous projects that also involves Focus-particles and intonational meaning and is related to this one can be found in the “Additive Particles” section.

Presuppositions are known to vary in how strictly they impose their requirements on the context and result in infelicity if not met, i.e. how easily they can be accommodated. As part of my dissertation, I provide experimental evidence that one relevant factor is Focus-sensitivity, with Focus-sensitive triggers being harder to accommodate than triggers lacking Focus-sensitivity. I relate this contrast to a difference in the underlying representations that these two classes access. While Focus-sensitive triggers require a linguistic antecedent in the discourse model, triggers that are not Focus-sensitive need their presupposition to be entailed by the Common Ground.

• Göbel, A. (2020). Representing Context: Presupposition Triggers and Focus-sensitivity. PhD Thesis, University of Massachusetts, Amherst. [preprint]
• — Representing Discourse Context: Presupposition Triggers and Focus-sensitivity. Virtual talk at Language & Cognition (LangCog) @Harvard, October 2020. [slides]
• — The Common Ground is not enough: Why Focus-sensitivity matters for Presupposition triggers. Virtual poster at Experiments in Linguistic Meaning (ELM), September 2020. [lightning talk slides, poster slides]
• — Representing Context: Presuppositions and Focus-sensitivity. Virtual talk at CompLang @MIT, April 2020. [slides]
• — Accommodating presuppositions: it’s focus-sensitivity that makes it hard. Virtual poster presented at the 33rd Annual CUNY Human Sentence Processing Conference (CUNY33) @UMass, Amherst, March 2020. [OSF Repository]

I have recently started on extensions to this work in collaboration with Thuy Bui, adding a cross-linguistic component by comparing English, German and Vietnamese. We plan to compare local and global accommodation for a broader range of triggers, as well as investigate additional factors that might play a role for accommodation, such as the event structure of aspectual triggers, what constituent a Focus-sensitive trigger associates with, and plausibility.

I have also been interested in the role of trigger differences with respect to accommodation for first language acquisition, which I have been working on in collaboration with Emma Nguyen. We compare two types of triggers – still as a trigger which has been argued to be independent of the assertion and continue as one that contains both presupposed and asserted content – and test both children’s and adult’s comprehension of them. The hypothesis we want to test is whether children are more likely to ignore a presupposition failure than adults, and whether they do more so for still than for continue assuming that the former is independent of the assertion.

• Nguyen, E. & A. Göbel (2018). Still true? How do children deal with presupposition failures across triggers. Talk at Language Acquisition Workshop in New England (LAWNE) @UConn, May 2018.

As a more general overview of research on trigger differences, I have put together a manuscript that provides an exhaustive review of all trigger classifications in the literature and related experimental work. Write me if you’d like a copy.

As part of my dissertation, I investigated how different presupposition triggers, namely also and again, find the material that satisfies their presupposition in short dialogues. The hypothesis I tested was that triggers differ in the cognitive representations they access based on Focus-sensitivity, with Focus-sensitive triggers requiring a linguistic antecedent in the discourse model, while triggers that are not Focus-sensitive need their presupposition to be entailed by the Common Ground. One source of evidence in favor of this hypothesis (see “Presupposition Accommodation” for the other) comes from the finding that also seems to be subject to interference-type effects relative to the QUD when again is not.

• Göbel, A. (2020). Representing Context: Presupposition Triggers and Focus-sensitivity. PhD Thesis, University of Massachusetts, Amherst. [preprint]
• — Representing Discourse Context: Presupposition Triggers and Focus-sensitivity. Virtual talk at Language & Cognition (LangCog) @Harvard, October 2020. [slides]
• — The Common Ground is not enough: Why Focus-sensitivity matters for Presupposition triggers. Virtual poster at Experiments in Linguistic Meaning (ELM), September 2020. [lightning talk slides, poster slides]
• — Representing Context: Presuppositions and Focus-sensitivity. Virtual talk at CompLang @MIT, April 2020. [slides]
• — Representing Context: Focus Alternatives, Common Ground and the QUD. Poster presented at the 94th Annual Meeting of the Linguistic Society of America (LSA2020) @New Orleans, January 2020. [poster]
• Göbel, A. (2019). Representing discourse: Focus, presuppositions and the QUD. Talk at Psycholinguistics/Neurolinguistics Seminar @UCLA, May 2019. [slides]

I am currently working on further extending this work by examining additional discourse factors contributing or ameliorating this effect. My hope is that this research will elevate the status of discourse processing informed by linguistic theory in psycholinguistics by drawing connections to prominent ideas on syntactic memory such as cue-based retrieval, as well as illustrate the relevance of discourse processing for semantic theory.

This first project is concerned with cases where also and too differ in acceptability. I argue that this contrast comes about due to an interaction with an intonational contour that mirrors the rise-fall-rise contour, pointing towards a common core of the L*+H pitch accent that is common among both.

• Göbel, A. (2019). Additives pitching in: L*+H signals ordered Focus alternative. Poster at Semantics and Linguistic Theory 29 (SALT29) @UCLA, May 2019.
• Göbel, A. (2016). Dividing Additives: On too, also and German auch. Talk at Southern New England Workshop in Semantics (SNEWS2016) @Brown University, November 2016. [handout]

I have also worked on the obligatoriness of additive particles in certain contexts, specifically in but-clauses, and argued for an analysis of additives as the negation of the exhaustivity operator and to relate its contribution to the Question Under Discussion.

• Göbel, A. (2018). Additives in ‘but’-clauses: ‘also’ as the Negation of EXH. Poster at 36th West Coast Conference on Formal Linguistics (WCCFL36) @UCLA, April 2018. [poster]

Finally, for my first Generals Paper, I did psycholinguistic work testing the hypothesis that additive particles qua anaphoricity show the processing profile of anaphors, focusing on parallelism effects.

• Göbel, A., B. Dillon & L. Frazier (2018). Investigating the Parallelism Requirement of ‘too’. Poster at the 92nd Annual Meeting of the Linguistic Society of America (LSA2018) @Salt Lake City, January 2018. [poster]
• Göbel, A. (2017). On Processing Anaphoric Presuppositions. Talk at UPenn Experimental Lab Meeting, March 2017. [slides]

A conglomeration of these different pieces was presented in the talk below.

• Göbel, A. (2018). Contrasting additives: on obligatoriness, coherence and the QUD. Invited talk at the Departmental Colloquium (“Linguistischer Arbeitskreis”), University of Cologne, June 2018. [handout]

Other phenomena related to additive particles I hope to investigate in the future are what I call double additives – seemingly redundant uses of a preverbal and a sentence-final additive associating with the same constituent – as well as scalar uses of too that don’t seem to be compatible with also, constituting the flip-side to the first project listed here.

It has been observed that the accessibility of parts of a multi-sentence/clausal discourse, as tested via anaphoric devices, is affected by the semantic relations that the respective discourse units stand in, the so-called Right Frontier Constraint (RFC). In a case study that picks up findings by Syrett & Koev (2015) on the special status of sentence-final appositive relative clauses, I provided experimental evidence for the RFC, as well as the need for a theory of discourse structure to take syntactic-pragmatic information of the linked clauses into account.

• Göbel, A. (2019). Final Appositives at the Right Frontier: An Experimental Investigation of Anaphoric Potential. Proceedings of Sinn und Bedeutung (SuB) 23. [final draft] [slides]
• Göbel, A. (2018). Pronouns at the right frontier: discourse structure affects accessibility of final appositives. Talk at 2nd International Conference on Prominence in Language (ICPL2) @Cologne, July 2018. [slides]
• Göbel, A. (2018). Pronouns at the right frontier: discourse structure affects accessibility of final appositives. Poster at the 31st Annual CUNY Sentence Processing Conference (CUNY31) @UC Davis, March 2018. [poster]

Since then, I have been exploring the question of what’s underlying the RFC with an attempt to connect QUD-based theories of discourse structure with relational approaches such as SDRT.

• Göbel, A. (2018). Exploring a Connection between QUDs and Discourse Structure. Talk at Workshop on “At-Issue, Scope and Coherence” @Cologne, July 2018. [handout]

In collaboration with Lyn Frazier and Chuck Clifton, I have been investigating how discourse relations affect syntactic dependencies. Based on a set of acceptability studies, we argue for conceiving of effects in line with the RFC in terms of how events are represented on the narrative timeline. A manuscript is currently under review at Discourse Processes.

For my second Generals Paper, I investigated the behavior of German D-pronouns in attitude contexts. I argue that co-reference of an embedded D-pronoun with an attitude holder is only possible (i) in a de re context or (ii) if the attitude holder is herself instantiated as a D-pronoun or epithet. These properties are derived by appealing to the expressive, epithet-like component of D-pronouns and feature transmission respectively.

• Göbel, A. (2019). Don’t give me that attitude! Anti-De Se and Feature Matching of German D-Pronouns. Proceedings of North East Linguistic Society (NELS) 49. [poster]
• Göbel, A. (2018). On German D-pronouns as Anti-Logophoric: Limiting a Competition-Based Account. Poster at Workshop on “Pronouns in Competition” @UC Santa Cruz, April 2018. [poster]

During my XPrag.de internship at the University of Cologne under the supervision of Petra Schumacher, I started investigating the relationship between D-pronouns and attitude ascriptions experimentally, which showed that de re contexts incur a processing difficulty offline and provided some tentative evidence for the importance of the expressive component of D-pronouns.

• Göbel, A. & P. Schumacher (2019). Self-awareness matters: co-reference in German attitude clauses. Poster at the 32nd Annual CUNY Sentence Processing Conference (CUNY32) @University of Colorado, Boulder, March 2019. [poster]

In a separate project in collaboration with Chris Hammerly on the obviation system in Ojibwe, we are arguing that obviation is governed by perspectival properties and provide novel data from attitude ascriptions in favor of this hypothesis.

• Göbel, A. & C. Hammerly (2019). A new perspective on obviation from attitude contexts. Talk at Person & Perspective Workshop @USC, May 2019.
• Hammerly, C & A. Göbel (2019). A new perspective on obviation in Ojibwe from attitude contexts. Poster at the 93rd Annual Meeting of the Linguistic Society of America (LSA2019) @New York City, January 2019.
• Hammerly, C & A. Göbel (2018). A “perspective” on obviation from attitude contexts. Talk at 50th Algonquian Conference @University of Alberta, October 2018.
• Göbel, A. & Hammerly, C. (2017). A new perspective on obviation in Ojibwe from the German pronominal system. Talk at Southern New England Workshop in Semantics (SNEWS2017) @MIT, December 2017.

Finally, in collaboration with Isabelle Charnavel and Brian Dillon, we have been investigating the role of perspective in implicit causality phenomena, as well as how (anti-)logophoricity may be processed. (Email me for preliminary results.)

This project is concerned with the semantics and pragmatics of wohl, with a special focus on its effects in interrogatives. For declaratives, I have argued that wohl contributes evidential meaning, contrary to the account by Zimmermann (2008), who analyzes it as indicating uncertainty. In interrogatives, wohl seems to give rise to undirected questions which cannot be used to address someone directly but are best translated as being prefaced by an implicit I wonder… . Moreover, these questions interact with the syntactic structure insofar as the verb position seems to indicate whether the speech-act is anchored to the speaker’s epistemic state or to an addressee. Interestingly, a similar ambiguity of evidential markers in questions has been observed in languages with evidential systems, for example Cheyenne (Murray 2010), which may open up a promising line of research on the cross-linguistic connection.
I am also working on discourse particle combinations, more specifically on an explanation for corpus data which provided counter-evidence against the rigidity of the combination ja wohl, but rather suggests that the order is reversible under certain information-structural conditions.

• Göbel, A. (2018). Evidentiality and Undirected Questions: A New Account of the German Discourse Particle wohl. University of Pennsylvania Working Papers in Linguistics 24.1, 77-86. [pdf] [handout]
• Göbel, A. (2016). On Evidentiality and Undirected Questions: Some Puzzles of the German Discourse Particle wohl. Talk at the UConn Meaning Group, November 2016. [handout]

This was the topic of my Masters Thesis and starts with a simple observation illustrated in (1) below: Subordinate clauses introduced by als ob (‘as if’) in German can be interpreted in two different ways. They can either function as something very close to a manner adverbial such that they modify the manner of the activity expressed by the verb of the matrix clause (1a), or they can refer to the whole proposition of the main clause (1b). (I believe this ambiguity is also present in English.)

(1a) Emma hat eine Auszeichnung gefeiert, als ob sie schüchtern wäre.
(1b) Emma hat eine Auszeichnung verschwiegen, als ob sie schüchtern wäre.
‘Emma {celebrated/concealed} an award as if she was shy.’

In my thesis, I investigated syntactic and semantic factors influencing the interpretation of this ambiguity – for instance the lexical aspect or aktionsart used in (1) – as well as the relation between this type of adverbial clause and non-sentential adverbials using experimental methods.

• Göbel, A. (2016).  Ambige Adverbialsätze an der Syntax-Semantik Schnittstelle: Eine psycholinguistische Fallstudie zu ‘Als ob’. [slides (only in German)] (‘Ambiguous adverbial-clauses at the syntax-semantics-interface: A psycholinguistic case study on als ob (‘as if’)’) Talk at the Workshop “Position and Interpretation: Syntax, Semantics and Information-Structure of Adverbial Modifiers”, June 2016.