Alex Lascarides (Edinburgh) will give a colloquium today (Friday Oct. 18) at 3:30PM in the Greenberg room, followed by a departmental social.
Abstract: In this talk I will present joint work with Nicholas Asher. Our aim is to model conversation when people have conflicting goals; for instance, courtroom cross examination and political debate. I’ll use naturally occurring dialogues to introduce a particular kind of deception, where the speaker doesn’t explicitly lie, but he implies a falsehood that he can subsequently deny was a part of his message. This motivates a new concept, that of SAFETY in discourse interpretation: one needs the means to test whether an implicature in non-cooperative conversation can be treated as a matter of public record.
We’ll propose two alternative ways of modelling both cooperative and non-cooperative conversation, both of which offer an analysis of safety. The first model uses standard techniques from game theory. The second is a proof theoretic analysis which improves on the standard game theory version by fully supporting reasoning about what utterances need to be a part of the definition of the `interpretation game’. In effect, it can distinguish between an utterance that a speaker contemplated but chose not to perform from an utterance that he didn’t contemplate performing at all.
This proof theoretic analysis not only models safety but also provides the means to prove a correspondence Gricean principles of cooperative conversation (e.g., Sincerity) and a situation where the preferences of the conversational participants are normally aligned. Thus Gricean models of implicature are a special case in our framework.
Next colloquium: Esli Kaiser, November 1.
Please join the Phonology Workshop today (Friday October 18) at noon in the Greenberg Room for a talk by Sarah Bakst, (UC Berkeley).
A phonetic basis for the patterning of [χ]
The sonority hierarchy determines a segment’s sonority by its natural class, with obstruents registering low on the scale, followed by nasals, liquids, glides, and finally high-sonority vowels. The definition of sonority and existence thereof remain in dispute, but most definitions relate to syllable structure and phonotactics. The phonetic definition in Wright (2004) ranks segments based on the robustness of formant transitions. Other definitions rely on phonological patterning; Clements (1990) relates sonority to the ability of a segment to be a syllable peak.
Some phonetic realizations of the French rhotic are problematic for the sonority hierarchy. When the rhotic occurs in onsets following a voiceless stop, it is realized as a voiceless uvular fricative [χ]. French rhotics, regardless of the phonetic realization, pattern as high-sonority liquids and are one of only a few French segments allowed to occur between a consonant and a vowel; because of its rhotic status, [χ] is the only fricative in French that may occur in this position. There are two possibilities for the analysis of this segment: either the French sonority hierarchy is phonological and abstract, or there is some phonetic property of the voiceless uvular fricative, such as the ability to bear more robust perceptual cues to preceding segments, that allows this peculiar patterning. The present experiment tests whether [χ] is better than another fricative found in French, [f], at providing cues of a preceding stop in stop-fricative clusters.
Geoffrey Pullum posted a very nice obituary for Ivan this week on Linguist List, worth reading for both personal and historical/intellectual value.
On Friday, October 25 at 3:30PM in the Greenberg Room, Scott Martin (OSU Linguistics Ph.D. 2012, now a researcher at Nuance Communications’ NLU lab) will give a talk at the Construction of Meaning Workshop. All are welcome!
“On the Dynamics of Sense and Implicature: Anaphora, ‘Presuppositions’, and CIs”
Abstract: In this talk, I’ll discuss some of the highlights of my recent dissertation, which is both a descriptive and formal effort to characterize implicatures, parts of the meanings of utterances that are separate from their sense, their main point. From the descriptive angle, I’ll lay out a new taxonomy of implicature that extends Grice’s seminal work based mainly on two criteria: (1) whether an implicature is conventional, and (2) whether it must be anchored to the speaker. In this taxonomy, some lexical items often treated as being presuppositional (factives, aspectuals, achievements) are characterized as merely giving rise to entailments. As corollaries, ‘presupposition’ and ‘anaphora’ become synonyms, and the process of accommodation takes on a much more limited role than is often assumed. I’ll then discuss a formal account of implicature in a compositional semantic framework that blends ideas from Karttunen and Peters, Potts, Beaver, and the dynamic tradition. This semantics uses the same mechanism to model implicatures associated both with anaphora and with Potts’s “CIs,” and allows the two types of implicature to interact, an advance from previous theories. It also provides a unified account of the (in)felicity of anaphora and CIs.
We inadvertently missed an important item in last week’s Look Who’s Talking! segment:
- At NWAV43 this weekend, John R. Rickford and Sharese King will present “Rachel Jeantel’s Testimony in the Zimmerman case: Descriptive, Theoretical and Applied Linguistics Perspectives”.
Tom Wasow will be presenting at Princeton University this Thursday, October 24 on “The Do-Be Construction.”
Lauri Karttunen and Annie Zaenen will be giving two talks at the University of Edinburgh this weekend:
- “The multiple interpretations of be lucky to VP”
- “When you aren’t stupid you do not do stupid things: implicative uses of evaluative factive adjectives”
These presentations result from joint work with Cleo Condoravdi and Stanley Peters and their 2012 interns Miriam King Connor, Marianne Naval, Tania Rojas-Esponda, and their 2013 interns, Stuart Melton, Kenny Moran, Dasha Popova, in the Language and Natural Reasoning group at CSLI.
Stephanie Shih and Sharon Inkelas will present “Unstable surface correspondence as the source of local conspiracies” at NELS 44 at the University of Connecticut on October 18-20.
A number of current or recent Stanford linguists and affliates will present their work at EMNLP 2013 in Seattle, WA on October 18-21:
- Sida Wang, Mengqiu Wang, Chris Manning, Percy Liang and Stefan Wager: “Feature Noising for Log-linear Structured Prediction”
- Aju Thalappillil Scaria, Jonathan Berant, Mengqiu Wang, Peter Clark, Justin Lewis, Brittany Harding and Christopher Manning: “Learning Biological Processes with Global Constraints”
- Richard Socher, Alex Perelygin, Jean Wu, Christopher Manning, Andrew Ng and Jason Chuang: “Recursive Models for Semantic Compositionality Over a Sentiment Treebank”
- Will Zou, Richard Socher, Daniel Cer and Christopher Manning: “Bilingual Word Embeddings for Phrase-Based Machine Translation”
- Valentin Spitkovsky, Hiyan Alshawi and Daniel Jurafsky: “Breaking Out of Local Optima with Count Transforms and Model Recombinations: A Study in Grammar Induction”
- Joo-Kyung Kim and Marie-Catherine de Marneffe: “Deriving adjectival scales from continuous space word representations”