What do I study?
My work is in the domain of experimental semantics and pragmatics. Overall, I am interested in how presupposition, implicature, and inference more generally work to shape the way we communicate using language. I am especially interested in our understanding of other's knowledge states, and how that understanding shapes the way we refer to things and the way we interpret referring expressions in conversation and in other language situations.
My current focus is on common ground: how we make inferences about what is in common ground, how we represent common ground, and whether and how we update those inferences and representations on the basis of information exchanged during interactive conversation.
As part of my work on common ground, I am interested in the production, interpretation, and mental repesentations of different types of referring expressions such as common nouns, definite descriptions, proper names, and things which appear to behave like proper names, such as the entrained-upon referring expressions developed collaboratively between interlocutors in interactive dialogs. In addition, I am interested in the interpretation of so-called "weak definite" noun phrases (such as "the trash", in "Mary took out the trash last weekend, and so did Jim", where Jim and Mary don't necessarily have to have taken out the same trash), which do not seem to carry the same sorts of presuppositions relating to unique identifiability as do ordinary definite NPs, and appear to be associated with enriched or event-based interpretations.
My work is in the domain of experimental semantics and pragmatics. Overall, I am interested in how presupposition, implicature, and inference more generally work to shape the way we communicate using language. I am especially interested in our understanding of other's knowledge states, and how that understanding shapes the way we refer to things and the way we interpret referring expressions in conversation and in other language situations.
My current focus is on common ground: how we make inferences about what is in common ground, how we represent common ground, and whether and how we update those inferences and representations on the basis of information exchanged during interactive conversation.
As part of my work on common ground, I am interested in the production, interpretation, and mental repesentations of different types of referring expressions such as common nouns, definite descriptions, proper names, and things which appear to behave like proper names, such as the entrained-upon referring expressions developed collaboratively between interlocutors in interactive dialogs. In addition, I am interested in the interpretation of so-called "weak definite" noun phrases (such as "the trash", in "Mary took out the trash last weekend, and so did Jim", where Jim and Mary don't necessarily have to have taken out the same trash), which do not seem to carry the same sorts of presuppositions relating to unique identifiability as do ordinary definite NPs, and appear to be associated with enriched or event-based interpretations.
How did I get here?
I once was a linguist of the computational persuasion, writing algorithms to match pronouns up with their antecedents, but then I started caring too much about dialog and implicature and inference and all that sort of stuff, and thus started drifting into Cognitive Science and AI, which is how I wound up here.
I have always been fascinated by linguistic communication. The notion that we can have an idea in our head, decide we want to communicate that idea to someone else, produce a bunch of sounds, and by virtue of doing so, put that idea (or some other idea, depending on how successful we are) in someone else's head, is to me absolutely magical. Except that, unlike magic, it works, here in the real world. It's even more amazing when you think about how much of the idea that we're trying to convey is typically underspecified by the particular series of sounds (or more precisely, the words and structures made up by those sounds) we produce. How on earth does this work? It works because we know so much: about our language, about the world, about how we (and other people) think. I am interested in understanding how all of these sources of knowledge come together to allow us to communicate via language, and how we might be able to endow artificial agents with the knowledge necessary for linguistic communication.
I once was a linguist of the computational persuasion, writing algorithms to match pronouns up with their antecedents, but then I started caring too much about dialog and implicature and inference and all that sort of stuff, and thus started drifting into Cognitive Science and AI, which is how I wound up here.
I have always been fascinated by linguistic communication. The notion that we can have an idea in our head, decide we want to communicate that idea to someone else, produce a bunch of sounds, and by virtue of doing so, put that idea (or some other idea, depending on how successful we are) in someone else's head, is to me absolutely magical. Except that, unlike magic, it works, here in the real world. It's even more amazing when you think about how much of the idea that we're trying to convey is typically underspecified by the particular series of sounds (or more precisely, the words and structures made up by those sounds) we produce. How on earth does this work? It works because we know so much: about our language, about the world, about how we (and other people) think. I am interested in understanding how all of these sources of knowledge come together to allow us to communicate via language, and how we might be able to endow artificial agents with the knowledge necessary for linguistic communication.
Writing:
I also have a very strong interest in writing. Not just the act itself (though if my attempts to summarize my interests are any indication, I do like the act of writing quite a bit!), but in the various ways written language is and is not like spoken language, and what it is that makes writing "good", and what makes some people better at producing "good" writing than others. When we write, we are able to make choices about the way we express a particular idea in a way that the fast pace of spoken language does not allow, and we are also able to edit what we have said in a way that is impossible in spoken language. What kinds of effects might these differences have on the cognitive processes involved in communicating via language?
We tell student writers that one of the most important things they can do to make their writing better is to write for a reader, to take their audience into account - but as writers, we cannot collaborate with our audience to ensure understanding, there is no "grounding process", and we do not get anything like the kind of immediate feedback that we get when speaking regarding how well we are being understood. But of course, as writers, we also have more time than speakers have to consider what our audience knows, and even how they might interpret a particular phrase, sentence, or passage, and we can revise to avoid ambiguity. Should we expect that students who are particularly good at taking their interlocutor's perspective into account during conversation would also be good at taking their intended audience into account during writing, or are these separate skills? Would training language educators in linguistics/psycholinguistics/cognitive science be a good thing to do? Would training student writers in those very fields be a good thing to do? What kinds of knowledge are most crucial for a student writer to obtain?
I also have a very strong interest in writing. Not just the act itself (though if my attempts to summarize my interests are any indication, I do like the act of writing quite a bit!), but in the various ways written language is and is not like spoken language, and what it is that makes writing "good", and what makes some people better at producing "good" writing than others. When we write, we are able to make choices about the way we express a particular idea in a way that the fast pace of spoken language does not allow, and we are also able to edit what we have said in a way that is impossible in spoken language. What kinds of effects might these differences have on the cognitive processes involved in communicating via language?
We tell student writers that one of the most important things they can do to make their writing better is to write for a reader, to take their audience into account - but as writers, we cannot collaborate with our audience to ensure understanding, there is no "grounding process", and we do not get anything like the kind of immediate feedback that we get when speaking regarding how well we are being understood. But of course, as writers, we also have more time than speakers have to consider what our audience knows, and even how they might interpret a particular phrase, sentence, or passage, and we can revise to avoid ambiguity. Should we expect that students who are particularly good at taking their interlocutor's perspective into account during conversation would also be good at taking their intended audience into account during writing, or are these separate skills? Would training language educators in linguistics/psycholinguistics/cognitive science be a good thing to do? Would training student writers in those very fields be a good thing to do? What kinds of knowledge are most crucial for a student writer to obtain?