Semantics
and
Psychological
Evidence
April 15-16, 2005
Security/access: 314-615-8000
AIMS
The aim of the workshop is to
bring together researchers from a range of disciplines to discuss prospects for
the use of empirical evidence in the investigation of the semantics of natural
languages. Invited speakers have been asked to reflect on the general prospects
for the incorporation of psychological and neurological evidence into the study
of natural language semantics, as well as to comment on specific examples of
the use of psychological or neurological evidence in semantic theorizing,
drawing on their own work or the work of others.
Historically, the study of
natural language semantics in linguistics and philosophy has proceeded without
the use of empirical evidence from psychology or neuroscience. Some of the
reasons for this have been purely practical – e.g., the lack of availability of
such evidence, the difficulty of applying existing psychological evidence to
semantics, and the inability of semanticists to conduct psychological studies
of their own. But some have advanced principled reasons for excluding such
evidence – e.g. Katz argued that language is an abstract object rather than a
psychological one, Quine precluded the incorporation of any data other than the
behavioral data necessary to establish stimulus meaning, and Wittgenstein
argued that linguistic meaning derives solely from the use of language in
social practices.
But since the cognitive
revolution, linguists and philosophers of language often take themselves to be
studying a faculty of the mind rather than an abstract object, or a set of
behavioral responses or social practices. Many linguists now conceive of
linguistics as the study of a psychological faculty implemented in the brain,
and much work in semantics now aims to be continuous with the new cognitivist
science of linguistics. Recently, following this cognitive turn to its logical
conclusion, a few researchers have begun to look to psychology and other
empirical sciences to evaluate theories in semantics and pragmatics. Is this
turn towards psychology justified? When, if at all, can psychological evidence
be used to help settle issues in semantics?
Similar questions have been
asked numerous times about linguistic knowledge in general. The goal of this
workshop is to advance the discussion in two ways: first, by discussing the
problem as related to natural language semantics in particular
(including reference and intentionality) rather than linguistic knowledge in
general; second, by asking the presenters to comment on specific examples
of the use of psychological data in semantic theory, drawing on their own work
or the work of others.
The specific research
questions addressed by the workshop are:
- What can psychological evidence tell us about the
semantics of natural language?
- What can psychological evidence tell us about the
referential/intentional properties of language and the mind?
- What lessons can we draw from specific examples
of the use of psychological data to support or criticize theories in
formal semantics?
speakers:
Ana Arregui, Assistant Professor of Linguistics,
David A. Balota, Professor of Psychology,
Steven E. Petersen, Professor of Neurology and Radiology, Associate Professor of Anatomy and Neurology, Neurosurgery, and Psychology, Washington University in St. Louis, sep@npg.wustl.edu; http://www.neuro.wustl.edu/people/petersen.html
Philip Robbins, Assistant Professor of Philosophy,
Robert J. Stainton, Associate Professor of
Philosophy,
Kenneth A. Taylor, Professor of Philosophy,
PROGRAM
Friday,
April 15, 2005
1:00 – 1:30 Sam Scott, “Introduction to the Workshop”
1:30 – 3:00 Robert J. Stainton, “Three Familiar Reasons for Restricting the Evidence Base for Semantics,
and Why Each Fails”
3:00 – 3:30
Refreshments (Foyer of Room 1013,
3:30 – 5:00 David
A. Balota, “
Saturday,
April 16, 2005
8:30 – 9:00 Breakfast (Foyer of Room 1013,
9:00 – 10:30 Ana Arregui, “Syntax Vs. Semantics During Parsing”
10:30 – 12:00 Steve
E. Petersen,“Exploring Issues in Semantic Processing with Functional
Neuroimaging”
12:00 – 2:00 Lunch
(on veranda, top floor, Clayton on the Park Hotel)
2:00 – 3:30 Kenneth
A. Taylor, “Pragmatics Everywhere”
3:30 – 4:00
Refreshments (Foyer of Room 1013,
4:00 – 5:30 Philip
Robbins, “Is Semantics Modular?”
7:00 Workshop Dinner (Cardwell’s Restaurant,
ABSTRACTS
Ana
Arregui
In this paper I investigate the interaction of syntax and semantics during sentence parsing. I argue that the human language processor uses both syntactic and semantic evidence during the online construction of sentence representations. In particular, I argue that formal semantic properties are relevant in the process of structure-formation, though in a restricted manner. The exact interaction between syntax and semantics during sentence processing is to be most fruitfully investigated on a case-by-case basis. I present experimental evidence supporting the view that semantics can help with the construction of syntactic structure. The evidence comes from the investigation of the online processing of quantifiers in multiple center-embedding constructions. I also present experimental evidence suggesting that semantics alone cannot justify the construction of syntactic structure. The evidence comes from the online processing of elided VPs without syntactically parallel antecedents.
David A.
Balota
Evidence for semantic processing in some experimental paradigms is controversial. I will discuss some recent evidence that suggests that apparent semantic effects may actually have associative underpinnings. The influence of attentional settings on the type of information necessary to perform a given task will be highlighted. Comparisons of individuals with distinct types of dementia, i.e., Alzheimer's Dementia vs Semantic Dementia will also help illustrate this point. Finally, discussion of the underlying structure of semantic memory will focus on recent arguments regarding small-world network structures that occur across naturally occurring complex systems.
Philip
Robbins
According to minimalist semantics, truth-conditional interpretation of utterances proceeds independently of pragmatic processing. Contextualist semantics denies this: it claims that assigning truth-conditions to utterances draws on a range of extralinguistic skills essential to pragmatics, including causal inference and mindreading. Recent proponents of minimalism have appealed to psychological evidence for the modularity of semantic processing to support their view. I review the evidence for semantic modularity and argue that it adds up to less than meets the eye. I then suggest what sort of evidence would be needed to support the modularity hypothesis, and what the available evidence tells us about the debate between minimalists and their contextualist rivals.
Robert J. Stainton
This talk is intended as stage setting for the workshop as a whole. I will investigate the appropriate evidence base for the study of language: should it be restricted in any way? In particular, I will be interested in the status of evidence from (mentalistic) psychology and cognitive science - does evidence from these domains bear on or illuminate our understanding of language? I will pursue these two questions in three steps.
I begin by considering three pro-restriction answers to them, answers that favour restricting the evidence base for linguistic enquiry, specifically setting aside psychological evidence. The goal in discussing these negative answers will be to clarify two things: first, the views of language (and language study) underlying each answer, and second, how each of those views leads to the restriction of the evidence base.
I then examine an anti-restriction answer to the original questions, according to which there are no limits in principle to the evidence base for language study, and more specifically, that evidence from mentalistic psychology and cognitive science in fact make very helpful contributions to the theorist's understanding of language. I describe the view of language from which this answer derives, and illustrate a few of the unexpected ways in which evidence from mentalistic psychology and cognitive science have illuminated our understanding of language.
In the final section of the paper, I respond to the pro-restriction answers in section 1, showing why they are less compelling than the positive approach described in section 2. Given the problems with the views of language that advocate restricting the evidence base, I conclude that such restrictions are not justified. Instead, I suggest that the study of language should proceed in the same way as other scientific endeavours, namely, by allowing that evidence from any quarter, and specifically from the domains of psychology and cognitive science, might in principle prove to be illuminating.
Kenneth
A. Taylor
In this paper, I argue that there is no pragmatics/semantics interface. In particular, there is no one place where the work of semantics ends and that of pragmatics begins. Pragmatics and semantics are everywhere intertwined because pragmatics happens everywhere. The lexicon itself directly licenses pragmatic intrusion from the very start of the journey from sentence meaning to narrowly semantic utterance content. Even before that journey is completed pragmatics can generate what I call one and half stage pragmatic externalities. Nor is pragmatics done when the journey from sentence meaning to utterance content is complete. Then pragmatics begins again to generate pragmatic externalities of a dizzying variety. Nonetheless, I shall argue that despite the fact that the semantic and the pragmatic are everywhere enmeshed, there is a distinction in both concept and principle to be had between the semantic and the pragmatic.
**************************************
PNP is an innovative program that fosters engagement
between different disciplines addressing the mind-brain. It draws together a
number of areas in which
The PNP program was originally set up as a graduate program in 1993 with a grant of $1,320,000 from the James S. McDonnell Foundation (a further grant of $880,000 was made in 1995). Subsequently the program was expanded to include an undergraduate second major, with the first students graduating in 1999. The PNP Program was approved as a first major program in 2004. The PNP Research center was founded in 2003.
Research
The PNP
Undergraduate Teaching
Undergraduates at
Graduate Teaching
At the graduate level, PNP sponsors a Ph.D. track within the Ph.D. program
in the Department of Philosophy. Students in the program complete the usual
requirements for a philosophy Ph.D. plus 16 hours of course work in cognitive
neuroscience and/or psychology. PNP students are able to tailor their emphases
within neuroscience or psychology to fit with their backgrounds and interests.
There are currently 10 students studying for the Ph.D. in the PNP program.
BACKGROUND
This workshop is organized under the auspices of the
Philosophy-Neuroscience-Psychology Research Center (PNPRC) at