Epistemology Brownbag: Nevin Climenhaga

Date: Tuesday, September 25th, 2018

Time: 12:30-2:00 PM

Location: Kresge 3438

Title: (Epistemic) Probabilities are Degrees of Support, not Degrees of (Rational) Belief

Abstract:

 I argue that when we use ‘probability’ language in epistemic contexts – e.g., when we ask how probable some hypothesis is, given the evidence available to us – we are talking about degrees of support, rather than degrees of belief. The probability of A given B is the (mind-independent) degree to which B supports A, not the degree to which someone with B as their evidence believes A, or the degree to which someone would or should believe A if they had B as their evidence. My central argument is that degrees of support can correctly model good reasoning in cases of old evidence where degrees of belief cannot. I consider ways to revise orthodox degree-of-belief Bayesianism in order to account for this kind of case, and argue that such revisions end up unable to account for reasoning about probabilities conditional on evidence that it is impossible not to possess.

PhLing: Chris Kennedy & Malte Willer (University of Chicago)

Date: Friday, May 12, 2017

Time: 1:30-3:30

Location: Kresge 3364

Title: Have you ever been experienced? The evidential basis for subjective judgment

Abstract:

Consider a situation in which our friend Kim presents her two cats with a new brand of cat food. Hoshi, who eats anything, devours the food, but Nikko, who is very picky, takes one sniff and walks away. Observing this behavior, Kim says “This new food is not tasty.” We can report on this episode by uttering either (1a) or (1b), but not (1c). 

(1) a. Kim doesn’t believe the new food is tasty, because Nikko won’t touch it.

b. Kim doesn’t consider the new food tasty, because Nikko won’t touch it.

c. ??Kim doesn’t find the new food tasty, because Nikko won’t touch it.

The difference between (1a-b) and (1c) is that the latter but not the former presupposes that Kim has tasted the food.  This is a special case of a more general requirement associated with “subjective” predicates like ‘tasty’ and ‘beautiful’ that the individual whose judgment provides the basis for claims about whether an object satisfies these predicates must have experience of those features of the object that are relevant to the judgment:  how it tastes, how it looks, etc.  Our goal in this talk is to show that the experience requirement can be derived as an evidential condition on subjective judgments, given a pragmatic model of subjectivity as sensitivity to what Kennedy and Willer (2016) call “counterstances:”  alternative ways of resolving uncertainty about meaning.  

References:

Kennedy, C. and M. Willer. 2016. “Subjective attitudes and counterstance contingency.”  Proceedings of Semantics and Linguistic Theory 26. 

PhLing: Michael Glanzberg (Northwestern)

Date: April 14 2017

Time: 1:30-3:30

Location: Kresge 3354

Title: The Cognitive Roots of Adjectival Meaning

AbstractIn this paper, I illustrate a way that work in cognitive psychology can

fruitfully interact with truth-conditional semantics.  A widely held view takes

the meanings of gradable adjectives to be measure functions, which map objects

to degrees on a scale.  Scales come equipped with dimensions that fix what the

degrees are.  Following Bartsch and Vennemann, I observe that this allows

dimensions to play the role of lexical roots, that provide the distinctive

contents for each lexical entry.  I review evidence that the grammar provides a

limited range of scale structures, presumably dense linear orderings with a

limited range of topological properties.  I then turn to how the content of the

root can be fixed.  In the verbal domain, there is evidence suggesting roots are

linked to concepts.  In many cases for adjectives, it is not concepts but

approximate magnitude representation systems that fix root contents.  However,

these magnitude representation systems are approximate or analog, and do not

provide precise values. I argue that the roots of adjectives like these

provide a weak, discrimination-based constraint on a grammatically fixed scale

structure.  Other adjectives can find concepts to fix roots, which can support a

well-known equivalence class construction which can fix precise values on a

scale.  I conclude that though adjectives have a uniform truth-conditional

semantics, they show substantial differences in the cognitive sources of their

root meanings. This shows that there are (at least) two sub-classes of

adjectives, with roots fixed by different mechanisms and with different degrees

of precision, and showing very different cognitive properties.

PhLing: Rachel Rudolph (Berkeley)

Title: Searching for the Perceptual Source

Date/Time: Friday, April 7, 1:30-3:30

Location: Kresge 3364

AbstractSome appearance reports specify a “perceptual source” — that is, an individual that must appear a certain way, for the report to be true. For instance, Asudeh and Toivonen (2012) hold that the copy raising sentence (1) requires that Tom himself look a certain way, while the null-subject variant (2) doesn’t require this, and could, for instance, be true in virtue of the state of the kitchen, even if Tom isn’t present.

(1) Tom looks like he’s cooking.

(2) It looks like Tom is cooking.

A number of analyses have been offered for appearance reports like these, making different predictions about when a perceptual source is specified and when not. I will claim, however, that all are unsatisfactory, and this is because they focus solely on matrix-level argument structure. By going through a variety of data, I’ll motivate the sensitivity of the perceptual source interpretation both to the fine-grained semantics of the embedded clause, as well as to the broader conversational context. This, I suggest, follows plausibly from the evidential role of appearance reports; and I’ll consider how some previous analyses might be elaborated to capture this idea.

PhLing: Christopher Vogel (University of Maryland)

DateMarch 10, 2017

Time: 1:30-3:30 pm

LocationKresge 5531

TitleExternalism’s Vagueness Problem

Abstract:

Many natural language expressions seemingly lead to Sorites paradoxes, in that we can construct a series of claims that make use of, or relate to, a given expression, each of which is intuitively plausible but cannot be mutually true. In this sense, many natural language expressions are vague.  If one accepts that the truth of `Donald is bald’ relates to the number of hairs on Donald’s head, one seems likewise committed to the implausible claim that the loss of a single hair can make someone bald.  But, to deny that baldness pertains to head hair count seems equally implausible.  I’ll argue that problems of this sort are merely apparent, and indeed not a consequence of natural language meanings.  Rather, these difficulties arise for a particular externalist commitment to the relationship between linguistic meaning and truth. That is, the long standing problems pertaining to vagueness are the consequence of a (likewise long standing) semantic theory which holds that the meaning of a linguistic expression determines the truth-conditions for that expression. Alternatively, if we deny that natural language expressions have truth-conditions, many of the problems posed by vagueness do not arise in the first place.  I’ll illustrate how an internalist semantics can account for natural language speaker judgments pertaining to vague expressions without generating Sorites paradoxes.

Epistemology Brownbag: Jonathan Jenkins Ichikawa (UBC)

Date: Wednesday, February 22

Time: 12-1:30 PM

Location: Kresge 3438

Title: “Positive Epistemic Norms”

Abstract:

If you’re considering a question, you have three choices: believe, disbelieve, or suspend judgment. Of these three, suspension tends to enjoy, implicitly or even explicitly, the privilege of a perceived ‘default’ status. Epistemologists are quick to emphasise respects in which judgments can be too hasty, or when a combination of attitudes would be irrational. Descartes starts his Meditations with the worry that some of his beliefs may be wrong—so he shifts into suspension, until he can certify his methods as trustworthy. Descartes’s project is familiar, and by and large, analytic philosophy has mostly worked in that paradigm. Until you have enough evidence, play it safe, and suspend judgment.
I will suggest that the neutrality often attributed to suspension is often unwarranted. Suspension is not epistemically best by default. Failure to believe—undue skepticism—can be just as epistemically erroneous as can hasty belief—undue gullibility. (What if Descartes were motivated by the idea of not letting any truths get past him?) I’ll work towards this case from three perspectives: the epistemology of the a priori; the epistemology of testimony; and pragmatic encroachment. The aim is a reorientation of epistemology, away from emphasising negative, restrictive norms, and towards positive ones. I aim to vindicate in a more serious way the natural thought that we often ought to believe things.

 

PhLing: Alexis Wellwood (Northwestern)

Date: February 17, 2017

Time: 1:30-3:30 pm

LocationKresge 4354

TitleWhat do comparatives with plurals mean?

Abstract:

Many authors have explored the syntax and semantics of sentences like (1), and overall their compositional semantic properties seem well-understood. Sentences like (2) have received less attention, and so far nothing approaching a consensus about their meaning exists. Early hypotheses have been judged too strong, or too weak, and recent approaches raise issues of their own. This is puzzling in light of broad agreement about the properties of the critical parts of (2).

(1) The red dot is bigger than the blue dot.

(2) The red dots are bigger than the blue dots.

Some authors have proposed determinate truth conditions for (2), while others have suggested it is somehow indeterminate. I provide experimental evidence that tells against both kinds of proposals, while suggesting considerable variation across (and within) speakers. This raises questions about why (2) has such an enigmatic character.

 

PhLing: Bernhard Nickel (Harvard)

Date: Friday, February 3

Time: 1:30 – 3:30 pm

LocationKresge 3364

TitleGenerics and Conservativity

Abstract:

Generic sentences, such as “ravens are black” or “tigers have stripes” seem to express a generalization of some sort. This suggests the hypothesis that generics contain an unpronounced quantificational element at LF, gen. However, this hypothesis faces several problems, too, including the apparent failure of conservativity. For example, ”ravens are black” does not seem to be equivalent to “ravens are ravens that are black”. I argue that the apparent failure of conservativity is merely apparent, and that sentences like “ravens are ravens that are black” or “ravens are black ravens” are unacceptable for reasons that are orthogonal to conservativity: semantic interpretation breaks down. This analysis can be extended to deal with some famously troublesome examples for normality-analyses of generics, including “books are paperbacks” and “prime numbers are odd.”