Diversity Reading List

Abstract:

In response to widespread use of automated decision-making technology, some have considered a right to explanation. In this paper I draw on insights from philosophical work on explanation to present a series of challenges to this idea, showing that the normative motivations for access to such explanations ask for something difficult, if not impossible, to extract from automated systems. I consider an alternative, outcomes-focused approach to the normative evaluation of automated decision-making, and recommend it as a way to pursue the goods originally associated with explainability.

Summary: Narrow content functionalism claims that the contents of beliefs are determined by their causal profile. If two belief tokens are of the same causal type, they are of the same semantic type. However, Taylor argues that de dicto semantic types do not supervene on causal types, due to multiple realizability. He establishes this with the thought experiment of "fraternal twin earth", where things are functionally identical but molecularily different.

Abstract: This paper examines and rejects some purported refutations of eliminative materialism in the philosophy of mind: a quasi-transcendental argument due to Jackson and Pettit (1990) to the effect that folk psychology is "peculiarly unlikely" to be radically revised or eliminated in light of the developments of cognitive science and neuroscience; and (b) certain straight-out transcendental arguments to the effect that eliminativism is somehow incoherent (Baker, 1987; Boghossian, 1990). It begins by clarifying the exact topology of the dialectical space in which debates between eliminativist and anti-eliminativist ought to be framed. I claim that both proponents and opponents of eliminativism have been insufficiently attentive to the range of dialectical possibilities. Consequently, the debate has not, in fact, been framed within the correct dialectical setting. I then go onto to show how inattentiveness to the range of dialectical possibilities undermines both transcendental and quasi-transcendental arguments against eliminativism. In particular, I argue that the quasi-transcendentalist overestimates the degree to which folk psychology can be insulated from the advance of neuroscience and cognitive science just in virtue of being a functional theory. I argue further that transcendental arguments are fallacious and do not succeed against even the strongest possible form of eliminativism. Finally, I argue that that transcendental arguments are irrelevant. Even if such arguments do succeed against a certain'very strong form of eliminativism, they remain complete non-starters against certain weaker forms of eliminativism. And I argue that if any of these weaker forms is true, folk psychology is in trouble enough to vindicate Paul Ckurchland's claim that our common sense psychological framework is "a radically false and misleading conception of the causes of human behavior and the nature of cognitive activity".

Summary: This lucid and wide-ranging volume constitutes a self-contained introduction to the elements and key issues of the philosophy of language. In particular, it focuses on the philosophical foundations of semantics, including the main challenges to and prospects for a truth conditional semantics. Since the book is neither single-mindedly philosophical, nor single-mindedly technical, it is an accessible introduction to the philosophical foundations of semantics, and will provide the ideal basis for a first course in the philosophy of language and philosophical logic.

Abstract: Consider utterances of the following two sentences: (1) Have you had breakfast? (2) Have you had sex? Utterances of (1) and (2) typically differ in temporal import. An utterance of (1) raises a 'this morning' question. An utterance of (2) raises an 'ever' question. The difference in felt temporal import clearly has something to do with the difference between our more or less shared breakfast eating practices and our more or less shared sexual practices. People tend to eat breakfast daily - though there are, of course, exceptions. People tend not to have sex daily - though here too there are exceptions. Moreover, people by and large mutually know these facts. The first goal of these remarks is to explain how our mutual knowledge of such shared practices influences the perceived temporal import of utterances like (1) and (2). The explanation is not terribly surprising, but this unsurprising explanation reveals something significant about the nature of the great divide between pragmatics and semantics. In particular, I'm going to argue that Grice got it pretty close to right. The explanation of this phenomenon, and certain others like it, turns out to be roughly, but still deeply Gricean. I say 'roughly' Gricean because the account I offer does not entail that the difference in temporal import between (1) and (2) is a difference in conversational implicature strictly so-called. But for reasons that will become clear in due course, the explanation I offer even if not strictly Gricean is nonetheless deeply Gricean. Armed with our roughly but deeply Gricean understanding of this easy case, I turn to the somewhat more challenging and controversial case of incomplete definite descriptions. Imagine an utterance of: (3) The cat is on the couch again. In uttering such a sentence, a speaker commits what we might call descriptus interruptus. The context independent meaning of the uttered sentence is insufficient to fix a fully determinate descriptive significance for the contained descriptions. Though we may justly infer that a speaker who utters such a sentence intends thereby to communicate some proposition or other to the effect that some unique cat or other is once again on some unique couch or other, nothing more determinate may be inferred on the basis of sentence meaning alone about the relevant cat and the relevant couch. But the speaker's act of descriptus interruptus does not prevent speaker and hearer from enjoying a mutually consummated communicative exchange. The roughly though deeply Gricean approach I outline explains how such consummation is possible in a relatively straightforward way.

Abstract:

Logicians have always found inspiration for new research in the ordinary language that is used on a daily basis and acquired naturally in childhood. Whereas the logical issues in the foundations of mathematics motivated the development of mathematical logic with its emphasis on notions of proof, validity, axiomatization, decidability, consistency, and completeness, the logical analysis of natural language motivated the development of philosophical logic with its emphasis on semantic notions of presupposition, entailment, modality, conditionals, and intensionality. The relation between research programs in both mathematical and philosophical logic and natural language syntax and semantics as branches of theoretical linguistics has increased in importance throughout the last fifty years. This chapter reviews the development of one particularly interesting and lively area of interaction between formal logic and linguistics—the semantics of natural language. Research in this emergent field has proved fruitful for the development of empirically, cognitively adequate models of reasoning with partial information, sharing or exchanging information, dynamic interpretation in context, belief revision and other cognitive processes.

Abstract: While the concept of adaptive preferences is an important tool for criticizing injustice, it is often claimed that using the concept involves showing disrespect for persons judged to have adaptive preferences. In this paper, I propose an account of adaptive preferences that does the relevant political work while still showing persons two centrally important kinds of respect. My account is based in what I call an indirect substantive account of autonomy, which places substantive requirements on the options available to a person, rather than on the option that she ultimately prefers. This allows us to pinpoint cases in which a person's circumstances have rendered her insufficiently autonomous, without saying that any conception of the good must be non-autonomous tout court.

Abstract: Social sciences face a well-known problem, which is an instance of a general problem faced as well by psychological and biological sciences: the problem of establishing their legitimate existence alongside physics. This, as will become clear, is a problem in metaphysics. I will show how a new account of structural explanations, put forward by Frank Jackson and Philip Pettit, which is designed to solve this metaphysical problem with social sciences in mind, fails to treat the problem in any importantly new way. Then I will propose a more modest approach, and show how it does not deserve the criticism directed at a prototype by Jackson and Pettit

Abstract: Science seems generally to aim at truth. And governmental support of science is often premised on the instrumental value of truth in service of advancing our practical objectives, both as individuals and as communities, large and small. While there is some political expediency to this view, it is not correct. The value of truth is nowise that it helps us achieve our aims. In fact, just the contrary: truth deserves to be believed only on the condition that its claim upon us is orthogonal to any utility it might have in the service of (any and all) practical ends

Abstract: I shall endeavor to show that every physical theory since Newton explainswithout drawing attention to causes-that, in other words, physical theories as physical theories aspire to explain under an ideal quite distinctfrom that of causal explanation. If I am right, then even if sometimes theexplanations achieved by a physical theory are not in violation ofthe standard of causal explanation, this is purely an accident. For physicaltheories, as I will show, do not, as such, aim at accommodating the goals oraspirations of causal explanation. This will serve as the founding insightfor a new theory of explanation, which will itself serve as the cornerstoneof a new theory of scientific method.