DISSERTATION
SINGULARITY: REFERENCE, THOUGHT, & BELIEF

Long Dissertation Abstract

It is a common view that thinking singularly about something involves being acquainted with it, but acquaintance-based theories of singular thought have come under criticism for being too demanding. It seems that there are plenty of things we might think about but lack acquaintance with such as the number 2, Harry Potter, or next Halloween, but it seems that we can refer directly and think singular thoughts about these. We might also introduce names for individuals we can only describe generally, such as the tallest female spy or the first child to be born in the next century. We may then think singular thoughts about those individuals using the introduced names, or so it seems. These cases are complex and show that it is unclear what constrains singular thought, if anything does. It is also not entirely clear, given the criticisms of acquaintance-based views, what is at stake if acquaintance fails to be a requirement for singular thought. Nevertheless, acquaintance is certainly involved in much if not most of our singular thinking and is worth serious consideration as constraining such thought. I defend acquaintance as a constraint on singular thought by arguing that critics of the view fail to appreciate how much of our thought is ultimately singular in one way or another.

​

Many philosophers of mind and language take a Fregean approach to reference in cognition. This view takes the content of a thought to be akin to a Fregean sense, an intermediary mode of presentation between a thinker and thing thought about. On this view, reference in cognition is always mediated by the content of the thought. The contrary Russellian view takes the object of thought to constitute the content of the thought. Allowing for the possibility of direct reference in cognition. On a Russellian view a thought can be directly about something without having an intermediary content. The most obvious cases available to the Russellian are those of perceptual judgments. When I see the red plate on the table my thoughts about it are directly about it. There is no intermediary content that must be grasped for my thought involve reference to the plate. Tyler Burge has defended a Fregean view against the Russellian, claiming that perceptual cognition is always guided by general abilities of the perceiver such as perceptual constancies, which act as the intermediaries in securing reference. However, as I argue, sensation itself, a passive causal process, must underlie all successful perception. One cannot sense what one is not acquainted with, and this undergirds an account of basic reference in cognition.

​

Basic reference is a form of reference that is passive. It is involved only in representational processes that must be impinged on by the world in order to represent. We may distinguish this passive form of reference from the active form which encompasses most of the interesting forms of reference studied today. Often reference is active and involves some cognitive work. Basic reference is important to all other forms of reference as it makes it possible for our thoughts to have singular content and not only be singular in form. This claim is contrary to a recently revived thesis about singular thought known as semantic instrumentalism, which takes our ability to introduce names that co-refer with uniquely satisfied descriptions to be sufficient for having singular thought. Instrumentalists claim that singular thought may be had on the cheap by manipulating the apparatus of direct reference. Elaborating on a point hinted at by Robin Jeshion, I argue that semantic instrumentalism fails in the face of a dilemma. The view must either accept that introducing a name depends on a singular intention or a general intention that the name co-refer with some description. It cannot depend on a singular intention on pain of circularity. If singular intentions are a form of singular thought, it must have acquired its singularity by some means. A general intention however cannot adequately specify an individual and thereby make a thought singular. Semantic instrumentalism fails to secure genuinely singular thoughts through introducing names by description.

​

Underlying the failure of semantic instrumentalism is a distinction between properties of a general qualitative nature and those that are relational. When we refer to something, we often specify a property of it in order to pick it out, but there are many kinds of specification. Some properties used to specify objects are had in virtue of the relations those objects have to other things with which we are acquainted. Most forms of non-basic reference take advantage of this fact. Very rarely does a case arise in which no relational condition guides reference in thought and language. This allows for a defense of an acquaintance constraint on singular thought. All genuinely singular thoughts have singular content in virtue of invoking basic reference or relational conditions that individuals stand in to other individuals with which one is acquainted. Thoughts may be classified then as general, directly singular, or relationally singular. Getting clear on this difference allows us to differentiate between reference in thought, direct and mediated, and the content of thought, singular and non-singular. Some singular thoughts are mediated, but it is argued, no genuinely singular thoughts lack acquaintance entirely.

​

Finally, distinguishing between basic and non-basic reference and different forms of singular thought allows for insight to be shed on how thinkers can have de re propositional attitudes toward numbers, fictional characters, and the future. It paves the way for a new understanding of the distinction between de re and de dicto attitudes in which de re attitudes play a fundamental explanatory role. This allows for an account of de dicto attitudes given in terms of de re attitudes, and this reduction helps explain intuitions about valid forms of quantification into and out of de dicto attitudes. This provides insight into the logic of propositional attitude attributions and other related non-extensional contexts.