Diana Raffman

The thesis develops a cognitivist account of the supposed ineffability of musical experience. It is contended that, when the ineffability is viewed as adhering to a certain kind of perceptual knowledge of a musical signal, its nature can be illuminated by the adoption of a recent cognitivist theory of perception in conjunction with a generative grammar for tonal music . On this two-headed view, music perception consists in a rule-governed process of computing a series of increasingly abstract mental representations of a musical signal. In this framework the investigation of musical ineffability is recast as the search for a level of mental representation of whose content we are conscious but cannot make verbal report. Two types of report are distinguished--one consisting in so-called absolute identification, the other in comparative identification or discrimination. It turns out that certain features of the signal are likely to be computed at such superficial or "shallow" neuronal levels that they fail to be mentally categorized in the manner thought necessary for the learning of verbal labels and ipso facto for report of the first kind. Therein, on the present account, lies their ineffability, called 'perceptual ineffability'. These shallow features do, however, admit of comparative identification and ipso facto of the second kind of report. Hence a principal task of the thesis is to show how a kind of ineffability and a kind of report are compatible. ;The account of perceptual ineffability is then set against two well-known texts in contemporary philosophy of art in order to show that it squares with more standard approaches and makes good on its explanatory promises. It is hoped that, over and above its proposed solution to a longstanding problem in aesthetics, the present study may shed some light on the relationship between philosophical and scientific theorizing

Of the many families of words that are thought to be vague, so-called observational predicates may be both the most fascinating and the most confounding. Roughly, observational predicates are terms that apply to objects on the basis of how those objects appear to us perceptually speaking. ‘Red’, ‘loud’, ‘sweet’, ‘acrid’, and ‘smooth’ are good examples. Delia Graff explains that a “predicate is observational just in case its applicability to an object (given a fixed context of evaluation) depends only on the way that object appears” (2001, 3). By the same token observational predicates are, as Crispin Wright observes, terms “whose senses are taught entirely by ostension” (1976). Like other vague predicates, observational words appear to generate sorites paradoxes. Consider for example a series of 20 colored patches progressing from a clearly red one to a clearly orange one, so ordered that each patch is just noticeably different in hue from the one before. The following argument then seems forced upon us: (1) Patch #1 is red. (2) Any patch that differs only slightly in hue from a red patch is itself red. (3) Therefore patch #20 is red. Premise (2) expresses what Wright has called the tolerance of ‘red’: the application of the predicate tolerates small changes in a decisive parameter (here, hue). Of course, most vague predicates, hence most versions of the sorites, are not observational. For instance, given a series of.

Worries about the artistic integrity (for lack of a better term) of twelve-tone music are not new. Critics, philosophers, musicians, even composers them- selves have assailed the idiom with a fervor usually reserved for individual artists or works. Just why it is supposed to be defective is not entirely clear, however. I want to revisit these questions by way of putting some insights from music history and theory together with some insights from the philosophy and psychology of music. To find out whether twelve-tone composition is defective we will need to reflect carefully upon our concepts of music and art in general, so if all goes well our conclusions should have some broader import. Because I will be pulling together considerations from several domains, I will need to lay a fair amount of groundwork at the beginning; but once that is done, we will have a vantage on the present issues that would not otherwise be open to us.

Higher-order vagueness is widely thought to be a feature of vague predicates that any adequate theory of vagueness must accommodate. It takes a variety of forms. Perhaps the most familiar is the supposed existence, or at least possibility, of higher-order borderline cases—borderline borderline cases, borderline borderline borderline cases, and so forth. A second form of higherorder vagueness, what I will call ‘prescriptive’ higher-order vagueness, is thought to characterize complex predicates constructed from vague predicates by attaching operators having to do with the predicates’ proper application. For example, the predicates ‘mandates application of “old”’ and ‘can competently be called “old”’ are prescriptively higher-order vague. Higher-order vagueness appears in other guises as well,1 but these two have been of particular interest to philosophers and will be my target here. I want to expose some misconceptions about them. If I am right, higher-order vagueness is less prevalent, and less important theoretically, than is usually supposed.2 In what follows I am going to assume that vagueness is a semantic feature of natural language. For the most part I won’t discuss epistemic or pragmatic views, and I will say nothing about so-called metaphysical vagueness.