Consciousness and Cognition 9, 363–369 (2000) doi:10.1006/ccog.2000.0465, available online at http://www.idealibrary.com on Reply to Thomas Metzinger and Bettina Walde Zolta ´n Jakab Institute of Interdisciplinary Studies, 2218 Dunton Tower, 1125 Colonel by Drive, Ottawa, Ontario K1S 5B6, Canada As I understand them, Metzinger and Walde make the following main points: (1) It is not lack of syntactic structure that is responsible for the ineffability of qualia. (2) The very notion of a representationally atomic perceptual state (and that of an elementary perceptual state) as I construe it is problematic as there is reason to sup- pose that such states do not exist. (3) Classical cognitive architecture is inadequate for framing a theory of ineffability. (4) In speaking about simple sensory experiences, a more fine-grained classification is needed. If all these objections were correct, that would be pretty devastating for my concep- tion of ineffability. My aim in this response is to argue that the first three points go a little too far, in that (a) syntactic structure does indeed play an interesting role in ineffability, (b) my notion of a representationally atomic state does describe some- thing real, and (c) the classical computational view is not so bad a choice for talking about perceptual representations, hence ineffability (although, I agree, there might be interesting alternatives to it). I accept the fourth point (d), but argue below that the required refinements can be accommodated in my account. Metzinger and Walde also make some smaller points which I will briefly address at the end. 1. THE FOUR KEY ISSUES (a) Even if an experience had syntactic structure this would not be enough to get expressibility of its unstructured constituents (or their phenomemal content, if you like). This much is true; however, if an experience has syntatic structure, this does confer to it expressibility in the sense I provided. The following is still true: if CN, SN3 and SN4 are satisfied, the experience under consideration is expressible. If the three conditions aren’t satisfied, then expressibility does not obtain. Since, as I argue, no imagination-driving information can be extracted from unstructured experiences by the converter, expressibility (the generation of an informative description) is based merely on structural information. If we stop the syntactic regress at some level (as we should), we get the atomic constituents, and ineffability arises. I still think that the reason why unstructured experiences are ineffable is precisely their lack of con- stituent structure. Here’s a little more: Format converters can only handle syntactic properties, and the phenomenal character of experiences is not a syntactic property. (Dimensional positions and constituent structure are syntactic properties of experi- Reply to Commentary on Z. Jakab (2000). Ineffability of qualia: A straightforward naturalistic explana- tion. Consciousness and Cognition, 9(3), 329–351. 363 1053-8100/00 $35.00 Copyright  2000 by Academic Press All rights of reproduction in any form reserved.