Are Salty Experiences Salty? Are Square Experiences Square?

It has often been noticed -- I won't try to track down the history of this observation, but Ned Block and David Chalmers come to mind, and Ignacio Prado states it nicely in a comment on my Aug. 7 post about Dennett on fictions -- that the language in which we describe sensory experience is often derivative of the language we use to describe the objects sensed. Asked to describe the taste experience I have biting into a potato chip, I might say "salty". Now probably I don't (or shouldn't?) mean that my experience itself is salty. Rather, it's the potato chip that's salty. My experience is only "salty" in the sense that it's the kind of experience typically caused by salty things -- it's an experience of the saltiness of something else. Likewise, I might say that my tactile experience of the carpet is "rough" and my olfactory experience of my wife's cooking is "oniony".

Asked to describe such experiences without recourse to the language of external objects, I'm stumped. I could try analogy: "It tastes like a rushing elephant!" -- but the metaphor is weak, and we're still in the language of external objects, anyway. I could say a few things using concepts that apply literally both to experiences and to external events: The taste experience had a sharp onset, then gradually faded. But this won't get us far. Likewise, I can say things like that it was pleasurable and that it made me want to eat more, but such remarks, again, don't have much specificity. In attempting to describe sensory experience accurately and in detail, we depend essentially on our language for describing sensed objects.

I've used tactile, gustatory, and olfactory examples so far. What about vision and hearing? It's clear -- I think it's clear! -- that taste experiences aren't literally salty, etc., in the same way external objects can be salty. The parallel observation is not as clear for auditory and visual experience. Are experiences of square things literally square? Are experiences of loud things literally loud? By analogy, it seems we should say no; but on the other hand I, at least, feel some temptation to say that there is something literally square in my visual experience of a square object -- a squarish sense-datum? And can auditory experiences literally differ in volume in the way external sound events can differ in volume (louder and quieter sense data)? My taste experiences plainly don't literally contain salt, but maybe my visual experiences of squares do literally contain four equally sized edges set a right angles to each other?

Aren't the senses all ontologically parallel? Why, then, is it so much more tempting to say that square experiences are square than that salty experiences are salty? (Color introduces a separate batch of primary-quality/secondary-quality confusions, so I'm avoiding it as an example.)

The description of emotional experience, I might add, seems to work a little differently, though it presents its own problems. We can try to locate it in space and associate it with bodily conditions (I felt a twisting in my gut); we can fit it into an emotional category (it was a feeling of sadness); but most people I've interviewed think such descriptions don't do full justice to the phenonema and feel frustrated when asked to describe their emotional phenomenology very precisely and accurately.