Experience and Fact
We normally suppose that experience and fact are separate entities. Suppose I observe my cat chasing a lizard: on the one hand, there is the fact of my cat chasing a lizard; on the other, there is my experience of my cat chasing a lizard. These could exist separately—there is a dualism of experience and fact. I might report the fact in question by saying, “My cat chased a lizard”, and I might report the experience by saying, “I had a visual experience of my cat chasing a lizard”. But these sentences mix up fact and experience, because the first refers to me in stating the fact in question, while the second refers to my cat in stating what I experienced. Shouldn’t we be able to describe fact and experience in their own terms? So let’s try again: suppose I say, “This cat chased a lizard” in an effort to keep myself out of the picture, sticking only to the fact itself. That looks better, but what does “This cat” mean? Doesn’t it mean, “The cat I am experiencing”? I am experiencing a cat and I use this fact to refer to the cat by employing a demonstrative: the demonstrative refers back to my experience in using it. So I have not succeeded in keeping myself out of the picture; the picture still contains a reference to myself in the corner. Worse, I have referred to my experience in trying to report a fact about the cat and the lizard. On the other hand, I scarcely remove the reference to the world in describing my experience by substituting “this cat” for “my cat”: now I am saying that I had an experience of this cat chasing a lizard not my cat—still a living breathing cat. I refer to experience in reporting the fact and to the fact (or a component of it) in reporting the experience. But if there is an ontological dualism here, why am I mixing up the two categories? Can’t I report the fact as it objectively is and the experience as it subjectively is?
It might be thought easy to achieve that ideal: I just substitute a name for the cat in the former case and an indefinite description in the latter case. Thus “Blackie chased a lizard” and “I had an experience of a cat chasing a lizard”: in the former there is no reference to my experience, and in the latter there is no reference to a particular cat. But matters are not so simple. How does “Blackie” refer to its feline bearer? Plausibly by trading on a prior demonstrative reference–either at an initial baptism or maybe as an ongoing prop. A perfectly general description is seldom if ever available and is certainly not the typical case; we generally rely on a bedrock of demonstrative reference to secure the reference of names. At root, then, our names refer back to our experiences—as in “the object I am seeing”. So we are still not succeeding in describing the facts purely, as they objectively exist. In the case of descriptions of experience we have a different problem: I am not experiencing any old cat as chasing a lizard—some cat or other—but a specific cat, viz. Blackie, my cat, this cat. There is a particularity to the experience that is not captured by the indefinite description “a cat”; so we need a singular term to do justice to this aspect of the experience. The natural choice is a name (if you know it) or a demonstrative (if you don’t): “I had an experience of Blackie chasing a lizard” or “I had an experience of this cat chasing a lizard”. But then we are back to referring to an actual singular cat. We can’t keep the world out of descriptions of the experience, as we couldn’t keep experience out of descriptions of the world. Yet aren’t these separate domains?
One response is to accept that they are not separate domains. Idealism says that facts are experiences, so it isn’t surprising that we can’t describe the one except by reference to the other: the fact of a cat chasing a lizard is just the occurrence of an experience of a cat chasing a lizard. On the other hand, externalism maintains that experiences are constituted by relations to worldly objects, so that my experience just was composed of a particular cat (inter alia): I wouldn’t have that experience unless it was directed at a particular individual cat. The reference to a specific cat simply reflects the actual nature of the experience, according to externalism. Thus there is no dualism of the sort described earlier: facts and experiences are inextricable. I won’t try to adjudicate this issue now; I have merely indicated a possible route to the doctrines in question. Our habitual modes of description can encourage the rejection of a strict dichotomy between facts and experiences. What I want to suggest is that there is room for an alternative view: granted we can’t find a way to exclude reference to experience from descriptions of fact, and reference to fact from descriptions of experience, but that is really a point about us not about reality. From the fact that we can’t describe facts and experiences without importing alien elements from the other side, it doesn’t follow that these things can’t exist without each other or that they have dependent natures. Ontological dependence doesn’t even follow from the fact that no one could describe the two independently. For it may be that we are limited by our language and cognitive resources to describing things in ways that don’t reflect their real nature. A cat could be chasing a lizard even if no one ever experienced that fact, though reports of the fact inevitably introduce reference to the reporter’s experience (even if only implicitly). And it could be that someone has an experience just like mine when I see my cat chasing a lizard, even though there is no cat there and hence there is no reference for “this cat”. Maybe I can’t capture the singularity of the experience without employing an existentially committal singular term, though the experience itself has a nature that is not dependent on the reference of such terms. So the dualism of fact and experience is not compromised by constraints on description, assuming such constraints to exist. Language is designed to fulfill certain practical purposes and these may not include capturing the objective nature of things (completely, purely). We talk about facts by relating them to our experiences, and we talk about experiences by relating them to facts (objects, properties), but that need not imply that reality itself is similarly structured. Perhaps we could invent new ways of talking that capture things more accurately (objectively, intrinsically), but as things stand we are apt to describe the world in something short of a logically perfect language.
 There is much to be said about the relation between demonstratives and perception, but it is clear that in standard cases one refers to the object that one is perceptually attending to, as if with an act of inner pointing. Without sense perception demonstratives would have little use—the audience needs to perceive what the speaker is referring to in order to understand the utterance. The mode of presentation associated with a demonstrative is perceptual, i.e. involves an experience of a certain object. But then demonstratives embed in their meaning acts of perceptual acquaintance—while the fact being reported is not itself a fact of perception.
 Physics might be regarded as the right kind of language to use to describe physical reality as it is in itself, using no indexical language to do so; but there doesn’t appear to be anything comparable that rids psychology of all reference to things outside the mind. It is hard to reconstruct commonsense psychology in purely non-externalist terms. Could there be a pure phenomenology that described experience without employing any terms applicable to physical objects? This would include even general terms for types of physical object such as “cat” or “square”.