Martin’s paper contains two main arguments in favor of disjunctivism. We’ll discuss these in turn, and then move on to discussing some objections to Martin’s view and disjunctivism more generally.
The main argument that we get for disjunctivism in McDowell is that it is a way of avoiding the incoherences of sense datum theory. Martin’s motivation for disjunctivism is similar; he thinks that it is the only plausible way of maintaining a few which he calls ‘naive realism.’ But he lays out the argument more explicitly. In outline, the central argument of §1 is supposed to work like this:
| 1. | Naive realism |
| 2. | Experiential naturalism |
| C. | ¬ (The common kind assumption) |
Disjunctivism is supposed to be equivalent to the negation of the common kind assumption, so the argument is supposed to deliver the truth of disjunctivism. To evaluate the argument, we’ll have to first figure out what Martin designates by these labels.
He describes naive realism as follows:
“The Naive Realist thinks that some at least of our sensory episodes are presentations of an experience-independent reality. ...Mind-independent reality can form the subject matter of sensuous experience.
The Naive Realist, however, [also] claims that our sense experience of the world is, at least in part, non-representational. Some of the objects of perception --- the concrete individuals, their properties, the events these partake in -- are constituents of the experience. No experience like this, no experience of fundamentally the same kind, could have occurred had no appropriate candidate for awareness existed. ...what is distinctive of sensing as opposed to thinking is that one cannot really sense in the absence of an object of sensing.” (38-9)
This makes it seem as though naive realism is the conjunction of the following two theses: (a) sometimes the object of experience is a bit of mind-independent reality, and (b) if an experience has some object o as its object, then no experience of fundamentally the same kind can occur in the absence of o.
Experiential naturalism is the view that “ sense experiences, like other events of states within the natural world, are subject to the causal order, and in this case are thereby subject just to broadly physical causes ...and psychological causes.” (39-40)
From these is supposed to follow the falsity of the common kind assumption, which is “the view that whatever kind of mental event occurs when one is veridically perceiving some scene ...that kind of event can occur whether or not one is [veridically] perceiving.”
If we fill out Martin’s argument in this way, then it looks like this:
| 1a. | Sometimes the object of experience is a bit of mind-independent reality. |
| 1b. | If an experience has some object o as its object, then no experience of fundamentally the same kind can occur in the absence of o. |
| 2. | Sense experiences are subject just to broadly physical and psychological causes. |
| C. | ¬ (Whatever kind of event it is when I veridically perceive an object o, an event of the same kind can occur in the absence of o.) |
The most puzzling thing about this argument is that the conclusion seems to be more or less equivalent to thesis (1b). This can’t be what Martin had in mind; on this interpretation, the second part of the definition of naive realism just is the negation of the common kind assumption. So on this interpretation it can hardly be necessary to bring in experiential naturalism, or indeed premise (1a), to show an inconsistency between naive realism and the common kind assumption.
We get a different version of the argument on pp. 40-1:
“...assume that we have ...an awareness of some lavender bush which exists independent of one’s current awareness of it. By the Common Kind Assumption, whatever kind of experience that is, just such an experience could have occurred were one merely hallucinating. By Experiential Naturalism, we know that there are sufficient ...causes of it. If the hallucinatory experience were relational in the manner that the Naive Realist supposes the perception of the bush to be, then the causes sufficient to bring about the hallucination must also have been sufficient for some appropriate object to be present in the experience. ...Hence the bringing about of the experience must have been sufficient for the existence of its object. If the experience alone is ...sufficient for this object of awareness in the case of hallucination, then the object in this case is not merely non-physical but dependent for its existence on the occurrence of this experience. ...
...So, contrary to the Naive Realist’s starting assumption, if the hallucinatory experience is a relation to an object of awareness, it is to a mind-dependent one, and hence the perception is a relation to a mind-dependent object, not the mind-independent object that the Naive Realist hypothesises.”
This makes Martin’s argument seem more like the traditional argument from illusion, turned upside down. But there are some differences. Here it seems as though Experiential Naturalism is the view that if x and y are experiential events of the same sort, then if they have the same proximate causes, they have the same objects. The Common Kind Assumption is not quite the premise from the argument from illusion which says that the objects of experience are the same in the case of illusory and veridical experience; rather, it says that hallucinations and veridical experiences can be events of the same fundamental sort. This, along with Experiential Naturalism, then yields the conclusion, given as an independent premise in the argument from illusion, that hallucinations and veridical experiences have the same sorts of objects.
Next week we’ll return to the question of whether the common kind theorist can reject Experiential Naturalism, so construed.
Martin’s second argument is that common kind views are committed to false views about the nature of our knowledge of our own perceptual states. (For a clear discussion of this argument, see §3 of Siegel’s ‘Indiscriminability and the phenomenal.’)
The basic idea of the argument is that the common-kind theorist and the disjunctivist have the following disagreement about hallucinations and the veridical experiences from which they are indistinguishable: the common-kind theorist says that there is some property common to the hallucination and veridical experience in virtue of which they are indistinguishable, whereas the disjunctivist says that the essence of the hallucination just is its indiscriminability from the veridical experience. Further, the common kind theorist is supposed to think that all experiences, whether veridical or not, share some property in virtue of which they are experiences. (This does not follow from the above claim about veridical experiences and the hallucinations from which they are indiscriminable; but it’s supposed to be motivated by the same core idea -- which the disjunctivist rejects -- that hallucinations and veridical experiences are members of the same kind.)
Let’s say, following Martin (47), that the common kind theorist is committed to the view that every perceptual experience, whether veridical or hallucinatory, is a perceptual experience in virtue of possession of some common set of features E1 ...En. Then the argument against the common kind theorist is as follows: it seems that we can grasp the possibility that for at least many veridical perceptions, there could be hallucinations which are indiscriminable from them. The common kind theorist must suppose that in grasping this, we are aware of the possibility of experiences with E1 ...En which are not veridical. So just in virtue of grasping the idea of matching hallucinations, we must have a grip on the set of properties in virtue of which experiences count as such. But in general a grasp of the idea of a matching hallucination does not require this kind of knowledge of the nature of perceptual experience. It is in this sense that the knowledge of our own experiences which the common kind theorist is committed to attributing to us is ‘immodest.’
By contrast, the disjunctivist does not think that veridical experiences and their matching hallucinations are of the same fundamental common kind, and so does not believe in the existence of the set of properties E1 ...En. The disjunctivist thinks that the only thing that veridical and hallucinatory experiences have in common is their indiscriminability from veridical experience.
It is hard (for me at least) to read this argument without seeing it as committing the following fallacy:
| 1. | x is the case in virtue of y being the case. |
| 2. | A knows (believes, is aware) that x is the case. |
| C. | A knows (believes, is aware) that y is the case. |
I argued in the discussion of McDowell that we need from the disjunctivist more than the claim that appearances can be indistinguishable from facts; we need some account of what appearances are which would make this intelligible.
McDowell seemed to present disjunctivism as the view that hallucinations and veridical experiences are genuine relational episodes of sensory awareness, but that the objects differ between the two cases: appearances in the first, and facts in the second. This is what seems to give rise to the explanatory need mentioned above. But Martin presents disjunctivism differently. In a way, he says less than McDowell about the nature of hallucination. According to Martin, the following explains the nature of hallucinatory experiences: “there may be sensory states whose mental nature is characterisable in nothing but epistemological terms, in terms of their unknowable difference from cases of veridical perception.” I.e., what it is for some state to be a hallucination is for that state to be indistinguishable from some veridical perception. Unlike McDowell, Martin does not seem to commit himself to the view that such hallucinations are relations to appearances, and so does not commit himself to the category of ‘appearances.’
It seems to me that there are three problems with this view that the essence of hallucinations is exhausted by their indiscriminability from veridical experiences:
It is worth noting one important difference at this point between Martin’s disjunctivism and the presentation of the view we find in McDowell. Whereas McDowell seemed happy to regard all experiences as relational, and simply to regard hallucinations as relations to a different sort of object -- mere appearances rather than facts -- Martin is less committal. It is clear that he thinks that hallucinations and veridical experiences are not the same sort of sensory event. But this denial is open to (at least) two different interpretations:
Above I suggested several problems specific to the latter. One interesting question is whether they fare differently in making sense of experiences - which seem quite commonplace - which are partly veridical, and partly illusory. The former view seems committed to the idea that experiences are not relations to an object of perception, but to many different ones - some facts, some mere appearances. The latter view seems committed either to treating all of these as non-veridical vents whose essence is indiscriminability from a veridical experience, or to thinking of sensory experiences of this kind as actually, and unbeknownst to the perceiver, consisting in several concurrent perceptual experiences.