Johnston sets up the paper as an argument against two views of perception, which he thinks go wrong for different reasons: the conjunctive and disjunctive analyses. One way to think of the general outline of Johnston’s argument is as follows: Johnston thinks that philosophers of perception have spent two much time worrying about the argument from illusion and hallucination, and not enough time thinking about the nature of hallucination. Once we see the nature of hallucination, we’ll see that it is in some respects similar to, and in other respects dissimilar to, veridical experience. The conjunctive analysis cannot explain the dissimilarities; disjunctivism cannot explain the similarities. But we want a view which can make sense of both.
Johnston describes the conjunctive analysis like this:
“The Conjunctive Analysis of seeing has it that when a subject is seeing
(i) the direct object of her visual awareness is not some particular in the external environment, but something that she could be aware of even if she were hallucinating
and the crucial conjunction
(ii) By contrast with the hallucinatory case, her visual awareness is appropriately caused by some external particular in the scene before the eyes. Thanks to this causal connection the material particular counts as an indirect object of experience.” (114)
Sense datum theory as the paradigm instance of the conjunctive analysis. How some forms of adverbialism and other views can also take a conjunctive form.
The case for the conjunctive analysis (115-118): the argument from illusion, with the premise that nothing can be an object of perception which could not be an object of hallucination delivered by the argument from the supervenience of hallucinatory experience on brain states plus the supposed unintelligibility of two experiences underwritten by the same brain state being awarenesses of different things.
Objections to the conjunctive analysis: it’s view that we cannot be directly acquainted with external particulars makes our access to them implausibly indirect, and is unsupported by any phenomenological sense in which our experience of external particulars seems to be indirect.
Disjunctivists see their position as a way of solving each of these two problems with the conjunctive analysis. Johnston introduces disjunctivism as follows:
“They hold that there is nothing more to be said about pairs of indistinguishable veridical and delusive sensings of say, a dagger hanging in the air, than the following. In the veridical case one is seeing a dagger hanging in the air, while in the delusive case one merely seems to be seeing a dagger hanging in the air. On this view, there is no neutral condition the appearing of a dagger common to indistinguishable cases of seeing and hallucination.” (120)
Johnston also suggests (121) that disjunctivists have a response to the argument for the conjunctive analysis; this echoes Martin’s idea that, given certain theses about the relationship between immediate causes of perceptual events and their objects, only disjunctivism is consistent with Naive Realism. Some doubts about this.
Johnston’s first argument against disjunctivism: its inability to explain seamless transitions between veridical and hallucinatory experiences. The example of the patient in the operating theater. The question to which the disjunctivist has no answer: “What kinds of things can visual experience be a relation to so that in a case of hallucination and a case of seeing there need be no difference which the subject can discern?” (123)
An attempted response by the disjunctivist: a higher-order account of hallucination. For A to hallucinate something is for it to falsely seem to A as though he is seeing that thing (126). Some problems for this response: hallucinations without ‘falsely seemings’, and an application of the missing explanation argument.
At this stage, we have two views of perception on the table, and some reason for skepticism about each. Some further thoughts about the nature of hallucination itself strengthen the case against each.
The questions which Johnston addresses in §6 are: do hallucinatory experiences have objects and, if so, what sorts of things can be the objects of hallucinatory experiences?
In favor of the act/object analysis of hallucinations: hallucinations make available properties for demonstration. The example of supersaturated red (130).
The case of external particulars contrasts with the case of qualities. One cannot acquire the ability to have thoughts about external particulars by having a hallucination of them; indeed, if, as the case of Noddy (131-2) seems to show, the objects of hallucination are (in Johnston’s phrase) construal-dependent, then having a hallucination of a given object seems to depend on a prior capacity to have thoughts about the object. The contrast with the case of veridical perception of objects.
We want a view which can make sense of the fact that hallucination, like veridical experience, can acquaint us with new properties; disjunctivism fails this test. We also want a view which can make sense of the fact that hallucination, unlike veridical experience, cannot acquaint us with new particulars; the conjunctive analysis fails this test.
Johnston lays out a view which he thinks passes both tests in §7:
“Your seeing the scene before your eyes is your being visually aware of a host of spatio-temporal particulars instantiating parts of such a profile or complex of sensible qualities and relations. The suggestion is that in the corresponding case of a subjectively indistinguishable hallucination you are simply aware of the partly qualitative, partly relational profile. This means that the objects of hallucination and the objects of seeing are in a certain way akin; the first are complexes of sensible qualities and relations while the second are spatio-temporal particulars instantiating such complexes. The visual system is adapted to put us in contact with the scenes or visible instantiations around us. When the visual system misfires, as in hallucination, it presents uninstantiated complexes of sensible qualities and relations, at least complexes not instantiated there in the scene before the eyes.”
Why this view seems attractive: direct realism without the weirdnesses of disjunctivism.
In the discussion of the conjunctive analysis, Johnston runs through an argument in favor of that analysis which, if sound, would undercut his proposed view of the objects of perception. That argument can be broken down as follows:
| 1. | For any hallucination, there is act of awareness which is involved in having that hallucination. |
| 2. | The state of the agent’s brain in this case is sufficient for that act of awareness to occur. |
| 3. | There could be a veridical perceptual experience which involves the agent being in the same brain state. |
| 4. | So, such a veridical experience would involve the same act of awareness as in the corresponding hallucination. (1,2,3) |
| 5. | In such veridical experiences, there is no other act of awareness. |
| 6. | In such veridical experiences, an agent is aware only of what he could be aware of in cases of hallucination. (4,5) |
| 7. | All veridical experiences are such that it is possible that an agent be in the same brain state but having a hallucinatory experience. |
| C. | In veridical experience, agents can never be aware of anything more than what they can be aware of in hallucination. (6,7)) |
We have already seen evidence in favor of (1), and (3) and (7) do not appear to be open to question. So, if the argument is valid, Johnston’s view of perception requires that either (2) or (5) be false.
Johnston’s argument for (5); the implausibility of their being two distinct acts of awareness in cases of veridical seeing. That leaves (2) as the only disputable premise. Johnston expresses the intuition behind it well:
“In defense of the step, the following sort of thing can be said. If we were to hold that a different kind of act of awareness say, one with different kinds of objects was caused or constituted by the same kind of brain state when that brain state was itself caused in the standard way then we would be committed to something extremely odd. We should have to suppose something akin to action at a distance. For the brain state would have to look back and inspect its causal antecedents in order to see whatmental act to cause or constitute. Otherwise, how could brain state know that it should cause or constitute direct awareness of things in the environment when it was preceded by a normal causal chain going all the way out through the visual system to an external object? How could the brain state know to instead cause or constitute the kind of visual awareness involved in hallucination, awareness which is not of any object there in the scene before ones eyes, in the case where there was no such normal external connection? Obviously there is no such action at a distance or looking back. The brain state does not know such things. Irrespective of how it is caused, the brain state causes or constitutes an act of awareness with a certain character and directed at certain objects, the very type of act of awareness that occurs in the hallucinatory case.” (116)
How the conjunctivist can accommodate the conclusion of the above argument; direct vs. indirect objects of perception.
In the end, though, this is the step that Johnston would have us reject. The above argument for premise (2) is flawed because it takes the relationship between the physical and the mental to be one of causation, rather than material constitution:
“Yet behind this supposed proof of the Conjunctive Analysis is a picture of the relationship between the brain and the mind familiar from newspaper cartoons, where thought and experience are depicted as a sort of mental bubble secreted from the head. That is, the decisive move in the conjuring game is to slide in the supposition that the connection between a physical processes and awareness is itself process causation, as if energy leaked from the external world into the mind via the brain.
As against this picture, the relation between seeing an object and the long physical process involving first the light coming from the object and then the operation of the visual system is not the relation between a first mental effect and a prior physical process that causes it. Seeing the object is not the next event after the visual system operates. Seeing the object is an event materially constituted by the long physical process connecting the object seen to the final state of the visual system. Seeing the object is an event that is (as it actually turns out) constituted by a physical process that goes all the way out to the object seen. There is accordingly no looking back required by the last brain state or pattern of neuronal firing in order to determine whether to cause veridical awareness of external objects as opposed to the type of awareness involved in hallucination. There is no such last brain state that then causes seeing.
Seeing is an environment-revealing mental act that is materially constituted by a physical process that subtends the revealed environment.” (138-9)
A second kind of argument against Johnston’s view, which can also be seen as counting in favor of the conjunctive analysis, is based not on the supervenience of the objects of awareness on brain states, but not the supervenience of the objects of awareness on the phenomenal character of the experience. This is the argument from the ‘phenomenal bottleneck principle’:
The Phenomenal Bottleneck Principle: If two acts of awareness are qualitatively indistinguishable for their subject then objects of the very same type are directly presented in each act of awareness.
Counterexamples to the principle: cases of imperceptibly changing experiences, like Johnston’s case of watching the change from morning gloom to full daylight.
Suppose we grant much of what Johnston has to say about the objects of experience. Does this translate into a view about the content of experience? What’s the relationship supposed to be between the immediate objects of experience and how experience represents the world as being?