Johansen on Inner Sense and the 'Structure of Consciousness'

(Is it because I'm composing this from a borrowed Mac--my %#**!@% PC being once again on the blink--or that Blogspot is not fully up and functioning today? In any case, I still cannot compose in html or introduce formatting, and the font seems different. So apologies for that.)

I had attempted, in a sense, to split the difference (or preserve the endoxa), in saying that Johansen seemed right about the 'capacity' reading, but perhaps Caston was right, after all, about the nature of an act of perception. But Johansen has sent me thoughtful comments, challenging this approach. I'll post them all at once, because, although without doubt they are fascinating and deserve a full discussion, that should perhaps be left to graduate students writing dissertations on the subject, or to a blog devoted entirely to De Anima 3.2 (and are there really 7,999,999 subjects more interesting than that?). Also, I don't want to take away all of the drama from the publication of Johansen's paper in the Proceedings of BACAP!

Here then are my comments in quotation marks, followed by Johansen's rejoinders in square brackets:

"I should perhaps explain why I think that Caston's view of the 'structure of consciousness' still matches better what Aristotle says in DA 3.2 than the 'inner sense' view of Johansen. As I understand DA 3.2, Aristotle is claiming that every sense, by its nature as a sense faculty, carries with it the capacity to monitor its own activity."
[As you know I agree, given a certain understanding of this nature.]

"We should stress, with Caston, that this 'monitoring' is not merely propriaceptive, in the sense that it is not merely a sensing that the sense is 'on'. The sensing is somehow a sensing, too, of the content of what is sensed."
[Again I agree. This is implied by Aristotle's saying that there will be a sense both of sight and of colour.]

"Now consider the conditions under which any sense operates. It operates if its object is present, if the medium (if any) is unobstructed, and if the sense is in working order. As regards sensing that we sense, all of these will necessarily be in place, when we sense. Hence there will be no sensing, without sensing that we sense,"
[Again I tend to agree. There is a long section in my paper based on the admission that we sense that we sense whenever we sense. The question is what are the implications for the relationship between those two sensings. However, you go on, apparently, to agree with Caston]
"and any act of sensing will have just the structure that Caston ascribes to it."
[But that does not seem to be a necessary consequence. The second-order act of sensing can follow necessarily from the first-order act without the implication that the second-order activity is tokened by the same activity that tokens the first-order activity, and specifically without the implication that the contents of these two activities form a complex content with the particular reflexive structure that Caston identifies. The second-order act can follow necessarily from the first-order act if the latter provides sufficient causal conditions for the second-order act to occur - in fact, that seems very similar to to the sort causal story I thought you were suggesting when you said above 'It operates if its object is present, etc.' I develop at some length in the paper a story which would make second-order perception causally but not conceptually necessary given  first-order perception, though I concede that the story is not watertight given our evidence. I don't think, then, that there is any need to accept Caston's specific proposals about the 'two types in one token' and the reflexive structure of this token perception (which you have not discussed in the blog - as yet?), as if this followed from saying that there is a necessary conjunction of first and second order perception. As I read it, Caston's fascinating proposal is a specific way of making sense of this conjunction, but by no means the only possible one, contrast your claim 'any act of sensing will have *just* the structure that Caston ascribes to it'. Even on a capacity reading, as I tried to show, there is a possible explanation of the conjunction.]

"Johansen ascribes sensing that we sense to a distinct 'inner sense', but I don't see that there is warrant for this in DA 3.2." [That all depends on what you mean by 'distinct'. Again in the paper I am very keen to maintain that it is sight that is responsible for perceiving that we see, but that we should take this to mean sight as integrated with the other senses. That was the way in which I tried to make DA III.2 consistent with De Somno. In fact, I don't think I have any stronger notion of 'inner sense' in mind than the one I thought you agreed with, namely a capacity which we have to perceive the workings of our senses. It seems meaningful to call this 'inner' in two ways: it is a capacity to perceiver the workings of the senses themselves (as opposed to the external objects) and it is a capacity that is internal to the external senses insofar as it relies (as De Somno in particular shows) on the integration of the external senses in the common sense. But 'inner' here decidedly does not mean distinct from (and it shouldn't, because we'll then get into the regress territory of DA III.2)]

"Also, Johansen needs to speculate that there can, in principle, be a sensing, without a sensing that we sense. He takes David Armstrong's example of a truck driver who drives 'on automatic pilot' to be in fact an example of this. But I think this is incorrect. The truck driver, in my view, senses that he senses: he simply doesn't think about what he is sensing."
[Ok, we can dispute whether the truck driver example in fact illustrates what Armstrong wants it to illustrate; I don't think Aristotle can agree with Armstrong's reading of the example either, since  I agree that Aristotle probably does mean to say that we perceive that we see whenever we see, whereas Armstrong wants to say that the truck driver is not aware of seeing the traffic lights, etc. Yet the example is suggestive of a structure of perceptual consciousness that Aristotle on my reading would agree with, namely that being aware that one sees is a matter of a further act of perception, delivered to you by an inner sense, which for Aristotle is not distinct from the special senses but is not identical with the special senses as special either.]

"Johansen might deny this and say that, no, the truck driver isn't aware that he is seeing the road, hearing traffic noises, etc. This seems wrong to me. But in any case the crucial question is: how do we decide which description is correct? Not, it seems to me--if we accept the 'capacity' reading of DA 3.2--through phenomenological analysis, or through introspection. Rather, it seems to me that Johansen is obliged to say how a sense faculty, as Aristotle understands it, could ever sense, without sensing that it senses."
[I did try to say this in the paper, though perhaps not succinctly enough in those terms. The account I developed here was in terms of the different contents, and consequently different truth conditions, of first and second order perception: perceiving colour is an act of special perception, perceiving that one sees colour is an act of accidental perception, which could be wrong in ways in which first order perception could not. Of course, since I agree that Aristotle probably thinks that we perceive that we see whenever we see, you wouldn't expect me to give actual (Aristotelian) examples of how first-order perception sometimes happens without second-order perception; all, I take it, you can expect me to show is how first-order perception is sufficiently different as a type of perception from second-order perception for it to be intelligible how the one might occur without the other; moreover you might also expect me, given Caston's alternative reading, to indicate reasons why first and second order perception ought to be tokened by different acts: here my suggestion was that since the truth conditions for first and second order perception differ (by virtue of their different contents) it was difficult to see (pace Caston) how they could consistently be tokened by the same act of perception.]

"The account has to be given in terms of the nature of a sense faculty. If Johansen can't do this, then he has to concede that Caston's account is more accurate."
[This was what I tried to in the paper by saying that since second-order perception involved accidental perception of a content that seemed deliverable only by virtue of the special senses' integration with each other in the common sense (cf. De Somno's parallel between perceiving that we see and hear, on the one hand, and perceiving the difference between white and sweet, on the other), we needed to understand the relationship between our abilities to engage in first and second order perception in terms of the relationship between the capacity of sight and the common sense.]