I liked it that Johansen’s paper began by trying to locate 3.2 within the program of De Anima as a whole. Usually the best way to see the significance of part of an Aristotelian treatise is to discover its role or function in the entire treatise. I've found that interpreters often go astray by looking at an Aristotelian text in isolation.
But this gave rise to Johansen’s first argument against Caston. Johansen argued that Aristotle’s general procedure in the De Anima is to explain life in terms of different sorts of faculties. This is evident throughout book II and in the rest of book III. We should therefore expect that 3.2 would similarly be an inquiry into whether a new faculty needs to be postulated in order to account for ‘perceiving that we perceive’. Thus we should expect 3.2 to be fundamentally making a point about capacities.
But Caston has a reply. He's familiar with this argument (a common view) and summarizes it nicely in his paper:
The motivation for a capacity reading appears to come directly from context. Having finished the discussion of the individual senses, Aristotle begins Book 3 of On the Soul by considering whether the five individual senses are sufficient to account for other perceptual abilities we possess, such as perceiving common perceptibles, or discriminating the perceptible qualities of one modality form another, or (on the capacity reading) perceiving that we see or here. On this reading, Aristotle would be resisting here, as elsewhere, an unnecessary multiplication of capacities, preferring instead to ground different abilities in a single capacity. (763)
Caston seems to admit that the argument has some force, but he tries to deflect its force by appeal to an even wider context:
But construing the passage in this way puts it at odds, prima facie, with other parts of his psychology. In On Sleeping and Waking [De Somno], Aristotle expressly denies that we perceive that we see by the capacity of sight:
There is a certain common capacity that supervenes on the others, by which one perceives that one is seeing and hearing. For it is surely not by sight that one sees that one sees…[455a15ff]
The difficulty is only prima facie, and Caston admits that a defender of a capacity reading could handle it in a variety of ways—with a developmental theory, for instance, or by holding that “in the end the two passages amount to much the same thing, though expressed in different ways.”
And it’s this latter approach that Johansen favors: Take De Somno (he suggested) to be saying that it’s not by the sense of sight in its standard or principal use that we perceive that we see; take De Anima to be saying that it’s sight in the respect in which it has the same power as any sense faculty (that is, as a “common sense”) by which we perceive that we see. Then there is no contradiction.
Okay, then, this disposes of the prima facie difficulty. So then, now where do we stand? It would seem that the point about the immediate context of 3.2 returns now, with something like its original force, and we should (once again) expect that the passage is making a claim about faculties or capacities. So it doesn't seem that Caston really has deflected the force of this consideration.
And then, additionally: it’s not as if the De Somno passage isn’t equally a prima facie difficulty for Caston’s interpretation. Yes, on the capacity reading, DA 3.2 ends up claiming explicitly that nothing more than sight is needed to explain perceiving that we see. But note: on Caston’s interpretation, it would seem that DA 3.2 ends up maintaining this, even if it does not say it outright. Recall, on Caston’s interpretation of DA 3.2, perceiving that one sees is built into each act of seeing. An act of seeing, and an act of perceiving that one sees, are one and the same act. But then, it would seem (at least prima facie): if sight accomplishes the first, then sight is similarly responsible for the second, and “we perceive that we see by the capacity of sight”.
Actually, it seems easier to make the capacity reading consistent with De Somno. It makes sense to say that a capacity can be used in a non-standard or incidental way, or only in a certain respect (‘in the respect in which that capacity has the same powers as other sense capacities’). But it seems less plausible to say that a feature that is inherent in the activity of a faculty—always, inevitably, inseparably—is not to be attributed to that faculty.