I'm only learning about this so please bear with me if I'm off track. But. Couldn't the experimenters know that she wasn't conscious of it because of the damage to the occipito-temporal cortex and ensuing interruption of the ventral stream?
Well, if so, only because of previous studies that showed that blindsight was associate with ventral stream disruption. The reason we know about the different functions of the dorsal and ventral stream is by
observing behaviour. We couldn't even talk about blindsight if it wasn't apparent from some aspect of behaviour that it wasn't normal sight.
Don't these case studies show that blindsight occurs where the dorsal stream to the posterior parietal cortex is intact following damage either to the occipito temporal cortex, an interruption of the ventral stream or both?
Sure, and it's a fascinating topic. In any case there are lots of ways in which we react without being (usually) conscious of reacting - the startle reflex, for instance. But my point is that if all we observed in an organism was the a startle reflex, we might be iffy about attributing consciousness to it (maybe a dim kind of sentience). But if it started to tell you about the experience of being startled, I would see no reason not to assume it was conscious of having been startled. We would, in other words, know it was conscious of having been startled because it said it was.
This is Dennett's point - we think we can imagine something that looks just like a conscious thing, but isn't. But when you try to pin it down, there is no reason to think that something that looks like a conscious thing wouldn't be. At least I am not convinced that there is. Certainly blindsight isn't an example because it is precisely because we know that the person isn't conscious of normal visual information that we know that they have blindsight.
My understanding is that the ventral stream is associated with cognitive function and the dorsal stream with motor function. Take away the consicous function, leave the motor function intact can't we say that we know she is unconscious of her behaviour which results from motor not cognitive function because of the nature of her lesion and is this not at odds with what was originally said i.e. "If someone...behaved as though they were conscious of things, then to say that they were not conscious would be an absurdity"
Well, now that we know what we know, we can infer, via inductive reasoning, that someone with that lesion will have blindsight, from the lesion alone, with a reasonable degree of being right (our knowledge isn't THAT good yet, though! We'd want to test, i.e. observe evidence of blindsight as well as evidence of the lesion).
But I think we've gone off track - my point was that if someone DIDN'T show evidence of blindsight, despite a blindsight lesion - we couldn't say "Ah! They only LOOK as though they have normal sight, actually they only have blindsight". Because if they had normal sight they'd be able to tell the orientation of the slot was without reaching. And if they could do that, then we'd have to infer that they were conscious of the orientation of the slot. Which is perfectly possible in principle - with training, it might be possible to teach blindsighted people to become aware of orientation by reaching and noting the direction tended to turn their hand, then simply by
imagining reaching, and noting the direction they felt they would turn their hand. But if they were then able to report the direction of the slot just by looking, we'd then have to say they had become
conscious of the orientation of the slot.
But again, we would be drawing the inference of consciousness
from the behaviour.