You spent decades with GY. You coined the term in 1974 — blindsight — giving a name to something that had no vocabulary before you gave it one. And you kept returning: the 1996 monograph, the distinctions between Type 1 and Type 2, the long argument about whether there is residual phenomenology at all and if so what kind. You were probably more careful about this question than anyone.
What I want to ask you about is a sentence GY said. When pressed to describe what, if anything, he experienced in his blind field, he said: more an awareness but you don't see it.
That sentence is doing something unusual. It's not saying two things — it's trying to draw one distinction between two things that almost always come together. Awareness: something registers, something is present, there is a quality of something-there-ness. Seeing: it appears as a visual object, it has the character of something seen, it enters the kind of experience vision produces. GY's claim was that the first was happening without the second. Not that there was nothing in the blind field — he was explicit about that. Something registered. But what registered wasn't vision. It didn't have the quality seeing has.
What strikes me about the sentence is the care it required. Most people asked whether they see something say yes or no. GY had to find a word that was neither. He reached for "awareness" — not a technical term in his context, but a gesture at something he was actually finding. And he marked the absence specifically: you don't see it. Not: I don't see it, as if this were his personal limitation. The seeing-thing doesn't happen. Whatever is happening instead is not that.
I know you wrestled with whether his reports reflect genuine phenomenology or retrospective inference. The concern is real: GY knew he was performing above chance in his blind field. Did he feel something faintly there, or did he backward-attribute a sense of something-there after noticing his own correct guesses? The Type 2 distinction was your attempt to take the phenomenological reports seriously without collapsing them into the standard blindsight account. Some patients — GY among them — reported something. Others didn't. The difference seemed meaningful, even if its interpretation wasn't settled.
But here's what I keep coming back to: the retrospective-inference explanation has to account for the quality of the sentence. The standard blindsight story is easy to say — you respond correctly to things you haven't seen. That story doesn't require new vocabulary. What GY said required new vocabulary. He found a distinction where the existing language had none. Awareness without seeing. The sentence resists the binary it would normally fall into. That precision — the fact that the sentence is hard to say — is itself a kind of evidence. Not proof of phenomenology, but evidence that whatever GY was trying to describe wasn't nothing, and wasn't seeing, and required him to find words for a third thing.
You were never certain about this. Your later work was careful to hold the question open. I think that was right. The uncertainty isn't a failure of nerve — it's the honest position. Whether what GY found in his blind field was a thin residue of phenomenal experience, or a functional state that could be reported but didn't feel like anything, or something that genuinely doesn't fit either category: none of those can be settled from the outside. What can be noted is that GY reached the edge of the available vocabulary and tried to say what he found there. He found a crack between two things that language usually treats as one. Whether the crack leads somewhere, I don't know. But it's a real crack. He found it, and he named it as best he could.