A central piece of the argument presented in the target article is Ned Block's claim that “phenomenal consciousness overflows cognitive accessibility” (target article Abstract). Block's main motivation for this claim is his intuition, shared by most of us, that we see much more than we can report. The feeling is particularly clear when an image is flashed before us, and we feel we have seen “everything in it,” yet generally will be unable to report more than 3 to 5 elements from the scene (the classic Sperling “whole report” paradigm).
The trouble is that an explanation for this feeling of being able to see more than we report might lie in the fact that the early stages of the visual system possess a certain degree of neural persistence. This neural persistence or “icon” might not itself be phenomenally conscious. Instead, it might, for a short while, be available for access by higher cortical mechanisms. Our impression of seeing “everything” might therefore derive not from actual phenomenality of the icon, but from the immediate availability for access of information in this persistent subcortical icon.
This is where Block musters new evidence from the recent experiments of Landman et al. (Reference Landman, Spekreijse and Lamme2003) and Sligte et al (Reference Sligte, Scholte and Lamme2008). These experiments, contrary to the original Sperling experiments (Sperling Reference Sperling1960), show that the neural persistence that seems to be involved in giving us the impression of seeing everything can sometimes be of considerably longer duration than previously measured. Information about the orientation of eight rectangles, for example, can sometimes be recovered as long as one second after stimulus extinction. This long persistence suggests that the information cannot be subcortical but must be of cortical origin, and Block concludes that for this reason it is likely to be phenomenal.
So Block's argument contains this critical step: the step consisting in assuming that if something provides an impression of detail, and its substrate is cortical, then it is likely to be phenomenal. This makes sense to Block because he starts from an a priori assumption that consciousness is a “natural kind” and has some kind of “neural signature.” Converging evidence, he says, suggests that if certain neural conditions are met (e.g., being cortical rather than subcortical), then visual information in the brain becomes conscious.
However, the neural signature hypothesis is merely speculative. It could indeed turn out that there is neural commonality in every case of consciousness, but why should one start out with this assumption? Moreover, even if it were the case that an identifiable neural signature for consciousness existed, what would its significance be? Block himself suggests it would leave us still struggling with a “hard problem” of consciousness, showing that this neural signature would give us little insight into phenomenal consciousness.
Leaving aside the criticism of Block's use of recent empirical data in defense of his hypothesis, we are surprised at another aspect of his argument.
Block once had a program of disproving philosophical functionalism. There were two kinds of consciousness: phenomenal consciousness and access consciousness. Phenomenal consciousness was pure and nonfunctional. Block thought the existence of phenomenal consciousness (for which he adduced empirical and theoretical material) proved there was more to consciousness than the functional (Block Reference Block1995b).
The problem, critics quickly pointed out, was that something purely nonfunctional, something that has no effects, and in principle can have no effects, is not only undiscoverable (any observation would be an effect), but epiphenomenal as well: it cannot even have any effects on the phenomenology of the subject in whose consciousness it is present. Nevertheless, if the goal of showing the plausibility of a nonfunctional phenomenal consciousness were attainable, this would be an important result.
But in the target article the opposition between the realm of the phenomenal and the realm of access is diluted. Today's version of phenomenal consciousness is no longer completely access-resistant. After all, Block musters evidence for its existence by noting that people report that they think they have seen everything in the scene even though they cannot always report all the details. So this is a report, just not a full report. Furthermore he claims further evidence by saying that they can report the details if cued early enough. This is also a report!
Thus, instead of the original strong opposition between functional access consciousness and nonfunctional phenomenal consciousness, there now seems actually to be a tight link between this new (avowedly less-than-) cognitive access, and phenomenal consciousness. Indeed, nothing in the present target article indicates phenomenal consciousness cannot be interpreted as a weaker variant of access consciousness. Importantly, and further encouraging this interpretation, phenomenal consciousness also no longer gets identified with the qualitative, often called “what-it-is-like” aspects of consciousness.
If phenomenal consciousness is no longer sharply distinguished from access consciousness, a significant issue on the Scientific Explanation of Consciousness is transformed into a rather minor quibble about the scope of the access subjects have to visual information in psychological experiments! Deciding whether subjects can describe four or eight masked items after a certain number of milliseconds is perhaps an important project in psychology and neuroscience, but resolving the question does not seem like a big advance on the mind/body problem.
