How to create a robot that has subjective experiences: Appendix
This post is an appendix to another one on subjective qualia in AI, and is not intended to be a standalone post. In this post we look at more detailed philosophical questions related to consciousness, reserved for readers interested in topics like objectivity and epiphenomena. This post is not a requirement if you want to understand the original.
Objective subjectivity
I find myself absorbed in an orange sensation, and something is going on. There is something that needs explaining, even after we have explained the processes of discrimination and action: there is the experience. — Chalmers, The Conscious Mind
The point of view in question is not one accessible only to a single individual. Rather it is a type. — Nagel, What is it like to be a bat?
When people ask questions like “is my experience of red the same as yours, or is it like your experience of blue?”, they are assuming that qualia are “static” entities that can in some sense be compared across people (substrate independence). Both Chalmers and Nagel treat various conscious experiences as “types”, that are consistent across an individual’s life, and perhaps even across a species, such as when you compare your experience of the colour red to others’. For example, although Nagel asserts that you could not imagine what it is like to be a bat, he concedes that you can imagine what it is like to be another human. Given that I cannot even imagine what it is like to be myself at a previous time (see the main post), I find this hard to accept.
Qualia are by definition subjective and momentary, and cannot be transposed outside the mind that is currently experiencing them, nor even between two moments in the same mind. The irony is that Chalmers himself, perhaps against his own wishes, makes great efforts to push the experience of qualia into the realm of objective, real facts that must be accounted for, and that are separate from their momentary interpretation:
It is often hard to pin down just what the qualitative feel of an occurrent thought is, but it is certainly there.
There is nothing we know about more directly than consciousness, but it is far from clear how to reconcile it with everything else we know. Why does it exist? What does it do? [emphases added]
He speaks about qualia as though they were objective. You are seeing them because they are there. You are analyzing and gathering facts about qualia as objects present to your mind. At the same time he insists that they are exclusively subjective, and can never be explained objectively:
The subject matter is perhaps best characterized as “the subjective quality of experience.”
when I talk about consciousness, I am talking only about the subjective quality of experience: what it is like to be a cognitive agent.
One can see how such a contradiction can lead to a “hard problem”. An internally inconsistent question would indeed be a “hard” one to solve. Do qualia exist as self-consistent entities in their own right, apart from how you perceive and interpret them, or are they a momentary interpretation? Should “qualia” be a verb and not a noun?
Our language and our thinking has always tended to frame and interpret experiences as “things”. This is the best and perhaps only way for the mind to think about anything, and qualia are no exceptions. For example, Frank Jackson’s thought experiment, Mary’s Room, describes how Mary, raised in a room with no red objects, suddenly sees a red item for the first time, and learns something “new”, something that could not be described to her objectively before. This frames knowledge of “red” as knowledge of an irreducible entity, as a “fact” about things, instead of a newfound mental channel through which she can now interpret and evaluate the world.
One reason it appears to your judgment that qualia are a constant feature of all experiences is because you can always perceive them when you want to, just as you can with physical objects. The “illusion” of qualia, if there is such an illusion, is that they are present and remain the same when you are not actively aware of them. But qualia don’t exist, free-standing, waiting for some function of your mind to “glance at” them. Although the physical inputs of colour, sound, etc. from your body are consistently present (more or less), their qualia, their subjective interpretations are instantiated the moment you look for them, by the very processes that is looking.
The epiphenomenon fallacy
To be clear, I am in no way arguing that qualia themselves are an illusion. Or, more accurately, they are as much an illusion as your belief in cars or beauty.
Some say that consciousness is an “illusion,” but I have little idea what this could even mean. It seems to me that we are surer of the existence of conscious experience than we are of anything else in the world. — Chalmers
Many philosophers have tried to explain away qualia as an epiphenomenon, suggesting that it’s merely a different perspective of the same mechanical processes of the brain, seen from the subjective side. For example, as you are thinking about an apple, and how to eat it, they suggest that the qualia of this experience is being “released” or “reflected” from mental interactions at all times, like a halo of radiation around a hot poker.
This perspective denies those experiences a proper seat of their own in the catalogue of mental events. It likens qualia to a reflection of other mental events in a parallel, subjective mirror — a superficial, reflected, or emergent property — whereas the experience of qualia must be granted its own seat and category as a mental event. Chalmers is justifiably indignant when people try to reduce consciousness to a mere echo of physical events, accusing them of avoiding the hard problem. He rightly recognizes that there are distinct processes involved in the experience of qualia that are not reducible to how they manifest in physical behaviour like writing a book.
There are two causes for the error of epiphenomenalism. The first is the behaviourist stance cognitive scientists tend to take towards consciousness. If you assume that all thinking ultimately serves behaviour, then there is no other place to situate qualia except as epiphenomena. and since nothing about the machinery of behaviour could ever lead one to propose that there is a subjective experience of consciousness, the misalignment must either be papered over metaphysically, or it will have to be ignored.
Consciousness is not an explanatory construct, postulated to help explain behavior or events in the world. Rather, it is a brute explanandum, a phenomenon in its own right that is in need of explanation. — Chalmers
But this is an unnecessary constraint, as everyday experience shows that you often contemplate a subject matter for its own sake, with nothing to show for it but the memories of your own thoughts. Most experiences of qualia fall into this category of contemplation without action, contemplation that is its own ends, which is why they get excluded from purely behaviourist interpretations.
The second cause is the objectification of qualia discussed above. If you treat qualia as subjective “entities”, then the obvious parallel to draw is to physical events in the brain. Both now become “properties” that must be matched against one another:
The properties of experience (phenomenal properties, or qualia) systematically depend on physical properties according to some lawful relation. […] precisely how do phenomenal properties depend on physical properties?
Even once it is accepted that experience arises from physical systems, the question remains open: in virtue of what sort of physical properties does conscious experience arise? Some property that brains can possess will presumably be among them, but it is far from clear just what the relevant properties are.
But as phenomenal “properties” they are basically supernatural, or at least non-scientific. So one must either confront or side-step the issue of how they correlate with their physical counterparts. This rightly incurs Nagel’s displeasure:
I find the hypothesis that a certain brain state should necessarily have a certain subjective character incomprehensible without further explanation. — Nagel, What is it like to be a bat?
Chalmers is justified in saying that subjective experiences can never be reduced to objective descriptions. This, however, is not a “problem” per se, but merely a difference. Subjectivity is specific to the subject; another person’s subjective experience will never be objective for you, and vice versa. Nagel recognized this:
Every subjective phenomenon is essentially connected with a single point of view, and it seems inevitable that an objective, physical theory will abandon that point of view. — Nagel, What is it like to be a bat?
Chalmers’ own interpretation on the other hand handcuffs the processing of qualia to the subjective experience of qualia. And since the latter is by definition inaccessible from the outside, this makes both unattainable. This is like saying that all caged birds are flightless birds; whereas the cage and the bird’s ability to fly must be considered separately. The subjectivity of qualia and the machinery of qualia must be kept as two distinct questions. Once you separate them, and pare away all the explainable machinery of qualia and consciousness, the only unexplained residue left is the question of why you, as a subject, exist at all; and that is not a question for psychology.
