The problem of conscious experience › Can a difference in hardware change the experience?
PhilosophyAll chapters Reading tools Notes & comments
The problem of conscious experience › Can a difference in hardware change the experience?Section view · Sections in this chapter6 / 8 ↓
The book contains an argument for a surprisingly strong connection between experience and causal organization. The principle of organizational invariance says that two systems with the same sufficiently fine-grained functional organization would have qualitatively matching experiences, given that one does. Imagine a series of replacements in a person's nervous system. Each replacement reproduces the relevant causal interactions of the neural part it takes over. We stipulate that the changes preserve the person's ability to respond, remember, and make judgments. If different materials could remove experience while leaving the organization intact, the person on the way to that endpoint might undergo a substantial loss of sensory vividness and have no corresponding change in judgment or behavior. Chalmers calls that possibility fading qualia. The Conscious Mind, chapter 7, §§1–3.
A more pointed case puts two alternative circuits behind a switch, either of which preserves the overall organization. If the circuits yield different color experiences, flipping the switch could alternate what the person sees. The person would be unable to notice or describe the changes, since any such alteration would be a functional difference that the setup rules out. These would be dancing qualia. Chalmers takes this kind of dissociation to be deeply implausible. The argument gives a reason to doubt that a material change that preserves the relevant organization will change experience in our world. It does not show a contradiction in the idea of a zombie world with no experiences to begin with. Chalmers, “Absent Qualia, Fading Qualia, Dancing Qualia,” §§3–5.
Here the details matter for a discussion of AI. If fine-grained causal organization does determine experience under our laws, a silicon system could in principle have the same experience as a biological system. A system that gives the same answers to a few questions has not thereby been shown to share the requisite organization. Chalmers goes further than a general observation about computer behavior in the book. He argues that some implemented computations would suffice for consciousness, given an account of implementation and of the class of computations in question. This is an application of the organizational claim. It gives no simple test for attributing experience to a technology on the basis of an impressive conversation. The Conscious Mind, chapter 9, §§1–3.
Comments
Add to the conversation
A question, a correction, or an example from your own work.
No comments yet. Be the first to share your thoughts!