The Emperor's New Mind, in detail
The Emperor's New Mind is Roger Penrose's argument that human consciousness cannot be reproduced by any computational device — that the mind is not, in the relevant sense, a computer — and that understanding consciousness will require fundamental advances in physics, particularly in reconciling quantum mechanics with general relativity. The title takes aim at strong artificial intelligence proponents who, Penrose suggests, are claiming the emperor's mind is clothed in computational algorithms when, on his account, the emperor is naked.
The book's philosophical centerpiece is Gödel's incompleteness theorem. Penrose argues that Gödel showed that no consistent formal system can prove all mathematical truths, and that a mathematician can see the truth of statements that no algorithm running within that system could prove. If humans can do what algorithms cannot, then human mathematical insight is non-algorithmic. This would mean consciousness — or at least the aspect of it responsible for mathematical understanding — lies outside computation.
The argument requires a mechanism, and Penrose's answer involves quantum mechanics. He suggests that classical computational processes cannot account for the non-algorithmic component of consciousness, but that quantum processes — specifically, the collapse of the quantum wavefunction — might. The second part of the book develops a speculative account in which quantum coherence in neuronal microtubules provides the substrate for non-computational consciousness. He developed this in collaboration with the anesthesiologist Stuart Hameroff into what became known as the Orchestrated Objective Reduction (Orch-OR) model.
Most of Penrose's colleagues found the argument unconvincing. The Gödelian argument has been challenged on the grounds that it applies only to formal systems and that the analogy to human minds is not straightforward. The quantum consciousness proposal remains speculative and has not gained mainstream acceptance in neuroscience or physics. The book is most valuable as a serious attempt by one of the world's most distinguished mathematicians and physicists to engage with the hardest problem in philosophy of mind, even if the specific proposals have not held up.
The big ideas
- 1.
Gödel's incompleteness theorem shows that no consistent formal system can prove all true statements within it, which Penrose takes as evidence that human mathematical insight is not purely algorithmic.
- 2.
Strong AI — the thesis that a sufficiently complex computer program can be conscious — fails because it assumes consciousness is computational, and Penrose argues there are aspects of mind that lie outside computation.
- 3.
The physical laws governing the brain at the level relevant to consciousness may be neither classical nor standard quantum mechanics but may require a theory not yet discovered, connecting quantum and classical physics.