In 1989, with the release of Roger Penrose’s book The Emperor’s New Mind, a squabble that had until then remained largely confined within the halls of academia reached the streets—though not with considerable uproar, which is understandable considering the topic and the means necessary to the discussion of that topic. What had been brewing in the academies was a controversy over the possibility of creating intelligence in machines. The proponents of artificial intelligence (AI for short) were multitudinous, hard scientists who did not bat an eye at reducing those most cherished of human attributes, intelligence and consciousness, to wires and binary notations. Aligned against them were the usual ragbag of anachronistic science-bangers— humanists, philosophers, literary critics, and the like—along with one or two bona fide scientists such as Penrose, who, unlike the humanists, did not manifest any distrust of the scientific method but nevertheless shared with them some discomfort as to the claims that method appeared so ready to make.
As could have been expected, Penrose’s book did little to soften anyone’s opinion in either direction. Indeed, the scientists appeared more convinced than ever of their claims because of certain weaknesses, acknowledged by Penrose himself, in the arguments he offered. The science-bangers, meanwhile, were generally grateful for help from so unexpected a quarter, even if that help came garbed in the language of the scientists and therefore lacked some of the more humanistic roundness and ring. Generally, the scientific community had two complaints: First, Penrose was too vague in his division between computational procedures and the noncomputational activity he assigned to the processes of consciousness, and second, his recourse to yet-to-be-discovered quantum-level effects to explain consciousness was not only vague but positively unscientific.
Shadows of the Mind: A Search for the Missing Science of Consciousness constitutes an attempt by Penrose to answer these two complaints, to fill in the holes visible to the scientific community. Accordingly, certain sections of the book are even more daunting to the layperson than similar sections of The Emperor’s New Mind. Nevertheless, a careful, thoughtful reading by the nonmathemetician can lead to comprehension of all but the most abstruse material Penrose presents. That is to say, Penrose succeeds in addressing scientist and nonscientist alike, with enough grist for both camps to grind.
The book is divided neatly into two parts, which address the two complaints leveled at the previous book. Part 1, entitled “Why We Need New Physics to Understand the Mind: The Non-computability of Conscious Thought,” explains fully what Penrose means by computation and why computation cannot elicit conscious thought. First, he breaks the various arguments regarding consciousness into four camps. Those in the first camp hold that all thinking is computation and that therefore, given the appropriate program and hardware, there is no reason a machine cannot have feelings of conscious awareness. This standpoint is often referred to as “strong AI.” If it is understood that a computer, through its digitalization of information about the physical universe, can present simulations of that universe, then, the strong AI standpoint holds, simulations of the physical activity of the brain will evoke in the simulation the same thing people in their brains call consciousness. The second viewpoint can be referred to as “weak AI.” Here, while all physical phenomena are computationally simulatable, a simulation will not itself be conscious. Consciousness is ascribed to the actual physical makeup of the brain. The philosopher John Searle, with his “Chinese room,” lends support to this viewpoint while attempting to undermine the strong AI claims. In the “Chinese room” a person who does not know Chinese waits for a card with a Chinese character on it to be passed through a slot in the wall. The person looks in a Chinese dictionary for the character, finds a definition in Chinese characters, and passes cards with those characters out of the slot. This person, Searle asserts, being unconscious of the meanings of the cards, is in the same situation as a computer simulation that, through its substitution of one symbol for another, accomplishes the same activity as the person in the Chinese room.
Penrose finds this argument inconclusive and...
(The entire section is 1823 words.)