To the editors:
Re “Northwestern’s New A.I. Hotshot” Dec. 15
Roger Schank, the Dr. Pangloss of Northwestern U., thinks Chicago “too slow”; he would do well to tarry: Judging by Harold Henderson’s adulatory article, he and his alter ego-tist accept the common fallacy that simulation is reproduction–not that a rose is a rose is a rose, but that a rose made of wire and tissue would smell as sweet.
Henderson writes, “you need to understand very clearly how people think in order to program a computer to do so–and at the same time, you can check up on your understanding by seeing whether the computer in fact does what you meant it to do.” This statement overlooks the fact that there is no reason on this green earth to assume that the living thought of a living mind is in any way similar to the mechanical regurgitations of Dr. Schank’s computer. To be sure, we have Schank’s word on it, but–you must pardon me–Schank is wrong! That the computer can be programmed to crudely simulate certain products of the human mind may give us no more idea of the psychology of thought than the insights afforded by the B-1 bomber into the metabolism and mating habits of hummingbirds.
But Schank’s mind is well wired, and Henderson is thrilled to point out the connections (“the psychology-department connection” “the education-department connection”), and as the synapses crackle and pop we become aware that for Schank intelligence is a simple thing, being no more than “the ability to say the right thing at the right time,” and that if you think of the right retort two hours or two days later, “That’s not as intelligent.”
Even if we accept this purely verbal definition, we are entitled to make the distinction between occasional wit and perennial wisdom; between J.M. Whistler’s witticism “you will, Oscar, you will,” and Wilde’s observation that “the tyranny of the weak over the strong is the only kind that lasts.” Whistler had wit, Wilde had more.
In my opinion Schank is most dangerous and needs most to be curbed and opposed in his self-appointed role as educational reformer–or, as Henderson would say, in his education-department connection.
Reform will require more hardware and more software to make it possible for students to re-invent for themselves the science of chemistry. Any sensible teacher who demurs is dismissed as an “education bureaucrat.” The bother with this scheme of self-instruction (enabling, in Schank’s words, “children to explore, to question, to create, or to fail in an interesting way”) is the assumption of a considerable amount of knowledge and sophistication–even genius–on the part of the students; else how would they know what to explore, what questions to ask, how to “create,” or how to avoid–dithering among Professor Schank’s story-telling machines–uninteresting failure.
Should this kind of mechanized quackery be adopted in the schools, you can kiss good-bye to whatever is left of American intellect.
But enough! Almost every statement made by Schank and his Boswell begs–nay, beseeches and implores–a litany of questions which they are too arrogant to acknowledge or too ignorant to explore. Henderson tells us that Schank doesn’t suffer fools gladly but seems to think the rest of us ought to.