The Rabbis and the First Thinking Machine

• Email
• Share
• Print

When the first thinking machine rolls out of the lab — and it will, one of these days — it will seem human. When it controls a sophisticated anthropoid robot, the robot’s behavior will be indistinguishable, by and large, from a human being's. It will converse and cope with the world as a human does. It will certainly be far more human-like than any ape or dolphin.

Most of us tend to anthropomorphize the natural world. We treat pets as quasi-human. We anthropomorphize animals we work with or rely on, as many people used to treat their horses. Small children treat dolls or stuffed animals as if they were human.

The first thinking machine or anthropoid robot will be far more human-like than any pet dog or workhorse. It will be natural to address such a machine as “you” and unnatural to smash it to bits with a sledgehammer. After all, you might have had detailed, intimate, confidential conversations with this machine about your love life or children or best friends — and the machine might have made witty comments or offered valuable advice.

Still: it is only a machine. It acts the part of an intelligent agent perfectly, yet it is unconscious (as far as we know, there is no way to create consciousness using software and digital computers). Being unconscious, it has no mind. Software will make it possible for a computer to imitate human behavior in detail and in depth. But machine intelligence is a mere façade. If we kick our human-like robot in the shin, it will act as if it is in pain but will feel no pain. It is not even fair to say that it will be acting. A human actor takes on a false persona, but underneath is a true persona; a thinking machine will have nothing “underneath.” Behind the impressive false front, there will be no one home. The robot will have no inner life, no mental landscape, no true emotions, no awareness of anything.

So how could we possibly have moral duties to a robot or a thinking machine any more than we do to a toaster? How could we possibly have moral duties to a mere it? Won’t it be ridiculous, morally speaking, to have any qualms about smashing a robot, replacing its head, or dumping it in the garbage?

How should we treat thinking machines and human-like robots? Jewish thought offers us a way to proceed.

One way to discuss the problem is in the terms developed by Martin Buber, who created an ethics and theology based on relations among I, you, and it. For Buber, I and you can enter sympathetically into each others’ lives; our mental worlds flow together. But I and it are permanently separate. When I converse with an it, I do not actually converse at all; I conduct a monologue in which one party is me and the other is also me. This “other” is my own private, personal conception of someone or something else.

Buber used these terms to describe relations among human beings and between human beings and God. But we can press them into service in a different, simpler context. We can say that an I always has moral duties to a you. But ordinarily, an I has no moral duties to an it.

Does a machine, once it has become intelligent, make the transition from it to you? Or do I sometimes have moral duties to an it as if it were a you? Could I have moral duties to a mere thing that is unconscious, has never been conscious, and never will be?

Here we can make use of Judaism's view of animals. The rabbis of the Talmud unequivocally condemn cruelty to animals. They could hardly do otherwise, given the Bible’s insistence that on the Sabbath, humans and their animals must rest. The Talmud requires that we feed our animals before we ourselves eat, and one of its “seven laws of the sons of Noach” — the rabbis’ view of the basic moral principles binding on all mankind — is a prohibition against eating flesh from a live animal. This particular revolting practice no longer exists, but the prohibition is a typical rabbinic synecdoche, condemning cruelty to animals in general by means of one flagrantly concrete example (just as we refer to Israel in general by the name of one mountain, Zion).

Judaism has no concept of rights and certainly no concept of animal rights. Rights are a Roman idea; Judaism deals instead with duties. Kindness to animals is not a right enjoyed by cows and elephants but a duty that binds every human being. Irrespective of the moral standing of animals, our own moral standing requires that we treat them kindly. To treat animals cruelly is inconsistent with the moral stature and dignity of human beings. And insofar as animals are human-like, treating them cruelly numbs us to the mistreatment of human beings.

Compare this view to that of today’s best-known animal-rights group. In 1989, PETA famously announced that it was opposed to drug testing on animals “even if it resulted in a cure for AIDS” (then a more relentlessly deadly disease than it is today). This view implies that killing or endangering animals is prohibited even if human lives are at stake.

Judaism, on the other hand, prohibits cruelty to animals because of the sanctity and ineffable value of human life. Allowing human interests to be compromised because of our duties to animals might be necessary at times, but the starting point in evaluating such cases must always be the well-being of our fellow men.

Thinking machines will present a new challenge. “Cruelty” to thinking machines or anthropoid robots will be wrong because such machines will seem human. We should do nothing that elicits expressions of pain from a thinking machine or human-like robot. (I speak here of a real thinking machine, not the weak imitations we see today; true thinking machines are many decades in the future.) Wantonly “hurting” such a machine will damage the moral atmosphere by making us more oblivious of cruelty to human beings.

Because we anthropomorphize so eagerly, the fact that thinking machines are unconscious will tend to disappear from human awareness. Only a few stubborn philosophers and technologists will even remember that a thinking machine is an it and not a you. When we reach this point, we will be in danger of succumbing to a pernicious incrementalism that could badly damage civilized morality.

Some people will say: If we must be kind to anthropoid robots, why not to robots that don’t think but are nonetheless complex and sophisticated? And what about other complex, sophisticated machines? And what about machines in general?

The idea that society will one day be faced with a “machine rights” campaign seems absurd. But it is no more absurd than PETA’s absolutist credo: “animals are not ours to eat, wear, experiment on, or use for entertainment” — which rules out beetle collections, pet dogs, and leather shoes, as well as any life-saving medical or surgical breakthrough based on animal tests.

We will need to be robust and sure-footed in our moral analysis of thinking machines. Our approach must be grounded in the Judeo-Christian tradition from which our civilization has grown, not on the polemical instincts of radicals and academic extremists. Above all, we must be prepared for this ethical challenge — and start thinking about it now.

David Gelernter is a professor of computer science at Yale University, a contributing editor at the Weekly Standard, and a regular contributor to the Frankfurter Allgemeine Zeitung. The most recent of his many books is Judaism: A Way of Being.