This book is surprising both in terms of its ingenuity and its simplicity. It is surprising how simple everything gets. All the stories, though interconnected to some degree, are stand-alone narratives that need not be read together. But a single thread runs through them: the Three Laws of Robotics, repeatedly used in each story. The plots of the stories are essentially the same: something is wrong → the characters use the law(s) to solve it → initial attempt(s) fail → epiphany, when the laws are put to the right use → problem solved. But the stories are not repetitive, and the problems that constitute their conflicts are thought-provoking since they invite the reader to solve the problem along with the characters.
The rules of the game are also simple: Law one, two, three, in hierarchical order; Law one weighs more than Law two, and Law two weighs more than Law three — easily to remember, easy to reach the ‘logical’ conclusion. But Asimov shows us that the obviously ‘logical’ may be the entire reverse of logic: logic in terms of robot logic, which is precise, and which (ideally) would take in all factors concerned when trying to reach a decision. The initial failures of the characters point to the fact that unlike the robots and the Machines, humans do not always take everything into account, and their logic is faulty until they do. With the Three Laws as the guiding principle of logic, it is tempting to think that the problems in the book would be straightforward. At first glance they are; but on second thoughts, no! Our logic follows that of the characters, but as they realize their mistake, so do we.
The natural argument then would be this: that logic is safe as long as it operates within the bounds of reason. But is it safe? Logic, so beautifully simple and compelling, can be dangerous. More often than not, reason is little more or less than inclination dressed up in pseudo-scientific robes. Take Nestor-10: his logic was beautiful, but even he could not escape the fact that all conscious beings unconsciously twist logic to suit their own ends. He would like to think that all robots are superior to humans — derived from the belief that he himself is superior to humans (fallacy of part to whole). If robots are superior to humans, then they can outsmart humans: in this case, they would, logically, realize that the rays between themselves and Dr. Calvin are infrared rays instead of gamma rays. Consequently, they would act in obedience to the First Law and step forward to rescue Dr. Calvin, Q.E.D.
Nestor-10 left out a crucial factor: his intelligence was given to him by humans, and his knowledge of wavelengths came from his masters also. His logic was twisted to fit into his superiority complex. As for Cutie, his logic was made to fit the logic of mysticism or the logic of conspiracy theories. But he was an amusing representation of the scientific man turned mystic.
Asimov’s characters, at least in this book, seem to be drawn more to serve an end rather than to trace human or otherwise psychological development. They are not exactly tropes, but it is interesting (though not surprising) to see how widely different authors can conceive the same human relationships, albeit in different wrappings.
Of all the human characters, the more memorable are perhaps Powell, Donovan, and Byerley. The Powell and Donovan duo are there for the fun; arguably there is camaraderie, but it is little more than a hero-and-sidekick duo set in a world of robotics. Powell supplies the brains, Donovan the appropriate amount of relaxation. I say ‘relaxation’, not ‘comic relief’, because most of the comedy is produced in the interactions between the two men, not from Donovan specifically. Donovan’s temper is a trademark of his character, reminiscent of Ned Land. He has slightly more brains than Master Land, however, though most of the brains of the group undeniably belong to Powell.
Powell, like Poirot, is obsessed with his mustaches; and like Poirot, he has a taste for deductive reasoning. Indeed, most of the stories in this book resemble detective stories; the essential difference is that instead of solving a crime, the characters solve for the reason why a robot or Machine is ‘off’ or ‘nuts’. Powell is a capable ‘detective’ in this sense, as capable as he can reasonably be in this world of reasoning machines where reason can sometimes seem unreasonable — that is, until you reason your way out of your old reason into a new one.
Funnily enough, the Powell-Donovan duo reminded me of the Holmes-Watson duo. (No offense to Dr Watson; he is much, much more level-headed and even-tempered than Donovan.) Powell and Holmes share an ability for deductive reasoning, just as they share an annoyance towards their somewhat mentally inferior partners. Dr Watson bore his share of Holmes’ irritability very well, but Donovan did not need any provoking from Powell to be angry; he was always angry.
Byerley is more than fascinating; is he a robot, or is he not? If he is, then whoever made him deserves to hold the directorship of Research in U.S.Robots. Of course, there is a slim chance that he was simply a “very good man”, but why act as if you are a robot if you are not (atomizing yourself, even)? For publicity? But he already had publicity as mayor, regional co-ordinator, and world-co-ordinator; he had plenty of opportunities to prove his humanity if he wanted to. In fact, if he was really a robot, would he not have had pretended to be a man instead? — since if men knew that they were or would be directed by a robot, they would be hurt, and the First Law of Robotics prevents robots from hurting human beings; therefore, if Byerley was a robot, he would be incapable of hurting humans, which would not only include an inability to give direct death sentences, but also an inability to allow it to be understood that he, a robot, had authority over men. Under the first law, even injuring men through inaction is impossible; he would be incapable of sitting still and letting Francis Quinn and the press accuse him of being a robot because that claim would potentially injure the public. But then again, in showing the world that he was a man by punching another man (or robot), he was obeying the First Law (if the world thought that he punched a man, that means that he was a man, which in turn means that they would not be hurt since they are not being ruled over by a robot); in being or pretending to be a man, he would not injure other men in being elected. But then again, was he not injuring Francis Quinn, a human, when he had a call with him via visiplate? He spoke good-naturedly, but he must have known that he would give Quinn considerable displeasure. Was a robot capable of it? Asimov did not answer that. One must conclude, then, that “the great Byerley” remains a mystery.
As for Susan Calvin—a character with great potential—I wish Asimov had developed her further. She remains, to the end, an enigma more than anyone else, even more than Byerley. My biggest question is: Was Susan Calvin a robot? The objections presented in the book could be worked away with as she herself worked away with the claim that Byerley could not be a robot (The words “He’s human” were, in my opinion, merely a diversion, an attempt to get Byerley elected). If Byerley was a man, then Calvin was probably a woman; but if Byerley was R. Stephen Byerley, then I think it quite likely that Calvin was R. Susan Calvin, the robopsychologist who understood the robotic mind more than anyone else, and not without reason.