Амброз Бирс - We, Robots

To declare that the personoids are "handicapped" with respect to us, inasmuch as they do not see or hear as we do, is totally absurd, because with equal justice one could assert that it is we who are deprived with respect to them—unable to feel with immediacy the phenomenalism of mathematics, which, after all, we know only in a cerebral, inferential fashion. It is only through reasoning that we are in touch with mathematics, only through abstract thought that we "experience" it. Whereas the personoids live in it; it is their air, their earth, clouds, water, and even bread—yes, even food, because in a certain sense they take nourishment from it. And so they are "imprisoned," hermetically locked inside the machine, solely from our point of view; just as they cannot work their way out to us, to the human world, so, conversely—and symmetrically—a man can in no wise enter the interior of their world, so as to exist in it and know it directly. Mathematics has become, then, in certain of its embodiments, the life-space of an intelligence so spiritualized as to be totally incorporeal, the niche and cradle of its existence, its element.

The personoids are in many respects similar to man. They are able to imagine a particular contradiction (that a is and that not- a is) but cannot bring about its realization, just as we cannot. The physics of our world, the logic of theirs, does not allow it, since logic is for the personoids’ universum the very same action-confining frame that physics is for our world. In any case—emphasizes Dobb—it is quite out of the question that we could ever fully, introspectively grasp what the personoids "feel" and what they "experience" as they go about their intensive tasks in their nonfinite universum. Its utter spacelessness is no prison—that is a piece of nonsense the journalists latched onto—but is, on the contrary, the guarantee of their freedom, because the mathematics that is spun by the computer generators when "excited" into activity (and what excites them thus is precisely the activity of the personoids)—that mathematics is, as it were, a self-realizing infinite field for optional actions, architectural and other labors, for exploration, heroic excursions, daring incursions, surmises. In a word: we have done the personoids no injustice by putting them in possession of precisely such and not a different cosmos. It is not in this that one finds the cruelty, the immorality of personetics.

In the seventh chapter of Non Serviam Dobb presents to the reader the inhabitants of the digital universum. The personoids have at their disposal a fluency of thought as well as of language, and they also have emotions. Each of them is an individual entity; their differentiation is not the mere consequence of the decisions of the creator-programmer but results from the extraordinary complexity of their internal structure. They can be very like, one to another, but never are they identical. Coming into the world, each is endowed with a "core," a "personal nucleus," and already possesses the faculty of speech and thought, albeit in a rudimentary state. They have a vocabulary, but it is quite spare, and they have the ability to construct sentences in accordance with the rules of the syntax imposed upon them. It appears that in the future it will be possible for us not to impose upon them even these determinants, but to sit back and wait until, like a primeval human group in the course of socialization, they develop their own speech. But this direction of personetics confronts two cardinal obstacles. In the first place, the time required to await the creation of speech would have to be very long. At present, it would take twelve years, even with the maximization of the rate of intracomputer transformations (speaking figuratively and very roughly, one second of machine time corresponds to one year of human life). Secondly, and this is the greater problem, a language arising spontaneously in the "group evolution of the personoids" would be incomprehensible to us, and its fathoming would be bound to resemble the arduous task of breaking an enigmatic code—a task made all the more difficult by the fact that such a code would not have been created by people for other people in a world shared by the decoders. The world of the personoids is vastly different in qualities from ours, and therefore a language suited to it would have to be far removed from any ethnic language. So, for the time being, linguistic evolution ex nihilo is only a dream of the personeticists.

The personoids, when they have "taken root developmentally," come up against an enigma that is fundamental, and for them paramount—that of their own origin. To wit, they set themselves questions—questions known to us from the history of man, from the history of his religious beliefs, philosophical inquiries, and mythic creations: Where did we come from? Why are we made thus and not otherwise? Why is it that the world we perceive has these and not other, wholly different properties? What meaning do we have for the world? What meaning does it have for us? The train of such speculations leads them ultimately, unavoidably, to the elemental questions of ontology, to the problem of whether existence came about "in and of itself," or whether it was the product, instead, of a particular creative act—that is, whether there might not be, hidden behind it, invested with will and consciousness, purposively active, master of the situation, a Creator. It is here that the whole cruelty, the immorality of personetics manifests itself.

But before Dobb takes up, in the second half of his work, the account of these intellectual strivings—these struggles of a mentality made prey to the torment of such questions—he presents in a series of successive chapters a portrait of the "typical personoid," its "anatomy, physiology, and psychology."

A solitary personoid is unable to go beyond the stage of rudimentary thinking, since, solitary, it cannot exercise itself in speech, and without speech discursive thought cannot develop. As hundreds of experiments have shown, groups numbering from four to seven personoids are optimal, at least for the development of speech and typical exploratory activity, and also for "culturization." On the other hand, phenomena corresponding to social processes on a larger scale require larger groups. At present it is possible to "accommodate" up to one thousand personoids, roughly speaking, in a computer universum of fair capacity; but studies of this type, belonging to a separate and independent discipline—sociodynamics—lie outside the area of Dobb’s primary concerns, and for this reason his book makes only passing mention of them. As was said, a personoid does not have a body, but it does have a "soul." This soul—to an outside observer who has a view into the machine world (by means of a special installation, an auxiliary module that is a type of probe, built into the computer)—appears as a "coherent cloud of processes," as a functional aggregate with a kind of "center" that can be isolated fairly precisely, i.e., delimited within the machine network. (This, nota bene, is not easy, and in more than one way resembles the search by neurophysiologists for the localized centers of many functions in the human brain.) Crucial to an understanding of what makes possible the creation of the personoids is Chapter 11 of Non Serviam, which in fairly simple terms explains the fundamentals of the theory of consciousness. Consciousness—all consciousness, not merely the personoid—is in its physical aspect an "informational standing wave," a certain dynamic invariant in a stream of incessant transformations, peculiar in that it represents a "compromise" and at the same time is a "resultant" that, as far as we can tell, was not at all planned for by natural evolution. Quite the contrary; evolution from the first placed tremendous problems and difficulties in the way of the harmonizing of the work of brains above a certain magnitude—i.e., above a certain level of complication—and it trespassed on the territory of these dilemmas clearly without design, for evolution is not a deliberate artificer. It happened, simply, that certain very old evolutionary solutions to problems of control and regulation, common to the nervous system, were "carried along" up to the level at which anthropogenesis began. These solutions ought to have been, from a purely rational, efficiency-engineering standpoint, canceled or abandoned, and something entirely new designed—namely, the brain of an intelligent being. But, obviously, evolution could not proceed in this way, because disencumbering itself of the inheritance of old solutions—solutions often as much as hundreds of millions of years old—did not lie within its power. Since it advances always in very minute increments of adaptation, since it "crawls" and cannot "leap," evolution is a dragnet "that lugs after it innumerable archaisms, all sorts of refuse," as was bluntly put by Tammer and Bovine. (Tammer and Bovine are two of the creators of the computer simulation of the human psyche, a simulation that laid the groundwork for the birth of personetics.) The consciousness of man is the result of a special kind of compromise. It is a "patchwork," or, as was observed, e.g., by Gebhardt, a perfect exemplification of the well-known German saying: "Aus einer Not eine Tugend machen" (in effect: "To turn a certain defect, a certain difficulty, into a virtue"). A digital machine cannot of itself ever acquire consciousness, for the simple reason that in it there do not arise hierarchical conflicts of operation. Such a machine can, at most, fall into a type of "logical palsy" or "logical stupor" when the antinomies in it multiply. The contradictions with which the brain of man positively teems were, however, in the course of hundreds of thousands of years, gradually subjected to arbitrational procedures. There came to be levels higher and lower, levels of reflex and of reflection, impulse and control, the modeling of the elemental environment by zoological means and of the conceptual by linguistic means. All of these levels cannot, do not "want" to tally perfectly or merge to form a whole.