Амброз Бирс - We, Robots

NAAD says that perhaps God does not wish a situation of such compulsion to believe in Him—a situation that would arise in a creation based on that other logic postulated by ADAN 300. To this the latter replies:

An almighty God must also be all-knowing; absolute power is not something independent of absolute knowledge, because he who can do all but knows not what consequences will attend the bringing into play of his omnipotence is, ipso facto, no longer omnipotent; were God to work miracles now and then, as it is rumored He does, it would put His perfection in a most dubious light, because a miracle is a violation of the autonomy of His own creation, a violent intervention. Yet he who has regulated the product of his creation and knows its behavior from beginning to end has no need to violate that autonomy; if he does nevertheless violate it, remaining all-knowing, this means that he is not in the least correcting his handiwork (a correction can only mean, after all, an initial non-omniscience), but instead is providing—with the miracle—a sign of his existence. Now, this is faulty logic, because the providing of any such sign must produce the impression that the creation is nevertheless improved in its local stumblings. For a logical analysis of the new model yields the following: the creation undergoes corrections that do not proceed from it but come from without (from the transcendental, from God), and therefore miracles ought really to be made the norm; or, in other words, the creation ought to be so corrected and so perfected that miracles are at last no longer needed. For miracles, as ad hoc interventions, cannot be merely signs of God’s existence: they always, after all, besides revealing their Author, indicate an addressee (being directed to someone here in a helpful way). So, then, with respect to logic it must be thus: either the creation is perfect, in which case miracles are unnecessary, or the miracles are necessary, in which case the creation is not perfect. (With miracle or without, one may correct only that which is somehow flawed, for a miracle that meddles with perfection will simply disturb it, more, worsen it.) Therefore, the signaling by miracle of one’s own presence amounts to using the worst possible means, logically, of its manifestation.

NAAD asks if God may not actually want there to be a dichotomy between logic and belief in Him: perhaps the act of faith should be precisely a resignation of logic in favor of a total trust.

ADAN: Once we allow the logical reconstruction of something (a being, a theodicy, a theogony, and the like) to have internal self-contradiction, it obviously becomes possible to prove absolutely anything, whatever one pleases. Consider how the matter lies. We are speaking of creating someone and of endowing him with a particular logic, and then demanding that this same logic be offered up in sacrifice to a belief in the Maker of all things. If this model itself is to remain noncontradictory, it calls for the application, in the form of a metalogic, of a totally different type of reasoning from that which is natural to the logic of the one created. If that does not reveal the outright imperfection of the Creator, then it reveals a quality that I would call mathematical inelegance—a sui generis unmethodicalness (incoherence) of the creative act.

NAAD persists: Perhaps God acts thus, desiring precisely to remain inscrutable to His creation—i.e., nonreconstructible by the logic with which He has provided it. He demands, in short, the supremacy of faith over logic.

ADAN answers him: I follow you. This is, of course, possible, but even if such were the case, a faith that proves incompatible with logic presents an exceedingly unpleasant dilemma of a moral nature. For then it is necessary at some point in one’s reasonings to suspend them and give precedence to an unclear supposition—in other words, to set the supposition above logical certainty. This is to be done in the name of unlimited trust; we enter, here, into a circuius vitiosus, because the postulated existence of that in which it behooves one now to place one’s trust is the product of a line of reasoning that was, in the first place, logically correct ; thus arises a logical contradiction, which, for some, takes on a positive value and is called the Mystery of God. Now, from the purely constructional point of view such a solution is shoddy, and from the moral point of view questionable, because Mystery may satisfactorily be founded upon infinity (infiniteness, after all, is a characteristic of our world), but the maintaining and the reinforcing of it through internal paradox is, by any architectural criterion, perfidious. The advocates of theodicy are in general not aware that this is so, because to certain parts of their theodicy they continue to apply ordinary logic and to other parts, not. What I wish to say is this, that if one believes in contradiction, one should then believe only in contradiction, and not at the same time still in some noncontradiction (i.e., in logic) in some other area. If, however, such a curious dualism is insisted upon (that the temporal is always subject to logic, the transcendental only fragmentarily), then one thereupon obtains a model of Creation as something that is, with regard to logical correctness, "patched," and it is no longer possible for one to postulate its perfection. One comes inescapably to the conclusion that perfection is a thing that must be logically patched.

EDNA asks whether the conjunction of these incoherencies might not be love.

ADAN: And even were this to be so, it can be not any form of love but only one such as is blinding. God, if He is, if He created the world, has permitted it to govern itself as it can and wishes. For the fact that God exists, no gratitude to Him is required; such gratitude assumes the prior determination that God is able not to exist, and that this would be bad—a premise that leads to yet another kind of contradiction. And what of gratitude for the act of creation? This is not due God, either. For it assumes a compulsion to believe that to be is definitely better than not to be; I cannot conceive how that, in turn, could be proven. To one who does not exist surely it is not possible to do either a service or an injury; and if the Creating One, in His omniscience, knows beforehand that the one created will be grateful to Him and love Him or that he will be ungrateful and deny Him, He thereby produces a constraint, albeit one not accessible to the direct comprehension of the one created. For this very reason nothing is due God: neither love nor hate, nor gratitude, nor rebuke, nor the hope of reward, nor the fear of retribution. Nothing is due Him. A God who craves such feelings must first assure their feeling subject that He exists beyond all question. Love may be forced to rely on speculations as to the reciprocity it inspires; that is understandable. But a love forced to rely on speculations as to whether or not the beloved exists is nonsense. He who is almighty could have provided certainty. Since He did not provide it, if He exists, He must have deemed it unnecessary. Why unnecessary? One begins to suspect that maybe He is not almighty. A God not almighty would be deserving of feelings akin to pity, and indeed to love as well; but this, I think, none of our theodicies allow. And so we say: We serve ourselves and no one else.

We pass over the further deliberations on the topic of whether the God of the theodicy is more of a liberal or an autocrat; it is difficult to condense arguments that take up such a large part of the book. The discussions and deliberations that Dobb has recorded, sometimes in group colloquia of ADAN 300, NAAD, and other personoids, and sometimes in soliloquies (an experimenter is able to take down even a purely mental sequence by means of appropriate devices hooked into the computer network), constitute practically a third of Non Serviam. In the text itself we find no commentary on them. In Dobb’s Afterword, however, we find this statement: