Omnipotence Paradox and Problem of Evil

Tintoretto, Moses Striking the Rock, 1555 - 1570


“Paradox is the philosophers’ enchantment, their fetish. It fascinates them, as a light does a moth. But at the same time, it cannot be endured. Every force available must be brought to bear to remove it. The philosopher is the shaman, whose task is to save us and rid us of the evil demon.” - Stephen Read, Thinking About Logic (1994)

Can an omnipotent being create a rock that nobody can lift? If he can, then he made a rock that he himself cannot lift. Then, there is at least one thing he cannot do. So, this being is not omnipotent after all. If he cannot, then there is again at least one thing he cannot do. Thus, this being is not omnipotent. Therefore, it is argued, omnipotence is impossible. Ergo, either God exists and he is not omnipotent or God is omnipotent and does not exist. This is the paradox of (the concept of) omnipotence. In its form, this paradox is akin to the Epicurean problem of evil which states that either God exists and he is not omnipotent, omniscient, or omnibenevolent or God is omnipotent, omniscient, and omnibenevolent, but does not exist.[1] The arguments of these paradoxes are valid. That is, these arguments are designed in such a way that, if one accepts the premises, he must accept the conclusion as well (or, differently put, if the premises are true, then it is impossible for the conclusion to be false), and the premises of these arguments are putatively intuitive. Thus, it seems that these paradoxes are inescapable. It is remarked that one has two choices in response to these paradoxes: either one accepts or dismisses the arguments. The former choice leads to some version of atheism or deflated version of theism. The latter choice bites the bullet at the cost of contrasting faith with reason. I think that this is a false dilemma and that there is the third way out. We can model this third way on how other paradoxes have been resolved in mathematics and logic, particularly Zeno’s paradox and Russell’s paradox.

Zeno’s paradox states that one can never arrive at a destination because the distance between the departing point and the destination is infinitely divisible. Russell’s paradox states that the concept of set or class is absurd because there can be self-contradicting sets such as the (super-)set of all sets that are not members of themselves: the super-set S is either a member of S or not; if SS, then S S, which implies ; if S S, then SS, which again implies . Obviously, we do not want to dismiss these paradoxes, but also need to find reasons not to accept them. If we accept them, we are basically accepting that no travel and grouping are possible. Yet, we do travel and group things all the time. If we dismiss them, we are admitting that many of our ordinary going-abouts are not backed up by reason. But, if so, why engage in them at all? A pragmatic reason is still a reason. Moreover, if we simply dismiss them without resolving the paradoxes, then we are admitting that all we are truly doing when we “think” we are traveling or grouping things is merely having the appearance of traveling or grouping things (if there is any sense left to the concept of travel and of group at all). But, in reality, we are not doing anything, or at least we are not doing what we think we are doing. Applied to the paradox of omnipotence above, if we dismiss its argument, then we are basically admitting that, while we “think” we believe in and pray to omnipotent God, what we are in fact doing (i.e., what we really believe in and whom we pray to) is something else. In other words, there is an unbridgeable gap between our concept (of omnipotence) and the reality (facts of matters)—for all we know, we could be making noises (such as “Omnipotence!”) while these noises fail to refer to anything. At worst, the noises do not stand for any concept to begin with. Thus, the paradoxes must be faced and be resolved lest our practices lose contact with reality.

Putting aside the mathematical details, there is a lesson to learn from the resolutions to Zeno’s paradox and Russell’s paradox. Theorists recognized that the paradoxes arise because our intuitive concepts of destination (or limit) and of set (or class) are too naïve or ambiguous to be capturing anything concrete; the paradoxes arise simply because our concepts themselves are intrinsically paradoxical.[2] Thus, the solution was to revise or improve naïve intuitions. The epsilon-delta definition of a limit formalizes the concept of limit by embedding it to the conceptual framework of mathematical analysis. The axiomatic set-theory (specifically, ZFC set theory) formalizes the concept of set by embedding it to the conceptual framework of first-order logic. As the concept of limit (or destination) and of set are rigorously defined or formulated in terms of function or by certain axioms, much of our naïve intuitions about these concepts are trimmed off, deflating the paradoxes stemming from the concepts. Furthermore, the epsilon-delta definition of a limit and the axiomatic set-theory, in each of their own way, have also formalized or contextualized the concept of infinity, preventing any paradox that could possibly arise from the ambiguity of the concept of infinity.

It is not difficult to find similar progresses in the history of mathematics and logic as well as of science. For instance, in mathematics, the problem of the trisection of an angle was solved by “discovering” (or formalizing) complex numbers by expanding the formal system of algebra. In logic, the problem of terms that denote nonexistent objects was solved by formalizing the concept of existence into a quantifier operator of predicate logic. In chemistry, the problem of some metals gaining mass after going through combustion was solved by replacing the conceptual framework of phlogiston theory with that of oxidation theory. In physics, certain problems involving high velocities, strong gravitational fields, etc., were solved by adopting non-Euclidean frameworks for construing space and time. In each of these historical instances, paradoxes, problems, and anomalies arose due to the lack of sufficient expressive powers of naïve concepts and intuitions, and resolutions were reached by formalizing the concepts or adopting some formal (or at least abductively better) system with more expressive powers.

What I propose in respect to the paradox of omnipotence and also to the Epicurean problem of evil is that the paradoxes must be taken seriously, but it should be hypothesized that the paradoxes arise because of the ambiguous nature of our naïve and intuitive concept of omnipotence, omniscience, and omnibenevolence. Here, I will only extremely briefly sketch the possible resolution for each paradox.

(1) The paradox of omnipotence

At the initial look, omnipotence has to do with what is possible or impossible for an entity with a certain kind of ability or disposition. What one is capable of doing is different from what one has in fact done as the former is counter-factual whereas the latter is factual. It is a factual matter that the matches that have been scratched so far have lighted, but a counter-factual matter that the unscratched matches have the dispositions to light or will light up once scratched simply because these scratching incidents have not yet happened. As Nelson Goodman contends in Fact, Fiction, and Forecast (1955), the problem of counterfactuals is one of many variations of the problem of projection: the old and new problem of induction, the problem of possibles, the problem of disposition, etc. Contemporary logic and metaphysics have been successful in developing modal logic and possible world semantics which could be used to formalize concepts of possibility, disposition, causation, counterfactual, etc. Thus, if the problem of omnipotence is the problem of what is possible for an entity with a certain kind of ability, then it is reducible to the problem of possibilities, in which case modal logic could be employed to formalize the notion of omnipotence. In fact, I think that this approach has already been amply debated and discussed among metaphysicians such as Alvin Plantinga. I am no expert on this topic and not too literate in the actual debates, but it is at least clear to me that there is a theoretical possibility to rigorously define omnipotence in such a way that the paradox of omnipotence may be resolved with relative ease.

(2) The Epicurean problem of evil

The problem of evil is intertwined with the problem of freewill: if God is omniscient, then everything is already known which implies determinism; consequently, freewill is incompatible with omniscience. In turn, the problem of freewill is intertwined with the problem of moral accountability: if God is omniscient and, therefore, determinism is true so that freewill is impossible, then no one is responsible for their actions. (Of course, one can reformulate the problem of freewill and of moral accountability in secular versions by replacing God with causation.) There have been attempts by compatibilists such as soft-determinists to make room for freewill (and moral accountability) despite (causal) determinism. However, these attempts usually redefine freewill in such a way that much of our original concept of freewill that gave rise to the problems is deflated or dismissed as unintuitive. Likewise, this deflation is not done by formalizing the concept of freewill, but by appealing to ordinary language or commonsensical notion of freewill. Such compatibilist arguments mimic arguments against skepticism in that both arguments try to turn the table around and shift the burden of proof to the concepts and intuitions from which the problems arise. For instance, skepticism arises because we associate knowledge with epistemic infallibility, and anti-skeptics argue that this association is unintuitive. However, approached this way, the debates on skepticism quickly meet a dead-end wherein the opposite sides beg the question to each other. In my view, the stalemate is anticipated because both sides appeal to commonsense in justifying their own perspectives as it often comes down to the competition of which view is more in tune with the hypothetically postulated sufficiently intelligent, but philosophically untrained ordinary person on the street. As David Lewis remarks in On the Plurality of Worlds (1986), although commonsense may be the initial starting point, it by no means has the absolute authority. Rather, from time to time, commonsense requires revisions and improvements by theorists within certain theoretical and pragmatic constraints such as conservatism, simplicity, etc., for otherwise no intellectual progress is possible. The same goes for both the concept of knowledge and of freewill (or moral accountability). In fact, the advent of the need to revisit the concept of freewill and moral accountability has arrived as new AI-based technologies blur the traditional line of moral agents and physical objects. AI-based technologies are definitely not commonsensical. Thus, all the more reason to engage in theoretical analysis of the concepts. In Freedom and Resentment (1963), Peter Strawson proposed a novel approach to the problem of freewill and moral accountability, inspiring the response-dependence theory of morality. This new approach shifted the metaphysical problem of freewill and moral accountability to the meta-ethical problem of personhood and normativity, focusing on the question of what it is to treat an individual as being subject to rules and what it is for a regular pattern (of social behaviors) to be normative—rather than on what are the necessary and sufficient conditions for freewill and moral accountability. In my view, if the problem of evil is essentially about what it is for something to be (judged as) good or evil and to what extent intelligent organisms such as humans have choices in response to such matter, it is reducible to the problem of freewill and moral accountability. If so, the problem is ultimately reducible to the problem of personhood and normativity, in which case the developments in contemporary meta-ethics could be employed to formalize the concepts of omnipotence, omniscience, omnibenevolence, and lastly evil.

The overall lesson is that it is too early to dismiss and give up the attempt to resolve the paradox of omnipotence and the Epicurean problem of evil. Surely, there is no guarantee that the paradoxes will be resolved or, at least, resolved in the way that is expected. Yet, we have sufficient inductive grounds from the historical progresses in logic, mathematics, and science to be optimistic about the possibility that our intuitive concepts may be revised or improved by formalization processes. For all we know, by dismissing the paradoxes, we may in fact be settling down in wilderness instead of charging into the Land of Promise. To confess that we do not know is one thing, and to make further claims about God (such as whether God exists) based on our ignorance is another thing. The former is Socratic apology, and the latter is a non sequitur. If God is truly beyond our understanding, we must rather keep quiet and abstain from making any gesture towards any direction when asked, “Where is God?” If quietism is the most appropriate response, then a true believer in God is paradoxically a full-blown agnostic.

“Whereof one cannot speak, thereof one must be silent.” - Ludwig Wittgenstein, Tractatus Logico-Philosophicus (1921)



[1] Evil exists. If God is omnipotent and omniscient, then he does not want to eliminate evil (not omnibenevolent). If God is omnipotent and omnibenevolent, then he does not know that evil exists (not omniscient). If God is omniscient and omnibenevolent, then he cannot eliminate evil (not omnipotent). If God is omnipotent, omniscient, and omnibenevolent, then evil can be eliminated. If so, evil should not exist.

[2] Without a resolution to the paradox, it is as if, in holding concepts such as of omnipotence, one believes in two contradicting propositions. If self-contradiction is how faith operates, then faith is at best a tic, not a statement of creeds. Creeds are propositions and, therefore, have inferential relations. Inferential relations are logically structured. Self-contradiction is illogical.

Comments

Popular posts from this blog

On the Meaning of Crucifixion

Market-Values and Responsible Consumption