A Conversation on Theodicy

A Conversation on Theodicy

Print Friendly, PDF & Email

Below the author engages himself in a conversation on theodicy—the conundrum of how it can be that (a) Evil exists, and yet (b) God is beneficent, and (c) God is omnipotent. 1


What is it about theodicy that is interesting? Isn’t it a rather old-fashioned subject?

It is old-fashioned, but I’ve never been new-fashioned. What interests me about it is that it tries to hold together the moral, the rational, and the transcendent. That’s an attractive combination.

How did you get to be interested in it? It’s not exactly a burning question of today or a subject one reads about in op-ed columns.

Some time ago, I was trying to understand, with both the acceptance and the criticism that goes with understanding, why it is that what is good oft goes awry. I had two different “goods” specifically in mind. One was a good in my life that wasn’t good enough, and the other was something in the world that aspired to good but produced evil.

You speak of this as if this effort to understand was in the past, and no longer continues. Is this so?

Yes. I arrived at a solution that seemed roughly satisfactory for my own purposes, but when I looked at what this solution was, it struck me that it was a theodicy, so I got interested in theodicy, and started thinking and learning about it.

Are you trying to develop a theodicy, or do you actually think that you have one?

First I found myself surprised that this was what I was doing, surprised in the silly way that the person who discovered that he was speaking “prose” was surprised. Then I started to enjoy the fact that my thoughts brought me into relation with the ideas of so many great philosophers: Plato, Kant, Hegel, and so on. But now I’ve gotten used to the exhilaration of this company, and the question occurs to me: do these thoughts I have actually cohere and do they amount to a real theodicy?

But do you really have the whole religious world view that theodicy presumes, and do you feel the need to “defend God”?

It would be a “secular theodicy,” a systems-theoretic theodicy. I arrived at that phrase by myself, but have since realized that it’s an obvious formulation, and that many have sought after such a goal. As a secular theodicy it wouldn’t concern the justification of God.

So you are not trying to find a way out of the classic conundrum of how Evil can exist if God is beneficent and omnipotent? If that isn’t the question, what is it that a secular theodicy is trying to achieve?

No, that conundrum doesn’t feel compelling to me. I’m also uncomfortable speaking about “God”—at least in this context. I’m rather interested in (a) understanding the existence of evil, and (b) finding some stance in relation to it.

No conundrum there, but I suppose that what you’re interested in is close enough. But that word “evil,” do you mean by it wrongful, intentionally harmful, human action?

No. I mean “evil” in the pre-modern sense that included not only bad human action but natural disaster, death, and the like.

Don’t you want to distinguish between moral evil and natural evil? Or do you think you can really get away with not doing this? And besides, you’re a Jew; isn’t the Jewish imperative le-havdil, to make a distinction? How can you start off refusing to make this distinction?

I guess I must have a pre-modern sensibility; this was how it occurred to me when I first started thinking about all this. I suppose the reality and inevitability of death struck me as an evil (I’m aware that to some people it doesn’t), and I realized that I was far from being able to accept the fact of it just because it’s natural.

But what about le-havdil; what kind of Jew would deny the distinction between moral evil and natural evil?

I don’t deny the distinction; I just put it in what I feel is its place. To me one can always talk about things “under the aspect of similarity” and “under the aspect of difference.” Under the aspect of difference, there’s an important distinction between moral and natural evil, but under the aspect of similarity they’re not different. A Talmudist would say haynu-hach: this is the same as that.

Sounds to me like you’re trying to have your cake and eat it, too.

It’s Buddhist logic, which is definitely non-Aristotelian. There’s the assertion of A. There’s the denial of A. And there’s the refusal to exclude the middle in the assertion of both A and not-A. That’s why they have such round tummies; they’re always eating their cake and having it, too. There’s also a fourth possibility of neither A nor not-A; that’s for Kant, Wittgenstein, and other lovers of inexpressibility.

You’re avoiding the issue. Just what is the sameness in moral evil and natural evil? Obviously the use of the word “evil” in both phrases isn’t an acceptable answer, since this is what is being called into question.

Moral evil and natural evil have their common ground in “metaphysical evil”: finitude, incompleteness, limitation, the constraints and flaws in existence itself.

Ah, this is Leibniz. So both moral evil and natural evil have their root in some basic evil that afflicts all of existence. Is this the best of all possible worlds, then? Is that your theodicy? That would be hard for me to believe, and anyway, it’s been done.

I agree with Leibniz about metaphysical evil as the ground of both moral and natural evil, but I don’t agree that this is the best of all possible worlds. This world is obviously not the best of all possible worlds. At this time in history I can’t imagine anyone thinking that it was.

I see three problems here. First, this would not do as a defense of God. I realize that this isn’t your purpose, but still you’re taking rather extreme liberties with Leibniz. Second, if evil is metaphysical, inherent in the fabric of existence, why couldn’t this be the best of all possible worlds? Third, if this isn’t the best of all possible worlds, how would such a theodicy fulfill the function of consolation, or is your secular theodicy going to abandon this purpose also?

Let me address the first and the third of these, which are related. That is, defending God might be considered an “objective” function of theodicy, while offering consolation is a “subjective” function. A successful defense would offer consolation, and effective consolation would contribute to a defense. My answer to these two questions lies in the differences between the noun, perfection; the verb, perfecting; and the adjective, perfectible. The world is not perfect, and is not even as perfect as it could be, given the metaphysical constraints that have to be satisfied. But perfecting is still possible. The world is perfectible. That’s a defense of God, and also a consolation.

So, you are saying that God made a perfectible world, but not a perfect one? Oh, sorry, I forgot that you said that you’re uncomfortable speaking about God. But can you answer this question anyway? Why should God make a perfectible world, and not a perfect one?

It’s always seemed to me absurd to talk about ‘God’ in an ordinary way, as if God was some guy who lived down the block, but I guess one has to accept reason taking on issues that transcend reason. About the substantive question, a perfect world is static. Nothing happens in it. There is no point in us (or anything) being in this world. So it isn’t even perfect. Theodicians must surely have pointed this out. The very notion of a perfect world is self-contradictory. And, given God as creator, it’s also redundant. Only God is perfect. If the world were perfect also, it would be the same as God, so we’d have two Gods. Absurd. But a perfectible world is coherent. Moreover, what we really want to be living in is a perfectible world, not a perfect world. There’s nothing to do in a perfect world. We’d be bored out of our minds. And God would be bored, too.

Oh, I don’t know. I might be able to get into the permanent absolute bliss that would follow from being in a perfect world. Isn’t that our idea of Heaven?

But wouldn’t you rather be in bliss half the time, and spend the other half experiencing the perfectibility of your golf game?

I’m not sure, but I don’t think this line of argument is productive. I think the point you made that the idea of a perfect world is incoherent is a better one. I can’t get rid of the feeling, though, that perfectibility is second best to perfection. But let’s leave it. We still have the second question from before: if flaws are inherent in existence, couldn’t this be the best of all possible worlds? Also, what exactly is the relationship between your position that there is metaphysical imperfection and your position that the world is perfectible? Is it perfectible because it is metaphysically imperfect, or despite this fact, or what?

Given metaphysical imperfection, our imperfect world could be the best of all possible worlds. I can’t prove that it isn’t. I just prefer to believe—and also find it much easier to believe, given the facts of ordinary experience—that the world is perfectible. By perfectible, by the way, I don’t mean “made perfect”; I mean just improvable. Perfectibility doesn’t follow from metaphysical imperfection, but it’s compatible with it.

I still don’t see it. If the world is inherently imperfect, then it is imperfect “by law”, so how can it also be perfectible? Wouldn’t the metaphysical law preclude the removal of the imperfection? I don’t see how the two are compatible at all.

The critical issue here involves the difference between the general and the specific. The metaphysical imperfection would represent a general problem, but general metaphysical problems can still have specific solutions, that are solutions only locally or temporarily. Since these solutions don’t solve the problem in general, they don’t contradict the lawfulness of the general problem. For example, one general problem is the need in all systems of both constraint and variety or order and disorder, yet the two are opposites. In the structure of DNA, for example, the two are reconciled in the fact that the DNA backbone is completely orderly and regular, while the sequence of the bases in the middle of the DNA that carry its information is completely disorderly, i.e., effectively random. But this solution to the lawful tension between order and disorder works well in this particular case, because of all kinds of other specific things that are true inside cells. It isn’t a general solution to the tension between order and disorder.

Wait. Before going into the details, I want to note that you are using a bunch of words roughly synonymously, and I want to understand what the implications are of this. You first spoke of “evil,” and then the word “imperfection” came up, and you’re also speaking about “problems.” Am I to assume that these are all the same? That’s one question. Another question is: evil, imperfection, problems, for whom? There seems to be an interested party lurking behind this discussion, despite the fact that in calling the ground of this evil (or whatever) “metaphysical,” you seem to be implying that there is something objective about it. Is what is evil or imperfect an objective fact or a subjective one, relative to this interested party?

Good points. First, I am using these three words synonymously. They’re useful because they offer different degrees of intensity. I’m not committed to the word “evil.” I don’t need all its connotations, and would be content to talk about imperfections or problems. Second, you are right to point to a veiled presence of an interested party. Here I can imagine taking either a “strong” or a “weak” position about the implicitness of such a party. In the weak position, I would accept that there is an interested party—this would be life, or more specifically organisms that are alive—so evil, imperfections, problems would be relative to the interests of life in general or of specific living organisms. There’s a certain appropriateness to this position, in that I will argue that life is assigned the task of doing the perfecting that I’ve spoken of above. So it’s life that has problems, and life that is charged with their solutions. Quite balanced. But I don’t want to completely preclude the strong position that might argue, for example, that even if there were no life in the solar system, the fact that planetary orbits are not perfect ellipses, but are probably chaotic (so the solar system might ultimately be unstable) is an example of “imperfection.” The strong position might require viewing the continued existence of even entities that aren’t alive as an abstract good, or might implicitly invoke the mind of God.

OK, let me see if I can sum up some of the recent additions to your argument. There are general problems, imperfections, evils, that are such because of their effect on life. These are the metaphysical evils you spoke of earlier. There are no general solutions to such problems, in that they reflect an inherent imperfection in the world, again from the vantage point of life (at least in your weak position). But these problems can be solved by life locally and temporarily, so the world is perfectible. Is that it? Sounds like you’re talking only about “natural evil”; where does “moral evil” fit in?

Yes, that’s basically it. And you’re right that this sounds more like natural evil than moral evil. In fact, I would subsume moral evil within natural evil. The idea of putting them side-by-side seems to me to be wrong. Human beings and social systems are part of the natural world. I’m a Darwinist here, and I think everyone should be. Not that Darwin had all of the truth, but that the truth that Darwin brought us must be incorporated in every philosophy. Moral good and evil are affected by our capacities and dispositions, which are part of the natural world. So in my view, moral evil is a “special case” of natural evil. It’s just one aspect of natural evil that is distinctive to us, but it isn’t a separate category outside of natural evil.

Then what about freedom? If our actions are viewed as totally within the natural order, and if moral evil is totally subsumed within natural evil, is there any place for free will? And without free will, is there any responsibility?

I would argue that saying that will is more free or less free, on a continuum, is more accurate and more useful than saying that there either is free will or there isn’t. Putting it on a continuum also connects it to the natural order. It’s probably an emergent of mind, just as mind is an emergent from life. So even if moral evil depends on this free will, I would still place it in the natural world. All this is under the aspect of sameness; under the aspect of difference, quantitative differences (in freedom) seem to result in qualitative differences, as Hegel might say, so I don’t disdain the distinction between natural and moral evil. We are responsible for moral evil, but we are not responsible for natural evil.

Ah, you’ve just introduced a new idea: responsibility. So even though, from one perspective, moral evil is just a manifestation of natural evil, from another point of view they’re different, especially in the fact that we human beings are responsible only for moral evil. But this could have different meanings. It could mean that we’re responsible in the sense of (a) causing, i.e., we don’t cause evils that aren’t moral, or in the sense of (b) being obligated to fix, i.e., we’re only obligated to fix moral evils.

I mean (a): we cause moral evils but don’t cause natural evils that are not moral evils (e.g., earthquakes), but we are responsible for all natural evils, in the sense of having the obligation to fix them if we can. It is the function of life, or at least a function of life, and in particular human life, to “repair the world,” in Hebrew, tikkun olam. But our responsibility to fix moral evils is “prior,” more urgent, more immediate.

A while back you took the position that the world is perfectible. Now you’re placing on life, and particularly human life, the obligation of doing the job of perfecting. So, at least one thing is clear: your theodicy isn’t subject to the criticism often made of theodicies, that they encourage quietism. Quite the opposite: not only does your theodicy declare that the world is imperfect, but it asks us to get busy in trying to perfect it.

Yes. It isn’t quietist at all. The key is replacing perfection with perfectibility.

But to get back to the original theological type of theodicy, doesn’t this let God off the hook completely? God created a world that is flawed metaphysically and it’s our job to fix it? Isn’t this a rather long Sabbath for God? And are we really up to the task?

Creation is a more difficult job than we can understand. Read Job. Consider Yitschak Luria’s Kabbalist account of creation. Even omnipotence cannot change the fact that one plus one is two. Or to be more precise, creation requires laws, and laws imply constraint. So there is no creation without constraint. So to imagine that omnipotence allows both the creation and overruling of constraint is contradictory. (To allow both, for example in accepting the possibility of miracles, is to abandon rational discourse.) Part of the problem with the traditional formulation of theodicy is that the premise of omnipotence, when taken to its limit, is incoherent.

Couldn’t God transcend such limitations of ordinary logic?

God might, but rational discourse can’t, so once we reach such contradictions, we have to stop talking. That is, unless we practice Buddhist logic, which would let us take one more step, to ‘both A and not-A,’ but it too just ends us up in silence, because that is what the fourth level of ‘neither A nor not-A’ amounts to.

I want to get back to this responsibility of life and humanity. Isn’t it much to put the whole task of fixing on life/humanity? Isn’t God or God’s religion(s) supposed to help? Aren’t we fallen?

It is a lot, you are right, but our obligation is limited by what is possible. As Pirke Avot says, “It is not for us to complete the task, but we have no right to abstain from trying.” Also, the division of labor between us and God is not inappropriate; and God helps. These two ideas are captured in Rosenzweig’s wonderful interpretation of the Star of David. We have two interlocking triangles: in one, the terms (points of the triangle) are God, Humanity (or perhaps Life), and World. In the second, the terms are Creation, Revelation, and Redemption. God in relation to World is Creation (transcendence); God in relation to Humanity is Revelation (immanence); Humanity in relation to World is Redemption. So, fixing the world is our (Life’s) job, and God helps through revelation. We are fallen, but not uniquely so; our fallenness is just metaphysical imperfection as it applies to us.

So God is off the hook. A perfectible world is a world that is coherent, and may even be even better, oddly enough, than a perfect world. Life, especially human life, has meaning in such a cosmos. But I notice that you speak of perfectibility and responsibility, but don’t speak about the actual prospects of success, the likelihood of fixing. The literature of theodicy has always been about the grounds for optimism vs. pessimism. Perfectibility sounds optimistic, but there is also pessimism in its implication that things are never as good as they might be, so this suggests two questions: (1) Will attempts at perfecting succeed? (2) And is there any rest from this task? It has a rather arduous feel to it. Isn’t it like the labors of Sisyphus, always to be perfecting, never to reach perfection?

Hegel thought that perfecting is guaranteed at least on a large scale, but this doesn’t seem plausible to me. I think there must be genuine hazard, and conversely, also genuine opportunity. But it’s hard to take this seriously. It’s hard to imagine that the destruction of life on this planet through nuclear war was a real possibility, and maybe still is. But many horrific things that are or would have been impossible to imagine have actually happened—it isn’t necessary to name them—so the idea that a good outcome is inevitable or even likely is impossible for me to hold, though I would certainly love to believe that some divine thumb is biasing the scales in the direction of a good outcome. Whether this is optimistic or pessimistic isn’t clear to me—it probably depends on temperament—but it ought to be sobering. And as to rest, we are granted moments of being, in love or beauty or simple presence or by other means; such moments and remaining in some connection with these moments are a rest, and they are also probably necessary for successful fixing.

I think you are trying to have your cake and eat it too about God. On the one hand you say you’re uncomfortable to use the word and everyone should at least be a Darwinist and now you also say that you can’t imagine God’s thumb on the scale helping us out, but on the other hand you kind of snuck in that word “revelation” before in talking about Rosenzweig. Redemption is a religious word, too, but secular equivalents are pretty obvious, with Marx showing us one such version. Creation presumably refers to the origins of the universe, and that’s shrouded in mystery, so I won’t object too strenuously to the word. But revelation? Are you talking about Sinai, or the incarnation of Christ, or angelic dictation to Mohammed, or the enlightenment of the Buddha, or what? Does revelation have anything to do with theodicy, and if so, is it still a secular theodicy?

Well, my main reason for invoking Rosenzweig was his notion of Redemption, which links Humanity and World, and is the same as tikkun olam, the point being made in this link that the obligation of perfecting is our job, not God’s. So far as this argument goes, the presence of Revelation in the symbol is not necessary relevant, but it does bear on the perfectibility of the world. And I’m actually not invoking Sinai, or the other examples you gave. I don’t mean Revelation as in “revealed religion.” What I mean by it (and I don’t actually know what Rosenzweig himself meant by it) is something more like those moments of being I spoke of before. But you know, now that this is on the table, maybe it is relevant in another way. If Creation leaves a world precariously balanced between hazard and opportunity, and if evil is given its Augustinian sense of privation of being, then maybe Revelation, interpreted as the experience and strengthening of being, is a divine thumb on the scale helping us out. Whether this is still “secular” depends on what the word means. Would it include the increase of being that can be the result of meditation or comparable practice?

Well, this kind of strengthening of being could be viewed as a natural process, so I guess “revelation” of this sort could qualify as secular. But why are you continually stepping in and out of religious language. If you say you are interested in a “secular theodicy,” why not restrict yourself to secular language?

If I were a novelist or poet, I might be able to, but I’m not, so religious talk is a way of trying to transcend the limitations of ordinary language, and the even greater limitations of scientific language. But there is an argument for religious talk. For me, and for many other people, it has great resonance and deep associations. Anyway, there isn’t a secular term that corresponds to theodicy. Why not? Because to a secular mind, the question of theodicy does not arise. Natural evil is separated from moral evil; eliminating natural evil is a task for technology, and eliminating moral evil is a task for psychology or politics.

Sounds reasonable to me. What’s wrong with that position?

It’s shallow and ineffective and also fragmenting and unscientific. It’s shallow because, for example, if you examine social problems that have the character of a prisoner’s dilemma (legislative log-rolling, arms races, the destructive exploitation of commons, etc.), to say that these are moral problems misses a critical aspect of their essence, and not understanding this aspect often results in ineffective solutions. It is fragmenting because it sharply separates the natural world from the world of human action, and to posit such an absolute separation is unscientific. In joining natural and moral evils together, religious and metaphysical thought reflected a position that is more scientific than the prevailing secular separation of the two.

You’re saying that religious thought is more scientific than modern secularism?! Will you argue next for the continued relevance of the conundrum of (a) Evil exists, (b) God is benevolent, and (c) God is omnipotent? Or, do you have anything to offer in its place?

Well, there’s an interesting contradiction in modernity. Modernization is differentiation, and the distinction between natural evil and moral evil is an aspect of this differentiation, but modernity also brings with it the hegemony of science, and science ultimately presupposes and strives for integration. Very dialectical. As for the relevance of the classic conundrum, if it is literally interpreted, it is truly obsolete. There is a possible replacement for it, but it is not a conundrum. It is (a) hazard and opportunity are inherent in the structure of the universe, (b) the universe is perfectible, and (c) perfecting is the task and meaning of life, and in particular human life.

I note that you left out that bit about assistance, and the providence of the divine thumb.

I accept your addition.


1 This dialog was inspired by Susan Neiman’s Evil in Modern Thought: An Alternative History of Philosophy, Princeton University Press, 2002. The idea of a systems-theoretic ‘secular theodicy’ can be found in expanded form in the author’s in-depth works:

“Towards an Ontology of Problems.” Advances in Systems Science and Applications (1995) Inauguration Issue I, pp. 37-42. http://www.sysc.pdx.edu/download/papers/mz_assa_met.pdf

“Understanding Imperfection.” Plenary talk given at the World Congress of the Systems Sciences & 44th Annual Meeting of the International Society for the Systems Sciences, Toronto, July 16-22, 2000. http://www.sysc.pdx.edu/download/papers/isss_plenary.pdf

“Incompleteness, Negation, and Hazard: On the Precariousness of Systems.” Nature and System, Vol.6. (1984). pp. 33-42. http://www.sysc.pdx.edu/download/papers/incomplete.pdf