Research Article

If God Didn’t Satisfice, We Could Still Exist

Rick Repetti

Department of History, Philosophy & Political Science, Kingsborough Community College, 2001 Oriental Blvd., D-223, Brooklyn, NY 11235, 718-368-5226, USA.

Editor | Gregg D. Caruso, Corning Community College, SUNY (USA).

Received | November 14, 2015; Accepted | Febuary 13, 2016; Published | March 06, 2016

Correspondence | Rick Repetti, Kingsborough Community College, 2001 Oriental Blvd., Brooklyn, USA; Email: rrepetti@kingsborough.edu

Citation | Repetti, R. (2015). If god didn’t satisfice, we could still exist. Science, Religion and Culture, 2(4): 115-124.

DOI | http://dx.doi.org/10.17582/journal.src/2015/2.4.115.124

Adams’s interesting argument – that God need not create the best, that in creating us lesser beings God was gracious, and that we should be grateful for the undeserved gift of life – is standardly cited as a starting point in many arguments for or against satisficing theodicies (see, e.g., Grover 2004, Kraay 2010, 2011, Langtry 2008, Rowe 2004, Tucker 2015, van Inwagen 2015). Weinstock (1975), for example, objected to the grace component of Adams’s satisficing theodicy. Weinstock argued, among other objections, that even if we assume grace is a virtue and its manifestation in a world w would make w superior to a world w* that lacked grace, God still could create a better world x (with better beings) with greater grace simply by enabling those better beings to benefit more than the lesser beings in w; if that makes sense, then Adams’s assumption – that the satisficing involved in creating lesser beings is justified by the presence of grace in a world – would be false. It would not be what Tucker calls “motivated suboptimization” (2015), by which he means satisficing that is justified because it maximizes from among sets of alternatives the optimal outcomes available under those constraints. Wielenberg (2004) argues along similar lines as Tucker, though not explicitly, in noting that the presence of grace in a world alone is insufficient to justify submaximizing because, even if Adams is right in (a), that God wrongs no non-existing beings in failing to create them, in failing to create optimal beings in optimal worlds, God manifests or exhibits imperfect motivational dispositions in not optimizing beings/worlds. Intuitively, if the presence of God’s grace in a world w would make w superior to a world w* that lacked God’s grace, then the absence of God’s perfect motivational dispositions in a world w would make w inferior to a world w* that exhibited God’s perfect motivational dispositions. (For other criticisms of Adams’s argument, see, e.g., Coughlan 1987, Grover 2003.).

Notwithstanding the interesting character of these objections, I do not wish to vet them here, as I think they may be analyzed as more relevant to (a) above, the idea that God wrongs no non-existing beings in failing to create them, than they are to (b) above, the claim that we would not exist if those better beings were created. I think (b) has an implicit existential, psychological, or emotional appeal – however difficult it may be to characterize – that survives these objections, at least for those of us who prefer that we exist than that God should have created other better beings in better worlds, with greater grace, which beings would not be us. More importantly, the objections above, of Weinstock, Wielenberg, and others, do not get to the unnoticed flaw in the heart of Adams’s argument connected with (b): the inconsistency of (b), the non-identity premise in his theodicy, with the identity premise in his broader theology. Though it may not worry those lacking commitments to the tenets in the broader theology, if the identity thesis is more plausible than the non-identity thesis within the (Judeo-Christian) theology that informs Adams’s theodicy, then Adams’s theodicy collapses for those for whom it was formulated – likely the only ones who might have been moved by it.

Adams’s satisficing theodicy is a response to the problem of evil within Judeo-Christian theology. The problem of evil consists in the recognition that there is a prima facie tension between:

1. the Judeo-Christian idea that God is all-loving and all-powerful, and
2. the obvious fact of the existence of massive evil and gratuitous suffering.

Hick’s classic theodicy is that we are better off as real persons with free will in a non-cartoon-like world with real consequences, such that our choices matter, than we would be otherwise as automata in a cartoon-like world, and evil and imperfection are necessary corollaries of real persons in a real world, engaged in a soteriological process or “soul-making” (Hick 1973). Hick acknowledges, however, that while his theodicy shows consistency between God and evil, it doesn’t justify the extent and magnitude of apparently gratuitous suffering; thus, it fails to render plausible the idea that they are perfectly compatible. For this, Hick admits one must hope that the soteriological process continues into the afterlife (Hick 1973, 102-3), something Adams also appeals to (1979, 56), but this appeal begs the question in the context of the problem of evil, and is perhaps even less plausible than the flawed general theodicy it attempts to patch (Grover 2003, 161; Grover 2004, 113-14).

Adams’s Theodicy

Adams’s theodicy might be thought to attempt to go beyond the mere consistency achieved by Hick’s theodicy, rendering it more coherent and, presumably, plausible, by situating it within a more comprehensive explanatory framework within which created beings, instead of complaining about the extent and magnitude of their suffering, are led to appreciate their lives as undeserved gifts of a gracious, loving creator. Intuitively, while God may have been under some moral constraints in choosing beings and worlds (Kraay 2010), e.g., perhaps He may only engage in motivated submaximization (Tucker 2015), we cannot help but value our lives, and thus our very existence evidences the value of satisficing, to the extent we deem that our lives are overall worth living. Adams appeals to Leibniz to support the idea that our lives are worth living, and if anything about the evils in the past that led to us changed, we would not exist, which implicitly supports a satisficing theodicy, but he obviously does not entirely agree with Leibniz’s views, since Adams’s main point is that this need not be the best possible world, contra Leibniz (Adams 1979, 54-5).

Whereas Hick and others propose theodicies centered on free will within the sort of consequences-involving world required for spiritual development, or attempt to render coherent the idea that a morally unsurpassable being could submaximize a world (cf. Rowe 1994) or do so with motivation (Tucker 2015), or create all minimally acceptable worlds in one best possible multiverse (Kraay 2010), among similar approaches, a dialectical virtue of Adams’s theodicy is that he focuses on another reason to think God need not have created us as better or perfect beings in a better world: we have no legitimate claim on Him that He should have done so, for if God were obliged to create better beings, we would not be those beings. Implicitly, to press the claim, from Adams’s perspective, is to argue against our own existence! (There is a self-stultifying element, however inchoate, in that realization, which Adams (1979, 54) attributes to Leibniz, who was much more explicit about it.)

Adams sidesteps the more pressing dimension of the problem of evil, namely, imagining a plausible, coherent justification for gratuitous suffering. In that regard, Adams’s grace argument functions, however unwittingly, as a distraction, if not a red herring. Wielenberg (2004) implies as much when he argues that the presence of grace does not necessarily compensate for or otherwise discount the imperfection of satisficing. Adams, instead of directly challenging the alleged inconsistency between a morally unsurpassable God and evil/imperfection, which is what Hick and many other theodicists do, indirectly does so by targeting a related idea:

(P) If a perfectly good moral agent created any world at all, it would have to be the very best world that he could create. (Adams 1972, 317)

P may be credited with generating a turn in theodicy literature that vets whether there can only be one best possible world, infinitely many unsurpassable worlds, or infinitely many surpassable worlds, which are a priori or logical questions, and which of these, if any, characterizes the actual world, which is at least partly an a posteriori or empirical question (Rowe 1979). (For discussions of these ‘worlds’ issues, see Grover 2004, Kraay 2010, 2011, Langtry 2008, Rowe 2004, Tucker 2015, van Inwagen 2015, Wielenberg 2004, and Weinstock 1975.)

To situate Adams’s critique of P within the context of his entire argument, let us sketch an abstract version of the argument, as follows:

1. If there were perfect worlds with perfect beings, we would not exist in those worlds because we are not perfect.
2. If God were morally obliged to create those worlds, we wouldn’t exist.
3. By creating this imperfect world, imperfect beings like us become possible.
4. By creating this imperfect world, God graced us with the undeserved gift of life.

Let us now consider how Adams seeks to undermine P. Adams thinks God could create a world in which:

1. None of the individual creatures in it would exist in the best of all possible worlds.
2. None of the creatures in it has a life that is so miserable on the whole that it would have been better for that creature if it had never existed.
3. Every individual creature in the world is at least as happy on the whole as it would have been in any other possible world in which it could have existed. (320)

   Call (1)-(3) Adams’s “minimally-morally-acceptable-world” (MMAW) criteria. Adams adduces two criteria to address whether an individual being satisfies MMAW-type criteria:

4. The creature is not, on the whole, so miserable that it would have been better off for him if he had never existed.
5. No being who came into existence in better or happier circumstances would have been the same individual as the creature in question. (327)

Call (4) and (5) Adams’s “minimally-morally-acceptable-created-being” (MMACB) criteria, implied by (2) and (3), respectively.

The implicit idea in making MMAW criteria abstract is that if a world relevantly similar to ours satisfies MMAW criteria, God may satisfice in general, though perhaps not specifically in the actual world: it is presumably enough for an abstract satisficing theodicy if some world roughly similar to ours satisfies MMAW criteria. This abstraction is useful to separate two issues: whether a satisficing theodicy model case is coherent per se, which is a version of the a priori question, and whether our actual world – with its peculiar array of egregious cases of gratuitous suffering – sufficiently approximates such a model world, which is the a posteriori question. Whether our world matches up to such a possible world can be set aside, arguendo, while attempting to answer the a priori question, but only provisionally, so long as our world conceivably falls within the range of nearby possible worlds that satisfy these criteria, or within the same “cluster” of worlds, as Kraay (2011) puts it, for if the distance between the actual world and such possible worlds is great, then the issue of such possible worlds cannot be used in support of a theodicy for the actual world. And it seems intuitive that this is a surpassable world.

One may thus reasonably resist the idea that, because a model-case world or world-cluster – in which a satisficing theodicy is coherent – significantly resembles the actual world, the satisficing theodicy is cogent in the actual world. However, the main objection I develop here does not rely on resisting this model-case-to-actual-world move. Adams could admit that our world is not a model-case world without contradicting his conclusion that P is false:

(P) If a perfectly good moral agent created any world at all, it would have to be the very best world that he could create. (Adams 1972, 317)

If any world satisfies MMAW criteria, regardless of how imperfect it may be, P is false. But P would be false even if the penultimate highest heaven world was not as perfect as the ultimate highest heaven world, but was reasonably construed to result from God’s motivated submaximization, to borrow Tucker’s (2015) idea. If so, however, then showing P to be false is facile, and proves little that may carry over into a plausible satisficing theodicy for this world. Thus, there is a potentially huge gulf between a world that could falsify P and significantly lower worlds that satisfy the MMAW criteria. To the extent our world satisfies MMAW criteria, Adams’s argument applies to it, but it is not at all obvious that our world does. That is, mere consistency between God and evil/imperfection in one very remote possible suboptimal heaven world – though arguably enough to show P is false – is not enough to qualify the extent and magnitude of evil and suffering in our world as justified satisficing. Thus, Adams’s argument would fare better if some model-case world or worlds that his MMAW criteria describe turn out to be closer to ours rather than remotely like it, and if the MMACB criteria are true of most if not all beings in our world. That is a very tall order, but even if our world and its beings satisfy these criteria, my objection to Adams’s non-identity thesis undermines his argument.

To sketch how a world similar to our own and beings similar to us might satisfy MMAW and MMACB criteria, which we may jointly call “Adams’s satisficing criteria,” Adams adduces three cases that function implicitly as analogies with our own case, Cases A, B, and C. In Case A, a mother intentionally genetically alters her fetus, producing a handicapped child she avowedly would not bring into existence without the handicapping genetic transformation. Case B involves a breeder capable of breeding superior beings, say, puppies, who only breeds goldfish. And Case C involves parents with access to genetic engineering that are not under an obligation to use it on their offspring.

Case A: the intentionally handicapped child

In Case A (Adams 1972, 326-29), we suppose parents who so love their adopted handicapped children that they want one of their own, and will not have a non-handicapped child, so they genetically alter their fetus. Adams argues that this child either would not have existed, or would exist handicapped. A non-handicapped alternate would not be created, but would be more like a sibling anyway, having different genes. As long as the handicapped child’s life is worth living, the parents did not wrong the child.

Bringing this child into existence seems consistent with Adams’s satisficing criteria, since it is plausible that:

(1*) This child would not exist in the best of all possible worlds

(2*) This child does not have a life that is so miserable on the whole that it would have been better for it if it had never existed

(3*) This child is at least as happy on the whole as it would have been in any other possible world in which it could have existed

These three statements are crafted as instantiations of the MMAW criteria. Clearly, (2*) is equivalent to (4), and (3*) is related to (5), which this child also arguably satisfies. We can imagine that (2*) or (3*) may be false of a particular child, but all that is needed for the argument is that (1*)-(3*) and (5) be true of some such child. Adams thinks it is intuitive that they are true of this child. If so, Case A would support Adams’s conclusion that P is false.

Nonetheless, Case A arguably conjures the opposite intuition: by analogy, God has intentionally handicapped us, and since what the mother has done in Case A is immoral, what God has done in handicapping us might be. Thus, Adams asks whether what’s wrong with the parents in Case A is that they violate the following principle (328):

(Q) It is wrong to bring into existence, knowingly, a being less excellent than one could have brought into existence. (Adams 1972, 329)

Adams thinks Q implies P because a world containing imperfect beings would not be a perfect world; P and Q imply each other, for P also implies Q insofar as a perfect world presumably contains only perfect beings. So, if God violates Q, He thereby violates P. But Adams doesn’t think Q is morally correct; given their equivalence relations, if Q is false, so is P. Thus, Adams argues that Case B shows that Q is false.

Case B: the goldfish breeder

Case B (329) involves a breeder who is able to breed superior species, but breeds goldfish, who cannot complain that the breeder did not breed puppies, because they would not be those puppies; nor can non-existing puppies be harmed. Adams considers the question: Does the breeder wrong non-existing pups by not creating them? No, only existing beings can be wronged. A related question is: Can the breeder transform the fish into dogs? Adams’ reply would be that, even if this were possible in some sense, this would produce different creatures. This conclusion accords with his non-identity thesis; Adams’s reasoning about changes in bodies entailing changes in identity is made more explicitly elsewhere (1979, 56), but may be surmised from his examples here.

To make the analogy more intuitive between goldfish and puppies, on one hand, and us and better versions of ourselves, on the other, respectively, let us consider beings like us, say, siblings with greater cognitive endowments. It seems obvious goldfish cannot survive as the same beings if turned into puppies, but we might think we can survive as the same beings if turned into superior versions of ourselves – versions the differences between which are not like those between members of radically different species, but more like the sorts of differences that might obtain between siblings with lesser and greater cognitive endowments. This makes some sense, given how we already have ways of making piecemeal changes to ourselves without altering our identity.

Thus, if my mother got pregnant one month earlier or later than when she got pregnant with me, then that pregnancy would not involve the egg and sperm that produced me, but it would produce instead someone that would count more as a sibling (if I could, per impossibile, be imagined as existing in some abstract sense though the egg that produced me was never inseminated): “My identity is established by my beginning. It has been suggested that no one who was not produced from the same individual egg and sperm cells as I was could have been me” (Adams 1979, 56). Even identical twins, beings from the same zygote with identical genetic material, are not the same person, numerically – not even conjoined twins. So, any deviation from the egg and sperm that produced me (or that specifically divided when it did into my hypothetical identical twin) would produce something more like a sibling, who would not be me. It follows we cannot complain we weren’t made as better beings, because such beings wouldn’t be us. If better beings were required to be created, we would not exist – they would. This conclusion supports the implications of Cases A and B.

If this does not reveal the violation of Q to be the problem with Case A, then what is? To answer this question, Adams turns to Case C (Adams 1972, 329-30).

Case C: genetically engineering our children

The atheist objects that if God exists, He’d have made us, and the world, better. Given God’s attributes, He should have, but His not doing so contradicts His omnibenevolence. Case C involves the parental availability of genetic engineering, which Adams claims nobody is obliged to use. If we fail to use it to enhance our offspring, we clearly violate Q. But thinking carefully about Case C suggests that it’s obviously not immoral to refuse to genetically engineer our children. In fact, it seems to be a prima facie case that it is more obviously morally permissible not to genetically engineer our children than it is morally permissible to genetically engineer them. If this is correct, Adams implies, Q is false. If Q is false, that would seem to vindicate God for not creating us as the best versions of ourselves while still being who we are, by divine means analogous to genetic engineering, since God does violate Q. Cases A, B, and C together show that in creating lesser beings, God need not have wronged those beings, so long as they satisfy MMACB criteria in MMAWs. Cases A-C suggest it is conceivable that a world like ours satisfies these criteria.

But then what does Adams think is wrong with the parents in Case A? They have done the reverse of genetically engineering their child, to be less than it would have been (setting aside Adams’s insistence that the pre- and post-genetically-altered cases are not the same individual, an unnoticed consequence of which is that when they altered fetus1, producing fetus2, they killed fetus1 for the sake of fetus2). Adams claims that what is wrong in Case A is that the parents violate R:

(R) It is wrong for human beings to cause, knowingly and voluntarily, the procreation of an offspring of human parents which is notably deficient, by comparison with normal human beings, in mental or physical capacity. (Adams 1972, 329)

R evades the question of identity in the pre- and post-altering fetus in Case A. Adams notes that R seems ad hoc, but emphasizes that R is consistent with Judeo-Christian ethics. According to that theology, God was gracious in creating us lesser, undeserving beings; this is presumably like the graciousness of parents who do not abort fetuses known to have debilitating congenital defects. Rather than complain about our suboptimal status, we ought to be grateful to receive the undeserved gift of life. Thus, the religious reason Case A is reprehensible is that the parents, violating R, took matters into their own hands and genetically reduced this potential person into one that possibly lacks abilities requisite to enter a two-way relationship with God, violating God’s plan for that being. (On this being God’s plan for us, see Hick 1973, 43-4.) (Again, technically, if Adams is right about the non-identity of fetus1 and fetus2, then the parents killed and sacrificed the first potential person to create a lesser being in its place.)

Some more serious problems with Adams’s argument

Adams’s R-violation account carries no weight with nonbelievers. For the nonbeliever, this case is reprehensible, so it cannot be solely for a reason she cannot recognize. We need not articulate nonreligious moral reasons that might explain what’s wrong with Case A, but Adams must, though he doesn’t. This matters because Adams’s theodicy is supposed to render coherent to believers and nonbelievers alike the idea that a perfect being created our imperfect world; thus, appeal to beliefs that cannot engage both disputants beg the question, and amount to preaching to the choir.

Additionally, the nonbeliever is not given a reason she can recognize as valid for why God – equivalent to the parents – is not violating a principle analogous to R. Thus, R remains ad hoc even if it works within the theology to which Adams appeals to support the other side of the analogy. R is ad hoc because God theoretically sets the standards for what R means regarding deficiencies relative to ‘normal’ human beings when He created us in this world, unlike the parents, since there are no God-independent ‘normal’ standards against which to compare us. Adams needs to adduce a Q-like principle, Q*, that God does not violate, but it is unclear he can.

If the reversal of the analogy (applied to God) cannot be blocked because of the asymmetry and the ad hoc issue, then Case C becomes suspicious because it is unclear it can apply on the other side of the analogy, with God’s not violating some Q* in failing to make us better beings. It makes sense for us to be leery of genetically altering our children to improve them, but God has no such reservations, so God is not immune to Q*, even if we are immune to Q. Thus, showing a case in which Q is false is as facile as adducing a case in which P is false, but therefore does not have the desired satisficing justification. Without the sort of principled grounds on which to block these objections, we can object now that we could have been made better beings, with more resilient bodies, brains, dispositions, etc., and God could figure out a way to do so and preserve our identities; we will make this objection forcefully in the remainder of this paper.

While Adams has not identified what is wrong with Case A from a neutral perspective or offered reasons to block the analogies in Cases B and C from turning back against God, Adams can still insist Case B establishes that were God to create better beings, they would not be us, as puppies cannot be goldfish. God is graceful, so we should be grateful for the undeserved gift of life. For Adams, following Leibniz (Adams 1979), the question is: Are our lives worth living or are we better off not existing at all?

The question is: Who are we?

I think the relevant question is: Were God to make us better beings, would they necessarily not be us? Recall Weinstock’s objection to Adams, that any world in which God created undeserving beings would be one in which God’s grace could manifest, not just satisficed worlds (Weinstock 1975). Thus, God could create perfect beings, arguably, with grace, since: any created beings cannot exert desert claims prior to existence; thus, they would be undeserving; and, thus, by creating them God would manifest grace. Likewise, they may be grateful for their lives. Given that their greater ability to benefit, which follows from their greater perfection, outweighs our lesser ability to benefit, it seems God should have created more perfect rather than less perfect beings. Additionally, a God who brings into existence the greatest beings is unsurpassable in dispositions towards perfection, whereas a God who fails to bring into existence the best possible beings would be surpassable insofar as dispositions toward ideal states of affairs are concerned, an imperfection for which grace is not compensatory (Wielenberg 2004, 52).

This leaves only one move for Adams’s satisficing theodicy: the idea that we would not be such beings. This claim has a subtle existential, psychological, emotional force insofar as we value our own lives and want it to be us who might have been more perfect beings, or insofar as we, believing Adams’s theodicy, are grateful for our imperfect lives. We would likely drop the complaint, as Leibniz implies (Adams 1975), if pressing it required God to create perfect beings but not us. Technically, however, if the argument that God must create better beings in a better world is cogent, but the facts support the idea that this world and its beings are not so, then our desire to exist cannot alter that qua objection to satisficing theodicy, despite how contradictory it may be from some vantage defined by our emotions and existential predicament. In other words, the handicapped child can complain that its parents shouldn’t have altered the fetus that produced him, despite the fact that, if they had adhered to his objection earlier on, he would not exist.

Adams argues that God need not have created us as perfect beings in a perfect world because such beings would not be us, any more than goldfish can complain that God could have made them as puppies: God could create perfect beings or puppies, but they wouldn’t be us or goldfish, respectively; they would be other beings. That argument purportedly grounds the dialectically non-neutral (theological) claim that we imperfect beings ought to be thankful for the gift of life.

The question here is not about whether dogs and goldfish could share diachronic identity, nor is it “What are identity conditions for human beings?” The question for Adams is: What are the identity conditions for any person within the Judeo-Christian theological system that frames his theodicy? A central premise of Adams’ theological framework is that who we are – our true identity from the Abrahamic theological perspective – is not our imperfect bodies, regardless of Adams’s (distracting) appeals to Leibniz’s unique metaphysics (Adams 1979). In Adams’s Judeo-Christian theology, who we are, essentially, is immortal souls. We need not advance here the many arguments that have been presented against the coherence of the concept of an immaterial, nonphysical essence that somehow causally interacts with our brains and bodies, how my soul doesn’t interact with your brain, and so on. These questions are the source of well-worn objections that date back to ancient India (Federman 2010), where non-orthodox philosophical schools such as Carvaka and Buddhism challenged similar Vedic beliefs in the ātman (soul) (Nagapriya 2004, Garfield 2015). But these questions are unnecessary here. For even if the soul-concept was coherent, this would not help Adams, for if who we are is souls, then God could have given us better bodies in a better world, and we would still be us.

This “we’re souls” objection undermines (5), the non-identity thesis. From this perspective, there is no way to block the inference that God violates P and Q, and some equivalent of R that may be adduced by extrapolation from P, Q, R, and the reasoning above, say R*:

(R*) It is wrong for a perfect moral agent to cause, knowingly and voluntarily, the creation of a human notably deficient by comparison with ideal human beings in mental or physical capacity

Note that R* may be considered a non-satisficing criterion for an omnibenevolent God, unless God can create supra-human beings with the same souls; if so, then R* could be a satisficing criterion, and some version of it, say R**, could capture the greater requirement.

(R**) It is wrong for a perfect moral agent to cause, knowingly and voluntarily, the creation of a being notably deficient by comparison with ideal beings in mental or physical capacity.

Since most humans appear not to be in the range of ideal human beings, God violates R* in most cases; God probably violates R** in the case of most sentient beings known to us, as the animal kingdom is a relatively ruthless hell by even human standards.

In affording Adams a charitable interpretation, I suggested earlier that it might be enough if a neighboring cluster of possible worlds and possible beings satisfies Adams’s satisficing criteria, though actual beings in our world may not satisfy those criteria in this or that – implicitly unrepresentative – case. This accession may now be retracted. For, in light of the fact that we are supposed to be souls, God violates R* and R**, this world contains beings the majority of which represent instantiations of God’s failure with respect to R* and R**, and thus God does not even satisfy Adams’s satisficing criteria.

The actual world is not neighboring some possible world – nor, as Kraay would put it, in a world-cluster (Kraay 2011) – in which God satisfies Adams’s satisficing criteria. As Grover argued, “unless we already have in place a robust theodicy that offers either explanations for or justifications of some of the world’s most significant evils, we have little reason to believe that this world satisfies [such] conditions” (Grover 2003, 146). Thus, the hedges put in place to afford Adams’s satisficing argument the most charitable interpretation cannot block the fatal implications, ironically, of the idea that we are souls.

Possible replies

To resist this fatal “we’re souls” objection, Adams would have to either deny the existence of the soul or its associated identity thesis. Although the Buddhist no-soul doctrine might be an interesting avenue to explore, I suspect Adams would be inclined if not wiser to let go of this particular attempt at theodicy than to let go of his belief in the soul, just as Euthyphro was inclined, and wiser, to let go of his definition of piety instead of his polytheism when the two were shown by Socrates to contradict each other. If there’s no soul, Adams’s theological framework collapses, with Hick’s fallback afterlife position. Thus, soul-denial is not a viable option for Adams, Hick, or like-minded theists.

One might argue that for the sorts of changes under consideration (e.g., radical change of fetal genetics), the theist can say that there are different souls in the pre- and post-altered fetal bodies. Perhaps it is an implausible or odd claim, but the theist can argue that it is less implausible or odd to say this than to say that the soul of a goldfish could have been in a bulldog. Unlike soul-denial, therefore, soul-difference cannot be rejected outright; thus, soul-difference deserves a developed response.

Counter-reply: stem souls, and what it’s like to be a transubstantiated bat

The theist can object that it seems odd to think the soul of a goldfish could be the same soul as that of a bulldog after some sort of phi-fi (philosophical fiction) soul-body-transfer; likewise, she can object that we could still be us in better bodies/worlds. She can argue that souls in radically different species could be constituted differently, somewhat loosely describable as “soul-species,” to back up that intuition. However, while species-type differences in soul-types might be metaphysically possible, there is little to no basis for it within the theology that frames Adams’ theodicy, in which case it would be a transparently ad hoc emendation. And while it occurs within a different theology, Hindu reincarnation theory rejects the intuition that the soul of a goldfish cannot subsequently inhabit the body of a dog or human; to the contrary, reincarnation is believed to involve embodiment up and down the evolutionary scale (Nagapriya 2004).

However, even if there are non-interchangeable soul-species, this would not help Adams. For even if there are non-interchangeable soul-species-based metaphysical constraints that render it metaphysically impossible for God to take the soul of a goldfish and place it in the body of a dog, the variation between ordinary and superior humans is not nearly as great as that between goldfish and dogs, and humans all – on this option – have human-species-souls. Thus, even if there were non-interchangeable species-souls, presumably all members of a species could be constituted by them, regardless of where their embodied attributes fell along the spectrum of deficient, normal, and superior members of that species. Thus, all human-species souls could maintain identity if placed in superior human-species bodies.

To block this, Adams would have to argue that each individual has not merely a species-soul but an individual-type soul, a unique soul identity. That means, not just a numerically distinct human-type soul, but a qualitatively distinct individual-type soul, a metaphysical signature above and beyond the numerical uniqueness of being this human-souled subject of experience versus that one. But the only way that could be cashed out would be if individual-type souls were not qualitatively equivalent. But if we weren’t, the (unequal) differences between us would seem arbitrary, given that they are undeserved. This might be a surmountable problem.

However, one might argue that while it is conceivable or possible that God could make individual-souls, that isn’t the same as God being restricted to doing so, and there are no reasons to think He is so restricted. To the contrary, if He could do that, He could also do this: He could make qualitatively identical but numerically distinct human-type souls. Within the broader dualistic metaphysics and theology that informs Adams’s theodicy, most importantly, God is even capable of creating what I’ll call “stem-souls,” the soul equivalents of stem cells. Like stem cells, stem-souls are proto-type souls composed of undifferentiated soul substance, numerically distinct but qualitatively identical until put through an individuation process and developmental sequence in which they become qualitatively differentiated when embodied in different psychophysical configurations (minds/bodies), only after which they acquire soul-species-differentiated characteristics. Such ontic essences acquire differentiation, but are otherwise equal as pure subjects of experience. If these are possible, and they seem possible within Adams’s theological metaphysics, then it seems Adams’s satisficing theodicy ultimately fails.

The theological metaphysics behind Adams’s theodicy is dualistic, and thus it is reasonable to think that God, in creating a stem-soul, could make it the subject of experience, that mysterious thing about which one might wonder what it is like to be it, such as the experiencing subject that is a bat (Nagel 1974). If God has the power to transubstantiate anything, such as to convert the metaphysically mundane essence of otherwise ordinary bread into the sacred Eucharist, or turn water molecules into wine molecules, inter alia, then God has the power to create a stem-soul that is the same subject of experience that could be made into an ordinary human or into an extraordinary one, and even into a goldfish, dog, or bat while remaining the same subject of experience. Arguably, this is plausible if the concept of a soul is plausible. Certain Hindu philosophies hold that every being is eternal stem-soul substance (the ātman) that reincarnates and is the bearer of its previous karma. It is the same subject of experience throughout its transmigrational journey through countless bodies/species: some of us may have actually known, therefore, in previous lives, what it is like to be a bat.

Arguably, to appeal to materialist intuitions in which identity conditions completely supervene on bodily ones – e.g., in doubting that a goldfish mind or soul could occupy a dog body – as a ground for rejecting identity conditions on the soul level is, first, dialectically asymmetrical, and second, from the theological vantage, to beg the question against God’s abilities and against the very notion of a soul. Thus, the attempt to protect Adams’s theodicy from the failure of his non-identity premise is at least prima facie inconsistent with the theological metaphysics that purports to ground that theodicy: the dualist cannot consistently appeal to materialist identity criteria to defend dualism against objections to its dualistic identity criteria. That amounts to losing track of whether an objection is a pro or a con, or upon whom the burden of proof falls.

Conclusion, for theists and non-theists

I have argued that Adams’s theodicy is subject to a number of problems, chief among them the fact that it collapses on its own reliance on the issue of what makes an individual who or what that individual is. Whereas Adams thought his non-identity condition could ground the claim that if God were required to make better versions of us, they wouldn’t be us and we wouldn’t exist, on analysis of the identity condition that constitutes a central tenet of the theology within which Adams’s theodicy is constructed, that identity condition is the soul. Once the soul is acknowledged as such, the rest of Adams’s satisficing theodicy of grace unravels.

But one need not buy into the theological metaphysics of the soul to reject Adams’s non-identity thesis and thus to reject his satisficing theodicy. For Adams’s genetic argument about pre- and post-altering fetal non-identity is weak even from a non-theological vantage: if the fetus before the alteration is a subject of experience, and remains a subject of experience through and after the alterations, then arguably it is the same subject of experience and thus the same being (not unlike the case in which a subject of experience remains throughout amnesia or other disruptions in its more complex sense of self), even if its acquired identity differs from its inherited identity. It is a biomedical fact that fetuses have the sort of brain activity associated with sentience fairly early on in the first trimester. It is also a fact that organisms in the actual world can undergo small-scale genetic engineering while remaining subjects of experience. Thus, not only is the answer to the a priori question unsupportive of Adams’s generic satisficing theodicy of grace, but simple facts about actual beings suggest that the answer to the a posteriori question does not support the idea that this is a MMAW or that we are MMACBs. There may be other ways to ground satisficing theodicies, but Adams’s non-identity thesis is not one of them.

References

• Adams, Robert Merrihew. 1972. Must God create the best? Philosophical Review 81:3 (July 1972), 317-332. http://dx.doi.org/10.2307/2184329
• Adams, Robert Merrihew. 1979. Existence, self-interest and the problem of evil, Noûs 13 (1): 33-65. http://dx.doi.org/10.2307/2214795
• Coughlan, M. J. 1987. Must God create only the best possible world? Sophia 26 (2): 15-19. http://dx.doi.org/10.1007/BF02781168
• Federman, Asaf. 2010. What kind of free will did the Buddha teach? Philosophy East and West 60, 1-19. http://dx.doi.org/10.1353/pew.0.0086
• Garfield, Jay L. 2015. Engaging Buddhism: Why it matters to philosophy. New York: Oxford University Press. http://dx.doi.org/10.1093/acprof:oso/9780190204334.001.0001
• Grover, S. 2004. Rival creator arguments and the best of all possible worlds, Sophia 43 (1): 101-114.
• Grover, S. 2003. This world, ‘Adams Worlds’ and the best of all possible worlds, Religious Studies 39 (2): 145-163.
• Hick, John. 1973. Philosophy of religion. Upper Saddle River, NJ: Pearson Education.
• Howard-Snyder, Daniel and Frances Howard-Snyder. 1999. Is theism compatible with gratuitous evil? American Philosophical Quarterly 36 (2): 115-30.
• Kraay, Klaas. 2011. Incommensurability, incomparability, and God’s choice of a world, International Journal of the Philosophy of Religion 69: 91-102. http://dx.doi.org/10.1007/s11153-010-9266-1 68
• Kraay, Klaas. 2010. Theism, possible worlds and the multiverse, Philosophical Studies 147: 355 http://dx.doi.org/10.1007/s11098-008-9289-y
• Langtry, B. 2008. God, the best, and evil. Oxford: Oxford University Press.
• Nagapriya. 2004. Exploring karma and rebirth. Windhorse Publications. http://dx.doi.org/10.1093/acprof:oso/9780199238798.001.0001
• Nagel, Thomas. 1974. What is it like to be a bat? Philosophical Review 83 (4): 435-50.
• Rowe, William L. 2004. Can God be free? Oxford: Oxford University Press.
• Rowe, William L. 1994. The problem of no best world, Faith and Philosophy 11 (2): 269-71. http://dx.doi.org/10.5840/faithphil199411224
• Rowe, William L. 1979. The problem of evil and some varieties of atheism, American Philosophical Quarterly 16 (4): 335-41.
• Tucker, Chris. 2015. Satisficing and motivated submaximization (in the philosophy of religion), Philosophy and Phenomenological Research 1-17. http://dx.doi.org/10.1111/phpr.12191
• Van Inwagen, Peter. 2015. The problem of evil. In Oxford handbooks online. www.oxfordhandbooks.com
• Wielenberg, E. 2004. A morally unsurpassable God must create the best, Religious Studies 40 (1): 43-62. http://dx.doi.org/10.1017/S0034412503006796
• Weinstock, Jerome A. 1975. Must God create the best world? Sophia 32-9. http://dx.doi.org/10.1007/BF02804260