The privation theory of evil and the evil-God challenge

Abstract

Can the best arguments for a privation theory of evil be parodied, with equal plausibility, as arguments for a privation theory of good? The privation theory of evil claims that evil has no positive existence, and it is but a privation of good. The privation theory of good claims the opposite. I approach this topic as one element in the so-called evil-God Challenge. Stephen Law has argued that the epistemic support for belief in an omniscient, omnipotent, and morally perfect God (theism) is on a par with the epistemic support for belief in an omniscient, omnipotent, but completely evil-God (maltheism). In fact, he concludes, the arguments for an evil God are symmetrical with, and isomorphic to, those for a good God. The privation theory of evil has often been used to defend theism against the argument from evil. Thus, part of the evil-God Challenge is to evaluate arguments for the privation theory of evil for their vulnerability to maltheist parody. I consider a broad range of arguments for the privation theory of evil, and I argue that most of them are vulnerable to parodic neutralization. Furthermore, I argue that although the thesis of the convertibility of being and goodness is often held to entail the privation theory of evil, or to be entailed by it (or to be equivalent to it), it is independent of the privation theory. I do find that David Oderberg’s recent argument for the privation theory of evil resists any easy maltheist parody, but I argue that it has a defect. I sketch an argument according to which his good-as-fulfillment account is compatible with a perfectly evil god. My tentative conclusion is that the privation theory of evil enjoys little more plausibility than does the privation theory of good.

Introduction

My article addresses this question: Can the best arguments for a privation theory of evil be parodied, with equal plausibility, as arguments for a privation theory of good? That is, are the rational cases for each privation theory approximately on a par with each other? My tentative answer is ‘not perfectly, but mostly, yes’.

I come to this topic via an interest in the so-called evil-God Challenge. Law (2010) has argued that the epistemic grounds for belief in an omniscient, omnipotent, and morally perfect God (theism) is on a par with the epistemic support for belief in an omniscient, omnipotent, but completely evil God (or maltheism¹). In fact, he concludes, the arguments for an evil God are symmetrical with, and isomorphic to, those for a good God. In my 2019 article, I attempted to extend the evil-God Challenge and to defend it from objections. The privation theory of evil has often been used to defend theism against the argument from evil. Thus, part of the evil-God Challenge will be to evaluate arguments for the privation theory of evil for their vulnerability to maltheist parody. In my 2019 article, I briefly addressed the privation theory of evil, and I argued that it was subject to parodic neutralization. In this article, I tackle that issue in much greater depth, by considering a profusion of arguments for the privation theory of evil, from the classical to the contemporary, and I argue that most can be neutralized by parody arguments supporting the privation theory of good. Furthermore, I argue that although the thesis of the convertibility of being and goodness is often held to entail the privation theory of evil, or to be entailed by it (or to be equivalent to it), it is independent of the privation theory. Thus, a privation theory of good need not be saddled with the convertibility, or identity, of non-being and goodness. I do find that David Oderberg’s (2020) argument for the privation theory of evil resists any easy parodic neutralization, but it has some defects, including an inability to account for the intrinsic badness of pain. I sketch an argument according to which his good-as-fulfillment account is compatible with a perfectly evil god. I conclude by considering but ultimately dismissing further possible asymmetries between good and evil relevant to the privation theories.

The privation theory of evil

The privation theory of evil is very old, and after some centuries of doldrums, it is enjoying a comeback in recent decades. According to the privation theory of evil, as developed by St. Augustine and, later, Avicenna (or Ibn Sna), and Aquinas,² evil – whether it is moral evil, like vice, or natural evil, like illness – has no positive existence. Rather, it is a privation of goodness, which is an absence of due goodness (or, alternatively and allegedly equivalently, evil is a privation of being). Thus, for a human, blindness is a privation, but a lack of wings is not. Evil, according to this view, is metaphysically and perhaps epistemologically parasitic on good. The privation theory does not claim that evil does not exist; privations and absences do exist, but they lack positive existence. Note that although some reserve the term ‘evil’ only for the most egregious or heinous things, the theory under discussion is about anything bad, including minor ills, irritations, and defects. Explanations of the privation theory often analogize it with light and dark, or silence and sound. Light has a positive existence, and dark does not, being only an absence of light. Note that there is a disanalogy, though: any absence of light is darkness, but, according to the privation theory, not just any absence of good is evil.

While the privation theory of evil per se isn’t a thesis about the problem of evil, it has important implications for the problem of evil, and for philosophy of religion and theology generally. The privation theory of evil is in part motivated as a partial solution to the problem of evil. If evil is a privation of good, then God creates no evil but merely allows it to exist.

Some, like G. Stanley Kane (1980, 52–56), argue that the privation theory of evil is not necessary for the solution to the problem of evil, since God would be perfectly good in creating positively existing evils when they are necessary for greater goods. Todd Calder (2007, 376–377) has argued that, following James Rachels (1975), there is no moral distinction between causing and allowing evil, so if causing evil involves moral imperfection, so would allowing it, as privation. Thus, the privation theory solution would fail to leave God with clean hands. I will pass over these objections that the privation theory of evil is irrelevant to the problem of evil, as they are largely irrelevant to the evil-God Challenge. If true, the objections would be equally applicable to a maltheist’s use of the privation theory of good, as a solution to the problem of good.

The evil-God challenge

The evil-God Challenge, first formulated by Law (2010) proposes that the best arguments for theism – the thesis that there is one God, creator of the universe, omnipotent, omniscient, and perfectly good – provide no better epistemic support for it than the parodies of those arguments do for maltheism, the thesis that there is one being that is creator of the universe, omnipotent, omniscient, and perfectly evil, and thus that the cases for theism and maltheism, respectively, are symmetrical; it also challenges the theist to show otherwise.³

Calum Miller (2021, 449) categorizes the loci of symmetry or asymmetry in the evil-God Challenge as relating to (1) responses to the problem of evil, such as theodicy or skeptical theism; (2) positive arguments for theism, such as the Ontological Argument or the Argument from Miracles; and (3) general, theoretical considerations, such as simplicity, or the metaphysics of good and evil.

For examples of the first sort of alleged symmetry, the free will theodicy might be parodied as follows as a solution to the problem of good (in brief): evil-God created people, and gave them freedom, which allows for the possibility of moral evil, although some will use their freedom to do good. This will be a worse world than one in which people treat each other badly but with no moral responsibility for their actions. This freedom is a great harm to them, in that it makes them culpable and deserving of punishment. A parody of skeptical theism would be skeptical maltheism, according to which our lack of knowledge of a morally unjustified or evil reason for evil-God to permit a good is not a strong reason to conclude that evil-God lacks such an evil reason; such no-see-um inferences – the good is inscrutable, thus it is gratuitous – are unjustified, perhaps due to our lack of complete moral knowledge or our imaginative poverty. For a possible example of the second category of symmetry, Law offers a parody of the Ontological Argument,⁴ according to which no being could be the worst being conceivable if it existed only in the imagination (2010, 371). This article’s topic is squarely in Miller’s third category, as it relates to alleged ontological or metaphysical asymmetries between good and evil, though it has implications for the first.

Parody of theistic argumentation, as a means of neutralizing the argument, has long played a significant role in philosophy of religion, dating back at least to Gaunilo, and his ‘lost island’ parody, on behalf of the fool, of St Anselm’s Ontological Argument. One thing distinctive about the evil-God Challenge is that it relies almost fully on the neutralizing power of parody. So, while a typical skeptic about theism might reply to a theodicy based on the story of Adam and Eve by arguing, say, that the Garden of Eden account is inconsistent with highly confirmed science, a proponent of the evil-God Challenge might instead construct a similar story – perhaps a very similar story⁵ – as a maltheist solution to the problem of good. (That said, the weaker an argument for theism is, the less important it is to the evil-God Challenge whether an equally plausible maltheist parody can be fashioned (as Law argues (2010, 368–372). For example, a bandwagon argument for theism based on the large number of people who are theists resists any maltheist parody, as almost nobody is a maltheist; however, such an argument seems fallacious.⁶) Another difference between the evil-God Challenge and other atheist challenges to theism is that the evil-God Challenge, if successful, could confer some probability on maltheism, as a legitimate competitor to theism or atheism.

Thus, the evil-God Challenge confronts solutions to the problem of evil premised on the privation theory of evil by considering whether the best arguments for the privation theory of evil can be adapted with equal plausibility in support of the rival privation theory of good (and thus in support of a maltheist reverse theodicy for the problem of good). If the best arguments for the privation theories of evil and good, respectively, are epistemically symmetrical, this seriously undermines both the privation theory of evil and the solution to the problem of evil it underwrites.

Privation theory of good

The privation theory of good holds that good has no positive existence, and that it is but a privation of evil (or the absence of evils that are supposed to be present). Health is the absence of sickness, perhaps, virtue merely the absence of vice, and happiness the absence of misery. Good is metaphysically, and perhaps epistemologically, parasitic on evil. This account is left intentionally vague for now, to accommodate parody of the variety of accounts according to which natural evil, moral evil, and badness of any sort, like a broken pencil or a blemish, are privations of one sort or another. The privation theory of good, like the privation theory of evil, can take many forms. While the privation theory of evil has quite the philosophical pedigree, the privation theory of good has barely any defenders at all. Arthur Schopenhauer is one of the few:

[T]he nature of man and animal is such that we never really become conscious of what is agreeable to our will; if we are to notice something, our will has to have been thwarted, has to have experienced a shock of some kind. On the other hand, all that opposes, frustrates, and resists our will, that is to say all that is unpleasant and painful, impresses itself upon us instantly, directly, and with great clarity. Just as we are conscious not of the healthiness of our whole body but only of the little place where the shoe pinches, so we think not of the totality of our successful activities but of some insignificant trifle or other which continues to vex us. On this fact is founded what I have often before drawn attention to: the negativity of well-being and happiness, in antithesis to the positivity of pain.

I therefore know of no greater absurdity than that absurdity which characterizes almost all metaphysical systems: that of explaining evil as something negative. For evil is precisely that which is positive, that which makes itself palpable; and good, on the other hand, i.e. all happiness and all gratification, is that which is negative, the mere abolition of a desire and extinction of a pain (1850/2020, 103–104).

Even if Schopenhauer is correct that evil is conceptually or epistemically prior to good, it would need to be shown that this entails evil’s ontological priority, though. After all, it is not implausible on similar grounds that darkness is conceptually prior to light – children may be afraid of the dark before they have the concept of light – but it is extremely implausible that light is ontologically dependent on the dark. It should be noted, though, that there is no evident parody of Schopenhauer’s argument in favour of the privation theory of evil.

Not only does the privation theory of good have few defenders; neither does it have many critics. Most defenders of the privation theory of evil do not discuss the privation theory of good even as an alternative to be rejected.⁷

The arguments for the privation theory of evil most vulnerable to maltheist parody

There are very many arguments for the privation view of evil. Many are self-evidently vulnerable to maltheist parody, while for others the matter is more complicated. Let’s begin with a survey of the arguments most easily parodied. (Parts of this section are lightly adapted from my brief earlier treatment of the subject (Collins 2019, 87–89.)

Augustine

In the Enchiridion Augustine argues that since when a diseased body becomes healthy the disease does not dwell elsewhere, and when vices become virtues, the vice is not driven to some other location, disease is a mere absence of health, vice a mere absence of virtue:

For what is that which we call evil but the absence of good? In the bodies of animals, disease and wounds mean nothing but the absence of health; for when a cure is effected, that does not mean that the evils which were present – namely, the diseases and wounds – go away from the body and dwell elsewhere: they altogether cease to exist; for the wound or disease is not a substance, but a defect in the fleshly substance, – the flesh itself being a substance, and therefore something good, of which those evils – that is, privations of the good which we call health – are accidents. Just in the same way, what are called vices in the soul are nothing but privations of natural good. And when they are cured, they are not transferred elsewhere: when they cease to exist in the healthy soul, they cannot exist anywhere else (421/1961, Ch. 3, section 11).

Of course, the maltheist can make the same point about what happens to health or virtue when a person sickens or becomes corrupted. There is no significant difference in plausibility between the two arguments; both are very poor.

Anglin and Goetz

More recently, Bill Anglin and Stewart Goetz (1982) have played a prominent role in reviving the privationist view. Some of their arguments assume as a premise the existence of a supremely good God, and some do not. An example of the former:

1. (1) Any function which something inherently [i.e., positively] evil might serve in the fulfillment of a surpassingly good divine purpose could equally well be served by a privative evil.

2. (2) The only way an omnipotent being’s goodness would not be compromised if he created something inherently evil would be if that thing were logically necessary to the fulfillment of every surpassingly good purpose.

3. (3) Necessarily, God is good.

4. (4) Thus, God does not create anything inherently evil.

5. (5) Thus, nothing is inherently evil (1982, 9).

Here, too, the maltheist may offer mirror arguments for a privation theory of good: evil-God’s evil would be compromised if it created a positive, or inherent, good, since a privative good could equally well serve the fulfillment of a surpassingly evil purpose. Thus, evil-God creates no positive goods and thus there are no positive goods. Whatever the plausibility of Anglin and Goetz’s argument, which I think is not high, it is matched by the parody.

Anglin and Goetz offer other arguments for the privation view, without explicitly endorsing them, including these that they say can be gleaned from Augustine’s The Writings against the Manichaeans and against the Donatists (1979, 145–150).

1. (1) Nothing is evil unless it is destroying (or corrupting) something.

2. (2) Thus, it is not possible that something evil not be destroying something.

3. (3) If there were something inherently evil it could exist apart from other things, destroying neither them nor itself.

4. (4) Thus, there cannot be anything inherently evil.

5. (5) Thus, evil is just privation (1982, 10, line numbers changed).

The maltheist parody of this will involve the premise that nothing is good unless it is improving something, and the premise that if something is inherently or intrinsically good, it could exist apart from other things. It might be objected that although the Augustinian argument can be rivaled by the maltheist, the parody argument is not as plausible as the original is.⁸ Both the argument and its parody are highly implausible. One’s intention or will can be good or bad intrinsically, even if it is wholly lacking in the power to accomplish its purposes. Also, even if evil is a relational or extrinsic property, it does not follow from this that it is a privation; nobody is intrinsically an uncle, but unclehood is not a privation. Miller (2021) and Ben Page and Max Baker-Hytch (2020), in their objections to the evil-God Challenge, do not challenge the aptness of the parody just described (as it first occurs in my (2019, 88); rather, they regard Anglin and Goetz’s Augustinian argument as weak. Miller describes it as ‘not immediately convincing’ and says that the authors failed to motivate it (2021, 454); Page and Baker-Hytch argue that my (2019) parodies of the Anglin and Goetz Augustinian arguments, even if successful, do not undermine the privation theory of evil, since I parody ‘some of the weaker arguments for the [privation theory of evil]’ (Page and Baker-Hytch (2020), 496). Later in the article we will consider arguments for the privation theory that Miller, and Page and Baker-Hytch, esteem more highly.

Another argument Anglin and Goetz draws from Augustine:

1. (1) A typical evil involves no more or less than the loss of some transitory good or the failure to get some object of desire.

2. (2) Whatever always accompanies a typical example of something is essential to it.

3. (3) Thus, privation and nothing but privation is essential to evil (10).

Though premise two is dubious – all typical airplanes have seats, but a seat is not essential to an airplane – the evil-God Challenge can help itself to the premise in its parody argument, according to which a typical good involves no more or less than the loss of some transitory evil, or the avoidance of some object of fear or hatred. Perhaps the Augustinian argument could be developed in such a way that it becomes immune to maltheist parody, but in its present form it seems vulnerable, as there is no evident significant asymmetry between the original and the parody.

David E. Alexander

David E. Alexander (2012, 98–100) argues that if ‘good’ and ‘bad’ are logically attributive, then the privation theory of evil is true. Following Geach (1956), Alexander distinguishes between predicative and attributive adjectives in terms of the sort of inferences each will permit. Consider a sentence with form ‘X is a A N’, where X is a subject term, A is an adjective, and N is a direct object noun. If it is valid to infer ‘X is A’ and ‘X is a N’ from ‘X is a A N’, the adjective is predicative. But if it is invalid to infer ‘X is A’, then the adjective is attributive.

Standard examples of predicative adjectives are literal uses of ‘blue’, ‘twelve feet in length’, or ‘square.’ It follows from this being a blue bicycle that it is blue, and that it is a bicycle. ‘Short’ and ‘young’ are attributive adjectives; from the fact that someone is a young great grandfather, it does not follow that he is young, or a young man (though being a grandfather entails being a man). Further, it is senseless to describe something as young, apart from any category of thing to which it belongs. ‘Smith is young’ must be either elliptical for ‘Smith is a young K’, where ‘K’ denotes some category of person, or it is semantically incomplete. For some attributive adjectives – the subset of attributive adjectives known as ‘alienans’ adjectives – it is invalid to split the noun term from the adjective in an inference. ‘Forged’, ‘alleged’, and ‘mock’ are all alienans adjectives. That one bought a forged Rembrandt does not entail that one bought a Rembrandt. Furthermore, nothing can be simply forged; it must be a forged item of a certain sort, such as a will, artwork, or signature.⁹ So, predicative adjectives pass the splitting test, but attributive ones do not.

Alexander accepts and further defends Geach’s attributive account of the adjectives ‘good’ and ‘bad’ (with moral evil being one type of badness). On this view, it will not follow from Smith being a good musician that Smith is good, or that she is a good person, as it would if ‘good’ were predicative. Nor does Smith’s being a bad singer entail that she is bad, or a bad person. Nothing can be good or bad simpliciter, but only a good or bad member of some class of things; ‘good’ and ‘bad’ are kind-relative, as the goodness of one thing need not resemble the goodness of another. Although different occurrences of ‘good’ pick out different qualities, ‘good’ is not an ambiguous term, like ‘bank’. Rather, ascriptions of ‘good’ to diverse things are semantically unified. On Alexander’s view, this is best explained by seeing ‘good’ as a predicate-forming functor or property marker. (Describing the same idea, Page and Baker-Hytch say that the terms ‘good’ and ‘bad’ do not pick out first-order properties. ‘Rather, the semantic content of “good” and “bad” varies depending on the nouns those terms modify. There are good books, good husbands, good cakes, good looks, good assassins, good cars and so on, but there is no particular first-order property that all these things share in common’ (2020, 496). As Patrick Lee puts it, good ‘signifies not directly a nature of property, but a way or extent of having other properties, different properties in different cases’ (2007, 487).) The metaphysical corollary is that there are no such things as the properties of goodness or badness.¹⁰

Alexander defends the attributive account of ‘good’ and ‘bad’ at length. Let’s assume the view is correct, although the conclusion that there is no such thing as the property of goodness would seem to amount to a considerable departure from a Scholastic general account of goodness, according to which good is fulfillment of a thing’s nature. Nevertheless, how do we get from the attributive account to the privation theory?

Although Alexander is not explicit about the premises of the argument, he highlights several implications of the attributive account of ‘bad’, by way of argument for the privation theory:

1. i. ‘x is bad’ either is elliptical for ‘x is a bad K’ or it is semantically incomplete.

2. ii. Badness is always attached to something substantive.

3. iii. To say that x is a bad K is to say that x is not a thriving member of the kind to which it belongs (or that it does not possess the relevant features of K to the relevant degree).

4. iv. It does not make sense to say that x is wholly bad.

5. v. The only way x could be wholly bad is if it were bad simpliciter, which is inconsistent with the attributive account.

6. vi. Thus, the badness of x is parasitic on its goodness.

Of course, on Alexander’s account, ‘good’ is attributive, too, so corollaries of i., ii., iv., and v., with ‘bad’ or ‘badness’ replaced with ‘good’ or ‘goodness’, also will be true, by his lights. The key claim, then, seems to be iii. Let’s accept iii. though it seems to go a bit beyond the attributive account. But iii. is a very slender basis for the privation theory of evil. Yes, being bad can be characterized in terms of what the bad thing is not, but that is tautologous, and it does not entail a substantive metaphysical claim about the nature of badness or evil. After all, to say that x is a good K is to say that it is not a bad K, and that it is not a defective member of K.

The maltheist parody of Alexander’s argument would then hold, on parallel grounds, that a good K is a K that is not floundering, that nothing can be wholly good or good simpliciter, and thus that the goodness of a thing is parasitic on its badness. Alexander’s argument and the parody seem equally unpersuasive, as in neither argument is the privationist conclusion supported by the premises. And, at the risk of tu quoque, it should be noted that Alexander sees his attributive account of good as providing the ‘starting point for a few different, though related, arguments for God’s existence’ (2012, 91). Though he considers several ways to make God’s goodness compatible with the attributive account (at 119–120), none of them would involve God’s not being wholly good, though by his lights that is an implication of his view.

David S. Oderberg

Typically, accounts that deny the privation theory of evil do not deny that some evils are privational. These mixed accounts of evil hold that some evils, like blindness and apathy, are privations, whereas others, like pain and malice, have positive existence. (Indeed, most such would offer a mixed account of good, too, with positively existing goods like kindness and pleasure, but also goods that are absences, like the absence of pain or malice.) David S. Oderberg (2020, 117) argues against such mixed accounts on grounds of simplicity, saying that they would make evil metaphysically heterogeneous in a way that good (supposedly) is not. The privation theory gives a unified account of both evil and good.

I will concede that a privation theory of evil, conjoined with a strictly positive account of good, does enjoy the advantage of simplicity over a mixed account of both evil and good. But the privation theory of good (again, conjoined with a strictly positive account of evil) will be preferable in the same respect to a mixed account, in terms of simplicity and unification. (And any privation theory – of good or evil – will face metaphysical challenges, as well, such as settling whether privations can be causes and effects, and what sort of truthmakers true claims about privations will have. See Oderberg (2014b; 2020 (ch. 6 and 7)) for an extended attempt to meet those challenges.)

N.B.: this point only addresses Oderberg’s argument that simplicity and unification favour a privation theory over a mixed theory. It is not intended by itself to show that there are no differences in the respective plausibility of the privation theory of good and the privation theory of evil. I think, simplicity aside, that mixed theories of both good and evil are far more plausible than either privation or strictly positive accounts are.

Christophe de Ray

One argument for the privation theory of evil, offered by de Ray (2024, 2–5), is that the privation theory accounts for why good and evil are ‘contraries’, or opposites. The first way for two things to be opposites is for one to be the absence of the other. The second is for each to be ontologically on a par yet still opposite, as perhaps with sweet and bitter. According to de Ray, nobody has yet supplied an explication of the second sort for the oppositeness of good and evil.¹¹

Let’s assume for the sake of argument that de Ray is right that no satisfactory account has been given as to why good and evil are opposites, if neither is an absence of the other. However, de Ray’s point clearly can be adapted in support of a privation theory of good, since this too will account for the contrariety of good and evil.

A nest of arguments dependent on the convertibility thesis

Many philosophers take the privation theory of evil to entail, or to be entailed by, or to be equivalent to, the thesis of the Convertibility of Being and Goodness, a hoary principle defended by such luminaries as Augustine, Boethius, and Aquinas. This is sometimes put variously as the thesis that:

• • Being and good are identical (Alexander 2012, 91).

• • Anything that has being is good, and anything good has being (Alexander 2012, 90).

• • Goodness and being are interchangeable (Keltz 2020, 137).

• • ‘Being’ and ‘good’ are interchangeable terms in predication. Wherever being is predicated of something, the predicate ‘good’ is involved as well (Aertsen 1985, 449).

• • Being, as such, is good, so the absence of one implies the absence of the other (Aquinas ST I, q. 48, a. 1).

• • Goodness and being are really the same, and differ only in idea…But goodness expresses the aspect of desirableness, which being does not express (Aquinas ST I, q. 5, a. 1).

• • The terms ‘goodness’ and ‘being’ are the same in reference and differ only in sense (Alexander 2012, 92).

• • Goodness is the same as being itself, but considered from a particular point of view – that of fulfillment of appetite (Oderberg 2020, 14).

• • The predicates [being and goodness] can be exchanged salva veritate but not salva sensu: their referents are, as the maxim goes, really the same albeit conceptually different (Oderberg 2014a, 345).

For the sake of settling upon a single formulation, I will understand the principle in its most basic form:

Convertibility of Being and Goodness: the thesis that being and goodness are identical.

But, as appropriate, I will consider the alternative formulations.

Some see the privation theory’s connection to Convertibility as key to refuting the privation theory of good, and hence also to refuting the evil-God Challenge response to privation-based solutions to the problem of evil. If the privation theory of good is committed to the convertibility of non-being and goodness, then it allegedly would have to hold, among other claims, that non-existence is the ideal state of all things in nature, that a purely evil being is impossible, or that all beings naturally seek their own non-being, propositions that the evil-God Challenge theorist might be loath to accept.

Thus, the evil-God Challenge response to the privation theory of evil must contend with the closely associated principle of Convertibility.

Note that while it is natural to suppose that Convertibility also identifies evil with non-being and non-good,¹² it must not do so, if it is to be consistent with the Privation Theory of Evil. According to the Privation Theory, some absences of good are not evil, as they are not the absences of natural and due goods. Some people have photographic memory; tetrachromats have a genetic mutation that allows them to perceive many more colours than typical people can; one woman can smell Parkinson’s disease. Since these goods are not due to humans, though my complete lack of them is an absence of good and an absence of being, that lack is, per the privation theory, not an evil. The absence of these goods will be morally neutral, and not a privation.

According to Convertibility, then, some non-being is evil, when it is the absence of a due good, but some non-being is neither evil nor good, when it is the absence of an undue good. So, the price of homogeneous and unified accounts of both good, evil, and being is that non-being is heterogeneous; non-being is always not good, but only sometimes evil.

So, how do philosophers argue that convertibility entails or is entailed by the privation theory?

Keltz’s defence

B. Kyle Keltz says:

Aquinas believes that goodness and being are interchangeable concepts in that something is good insofar as it is a perfect example of its kind. Since things in the world desire (or tend toward) their perfection, and since something is only perfect insofar as it exists as an ideal instance of its kind, being is interchangeable with goodness (Keltz 2020, 36).

From those claims it follows that everything tends toward its own good (as opposed to the good). But that is not equivalent to the claim that being and goodness are identical, as what something tends toward – as eyes tend toward sight, airplanes toward flight – is generally distinct from what is doing the tending.

Of the privation theory of good, Keltz says it would entail that existence is undesirable, so presumably he holds that the privation theory of evil entails that existence is desirable. However, in characterizing Aquinas’s view, Keltz says that existence is desirable because all things tend toward and/or will their own best existence, or the perfection of their natures (2020, 142). But even if we allow the equivalence between what is good and what is desirable, what we observe in nature is not that being as such is desired, but that each individual desires or aims at its own best existence. This is not equivalent to the claim that being and goodness are identical.

The inference from ‘each individual’s (best) existence is good for it’ to ‘existence is good’, is reminiscent of John Stuart Mill’s famous inference in Utilitarianism: ‘each person’s happiness is a good to that person, and the general happiness, therefore, is a good to the aggregate of persons’ (1861/2001, 35–36). Mill’s inference has been much scrutinized, and it has its defenders, but it is generally judged to be fallacious.

Aquinas’s own argument, from which Keltz’s is adapted, seems to involve the same problematic inference:

Now, we have said above that good is everything appetible; and thus, since every nature desires its own being and its own perfection, it must be said also that the being and the perfection of any nature is good. Hence it cannot be that evil signifies being, or any form or nature. Therefore, it must be that by the name of evil is signified the absence of good. And this is what is meant by saying that evil is neither a being nor a good (ST I, q. 48, a. 1).

Alexander’s defence

Alexander offers a compressed argument for the privation theory entailing the convertibility thesis.

If the privation theory is true, and evil is the absence of some feature that should be present, then if x is a bad K, x lacks some feature that Ks should have. Good, though, is the presence of such features. Alexander continues:

That is, since evil is the absence of some feature that should be present, the absence of evil is equivalent to the absence of an absence of some feature that should be present, which is equivalent to the presence of some feature that should be present, which is being and hence goodness (2012, 100, italics added).

But Alexander, like Keltz, is seemingly blurring or eliding the distinction between some feature that should be present (i.e., goodness) and what is present (i.e., being). Note also that the privation theory of evil sorts good into two categories: those due and those not due, with evil being an absence only of a due good, and thus a privation. If the privation theory of evil is true, then, on the face of it, it seems we have three categories of positively existing things: (a) due goods, such as vision in a human; (b) undue goods, such as perfect recall or savant skills, in a person; and (c) things that aren’t good, such as, to use Alexander’s own colourful examples: ‘dishes in my pocket, bullets in my mouth, spikes in my eyes, trees in my bed, unicorns, etc.’ (2012, 99). That might not be the case, but the privation of evil does not entail that it is false.

Note also that the passage quoted above commits Alexander to a privation theory of good: he says that the absence of evil is equivalent to something – the absence of an absence of some feature that should be present – which is equivalent to something – the presence of some feature that should be present – which is being and hence goodness. In short, given the transitivity of equivalence and identity, he says: goodness is the absence of evil, and being is the absence of evil.

More from Alexander on Convertibility:

[I]t is easy to see that while ‘goodness’ and ‘being’ may refer to the same thing they do so under different descriptions or with different senses … Goodness picks out features that a thing is supposed to have in virtue of the kind of thing it is, whereas being picks out the features that a thing has, independently of any considerations as to whether the thing is supposed to have those features (2012, 92, italics added).

It is plainly evident from this passage that ‘goodness’ and ‘being’ will pick out different features, rather than picking out the same features via different descriptions. This will occur when something lacks features it is supposed to have, or it has features it is not supposed to have. The passage is also puzzling in being difficult to reconcile with the claim that evils are privations of due goods, which suggests that there are undue goods, and thus goods that a thing is not supposed to have.

The inadequacy of the account can be seen most clearly by examining Alexander’s own choice of example to illustrate the Convertibility thesis: being a good US President.

For example, Obama is a good president just in case Obama is the president and Obama possesses the features necessary for being a president to such-and-such a degree. To be president, Obama must possess certain features that everyone else currently lacks. To be a good president Obama must possess those features to such-and-such degree. That is, and putting this in the terms of defenders of the convertibility thesis, Obama is the president just in case he has the potentialities relevant to being president, and he is a good president just in case he has actualized those potentialities to such-and-such degree (2012, 94, italics added).

The features necessary for someone to be US President, good or bad, are that they acceded to the office per the rules specified in the US Constitution, such as winning a presidential election, and that they have not vacated that office. But the features necessary for the US President to be a good president presumably are things like wisdom, decisiveness, diligence, justice, and so forth. However, Alexander’s account holds that what it is to be a good president is to possess the features necessary for being president at all, features had by nobody else, to such-and-such degree. But those features, of having acceded to the office per the rule of law, are not had in degrees; they are binary, and they are possessed by all US Presidents, and they are possessed by nobody not a US President. Thus, the account explains how ‘the goodness of a thing and the being of a thing are not distinct’ (2012, 94) at the cost of entailing the absurdity that all US Presidents were good, and equally good, presidents.

Alexander points out, approvingly, that his ‘account attempted to explain goodness not in terms of some sui generis property of goodness, but simply in terms of mundane, and seemingly non-normative, properties’ (2012, 94, italics added).But therein lies its inadequacy.

I conclude, then, that we may distinguish between the privation theory of evil and the convertibility of being and goodness, and that we may also distinguish between the privation theory of good and the convertibility of being and evil, as independent pairs of theses. When we set aside the convertibility theses, we find that the privation theory of good is not so easily saddled with implications any more questionable than those of the privation theory of evil.

The argument from good-as-fulfillment

Ben Page and Max Baker-Hytch (2020, 495–497) offer a more concise version of the argument from the functional/attributive account of ‘good’ to the privation theory of evil, which they credit to Alexander (2012, 95–110). If for something to be good is for it to fulfil its natural function, then bad or evil must be understood as a failure to fulfil, or a deviation from, that function. Thus, good is understood positively, and evil negatively.

David Oderberg (2014a, 2014b, 2020) provides a more developed defence of this line of argument. Following Aquinas, he argues at great length in favour of good as fulfillment – fulfillment of appetite or essential tendency – as applicable to anything with being, even to inorganic objects, like stones and electrons, and to destructive objects, like missiles, as well as applicable to human beings and other living things. While broadly sympathetic to an attributive account of good, he defends a predicative account of good simpliciter (i.e., good as fulfillment) under which the various attributive goods fall.

Oderberg’s account of evil, then, it that it is unfulfilled need, and he argues convincingly that needs need not be understood as that which, if unfulfilled, will result in serious harm. (Satisfying a decent number of one’s desires for innocent pleasures is part of a good life.) Unfulfilled need can be identified with a privation of a due good, hence it is a privation theory of evil. The strength of the privation theory of evil thus depends on (1) the soundness of the good-as-fulfillment account, (2) the absence of counterexamples and absurd implications, and (3) shortcomings of alternative accounts of good and evil.

It is beyond the scope of this article to assess the theory that good is fulfillment of a thing’s essence or nature. Gwen Bradford’s (2017) ‘deep problem’ for such an account is to explain why fulfillment of essence or nature is good. As far as I can tell, Oderberg does not attempt to meet this challenge. Perhaps he can, and if he cannot, perhaps any theory of good will struggle to meet its own version of Bradford’s deep problem.

Oderberg (2020, 128–136) argues that the privation theory of evil can accommodate its apparent counterexamples, like the evils that seem to have positive existence, such as pain and malice, wading into a deep literature on the question (Anglin and Goetz 1982, Gracia 1991; Lee 2000; 2007; Crosby 2002; Murphy 2002; Calder 2007; Alexander 2012, for a few). He takes issue with some privationist responses to the pain objection. For example, he rightly notes the inadequacy of the answer that pain is good since it is a ‘warning system’ for injury or illness, as in Lee (2000, 2007) and Anglin and Goetz (1982); some pains do not serve as such a warning, and even if all did, this would show that they are instrumentally good, but it would do nothing to demonstrate that they are not intrinsically bad (2012, 129).¹³

The privationist can deal with the pain objection in one of three ways. One is to argue that, contrary to appearances, pain is a privation of good, and thus is an evil. The second is to argue that pain has a positive existence, and thus it is not evil. The third approach is a combination of the first two, arguing that whether pain is good or evil depends on whether it promotes the fulfillment of the nature of the creature that has the pain (just as having scales is good for the fish, but bad for the human).

Unsurprisingly, Oderberg defends the third approach, although he rejects the position of Lee (2007, 476) that ‘useless’ pain – pain that fails to convey a warning of damage or that is unaccompanied by any damage – is a disorder, and thus a privation. Oderberg says that useless pain might be the result of a disorder, but that the pain must be distinguished from the underlying disorder.

Oderberg’s position is that when pain functions correctly, to call one’s attention to (actual) damage, it is good and not privative, though unpleasant. Pain also often involves a distraction, or loss of mental equilibrium. On his view, in the standard, functional cases, this loss of equilibrium is typically not a privation, and thus not an evil, because it is not an absence of a due good; one is often supposed to be distracted by pain and directed toward the damage that it indicates. Unless, that is, the distraction, functional as it is in directing one toward bodily damage, involves a separate and distinct lack of functionality, as with a driver who is distracted by a sharp pain and misses that the traffic light has turned red, putting herself at risk of an accident. That pain involves two disturbances or distractions, one of which is privative, and one not. Though Oderberg does not explicitly give an account of ‘useless’ pains, presumably he would say they often involve a lack of functionality, since they distract without directing the sufferer’s attention toward any real damage. Thus, pain is evil when and only when it is privative, and when not privative, when achieving what it is supposed to, it is good, though unpleasant.

However, Oderberg’s account is a modified version of the ‘warning system’ idea, and it suffers from some, though not all, of the defects of that theory. Oderberg’s version is more nuanced, which allows it to accommodate the fact that not all pains warn us of damage. But two related problems remain. One is that this still seems to show, at best, that pains are frequently instrumental goods, and occasionally instrumental evils (and sometimes both at once), but not that pain is ever intrinsically good. The second problem is that Oderberg misidentifies pain with a certain kind of distraction or loss of mental equilibrium (2020, 134–135), when pain is an unpleasant feeling that causes that disequilibrium. Thus, pain can cause an absence of equilibrium, an absence that sometimes will be a privation and sometimes not, but that does not show that the feeling of pain itself is a privation or absence of any kind. Oderberg thus falls into the same kind of error that he identified Lee (2007) as making, of failing to distinguish pain from something to which it is causally related.

Oderberg devotes the last three chapters of the book – sixty-nine pages – to solving other problems attendant upon an account of evil as privation of good. Chapter six offers a theory of truthmakers for claims about privations, explaining what could make claims about evil true if evil has no positive being. (Truths about privations will be grounded in being.) Chapter seven examines the problem of causal relations had by privations, and Oderberg offers an account of how evil can be a cause or an effect if it has no positive being. (Essentially, privations cannot be causes or effects; but beings with privations can.) The final chapter discusses the sense in which evil exists, as well as the sense in which it does not. (Oderberg 2014b pp. 65–86 also deals with these three problems.) I will set aside these chapters here, since presumably his solutions, however successful, to these problems for the privation theory of evil would be equally applicable to the same objections to the privation theory of good. Thus, they are not relevant to any asymmetry between the privation theories of evil and good, or to the evil-God Challenge.

What would a maltheist parody of Oderberg’s view look like? Perhaps it would involve the idea that things are supposed to be bad – chaotic, miserable, cruel, etc. – because the creator wanted things to be bad, and that good is thus the absence of evils that are supposed to be present. Pleasure will typically be a privation of evil, except when pleasure has as its object something bad, as with schadenfreude and sadism.

Admittedly, that does not sound like a very promising parody, in part because it depends on the coherence of due evils. Perhaps the maltheist will have to be content with pointing out the inadequacy of Oderberg’s account’s handling of the pain objection, which mitigates somewhat the difficulties in parodying it.

However, Oderberg’s defence of the privation theory of evil suggests an alternative parodical route for the evil-God Challenge theorist, which involves accepting Oderberg’s account and noting an interesting implication of it. I will briefly sketch the outline of this idea.

Oderberg’s account of goodness, at its fundamental and general level, identifies goodness with being, and it holds that there is nothing essentially organic, moral, or beneficial about goodness. On his view, a triangular patch on a tree is a good triangle in that it exemplifies a close approximation of triangularity, regardless of whether it is good for some purpose, like education or ornamentation (Oderberg 2014a, 346–348). A good concrete object, such as a stone, is one that fulfills its tendency to continue to exist, and to persist in its spatio-temporal properties (Oderberg (2014a), 353), even if neither the stone nor anything else benefits from its fulfilling that tendency; a good nuclear missile is a nuclear missile that functions correctly, even if it would be bad to make or use such a missile (2020, 37–38). These things can be good without being good for anything. (Recall that on Page and Baker-Hytch’s similar account of good, that there can be not just good books and good cakes, but also good assassins (2020, 495).)

Let us assume this account, for the sake of argument. Consider the concept of a demon. Let’s say that the essential nature or tendency of a demon is to harm and corrupt through supernatural means. If that is so, then a good demon will be one that best fulfils this tendency, or which best approximates perfection at this, rather than a demon who is occasionally moved to kindness, or one which lacks the means or inclination to maximize corruption and misery. Such a being might create things like plagues, floods, weapons, and bigotry, which are good, though not good to create. Though Oderberg’s privation account rules out a supremely, purely, and wholly evil being, it seems to be compatible with a being perfectly good at doing what it would be bad to do. The difference between the two seems largely semantic.

Perhaps this is good enough for the evil-God Challenge theorist. The rest of the evil-God Challenge, such as the reverse theodicies and skeptical maltheism, will be unaffected.

Other possible asymmetries between good and evil

There are other possible asymmetries between good and evil that might have implications for the comparative likelihoods of the privation theory of evil and the privation theory of good. In this section, I will consider two.¹⁴

The ‘Karenina Principle’

‘All happy families are alike; each unhappy family is unhappy in its own way.’ So famously begins Tolstoy’s Anna Karenina (1877/2002, 1), inspiring the ‘Karenina Principle’: generally, failure is ensured by a deficiency in one, or in a small number, out of many respects, whereas success requires sufficiency in all, or nearly all, respects, meaning that failure can happen in many ways, but successes must be alike. The principle has antecedent corollaries, such as the unity of the virtues – the idea that one must have all the moral virtues of the Greek canon to have any of them – which is defended by Socrates in Plato’s Protagoras (1997, 746–790),¹⁵ and this from Aristotle’s Nicomachean Ethics:

Again, it is possible to fail in many ways (for evil belongs to the class of the unlimited, as the Pythagoreans conjectured, and good to that of the limited), while to succeed is possible only in one way (for which reason one is easy and the other is difficult); for these reasons also, then, excess and defect are characteristic of vice, and the mean of virtue. (Bk II, ch. 6)

There are a million ways to be sick, but perhaps just one way to be healthy. For a more mundane sort of good, a good printer will have a print head, an ink cartridge, a sheet feeder, a power source, and several other components, but a printer missing any one of those will be a bad printer.

If the Karenina Principle is true, then good and evil are asymmetrical at least in that the former is more difficult to achieve than the latter, just as it is easier to miss a bullseye than it is to hit one. Furthermore, if the sufficient condition for being a good X is the presence of a conjunction of numerous individually necessary conditions, while it suffices for something being a bad X that it lacks just one or two – as a functional or attributive account of good might affirm – this seems to lend some support to a privation theory of evil, but not to a privation theory of good.

Likely, there is some truth to the principle and its corollaries, though they are far from self-evident. While some elements, such as trust, love, and mutual support, may be common to all happy families, happy families can vary greatly in how affectionate or close they are. Many, perhaps most, families and people are happy,¹⁶ suggesting happiness is not especially difficult to achieve, compared with unhappiness. Similarly, history books are full of people who achieved greatness – George Washington, Mother Teresa, Nelson Mandela, Dr Martin Luther King, Jr, and many others – despite having serious flaws. In fact, I submit that nearly all great people had significant shortcomings, as did all great nations, and all great states-of-affairs.

Furthermore, even if a properly qualified version of the Karenina Principle is true, it is unclear what support it offers for a privation theory of evil. Does the proposition that there are many ways to be evil or bad but just one way to be good entail that evil is a privation of good? There are many ways to be coloured, and only one to be colourless; there are many ways for a bag to be full, but just one way for it to be empty. But we do not conclude that colour is an absence of colourlessness or that fullness is an absence of emptiness. Nor does the proposition that goodness is more difficult than badness entail a privation theory of good; after all, walking on one’s feet is easier than walking on one’s hands, but it does not follow that walking on feet is best understood as an absence of walking on hands.

I conclude that the support the Karenina principle seems to offer the privation theory is dependent on the principle’s congruence with the already discussed good-as-fulfillment account.

Possible asymmetries of moral transformation

One view of the nature and limits of moral transformation holds that there is an important asymmetry between the sort of moral improvement that a good God would be able to inculcate, and the moral depravity that an evil God would foment. One could argue that no matter how much a human person’s moral character could be improved, whether it could be improved without limit or would reach a maximally good state, the resulting being would still be a person, but to make a person’s character worse and worse, the worst that it could become, would be to reduce the person to some kind of beast or monster – something sub-human. The utmost moral improvement of any X results in an ideal X, but the utmost moral corruption of X results in something worse than an X.

If this is true, there will be an asymmetry in the perfect goodness of God, and the perfect evil of an evil God, with respect to their effects on the world, and some asymmetry in the nature of moral improvement and moral corruption. While it may have implications for the respective merits of the privation theories of evil and of good, they are not immediately evident. It is not a hallmark of positively existing things and their absences – light and darkness, sound and silence – that they are characterized by this sort of asymmetry.

Nor is the claim of asymmetry in the limits of improvement and corruption itself self-evident. While perpetual moral corruption of a human being might result in something sub-human or demonic, so too might perpetual moral improvement of a human being result in something nobler than a human being, such as an angel, or a being that loses its identity and becomes one with God. That is, either sort of change might be substantial change, in which a thing ceases to exist, upon transformation into some other thing.

I conclude that this possible asymmetry between good and evil, while it might have implications for the evil-God Challenge, does not pose an identifiable threat to the Challenge that falls within the scope of this article, concerning asymmetries between the case for a privation theory of evil and that for a privation theory of good.

Conclusion

The Evil God Challenge is an enormous, ambitious, and probably quixotic project, since there are so many arguments in support of the existence of a supremely good God. According to Graham Oppy’s (1996/2024) taxonomy, there are six major types of Ontological Argument alone. There is a similar profusion of arguments for the privation theory of evil. I have reviewed about a dozen of them here, though that is not an exhaustive list. Most of the arguments can be repurposed as equally plausible arguments for a privation theory of good, but at least one, while arguably flawed, is more resistant to being adapted in support of a rival view.

Therefore, only a provisional conclusion is warranted. The conclusion I tentatively advance is that the rational cases for the privation theories of evil and good are approximately on an epistemic par with each other, and approximately symmetrical. While Oderberg’s Thomistic argument might not, in the end, be parody-apt, neither does Schopenhauer’s argument for the privation theory of good lend itself readily to being repurposed. Furthermore, both arguments are far from conclusive.