Holiness among the divine attributes

Consider the big three divine attributes – omniscience, omnipotence, and omnibenevolence. Analytic philosophers of religion, most of whom are theists of a fairly traditional sort, have over the past half-century produced a voluminous literature on these divine attributes, especially on the former two. Remarkably, though, the scriptural resources that inform the thinking of these traditional theistic philosophers are not as explicit and as obviously consistent as one might like about God's exhibiting these features. Consider, by contrast, the divine attribute of holiness. There is absolutely no doubt that holiness is ascribed to God in Scripture. It is attributed to God not only directly by the human authors of those texts: the very angels whose words and deeds are reported in Scripture declare of God, emphatically, that God is not just holy, but holy, holy, holy (Isaiah 6:3, Revelation 4:8); and in Scripture God describes Godself as holy (Leviticus 11:44–45, among many others), and even swears by God's own holiness (Amos 4:2). God is, throughout the Hebrew Scriptures, designated as ‘the Holy One’.Footnote ¹ One might, then, have expected that the believers in old-time religion who make up the bulk of the class of analytic philosophers of religion would have something to say about this divine attribute. The discussion of holiness in contemporary philosophy of religion has been, however, very thin.Footnote ²

The holiness that I aim to characterize is what we could call ‘primary’ holiness, the holiness that God exhibits. While creatures, places, ways of behaving, etc. are also characterized as holy, I assume that these things, insofar as they really are holy, have their holiness in a way that is derivative from this primary holiness of God, and this is a matter on which Scripture scholars seem to be in consensus.Footnote ³ I am also going to make certain realist assumptions about holiness: that it is a real property of God and of some created things, and that experiences of things as holy are, at least in some cases, veridical.Footnote ⁴

I immediately put to the side as at least initially implausible two popular understandings of primary holiness. On one popular understanding, primary holiness is to be identified with perfect moral goodness. To call God ‘holy’, on this view, is to affirm God's absolute moral perfection.Footnote ⁵ I have two main reasons for finding this understanding initially unpromising. The first is that there are important historical figures and contemporary philosophers and theologians who deny that standards of moral perfection even apply to God, yet these authors seem nevertheless to have a very keen sense of God's holiness, and they do not seem to be conceptually confused. On these authors’ views, morality is not a pure perfection, but an impure oneFootnote ⁶ – it presupposes the limitations endemic to creaturely kinds.Footnote ⁷ Nevertheless, on their views, God can exhibit other wondrous, awe-inspiring excellences of agency, such that our declining to ascribe moral goodness to God in no way should detract from our experience of God as a superlatively holy being. Given, then, the prima facie coherence of combining a rejection of God's moral goodness with an affirmation of God's supreme holiness, it seems a mistake simply to identify holiness with perfect moral goodness. The second reason is that it just seems very implausible that God as presented in Scripture is simply being characterized as morally faultless when God is characterized as holy. On the face of it, the holiness ascribed to God is presented as a global feature of God rather than but one among the many attributes ascribed to God in a superlative way. That is why it is peculiarly appropriate for the humans and angels in Scripture who have encountered God to designate God summarily as ‘the Holy One’, and why it is peculiarly appropriate for God to describe Godself using that appellation; and that is why some writers are tempted to treat being holy as just equivalent to being divine (e.g. Sproul (Reference Sproul1985), 38). ‘Holiness is’, on this view, ‘God's “innermost reality” to which all of His attributes are related’ (Harrington (Reference Harrington2001), 11–12, citing Muilenberg (Reference Muilenberg1962), 617). If holiness appears to involve or presuppose a more global assessment of God – if, in the numerous ascriptions and self-ascriptions of holiness to God, there is no suggestion that it is a term of praise restricted to some particular feature of God, but seems to ascribe a more encompassing feature to God than that – then it should not be identified with moral perfection, which is a particular divine excellence.

The other common view of holiness that I will not employ is what we might call holiness as ‘simple separateness’. It is frequently remarked that the term in the Hebrew Scriptures translated as ‘holy’, qadosh, means ‘separate’, ‘set apart’.Footnote ⁸ I of course have no objection to the notion that what is holy is in some way separate or set apart, but that point obviously provides not much of a basis, on its own, for an account of the holy. There are lots of ways of being set apart – the clean-up crew sets apart the toxic waste for removal, the cookies made with nuts are set apart from those without nuts, a philosopher sets apart initially unpromising conceptions of holiness from initially promising ones, etc. While allowing that being holy includes in some way being set apart, it is not very informative to be told this unless one supplements it with an account of what sort of separation is being invoked.

So I will move past, without further comment, any conception of primary holiness on which it is to be identified with moral perfection or with simple separateness. Of course, any adequate account of holiness will incorporate these inadequate understandings – a correct account of holiness should explain why we should (or why we would be tempted to) say that a fully holy being must also be morally perfect, and why there are very important senses in which a holy being is separate, set apart.Footnote ⁹

I will instead take as my entry point for the characterization of primary holiness Rudolf Otto's classic study of the phenomenology of holiness in his The Idea of the Holy. The justification for beginning with Otto is in part that the influence of his views in religious studies is indubitably strong and in part that the conception of holiness that his view suggests is not subject to the same initially damning objections to which these other conceptions (holiness as moral perfection, holiness as simple separateness) are subject.Footnote ¹⁰ But in the main the justification for starting with Otto will have to be retrospective, whether the view that results strikes the reader as a defensible account of the nature of holiness.

The central thesis that I want to defend is that Otto's account of the holiness encounter suggests that the nature of primary holiness is best approached normatively, in terms of the appropriateness of certain normatively laden responses to the holy being. While there are a number of interesting further theses about God and creatures suggested by such an account, I will focus on one of these: that an absolutely holy being – one who is holy, holy, holy – would have to be the greatest possible being, and so the holy, holy, holy God of Scripture should be conceived as the most perfect possible being.

Otto on the holiness encounter

Otto aims to characterize what it is like to have what we might call a ‘holiness encounter’, which he calls a ‘numinous’ experience (Otto (Reference Otto and Harvey1923), 7).Footnote ¹¹ My focus will be on Otto's characterization of the affective stances in which we humans find ourselves when encountering something as holy.Footnote ¹² Otto characterizes it as a dual response, in his well-known mysterium tremendum et fascinans formulation (ibid., 58–59).Footnote ¹³ That God is mysterium, something mysterious in the sense of being wholly other, I will not elaborate upon here. I want to focus, instead, first on the fascinans part and then on the tremendum part.

For one of us to experience something as holy is to experience it as something attracting one, as pulling one towards it. It is fascinans, fascinating, in the sense that it captivates one's attention and draws one powerfully in its direction; it is ‘the object of search and desire and yearning, and that . . . for its own sake’ (ibid., 32), and it has a ‘potent charm’ through which one has ‘the impulse to turn to it . . . even to make it somehow his own’ (ibid., 31). To experience something as holy is incompatible with experiencing it with indifference, as something that one can take or leave. (One's recognition that some being or thing is holy can coexist with an absence of such motivational pull, but such an absence rules out one's having an experience of it as holy.)Footnote ¹⁴

So far experiencing something as holy seems just like experiencing it as something very much worth pursuing and having. It would be implausible to think, though, that the encounter with the holy is aptly characterized wholly in terms of the experience of attraction, even overwhelming attraction. But it is also part of Otto's thesis – this is the tremendum part – that to encounter something in its holiness is not simply to be attracted to it, but also to be in a way repelled by it.Footnote ¹⁵ A characteristic response to the holy by beings like us is to draw back, to feel the need to get away from it. Otto characterizes it as a fear, or at least analogous to fear, but not a fear that is of any natural sort (ibid., 15). One cannot simply turn the volume up on the sorts of fears one has of bears or heights or public speaking to get to the sort of fear that we have in encountering something as holy.

Otto seems to me to be right on this point, that we cannot capture the repulsion side of the experience of the holy just by increasing the amount of fear that we have of it. But, in keeping with his view that the characterization of the encounter with the holy is starkly limited in terms of how far it can be captured in rational terms,Footnote ¹⁶ he does not seek to go much beyond this negative characterization. We can do better. It seems a mistake to characterize this repulsion side of the holiness response primarily in terms of fear or a fearlike state, where there is some object that is conceived as dangerous to one and one is experiencing coming to be related to it in a certain way as a realistic possibility. Consider what happens to Uzzah in 2 Samuel. Uzzah touches the ark of the covenant because, the author tells us, ‘the oxen stumbled’ (2 Samuel 6–7). The Lord is angered by this, and Uzzah is struck down by the Lord's wrath then and there. David witnesses this event and is now simply extra-scared of God, scared that the supernatural connections between the ark and the Lord make the ark a dangerous thing to have in the midst of his people. I don't think it is correct to say that David now has a deeper sense of the holy simply by becoming more afraid. Nor would it be sufficient for David's fear to be more clearly of something beyond the natural – even if his fears were focused on the ark's supernatural character, or on the fate beyond the grave of those who spur the Lord's wrath by doing something to the ark that displeases God, that would be insufficient to support a claim that David's heightened fear counts as his more deeply encountering the ark and the Lord as holy. Yet that is what we would expect were the holiness response to be aptly characterized as including fear or a fearlike response.

Yet it does seem right to think that fear is relevant here. But I want to say that the fear aspect of the holiness response is secondary, downstream from the more primary repulsion response that is constitutive of responding to something as holy. The primary repulsion response in holiness experience, I suggest, is a sense of being profoundly out-of-place. There is a weaker sense of being out-of-place that is not enough to capture what I have in mind, in which being out-of-place is just finding something to be deeply unfamiliar, unusual, unheimlich, so that one does not have one's bearings, so that one lacks a sense of what to do next. The notion I have in mind is rather that, even if there is some sense of unfamiliarity, one knows enough to know that one should not be here. One's being in contact with the holy being is out-of-bounds, inappropriate. Consider, as examples, Isaiah's response to being called by God to serve as God's messenger (Isaiah 6:1–9), and Peter's response to being summoned by Jesus to join him (Luke 5:6–10). Isaiah and Peter are not merely overcome by the overwhelming strangeness of it all. In spite of the strangeness, they grasp that, in some sense, they should not serve in the intimate roles with respect to the holy one that nevertheless are roles in which it would be very good for them to serve.

That things are in a relationship that they should not be in can itself be a source of distress. It can be worse if one is one of the beings in that relationship. There are two sorts of fear or distress relevant: one is just the horror of being out of place, of being where one should not be; the second, the one that is further downstream, is the fear of what relevantly interested agents might do to get one out of that relationship. In the passage from 2 Samuel, David's fear is characterized in the latter way, as a fear of what God will do to those unauthorized people who get ‘too familiar’ by touching the ark. For a more mundane instance of the former, think of someone who takes very seriously rules of property and thus has a very strong sense of the laws against trespass. If such a person finds herself on another's property without permission – perhaps having wandered onto it while bushwhacking – there will be a sort of primary distress that one is out of place, that one is where one does not belong. There may be, further, some worry about how the property owner or the police will enforce the law. But that is a secondary fear, and one will not understand at all how someone who thinks that ‘a man's home is his castle’ responds to finding him- or herself a trespasser if one tries to characterize that response wholly in terms of fear of some non-normatively characterized evil (e.g. getting shot, getting bitten by a dog) befalling one. The response is not just a fear of these further consequences, but a revulsion against being improperly positioned with respect to the normatively relevant features, with respect to being outside where one is permitted to be.Footnote ¹⁷ Thus this person's fear will not only be that of dealing with an angry property owner, but with a justifiably angry property owner; and the person's fear will not even simply be that of dealing with a justifiably angry property owner, but that of having done – having violated the law of trespass – what justifies the property owner's anger.

The notion that I appeal to is that of unfittingness, that one experiences as profoundly unfitting that one is standing in that relation to the holy in which one is standing, or one vividly apprehends that entering into that relation is an unfitting prospect.Footnote ¹⁸ Imagine another of Uzzah's acquaintances, who had already heard about the ark's history and role in the community's life, but really for whom the ark presented itself as one more piece of very bulky luggage for the tribe to carry around. For this acquaintance, though, the response to Uzzah's death is not just to be more scared of the ark or of God. Rather, the ark's relation to the almighty God becomes more vivid, so that it now seems more clearly out-of-bounds to treat the ark in certain ways; even looking at it seems, somehow, something that is above one's station, because the ark is of God. By contrast with David's response, this response seems to be that of someone who is experiencing these as holy.

So far I have argued that we should characterize the tremendum aspect of Otto's account of the holiness encounter normatively, in terms of a sense of the unfittingness of how one is related to the holy one. It is pretty clear that this will not be sufficient, even combined with the fascinans aspect, to count as a holiness experience. For consider again the person described above who has a keen sense of the norms regarding trespass, while now adding that this person longs to be inside the house on the property, perhaps due to its having been prominently featured on a popular home improvement show. The particular sense that it would be good to be in the house (because it was made famous on television) yet it is unfitting that one be there (because of the norm against trespass) is not an experience of the house as holy. But one could adapt the case so that it seems more analogous to what Otto has in mind: suppose that one longs to be in the house because it is so historically important, but one feels out of place being there due to one's own historical insignificance. Here we see a being-drawn-towards, while taking oneself not to belong there, that is analogous to the holiness encounter. The key difference here seems to me that the unfittingness of being related to the holy being takes a certain shape, rooted in the contrast between the value of the being treated as holy (in this analogous case, the monumentally historical importance of the house) and the value of the being who seems not to be appropriately united to the holy one (in this analogous case, the historically insignificant person standing on monumentally important historical ground). While there are numerous ways that relationships can be unfitting, I take this particular specification of the relationship of unfittingness for granted in what follows in the development of an account of primary holiness.Footnote ¹⁹

As noted above, it is frequently remarked that in the Hebrew Scriptures the term qadosh, translated as ‘holy’, means ‘separate’, ‘set apart’. Characterized simply in terms of otherness, the experience of holiness would simply be weirdness or uncanniness or mysteriousness. I claim that this alternative elaboration of Otto's view, on which one experiences oneself as unfittingly related to the holy one when one stands, or envisions the prospect of standing, in certain relationships with that being, is the relevant gloss. The holy being is encountered as separate, or set apart, in the sense that one sees oneself as unfittingly related to it if one is or were to be related to it in certain ways.

What are these ‘certain ways’? I will postpone discussion of this issue until the next section, but it follows from Otto's account of the dual response to the holy that at least some of those relationships with the holy that one craves are those very relationships that one cannot fittingly stand in with respect to the holy. For one of us – for one of us humans, in this lifeFootnote ²⁰ – to experience something as holy is to experience it as something with which a certain sort of relationship is exceedingly desirable for one and as something such that having that sort of relationship with it is exceedingly unfitting, inappropriate.

I take Otto's characterization of the way that we encounter the holy in terms of the mysterium tremendum et fascinans to be largely correct. He is right that our encounter with the holy is not simply an encounter with something experienced as attractive; that does not capture how the holy is in some ways terrifying, unsettling, how it motivates towards avoidance. And he is right that our encounter with the holy is not simply an encounter with something experienced as terrifying and unsettling; that does not capture why some of us press towards the holy, in spite of its terrifying and unsettling character, rather than just steer clear of it. Otto's view requires correction, or supplementation, though, in his characterization of the way in which the mysterium is tremendum; we should understand our experience as normative through and through, as fundamentally an experience of being unfittingly related to the holy being.

An account of holiness

I am taking Otto's account of the experience of the numinous as a starting point for an account of what it is for God to be holy. How to proceed from an account of what it is to experience something as holy to an account of what it is for that, or any other, being actually to be holy, and how to employ such an account to get clear on what God's holiness amounts to?Footnote ²¹

Begin with the question of how to move from a theory of the holiness experience to a theory of holiness. We obviously cannot make use of Otto's view simply by saying that to be holy is to be experienced in this distinctive way, for there is no reason to think that one can have a holiness experience as Otto describes it only in having a veridical encounter with a holy being. One might have the experience Otto describes when placed in relation to a turnip, if one is given the proper hallucinogenic drugs first, but that does not make the turnip holy. Nor can we say that to be holy is to cause us, or to have the power to cause us, to have experiences of the mysterium tremendum et fascinans variety, as Otto at one point (unguardedly, I think) suggests.Footnote ²² The hallucinogenic drug might cause one to experience the drug itself as holy, but that does not make the drug holy. So the holy is not to be identified either with that which is the cause of a holiness experience or with the seeming object thereof.

The funny is not what one is amused by or what causes one's amusement, and the desirable is not what one desires or what causes one to desire it. The funny is what it is appropriate (in the relevant way!Footnote ²³) to be amused by and the desirable is what is worth desiring. Similarly, one might say, to be holy is not to be experienced in a certain way or to cause a certain kind of experience, but rather to be worthy of being experienced in that way. Holy is a normatively charged notion, as is evident from the fact that the experience of something as holy is a normatively charged experience. So perhaps we need to say something like this: to be holy is to exhibit those features that make it appropriate for us to have the responses characteristic of the holiness experience.

Such an account may be perfectly adequate in getting the extension of the holy right, and in giving us some guidance towards grasping the features of the holy being that do the normative work in making it appropriate to respond in the dual way that Otto describes. But it would nevertheless be misleading in a certain way. There is no reason why we should privilege the responses that are appropriate for us – us humans, in this life – when we are providing an account of the nature of holiness. God would have been holy even if there had not been humans but Martians; it would be weird, though, if holiness in a world with no humans were nevertheless a property keyed to what relationships with God were both attractive and unfitting for humans. Sticking with the actual world, different sorts of being may appropriately experience the holy in different ways. Consider some particular relationship with God, say, a sort of intimate knowledge and communion with that being. It seems plain that the vivid prospect of such a relationship will not, or should not, generate the same response in us ordinary humans as it would in, say, the good angels or the blessed in heaven. That vivid prospect might be experienced by us this-worldly run-of-the-mill humans as mysterium tremendum et fascinans. But the good angels need not experience themselves – need they? – as terrifyingly out of place if standing in some such relationship to God.Footnote ²⁴ Nor must the blessed in heaven experience this relationship as one in which they are terrifyingly out of place; how could they be achieving such fullness of good as heaven is supposed to promise if they are quaking with fear and, correctly, take themselves to be massively inappropriately related to God? So our account of holiness should provide not only the basis for an explanation of why, for some relationships with God, our own human this-worldly experience is just as Otto characterizes it; it should also provide the basis for an explanation of why, with respect the very same type of relationships with God, the good angels’ and the blessed's similar type of encounter with God might not answer to that description.

The thought, then, is that while an account of holiness should make use of the dual response that is appropriate with respect to a holy one, we should allow some variation in how it is appropriate for beings of different sorts to have these responses to particular types of relationship with the holy one. Here is a schema for a view, which makes room for some substantive disagreement on what a more concrete account of holiness should look like. On any acceptable view, part of what it is for a being to be holy is for that being to exhibit a set of features F that makes it overwhelmingly desirable for those of a certain type T1 to stand in some sort of intimate relationship with that holy being, and the more intimate, the more desirable. But part of what it is for a being to be holy is that it be unfitting for at least some beings – those of type T2 – to stand in those intimate relationships with the holy being, and the more intimate, the more unfitting. Given my assumption that we humans do appropriately have responses to the prospect of certain intimate relationships with God to which Otto makes reference in his characterization of holiness encounters, then however we specify T1 and T2, we typical this-worldly humans will have to fall into T1 and T2 with respect to some salient prospective relationships with a holy being, though it may well be that there are some beings (e.g. the good angels, the blessed in heaven) who fall into T1 with respect to some of those salient relationships with a holy being but do not fall into T2 with respect to them.

In filling out this schema a bit, let me first say a bit more about the notions of desirable for and unfitting that I am invoking. On the view I am filling out, to be holy is in part to have those features that make entering into certain relationships with the holy one desirable for some class of beings. One way of reading this is simply as stating that it is desirable, from the point of view of the universe, that some class of beings be related to the holy one in a certain way. That may be true, but it is not what I have in mind. Its being very desirable full-stop that some agent occupy some position does not entail that it is, for that agent, in any way a position very much worth occupying. (If one is posing an innocent threat to a lot of other people, it may be desirable that this person be destroyed in order to eliminate the threat, but it may not be very desirable for that person that he or she be destroyed.) Instead, I mean the desirability notion here to be understood in an irreducibly subject-relative way, that it is good for these beings, that it is for their good to be in a certain relationship with the holy one.Footnote ²⁵ This reading fits more naturally with Otto's characterization, on which one is attracted, on one's own account and for one's own sake, to the presence of the holy.Footnote ²⁶

On the view I am filling out, to be holy is in part to have those features that make entering into a certain relationship with the holy one unfitting for beings of some class. Notions of fittingness have been invoked in trying to give accounts of what is valuable and of what one has reason to do (see, for a survey, Jacobson (Reference Jacobson2011) ). I do not aim to take a particular stand on how to think about the notion of fittingness except to say that fittingness is not reducible to the notion of being good for someone and unfittingness is not reducible to the notion of being bad for someone or not being good for someone. The notion of being good for is a welfare notion – what contributes to well-being – and such welfare notions cannot be analysed in wholly agent-neutral terms (Sumner (Reference Sumner1996), 20–21). As the notion of unfittingness that I am employing here is agent-neutral, it cannot be that the notion of being good for is analysable in terms of being unfitting.

And it would be implausible to claim that unfittingness reduces to not being good for or being bad for. Consider a stock character, the Nazi war criminal who has escaped capture. The sense that it would be unfitting for this Nazi war criminal to spend the remainder of her days in luxury on a remote South American beach is very clear, even if it is granted that spending the remainder of her days in this way is very good for her – better for her than the alternatives, at any rate. It is unfitting that she would be thus related to happiness and contentment. If she were, at long last, to have a recognition of the evil of her deeds, she would recognize that while she enjoyed a good deal of well-being over her post-war life, she shouldn't have; she shouldn't have because it is so profoundly unfitting that she would enjoy this good.Footnote ²⁷

So much, for the moment, for the notions of desirability for and unfittingness of which I am making use in this account of holiness. Let me now fill out the account a bit, beginning with the relevant relationships. Such are relationships of unity with the holy being. There are a number of ways that beings can be unified, but it may be helpful to be guided by those relationships that persons can enter into that can realize their good. Relationships of cooperation, alliance, friendship, and intimate knowledge are relatively uncontroversial ways of being unified with another cognitively and volitionally, and these are paradigmatic ways of being related to others that can constitute our well-being.Footnote ²⁸ There may be other ways that are not so standard: consider, for example, the unity realized between God and a concrete human nature when that nature is assumed by a divine person, as the human nature of Jesus of Nazareth was by the second Person of the Trinity, and the further relationship of sharing a species with that holds between each of us and the second Person of the Trinity due to the Incarnation. But we can stick for the moment with the analogues to more mundane, valuable relationships between humans that can make for greater unity between created persons and divine persons.

The thought, then, is that holiness is that set of features one (‘the holy one’) can exhibit that, for some class of beings (T1), makes unifying with the holy one overwhelmingly desirable, and the more unified, the more desirable. That set of features is also such that, for some class of beings (T2), it can make those beings’ unifying with the holy one unfitting, and the more unified, the more profoundly unfitting. How, then, should we characterize these classes of being T1 and T2?

I suggest that we characterize T1 as the class of welfare subjects who are not intrinsically incapable of having some such unifying relationship with the holy one. If a being is such that some things are good or bad for that being, and it is capable of entering in some way into unifying relationships with the holy one, then entering into such relationships with the holy one is desirable for that being, and the more unified, the more desirable for that being.

With respect to T2, the class of those for whom it can be unfitting to stand in the unification relation to the holy one, I suggest that we begin by noting a certain feature of the holiness experience: that the sense of being out of place in relation to the holy one takes the shape of shame, of seeing oneself as unworthy to be united to a being with the status of the holy one.Footnote ²⁹ This is no doubt due to the sort of unfittingness involved in being related to the holy one, that it is our inferiority that makes being unified with the holy one in certain ways unfitting, inappropriate.

So consider those features of oneself that one can appropriately consider to be an object of one's shame. Paradigmatic of such features is having a wicked character. Paradigmatic of features that are not such is having a good character. Paradigmatic of such features is having performed an evil action of which one has not fully repented and for which one has not made atonement. Paradigmatic of features that are not such is having acted well on some occasion.

When I offer these as paradigms, it may give the impression that all that matters with respect to the fittingness of one's relating to the holy is one's profile as a moral agent. It seems right that being ashamed of oneself as a moral agent is a clear source of one's sense of unfitness to be united to the holy. But I don't mean to say that one's failures as a moral agent are the exclusive sources of appropriate shame in the sight of the holy. First, I want to say of ways in which we fall short of the more general sorts of human goodness that are not themselves moral failures but are due to shortcomings in our profiles as moral agents – the broken friendships due to one's betrayals or mere inattentiveness, the ignorance due to sloth in pursuit of knowledge of important matters, philistinism due to a self-absorbed dismissal of the aesthetic sensibilities of others – that, while these are not themselves moral failures, they are absences of goods resulting from ways in which we are moral failures, and that we can appropriately be ashamed of them. There may be nothing to be ashamed of, in the abstract, about having a broken bicycle. But if it is broken because one is lazy and uninterested in taking the most minor steps to care for good things, then one can be appropriately ashamed of the broken bicycle. And this is more obvious when the goods at stake are more important, genuine aspects of human perfection or well-being. One's lacking the means to feed one's children may not be of itself appropriate cause for shame, but if one lacks such means because of one's vices it can pretty clearly be something to be ashamed of.

Second, I want to say, somewhat more controversially, that we can appropriately be ashamed of shortcomings that are not attributable to our moral profile. This requires us to say that not all appropriate shame is moral shame, but we should be fine with that.Footnote ³⁰ The shame that I have in mind is something like this. Consider how one thinks of oneself in light of one's own abilities relatively early in one's career, maybe as an undergraduate or early graduate student, participating in a workshop led by an accomplished senior figure and populated by a group of very serious, smart, and committed academics – people who are top-notch in their disciplines and intellectually virtuous. There is a sense of shame here, of thinking of oneself as inadequate and not belonging among them, that is not a matter of moral shame at all. It is important that the appropriateness of shame in being related to others is going to be in some ways fixed by the context of the relationship. If one feels non-moral shame by interacting in the imagined seminar when one is a neophyte and the other participants are seasoned, brilliant, intellectually accomplished, and so forth, that is perfectly appropriate, and it might token cluelessness on one's part if one does not have the relevant non-moral shame in that context. But if the context is shifted and one is now encountering these people on a basketball court, where they are, honestly, no better than average compared to the typical players hereabouts, shame would be entirely out of place.

As these cases suggest, the features that make shame of any of these varieties an appropriate response are, at least in the vast run of cases, comparative and contextual: these features make it appropriate for one to be ashamed to be in the presence of, or related in some special way to, persons of some class, in some context of interaction. I might be ashamed to be in the presence of the great scholars in the seminar room while a neophyte; that is one context of intimate interaction in which their being out of my league is vivid and makes me in this way unfit to be there. But I could be in even more intimate relationships with those same people if we were to meet as fellow players in an intense pickup basketball game, or as tenants in a common apartment building, dealing with common problems with common limitations, and these encounters would offer no occasion for shame. One might say that there are exceptions to this context-sensitivity of the appropriateness of shame: if you violate the moral law, one might object, that gives you reason to be ashamed, simply as such, regardless of context. But that seems wrong: if there are some violations of the moral law that aren't all that serious, that all neither-incarnate-God-nor-immaculately-conceived humans have a strong tendency towards, and that all neither-incarnate-God-nor-immaculately-conceived humans engage in, then it would be weird to be ashamed, in the sight of one's fellow neither-incarnate-God-nor-immaculately-conceived humans, to have performed those acts. (One might be ashamed in God's sight, of course, to have done such things.) To say that the appropriateness of shame is comparative and context-sensitive of course does not mean that it is a subjective matter, or entirely dependent on the contingent choices of others; it is just to say that being a source of appropriate shame is implicitly relational, depending on whose presence one might be ashamed to be in, and on the context of the relationship with that other.

So for a being to exhibit (primary) holiness is for that being to exhibit a set of features that (a) make it overwhelmingly desirable for all of those beings capable of certain unifying relationships with that being to stand in those unified relationships with that being and (b) make it unfitting for those who have grounds to be ashamed in the presence of that being – call such beings ‘impurely good’ – to stand in some such unified relationships with that being. I add that the greater the unity in the relationship, the more desirable that relationship is for all of those beings capable of it to enter into it and the more unfitting it is for a being exhibiting a certain level of impurity in its goodness to enter into it. I add also that the greater the level of impurity of goodness, the more unfitting it is for a being exhibiting that impurity of goodness to enter into that unifying relationship with the holy one.Footnote ³¹

Providing an account of primary holiness that allows that beings with different levels of impurity of goodness may appropriately respond to a holy being in different ways makes room for what strikes me as a sensible, perhaps even truistic account of God's holiness in relation to us creatures that we should want our theory of holiness to accommodate. The sensible, perhaps truistic view is this: all of us human beings both would find our good in intimate union with God and yet are deficient in ways that make it unfitting for us to be in that relationship with God. To experience primary holiness is to experience, at least in some limited manner, God's presence in such a way that one experiences both that incredible attractiveness and the utter unfittingness for one to be in that relationship. But the good angels and the blessed in heaven may experience God's holiness differently. They have nothing to be ashamed of save their necessary creatureliness, their dependence and smallness in God's presence. Nevertheless, such shame is inevitable for us creatures; there is no way to ‘shift the context’, as in the above-described case of shift from the seminar to the basketball court, where creatures would not have some occasion for shame in the divine presence.Footnote ³² The relationships with God that we humans in this life would see as attractive and yet for which we are totally unfit may include some relationships that the good angels and the blessed in heaven have been made fit for and so can enjoy without being out of place. There may, though, be relationships with God that exhibit such unity with God that they would always remain unfitting for any creature to enjoy.

God's holiness and God's perfection

We have not yet identified those features that render certain relationships of unity with a holy being overwhelmingly attractive while rendering those relationships of unity massively unfitting for beings like us. As is obvious from the schematic nature of the account I am developing, there is some openness regarding the range of intimate relationships a candidate holy being can have with others in which their dual fascinans and tremendum responses would be appropriate, and at what levels of attractiveness and unfittingness, in order for that being to count as holy, and it is also clear that filling in the schema in different ways would lead to somewhat different views on the features a being would have to exhibit to count as holy. But I think that matters become somewhat clearer when we are considering the God of Scripture, who is not just holy, but holy, holy, holy – meaning, that is, that God's holiness is complete, unsurpassable, absolute.Footnote ³³ While there is some vagueness concerning what features a being would have to exhibit to count as holy, due to the vagueness concerning how attractive a being must be and how unfitting it would be for some impurely good other to be intimately related to that being for that being to count as holy, at least some of this vagueness falls away when we are asked to consider what a being of absolute holiness would have to be like. In particular, I want to say that, given the Scriptural support for a high conception of God's holiness, we have a plausible path from Scripture to an AnselmianFootnote ³⁴ ‘perfect being’ conception of God, a path that is distinct from and in some ways more promising than other such paths from Scripture that one might take. I make no claim here to exhibit an entailment from one to the other, or to prove that the path from Scripture to the ascription of maximal perfection to God presupposes no axiological theses that one could rationally reject. But I do claim that the absolute holiness of God gives some support for an Anselmian conception of God.

The account of holiness defended here holds that to be holy is to exhibit those features that make a dual response appropriate – that any being capable of unifying itself in the relevant way with the holy one should be overwhelmingly drawn to that relationship and that all beings that are in any way impurely good are not fittingly related to the holy one in some of these ways. What are the features that make appropriate that dual response?

It is in principle possible that the features that make the two very different responses appropriate are distinct features of God, necessarily both realized in God. Holiness would then be just a conjunction of the features that make the attraction to God appropriate and the features that make it true that one would be out of place were the course of that attraction to be followed to completion. (We would have to insist, of course, that both of these features be extremely positive; it is not believable that this most central of attributes in terms of divine praise would have as a major component something negative (impossible), neutral, or even only mildly positive.) But while such a view is possible, it is more likely that holiness is more unified than that, that the very features of God that make the attraction an appropriate response makes the realization of the sought union deeply unfitting. What makes union with God so worth seeking is just what makes union with God so out of bounds for beings like us.

I aim to sketch a case that what makes union with God so worth seeking yet so out-of-bounds for beings like us is that God is maximally perfect.Footnote ³⁵ Here is how I will proceed. I will first argue for the sufficiency of maximal perfection to fulfil the dual role set by the concept of holiness, taking note of what it is about maximal perfection that ensures that a maximally perfect being will count as absolutely holy. I will then point out some difficulties with attempts to offer something short of maximal perfection to meet the relevant desiderata for absolute holiness.

For God to be completely, unsurpassably, absolutely holy, it must of course be true that for all actual creatures, union with God must be overwhelmingly attractive for those creatures – not that they must all be actually attracted, but that such union would be overwhelmingly good for those beings. And for God to be completely, unsurpassably, absolutely holy, it must be true that for all actual creatures, for any context of intimate relationship with God, there is a level of intimacy that is too close to be fitting for that being. But it seems to me that for God to be completely, unsurpassably, absolutely holy, it must not be simply contingent whether the creatures that populate a world are such that union with God is so good for those beings yet these unions are just unfitting for them to occupy with respect to God. It must be a necessary truth that creatures are related in this way to God, if God is holy, holy, holy. So I am going to rely on this stronger understanding of God's holiness.Footnote ³⁶

For God to be absolutely holy, it must be that necessarily, for all creatures, union with God would be overwhelmingly attractive for those creatures. And for God to be absolutely holy, it must be that, necessarily, for every creature and for any context of intimate relationship with God, there is a level of intimacy that is too close to be fitting for that being.Footnote ³⁷ Absolute holiness rules out the existence of any context in which a creature is entirely at home with the absolutely holy being. While the neophyte can be at home with the great scholars by shifting the context of relationship from the academic to the athletic context, no creature can be at home with an absolutely holy being by shifting the context of the relationship in some way. A being with whom we can simply seek a different sort of intimate relationship in order to avoid being unfittingly if intimately related to that being is not absolutely holy.

Now, it seems clear that – unsurprisingly – being maximally perfect would suffice to meet the ‘attractiveness’ condition for absolute holiness.Footnote ³⁸ What makes union with God so overwhelmingly attractive is that God is so marvellously great: that God is knowledgeable, and so beautiful, and so loving, and so on. The ways in which relationships with other beings can contribute positively to our well-being seem to depend on ways in which those other beings are good: the ways in which encountering a beautiful artwork or a brilliant sunset adds to our good depends on these exhibiting aesthetic value; the ways in which friendships make our lives better depends on the ways in which the friends themselves are good; and so on. So it is not at all surprising that to be intimately related to a being who is maximally perfect in every way that a being can be perfect would be overwhelmingly attractive, and for every possible creature who can be a subject of well-being.

It also seems to be clear that – unsurprisingly – being maximally perfect would indeed suffice to meet the ‘unfittingness’ condition for absolute holiness. Every possible creature is, I assume, in some way significantly limited with respect to the value that it exhibits, even if there is no upper bound on the value that creatures can exhibit. But since that is so, there will be a massive gap between the excellence exhibited by a maximally perfect being and that exhibited by any creature. Shame is appropriate when, in some relevant context, one party in the relationship is massively outstripped in value by the other party in that relationship. But there is no relevant context of relationship in which a maximally perfect being would not massively outstrip any possible creature. If a creature were overwhelmed by the power of the holy one, noting how obviously it would be unfitting to stand in some sort of cooperative relationship for bringing about some objective – that creature's power being as nothing compared to a maximally perfect being's power – that creature cannot, to avoid that unfittingness, shift to a relationship in which it is seeking knowledge of some deep or important truth in common with the maximally perfect being, or contemplating something beautiful together, without there being some unfittingness to that relationship. For maximal perfection precludes there being any such relationship that a creature could enter into with respect to the holy one in which that creature could be at home with the maximally perfect being. Thus a maximally perfect being would meet the condition on absolute holiness that, for every possible creature and every relevant type of intimate relationship, there is a point at which that creature is not fit to be in that relationship with God.

For those who have reasons to endorse or be sympathetic both to a perfect being conception of God and to a Scriptural conception of God, it is good to see how these two sit easily together: God conceived as maximally perfect is sure to be a God who is, as Scripture attests, absolutely holy. But one might doubt whether holding that God is absolutely holy, even in the sense I have described above, gives us reason to think that God must be maximally perfect. Could not God be absolutely holy but not maximally perfect? Is there anything about the conditions for being an absolutely holy being that precludes a less-than-perfect being from being absolutely holy?

What makes it clear that a maximally perfect being will count as absolutely holy is that maximal perfection ensures the presence of the relevant value gap between the absolutely holy being and creatures, across all relevant contexts of intimate relationship. The presence of that gap ensures that creatures will find their well-being in relationship with that maximally perfect being and that any creature will, in every context of intimate relationship with the maximally perfect being, have occasion for shame, for each such being is in some way not fit for a relationship of full intimacy. So the doubt can be framed as: cannot the relevant value gap be preserved between an absolutely holy being and creatures, even if the absolutely holy being is conceived as not maximally perfect? I will not pretend to offer an impossibility proof that rules out absolute holiness without maximal perfection. But it does seem to me that attempts either rest on dubious axiologies or fail, on closer inspection, to meet the relevant desiderata for absolute holiness.

Consider an axiology that aims to exhibit how to preserve the gap without the absolutely holy being's realizing maximal perfection. Imagine that there is a limit to the value that creatures can reach, that while there can always be a better creature, the value of creatures approaches that limit asymptotically. And now imagine that the absolutely holy one has a value that is far above that limit – by hypothesis, at whatever distance is necessary to preserve the necessary value gap between that absolutely holy being and creatures to make union with the holy one overwhelmingly desirable for creatures while making such union deeply unfitting. That is sufficient for absolute holiness by the desiderata laid down above, yet this proposal regarding the value of the absolutely holy one and creatures does not seem to entail that the absolutely holy one's value is the greatest possible. It looks like it is compatible with this picture that the absolutely holy one's value could have been a little bit greater, which is incompatible with the absolutely holy being's exhibiting maximal greatness.

The prospects of this reply depend both on whether this is indeed a plausible axiology and on whether it meets the dual conditions. Without aiming to reply decisively here, let me indicate a source of concern on each of these matters. The first is that it looks like this sort of axiology requires that the value of being uncreated be lexically ordered over the value of all other great-making features, and this seems to be a not particularly plausible axiology. For what sets the upper limit for the value of creatures on this view looks to be nothing other than their status as creatures. So we have to imagine that no matter how fully some creature realizes value along some other dimension – even if it could outstrip the creator in realizing some dimension of value – the value that the absolutely holy one has in virtue of being uncreated will preserve the relevant value gap. But this seems not particularly plausible.

Think of it this way. Say that A, B, and C are the relevant great-making features other than being uncreated. The doubter proposes that even if the absolutely holy one is not maximally valuable solely with respect to A, B, and C, no creature can even approach its value, because it is limited in value due to its being created. But on the face of it this seems an implausible axiology. As far as the objection goes, a creature could surpass the absolutely holy one with respect to the realization of A, B, and C, and no matter how far it does so, it would still be massively outstripped in value due to its being created.

So, first, this appears to me to be a prima facie implausible axiology. Second, even if we put concerns about the axiology to one side, it seems that it fails to meet both the dual conditions on holiness, the condition of overwhelming attractiveness, and the condition that across contexts relationships of intimacy with the absolutely holy one have a limit beyond which they are unfitting. For, first, if there are great-making features of God that are not maximal, then some possible creature might be, appropriately, only mildly moved or unmoved by God in that respect; there would be aspects of God's own nature to which one could be reasonably nearly indifferent, and that seems incompatible with the overwhelming attractiveness that is essential to absolute holiness. And suppose that an absolutely holy being is conceived as not maximally perfect, and so is limited with respect to some dimension of perfection. It will still be possible for there to be a creature who, though perhaps overall outstripped in value by the absolutely holy one, is not outstripped with respect to this great-making feature, and so with respect to relationships built around that feature, there is no value gap and thus no cause for shame. Remember the neophyte who has cause for shame in the presence of the great scholars, but who has no such cause for shame playing pickup basketball with them. If there is some dimension of greatness that God has that can be approximated by a creature, then that creature can be at home with God with respect to that feature, for they would share a common limitation.

Consider another way of trying to preserve a necessary value gap between God and creatures without ascribing maximal perfection to God. An objector might claim that the perfections that God exhibits, in total, ensure that God is infinite in value, whereas the value of any creature will be finite. Thus we have the needed value gap. But, the objector goes on, having infinite value in virtue of realizing some set of great-making features in some way does not mean that the being who has that infinite value is maximally perfect with respect to the realization of those great-making features. Having infinite value is compatible with there being some metaphysically possible bit of value that is not exhibited, just as being an infinite set of natural numbers is compatible with there being some natural numbers that are not part of that set. So we can imagine that God, while having, say, an infinitely valuable combination of knowledge, power, and benevolence, nevertheless is missing some knowledge or power or goodness, and if God is missing some knowledge or power or goodness, then God is not maximally perfect. We thus can have, on these assumptions, the value gap needed for absolute holiness without maximal perfection.

Again, the prospects of this alternative depend both on whether this is indeed a plausible conception of God's goodness and on whether it meets the dual conditions for holiness. And again, I have concerns on both counts. The first is just that the infinitely-valuable-but-non-maximal conception of God's value seems to be ad hoc and unnatural in comparison with the less gerrymandered maximalist conception of God's greatness. This is not meant to be more than a prima facie consideration. I only suppose that if we ask, ‘Given that God's absolute holiness entails that there is a necessary value gap of a certain sort between God and any possible creature, what shape would be expect that value to take?’, the maximal perfection conception seems a more natural configuration – less in need of further explanationFootnote ³⁹ – than the sort of mottled-but-infinitely-valuable view of the divine goodness that is being considered as an alternative.

The second is that, again, it is unclear whether it even meets the required value gap constraints. For even if the value overall of this being were infinite, it still has patches of non-excellence that can be regarded as unattractive, and there are some contexts of relationship – those built around the patches of non-excellence – in which creatures could be at home with that being, where it would not be unfitting to be intimately related to it. And so it does not seem to me that such a configuration of value meets the demands of absolute holiness.

So, on the face of it, it looks like there is a plausible route from Scripture to Anselmian perfect being theology that takes God's holiness as a starting point.Footnote ⁴⁰ What makes this route particularly promising, I say, are two things. First, holiness seems literally ascribed to God in Scripture, and in a supreme form. It would be an entirely implausible interpretation of Scripture that would suggest that God isn't really meant to be characterized as holy, or that the holiness of God is in some way less than complete. Second, holiness is the right sort of property to serve as a bridge to the perfection of God. It is a normative property, that God is to be responded to in a certain way, and by all possible non-divine beings. But if God is to be responded in those ways, we can ask: what are those features that God has that make that normative response fitting? The normative character of holiness, that to be holy is to be something that is to be responded to in a certain way, gives it a second-order character that enables us to show that exhibiting it entails exhibiting all of the first-order good-making features that together constitute maximal perfection. So while we should say that the first-order features that make for divine perfection are metaphysically prior to God's holiness – we explain why God is holy, holy, holy in terms of God's maximal perfection – we can treat the affirmation of God's supreme holiness as the, or an, evidential basis for affirming God's perfection. In the order of being, God's perfection is prior to God's holiness, but in the order of knowing, the order could be reversed.

While there is no incompatibility between this way of trying to build a bridge between Scripture and perfect being theology and other ways, there is reason to think that this way of proceeding avoids some of the difficulties that the alternatives face. One might look in Scripture for evidence that maximal perfection is explicitly ascribed to God there. But the evidence is extremely thin, as Diller (Reference Diller1999, 234) has noted. One might, as Leftow (Reference Leftow2011) proposes, look in Scripture for evidence that the various distinct perfections that an absolutely perfect being would have to exhibit (omnipotence, omniscience, and so forth) are all exhibited by or ascribed to God in Scripture. But there are multiple obstacles here. First, the evidence with respect to some great-making features is mixed: though God is described as all-knowing, nevertheless God is often described as not knowing or needing to find out certain things; and though God is described as all-powerful, nevertheless God is sometimes described as being able to do even more than what it is possible to do. We need some good reasons, then, to employ interpretative principles for taking some of these ascriptions to God non-literally while we take some of them – exactly those that attribute perfections to God – as the sober truth, and it is hard to find such principles without presupposing the aim of ascribing literal perfection to God.Footnote ⁴¹ Second, it is hard to see why one would think, even if some small set of perfections were clearly ascribed to God in Scripture, that that would justify holding that God has all of the perfections necessary to be an absolutely perfect being. That one cannot imagine others, as Leftow (Reference Leftow2011, 108) says he is unable to do, does not seem much of a justification; the richness of what is needed for divine perfection may far surpass what we can imagine, and we are given in Scripture only the tiniest bit of the divine story, and, after all, from human authors who have the same limitations of imagination that Leftow has.

What would be a better alternative in bridging the gap from Scripture to perfect being theology is to make use of an unambiguously ascribed second-order feature of God the ascription of which makes sense only if God has all of the first-order attributes that make for absolute perfection, whatever those turn out to be. I claim that holiness is such a feature, for it is unambiguously ascribed to God, and in its superlative form, and because it is to be characterized normatively, we can ask: what sort of being would merit the sort of response that a superlatively holy being must merit? As only a maximally perfect being — whatever the particular features are that make for maximal perfection – could merit the responses that the superlatively holy being must merit, and superlative holiness is clearly ascribed to God in Scripture, the God of Scripture must be maximally perfect.Footnote ⁴²