Introduction
The topic of divine omniscience is well-trodden ground, with philosophers and theologians having asked virtually every question there is to ask about it: what does it mean to say that God – the greatest actual being if not the greatest possible being – is omniscient? Is God's omniscience logically compatible with free will? Does God's omniscience include middle knowledge? Does it include knowledge of propositions involving indexicals? Is omniscience even logically possible? And so on.Footnote 1 The questions regarding God's omniscience to be addressed here are as follows. First, is omniscience best understood as maximal propositional knowledge along with maximal experiential knowledge (both to be defined shortly)? I argue that it is. Second, is it possible for God to be essentially omniscient? I argue that it is not.
Key concepts and assumptions
The key concepts of this article are propositional knowledge, experiential knowledge, and omniscience. Each will be discussed in turn.
Regarding propositional knowledge and experiential knowledge, a thorough discussion of the natures of these concepts as well as what, precisely, distinguishes them is beyond the scope of this article.Footnote 2 The following, then, will have to suffice.
By ‘propositional knowledge’, I mean knowledge of the truth-values of propositions. Statements in the form ‘S knows the truth-value of p’ and ‘S does not know the truth-value of p’ – where ‘S’ stands for an individual possessing the capacity for knowledge and ‘p’ stands for a proposition – are to be understood as statements of propositional knowledge. For example, ‘Joe knows the truth-value of “Barack Obama is President of the United States”’ is a statement of propositional knowledge.
The previous example may also be used to introduce an important distinction: the distinction between temporally variant and temporally invariant propositions.Footnote 3 Temporally variant propositions are propositions whose truth-values change with time, while temporally invariant propositions are propositions whose truth-values do not change with time. The understanding of propositional knowledge at work here is knowledge of the truth-values of both temporally variant and temporally invariant propositions.
With this understanding of propositional knowledge in mind, by ‘maximal’ propositional knowledge I mean knowledge of all the truth-values of all the propositions – temporally variant or invariant – it is logically possible to know.Footnote 4
By ‘experiential knowledge’, I mean knowledge of things other than propositions, such as knowledge of what it's like to surf, of what it's like to lead a platoon into battle, of what it's like to be in love, and so on. Statements in the form ‘S knows what it's like to x’ and ‘S does not know what it's like to x’ – where ‘x’ stands for an activity, the experiencing of an emotional state, the experiencing of a state of affairs, or anything else that may be known experientially – are to be understood as statements of experiential knowledge. That said, by ‘maximal’ experiential knowledge, I mean knowledge of all that it is logically possible to know experientially.
As for ‘omniscience’, there are at least two understandings of the concept. One understanding of ‘omniscience’ is maximal propositional knowledge; the other is maximal propositional and maximal experiential knowledge (maximal propositional/experiential knowledge, for short).Footnote 5 It is the latter understanding of ‘omniscience’ that is under consideration and will be defended here (though the former understanding will be given due consideration). Accordingly, unless otherwise noted, when I refer to God's omniscience, I refer to his maximal propositional/experiential knowledge. Moreover, given that God's omniscience is one of his essential properties, statements such as ‘God is omniscient’ are meant to convey that God not only possesses maximal propositional/experiential knowledge but has always possessed it.
Having addressed some of the key concepts of this article, let us turn our attention to two of the key assumptions. Both assumptions will be examined critically as the article progresses, but it should be noted now that the first may be derived from the understanding of ‘omniscience’ under consideration (maximal propositional/experiential knowledge). The key assumptions are as follows:
A1: Necessarily, if S is omniscient, then there is not a time t during S's existence at which (i) S does not know the truth-value of at least one proposition or (ii) S does not know what it's like not to know the truth-value of propositions.
A2: Necessarily, if S knows what it's like not to know the truth-value of propositions, then there is a time t during S's existence at which S does not know the truth-value of at least one proposition.
(By ‘necessarily’, I have metaphysical necessity in mind; while by a ‘time’ t, I mean physical if not metaphysical time.) Objections to these assumptions will be considered in due course.
God cannot be omniscient: the argument
It is not possible for God to be essentially omniscient – that is, it is not possible for God to possess maximal propositional/experiential knowledge and always to have done so, or so I shall argue. For, briefly, if God has always possessed maximal propositional knowledge, then there is a time t at which God does not possess maximal experiential knowledge; while if God has always possessed maximal experiential knowledge, then there is a time t at which God does not possess maximal propositional knowledge.
To see this, consider the proposition, p: ‘God knows what it's like not to know the truth-value of propositions’. If p is true, then there is a time t during God's existence at which God does not know the truth-value of at least one proposition and, thus, God is not omniscient. To see this clearly, consider the following:
P1: God knows what it's like not to know the truth-value of propositions. (p)
P2: Necessarily, if God knows what it's like not to know the truth-value of propositions, then there is a time t during God's existence at which God does not know the truth-value of at least one proposition. (A2)
C1: There is a time t during God's existence at which God does not know the truth-value of at least one proposition. (From P1 and P2)
P3: Necessarily, if God is omniscient, then there is not a time t during God's existence at which (i) God does not know the truth-value of at least one proposition or (ii) God does not know what it's like not to know the truth-value of propositions. (A1)
C2: God is not omniscient. (From C1 and P3)
So, again, if p is true, then there is a time t during God's existence at which God does not know the truth-value of at least one proposition and, thus, God is not omniscient.
If, on the other hand, p is false, then, once again, God is not omniscient. To see this clearly, consider the following:
P1: God does not know what it's like not to know the truth-value of propositions. (∼p)
P2: Necessarily, if God is omniscient, then there is not a time t during God's existence at which (i) God does not know the truth-value of at least one proposition or (ii) God does not know what it's like not to know the truth-value of propositions. (A1)
C: God is not omniscient. (From P1 and P2)
So, if p is false, then there is a time t during God's existence at which God does not know what it's like not to know the truth-value of propositions and, thus, God is not omniscient.
To sum up: the understanding of God's omniscience under consideration is maximal propositional/experiential knowledge. Now, either God knows what it's like not to know the truth-value of propositions or he does not. Either way, I have argued, it follows that God is not – indeed, cannot be – omniscient.
Possible solutions
There are two possible solutions I'd like to discuss here, and they may be distinguished in the following way: one solution retains the understanding of ‘omniscience’ as maximal propositional/experiential knowledge (for the most part, anyway), while the other foregoes this understanding and restricts ‘omniscience’ to maximal propositional knowledge. (I will refer to this latter understanding as the ‘restricted sense of omniscience’.) Each solution will be examined in turn.
Regarding the first solution, one could modify the understanding of God's omniscience by construing it in terms of maximal propositional and nearly maximal experiential knowledge, with the latter meant to convey that God has experiential knowledge of everything save for what it's like not to know the truth-value of at least one proposition (at a time t). An immediately identifiable problem with this solution, however, is that one could just as easily modify the understanding of God's omniscience by construing it in terms of maximal experiential and nearly maximal propositional knowledge, with the latter meant to convey that God has propositional knowledge of everything save for one proposition (at a time t). And it's not at all clear which solution is preferable. Whatever it is, it will have to be the one that, among other things, renders God the greater being, since God is understood here to be the greatest actual being if not the greatest possible being. So, assuming both propositional knowledge and experiential knowledge are great-making qualities (an assumption to be addressed shortly), the question is: which God is greater? The God who possesses maximal propositional and nearly maximal experiential knowledge, or the God who possesses maximal experiential and nearly maximal propositional knowledge? For present purposes, I'll simply submit that I haven't the faintest idea which God is greater. What's more, I do not see how anyone could have much confidence in any answer that may be proffered, as there are so many other difficult issues one would need to address before one can adequately answer this question.
To begin with, one would need to address the extent to which propositional knowledge and experiential knowledge are great-making qualities as well as the issue of whether the great-making quality of propositional knowledge is greater than the great-making quality of experiential knowledge (or vice versa), among other things. And doing these things would be no mean feat, as they would involve answering questions such as:
• Is propositional knowledge equally great-making regardless of what is known propositionally, or is some propositional knowledge more great-making than other propositional knowledge?
• Is experiential knowledge equally great-making regardless of what is known experientially, or is some experiential knowledge more great-making than other experiential knowledge?
• To what extent is knowing the truth-value of, say, ‘Some dogs are blind’ a great-making quality?
• To what extent is knowing what it's like, say, to snowboard a great-making quality?
• Is knowing the truth-value of ‘Some dogs are blind’ a greater great-making quality than knowing the truth-value of, say, ‘Some people dislike their siblings’?
• Is knowing what it's like to snowboard a greater great-making quality than knowing what it's like, say, to love someone?
• Is knowing the truth-value of ‘Some dogs are blind’ a greater great-making quality than knowing what it's like to snowboard?
• What renders God greater: possessing maximal propositional knowledge and nearly maximal experiential knowledge save for knowing what it is like to snowboard, or possessing maximal experiential knowledge and nearly maximal propositional knowledge save for knowing the truth-value of ‘Some dogs are blind’?
With questions such as these to be answered, I simply do not see how anyone could have much confidence in any answer that may be proffered to the question: which God is greater – the God who possesses maximal propositional and nearly maximal experiential knowledge, or the God who possesses maximal experiential and nearly maximal propositional knowledge?
Regarding the second solution, one could reject the view that God's omniscience is best understood in terms of maximal propositional/experiential knowledge and, instead, adopt the restricted sense of omniscience. In so doing, one would be rejecting A1, since A1 claims that if S is omniscient, then there is not a time t during Ss existence at which S does not know what it's like not to know the truth-value of propositions. But restricting God's omniscience to maximal propositional knowledge produces problems of its own.
First, unless there is good reason to restrict the concept of omniscience to maximal propositional knowledge, it seems the concept of omniscience should include all varieties of knowledge. The question, then, is whether there is good reason to restrict the concept of omniscience to maximal propositional knowledge. If there is, it has yet to be sufficiently articulated, as philosophers who restrict it in this way usually do so by stipulation. In his article ‘The Divine Attributes’, Nicholas Everitt epitomizes this approach of restricting the concept of omniscience to maximal propositional knowledge when he writes: ‘Must an omniscient being know everything? A first qualification is to set aside the so-called “knowledge how to”, and knowledge by acquaintance, and to restrict omniscience to propositional knowledge’ (Everitt (Reference Everitt2010), 81).Footnote 6
Second, there is at least one reason – and a compelling one at that – to include experiential knowledge in the understanding of God's omniscience. Consider that if God's knowledge were restricted to propositional knowledge, then it would be possible for there to be a being greater than God, namely, a being who shares all of God's great-making qualities – including maximal propositional knowledge – plus the great-making quality of experiential knowledge. But, God is understood here to be the greatest actual being if not the greatest possible being. Assuming a greatest possible being is indeed possible, then God must possess experiential knowledge in addition to maximal propositional knowledge. And, it would be odd if not arbitrary not to count God's experiential knowledge as constitutive of his omniscience.
Now, someone might object to my contention that such a being would be greater than God on the grounds that experiential knowledge is not a great-making quality. But, arguably, this would be a mistake. As Joshua Hoffman and Gary Rosenkrantz state, ‘Whether a particular quality should be included in a set of great-making qualities depends on the nature of the pertinent category’ (Hoffman & Rosenkrantz (Reference Hoffman and Rosenkrantz2002), 15). In the case of God, the pertinent category is that of entity and, as such, ‘the relevant great-making qualities pertain to an entity's worthiness for worship and moral admiration’ (ibid., 16). (Lest there be any misunderstanding, ‘worship’ means (minimally) ‘reverence’ or ‘adoration’. I will use ‘reverence’ or ‘adoration’ throughout to avoid confusion.) And, there is reason to think that experiential knowledge makes an entity more worthy of reverence and moral admiration than it otherwise would be.
To begin with, it's hard to see why having maximal propositional knowledge would contribute to an entity's worthiness for reverence and moral admiration – as most proponents of the restricted sense of omniscience believe it does – while having experiential knowledge would not. As indicated above, experiential knowledge differs from propositional knowledge with respect to the kinds of things that are known (and, perhaps, the way in which these things come to be known). In the case of propositional knowledge, what is known are propositions; while in the case of experiential knowledge, what is known are things other than propositions. But, in both cases something is known, and it's hard to believe that whether knowledge is a great-making quality turns entirely on the kinds of things that are known – whether what's known is propositional in nature or non-propositional in nature – and not that something is known as well.
To motivate this point, consider a possible entity, E, which may exist in one of two ways: E possesses propositional knowledge of, say, leadership (i.e. what it means to be a leader, what the capacity to lead involves, etc.) but no experiential knowledge of leadership whatsoever, or E possesses propositional knowledge of leadership plus experiential knowledge of leadership. In the first scenario, E is, if you will, ‘book smart’ when it comes to leadership but not ‘street smart’, while in the second scenario, E is both ‘book smart’ and ‘street smart’. That the kinds of things that are known about leadership experientially are different from the kinds of things that are known about leadership propositionally seems insufficient when it comes to determining whether E's overall greater knowledge of leadership in the latter case makes E more worthy of reverence and moral admiration. What seems relevant as well is that something is known about leadership, propositionally or otherwise. Furthermore, E's possession of experiential knowledge of leadership seems to make E greater – more worthy of reverence and moral admiration, all things considered – than E otherwise would be.
One might object to the preceding by claiming that experiential knowledge of leadership makes E greater qua leader, but not qua entity. But this is implausible, particularly if one believes that propositional knowledge of leadership makes an entity greater than it otherwise would be. To see this, consider God again. As omniscient, God has maximal propositional knowledge, which includes propositional knowledge of leadership, among other things. (Without propositional knowledge of leadership, after all, God would not have maximal propositional knowledge and, in turn, God would not be omniscient.) And God's maximal propositional knowledge makes him greater qua entity than he otherwise would be, or so proponents of the restricted sense of omniscience believe. Yet, if God's having maximal propositional knowledge – which includes propositional knowledge of leadership – renders him greater qua entity than he otherwise would be, why wouldn't his having knowledge of what it's like to be a leader likewise make him greater qua entity than he otherwise would be? There seems to be no principled reason for thinking it would not. What's more, there is at least one compelling reason to think that it would: we have more reverence and moral admiration for those individuals who have led and thereby know what it's like to lead than for those who simply have propositional knowledge about leadership. Consider, for example, the reverence and moral admiration we have for the likes of George Washington and Winston Churchill versus the likes of newly minted graduates of West Point or the Naval Academy. And the reverence and moral admiration we have for individuals who have led and thereby know what it's like to lead is qua entity, not simply qua leader – though, to be sure, our reverence and moral admiration for them qua leader contributes to our reverence and moral admiration for them qua entity.
(It should be added that having reverence and moral admiration for E qua entity in virtue of E's experiential knowledge of what it's like to lead does not entail having reverence and moral admiration for E qua entity full stop. In some cases, in addition to the great-making quality of experiential knowledge of leadership, E possesses a whole host of bad-making qualities which renders E – qua entity – unworthy of reverence and moral admiration, all things considered.)
To sum up: one possible solution to my argument for the impossibility of divine omniscience is to modify the understanding of omniscience by construing it in terms of maximal propositional and nearly maximal experiential knowledge. A problem with this solution is that one could just as easily modify the understanding omniscience by construing it in terms of maximal experiential and nearly maximal propositional knowledge, and it's not at all clear which solution is preferable since, among other things, it is not at all clear which possibility renders God the greater being. Another possible solution is to reject A1 and, instead, adopt the restricted sense of omniscience. But, if experiential knowledge is a great-making quality – as arguably it is – then understanding God's omniscience in the restricted sense would entail that God is not the greatest possible being (assuming a greatest possible being is indeed possible).
Objections
One objection has already been addressed above, namely, the objection to A1. Three more objections I'd like to discuss pertain to (1) A2, (2) whether all experiential knowledge is great-making, and (3) whether there really is something it's like not to know the truth-value of a proposition.
On A2
As for objections to A2, I will simply cut to the chase and consider the implications of A2 being false. If A2 is false, then it is possible for S to know what it's like not to know the truth-value of propositions despite S having always known all the truth-values of all the propositions it is logically possible to know. But it's hard to see how this could be the case. To begin with, S couldn't know what it's like not to know the truth-value of propositions by not knowing the truth-value of a proposition it is logically possible to know since, ex hypothesi, S has always known all the truth-values of all the propositions it is logically possible to know. If S is to know what it's like not to know the truth-value of propositions, then, it must be by not knowing the truth-value of a proposition that it is logically impossible for S to know. So, what might be an example of such a proposition?
One might think that contradictory propositions are propositions the truth-values of which it is logically impossible for S to know. But one would be mistaken in thinking so, as it is logically possible for S to know the truth-value of contradictory propositions, according to both classical and non-classical logics.Footnote 7
One might also think that nonsensical propositions – such as ‘The color of C major is anger’ – are propositions the truth-values of which it is logically impossible for S to know. But, once again, one would be mistaken in thinking so, for nonsensical propositions aren't genuine propositions; rather, they are pseudo-propositions.
One might think that propositions involving indexicals are propositions the truth-values of which it is logically impossible for S to know. Take the proposition ‘I am typing at this moment’ with the ‘I’ referring to this author, Rob Lovering. One might think it is logically impossible for S to know the truth-value of this proposition for all Ss that are not numerically identical with the individual referred to by ‘I’. But, again, one would be mistaken in thinking so, as it is logically possible for Ss not numerically identical with the individual referred to by the ‘I’ to know the truth-value of this proposition. It is logically possible, for example, for someone who is watching me type at this moment to know the truth-value of this proposition, though he/she is not numerically identical with me.
Finally, one might think that propositions contained in a Wittgensteinian private language are propositions the truth-values of which it is logically impossible for S to know for all Ss not identical with the possessor of the private language in question. Now, if private languages were possible, then propositions contained therein may be propositions the truth-values of which it is logically impossible for S to know for all Ss not identical with the possessor of the private language. The question, then, is: are private languages possible? I, for one, agree with Wittgenstein that they are not. If this is the case, then one cannot know what it's like not to know the truth-value of propositions by not knowing the truth-value of a proposition contained in the private languages of others.
But even if private languages are possible and propositions contained in one are propositions the truth-values of which it is logically impossible for S to know for all Ss not identical with the possessor of the private language in question, it's not clear how Ss not identical with the possessor of the private language could know what's it's like not to know the truth-value of such a proposition. For S to know what's it's like not to know the truth-value of a proposition contained in someone else's private language, S would have somehow to entertain that proposition mentally. But how could that ever occur if said proposition is constitutive of someone else's private language? I simply don't see how this could ever occur.
So, again, what might be an example of a proposition the truth-value of which it is logically impossible for S to know? I, for one, cannot think of one. And, if there aren't any, then S cannot know what it's like not to know the truth-value of propositions by not knowing the truth-value of a proposition it is logically impossible to know. Moreover, since S cannot know what it's like not to know the truth-value of propositions by not knowing the truth-value of a proposition it is logically possible to know, it follows that S cannot know what it's like not to know the truth-value of propositions by not knowing what it's like not to know either logically possible or logically impossible propositions. Given this, I do not see how S could know what it's like not to know the truth-value of propositions.
A few more possibilities regarding how S could know what it's like not to know the truth-value of propositions despite S having always known all the truth-values of all the propositions it is logically possible to know are worth considering here.Footnote 8
First, perhaps S could know what it's like not to know the truth-value of propositions by way of directly perceiving the content of the consciousness of someone else who, at a time t, does not know the truth-value of at least one proposition. So, perhaps God could know what it's like not to know the truth-value of propositions by directly perceiving the content of Barack Obama's consciousness who, at a time, does not know the truth-value of at least one proposition. But, it is not at all clear that, in so doing, God knows what it's like not to know the truth-value of propositions: is it that God knows what it's like not to know the truth-value of propositions, or that God knows what it's like for Barack Obama not to know the truth-value of propositions? As with a question raised above, I see no way of settling this dispute with any confidence. Perhaps, one might argue, that God's knowing what it's like for Barack Obama not to know the truth-value of propositions suffices for God to know what it's like not to know the truth-value of propositions. Perhaps. Then again, perhaps not. And, once again, I see no way of settling this dispute with any confidence.
Second, perhaps S could know what it's like not to know the truth-value of propositions by somehow borrowing memories from someone else who, at a time t, did not know the truth-value of at least one proposition. So, perhaps God could know what it's like not to know the truth-value of propositions by borrowing Obama's memory of not knowing the truth-value of at least one proposition at a time t. But, as with the preceding possibility, it is not at all clear that God knows what it's like not to know the truth-value of propositions rather than that God knows what it's like for Barack Obama not to know the truth-value of propositions.
Finally, perhaps S could know what it's like not to know the truth-value of propositions by simply imagining what it's like not to know the truth-value of propositions. But this will not do. For S's imagining what it's like not to know the truth-value of propositions to be a case of knowing what it's like not to know the truth-value of propositions, S must have reasons for believing that his imagining of an experiential event is accurate.Footnote 9 And it's very difficult to see how S could have such reasons if he has always known the truth-value of all the propositions it is logically possible to know. Indeed, the reasons would have to be a priori in nature, and it's very difficult to see how one could have a priori reasons for thinking that one's imagining of an experiential event is accurate.
Is all experiential knowledge great-making?
Previously, I defined maximal experiential knowledge as knowledge of all that it is logically possible to know experientially. Moreover, I have argued that having experiential knowledge is a great-making quality. But maximal experiential knowledge – if it's truly to be maximal experiential knowledge – must encompass knowledge of what it's like to do horrible things, such as what it's like to murder someone. Given this, a question naturally arises: is all experiential knowledge great-making? For example, is knowing what it's like to murder someone a great-making quality? Some think not. As one philosopher put it: ‘Consider knowing what it's like to murder someone in cold blood, or knowing what it's like to commit a heinous sexual crime … None of these strikes me as great-making qualities – in fact they strike me as worse-making qualities.’Footnote 10
(Before submitting my reply to this objection, it's worth noting here that if knowing what it's like to murder someone is always a worse-making quality, then a problem presents itself for some theistic philosophers' understanding of ‘omniscience’. Linda Zagzebski, for example, holds that omniscience includes what she calls ‘omnisubjectivity’. As she describes this property, ‘An omnisubjective being would know what it is like to be you, as well as what it is like to be your dog, the bats in the cave, the birds, the fish, the reptiles, and each human being yet to be born’ (Zagzebski (Reference Zagzebski and Kvanvig2008), 245). If omniscience does indeed include omnisubjectivity, then God knows what it's like to be Jack the Ripper, John Wayne Gacy, and Jeffrey Dahmer, among many other murderers. In turn, God knows what it's like to murder someone. If, then, omniscience does indeed include omnisubjectivity, and if knowing what it's like to murder someone is always a worse-making quality, then God possesses this and a whole host of other worse-making qualities. God, then – an omniscient and perfectly good being – is not a logically possible being. Zagzebski seems aware of and troubled by this implication, for she later writes: ‘I will not address the issue of whether omnisubjectivity is compatible with the other traditional divine attributes’ (ibid.).)
But even if knowing what it's like to murder someone or to commit a heinous sexual crime are worse-making qualities, it does not follow that experiential knowledge is not great-making. All that follows is that some experiential knowledge is not great-making. If this is the case, then the concept of maximal experiential knowledge, when applied to God – a perfectly good being – would need to be qualified. When applied to God, the concept of maximal knowledge would have to be limited not only to that which is logically possible to know experientially but to that experiential knowledge which is great-making. (We could refer to this as ‘maximal great-making experiential knowledge’.) Duly qualified, then, God would know what it's like to surf and to lead, but he would not know what it's like to murder someone or to commit a heinous sexual crime.
With the preceding in mind, the question for present purposes is: is knowing what it's like not to know the truth-value of propositions the kind of experiential knowledge that is great-making or the kind of experiential knowledge that is not great-making? At first glance, one might think it is the latter. But there is reason to think that it is the former.
To see this, let's begin with a claim that I believe most people would find uncontroversial: being good to others involves, among other things, being sympathetic to them when sympathy is called for. And, in some cases, sympathy for others is called for when the other does not know the truth-value of a given proposition. For example, suppose that, through no fault of his own, S does not know the truth-value of the proposition ‘The earth is roughly 4.5 billion years old’. In this case, being good to S would involve having sympathy for him, among other things – it would not involve, for example, ridiculing him for being an ignoramus. Since having sympathy for another involves sharing in (to some extent) the other's psychological state, knowing what it's like not to know the truth-value of propositions allows for a greater capacity for one to be good to others, for it allows for one to sympathize with them in such situations. Accordingly, knowing what it's like not to know the truth-value of propositions is a great-making quality, for knowing what it's like not to know the truth-value of propositions is inextricably linked with a greater capacity for one to be good to those who do not know the truth-values of propositions.
To see this more clearly, consider someone, S, who does not know the truth-value of a proposition, p, at a time, t1. At t1, then, S knows what it's like not to know the truth-value of propositions. Now, suppose at a later time, t2, S comes to know the truth-value of p. In so doing, S acquires a greater understanding of himself, for he now knows not only what it was like for him not to know the truth-value of p at t1, but what it is like for him to know at t2 the truth-value of p having previously not known the truth-value of p. And this greater understanding of himself allows for a greater capacity for S to be good to himself as well as to others. For this greater understanding of himself serves to encourage S to have greater acceptance of himself with respect to what he does not know propositionally. In turn, this greater understanding serves to encourage S to have greater sympathy for others with respect to what they do not know propositionally. By knowing what it's like not to know the truth-value of propositions, then, S is a greater being – a being more worthy of reverence and moral admiration – than he otherwise would be.
Of course, even if the preceding is correct, it may be that God can have such sympathy without knowing what it's like not to know the truth-value of propositions. So, can he? Can God have such sympathy without knowing what it's like not to know the truth-value of propositions? I don't see how he could. For, again, having sympathy for another involves sharing in (to some extent) the other's psychological state. In this case, having sympathy for someone who does not know the truth-value of p involves knowing what it's like not to know the truth-value of p (or, at least, knowing what it's like not to know the truth-value of one proposition or another). But, given A1, there is no proposition of which God knows what it's like not to know its truth-value. And, as I argued above, I do not see how anyone could have much confidence in proposals that God could know what it's like not to know its truth-value of a proposition by direct awareness of the content of another's consciousness, by borrowing another's memories, or by imagining what it's like.
An unsavoury dilemma, then, presents itself – unsavoury, that is, for theistic philosophers. For, either God can have such sympathy or he cannot and, either way, the logical possibility of an essentially omniscient and essentially perfectly good being is challenged. If God cannot have such sympathy, then he is not able to be good to others in this way; while, if God can have such sympathy, then he knows what it's like not to know the truth-value of propositions. God, then, cannot be essentially perfectly good and essentially omniscient.
Is there something it's like not to know the truth-value of a proposition?
In short, yes, there is. To be sure, there isn't just one thing it's like not to know the truth-value of a proposition, just as there isn't just one thing it's like to be in love, or to surf, or to lead a platoon into battle. But there is something it's like not to know the truth-value of a proposition. What, exactly, it's like varies from proposition to proposition, but that there is something it's like not to know the truth-value of a proposition is, it seems to me, all but incontrovertible. Consider, for example, the proposition ‘There is intelligent life on another planet’. When I reflect upon this proposition and, specifically, when I become aware of the fact that I do not know its truth-value, a qualitative experiential shift occurs, one perhaps best described as from a harmony of sorts to a disharmony of sorts. It isn't a dramatic shift, to be sure; it's not paralyzing, or depressing, or uplifting, or what have you. But it is just that: a qualitative experiential shift. And so it is with other propositions of which I do not know the truth-values, at least in many cases.
Does everyone undergo this qualitative experiential shift when they become aware of their not knowing the truth-value of a proposition? I imagine so, though I don't know for certain. But I, for one, do. Indeed, not to be overly clever, but when I reflect upon the proposition ‘Everyone undergoes this qualitative experiential shift when they become aware of their not knowing the truth-value of a proposition’ and become aware of the fact that I do not know its truth-value, I undergo this very qualitative experiential shift. And, importantly, only one instance of there being something it's like not to know the truth-value of a proposition is needed to run the argument above.
Conclusion
I have argued here that God is not – indeed, cannot be – omniscient. Specifically, I have argued that it is not possible for God always to have possessed maximal propositional knowledge as well as maximal experiential knowledge. If God has always possessed maximal propositional knowledge, then there is a time t at which God does not possess maximal experiential knowledge. If, on the other hand, God has always possessed maximal experiential knowledge, then there is a time t at which God does not possess maximal propositional knowledge. Indeed, even when maximal experiential knowledge is qualified – being limited to all that it is logically possible to know experientially and to that experiential knowledge that is great-making – the logical possibility of God, as an essentially omniscient and essentially perfectly good being, is challenged.
