Divine Eternity and Libertarian Free Will
March 12, 2006 — 20:13

Author: Kevin Timpe  Category: Divine Foreknowledge  Comments: 31

I used to reject the Boethian understanding of eternity (i.e., the claim that God is atemporal) because I thought there were good objections to the view. And since the truth of divine simplicity would entail the truth of divine eternity (though it isn’t clear that the entailment goes the other way), these objections would also be objections to divine simplicity. But I’m becoming less and less convinced that the concept of divine eternity is problematic. One of the objections that I used to think was problematic involves our having free will. One way of spelling out the objection in more detail is as follows:
Knowledge Objection:
If agents have libertarian free will, then it is not the case that all of their actions are determined by antecedent causes outside of their control (including God). But if God isn’t the ultimate cause of an agent’s action, then He does not know about that action in virtue of causing it. Instead, God actually has the knowledge that He does because of a free agent’s action. Thus, God is dependent on the agent’s action for His knowledge. However, if God’s knowledge is dependent on another agent’s action, then that agent’s action causally affects God. However, the doctrine of eternity rules out that God can be causally affected by anything outside of Himself, since to be causally affected it a kind of change, and change requires time.

In other words, the only way for God to have knowledge of our actions is to determine those actions. Garrigou-Lagrange, for instance, voices a version of this objection when he wrties that “the knowledge of God is the CAUSE of our free determinations, or else it is CAUSED by them.” Garrigou-Lagrange rejects the latter disjunct, and so favors the former–and theological determinism. I had a slightly different take. Given that I think that we are justly punished, but that we are deserving of punishment only if we are morally blameworthy for something, and that I think that determinism isn’t compatible with moral responsibility, the Knowledge Objection seemed to provide a good reason for rejecting divine eternity. But now I think that the objection is a bad one, becauses it confuses causation with truthmaking. I admit that there is intuitive plausibility to the association of being the truthmaker for a proposition about an action (such as I ate ice cream for dinner) and causally contributing to an agent’s knowing the truth of that proposition. An agent’s action often serves both as the truthmaker for a proposition about that action and as an efficient cause of another agent’s coming to know that proposition. And this association could be there if God is in time: if God is temporal, then He could be caused to have knowledge of a proposition by the truthmaker for that proposition.
But I think the Knowledge Objection fails insofar as it holds that being the cause of God’s knoweldge and being the truthmaker for the objects of God’s knowledge must always coincide. But according to the doctrine of divine eternity, God is outside of time, and thus His beliefs don’t change as a result of some action A; the occurrence of A doesn’t cause God to have different beliefs than He previously had. Whatever beliefs God has, He timelessly has. And this makes no claim about what serves as the truthmaker for the propositions that God timelessly knows. It is possible for God’s timeless beliefs to be made true by temporal truthmakers such as human actions. In other words, while truthmaking as applied to God’s knowledge requires a kind of dependency between God’s knowledge and the truthmakers for the propositions He knows, this dependency isn’t a causal dependency.
According to truthmaker theory, the truth of a true proposition p is counterfactually grounded in the existence of its truthmaker, A. Had A not existed, then p wouldn’t have been true. Likewise, God’s true belief can be counterfactually grounded in the existence of that same truthmaker. Had the truthmaker not existed, God wouldn’t have believed the proposition in question, because in that case it would have been false. In other words, had A not existed, then it would be false that God even believed p. But this in no way entails that God is caused to believe something that He didn’t previously believe. In other words, the doctrine of divine eternity only commits its proponents to the claim that God’s knowledge does not and cannot change over time, since such a being is not in time. The doctrine of divine eternity does not, however, commit one to the claim that God’s knowledge doesn’t depend ontologically on the existence of truthmakers.
What do you all think of this response to the Knowlege Objection?

  • Sounds good to me. This is pretty much what I think, except that I’m not averse to describing it as causation of God’s knowledge, because I’m not tied to thinking of causation as one earlier thing causing another later thing. If you remove that, why can’t it be timelessly true that an event in time causes God’s eternal knowledge? If causation doesn’t imply change, then I see no objection to calling it causation.

    March 13, 2006 — 12:56
  • christian Lee

    Here’s an objection:
    There are some propositions that change in truth value. Example: (P) It’s now raining in Paris.
    For any subject, if they believe (P) when it is false, then they are not God.
    For any subject, if they do not believe (P) when it is true, then they are not God.
    So, God believes (P) iff (P) is true.
    But, (P) is true at some times and not at others.
    So, God believes (P) at some times and not at others.
    So, God changes his beliefs.
    So, God exists in time.
    What do you think?

    March 13, 2006 — 15:34
  • (P) isn’t a proposition. It’s a sentence that expresses different propositions depending on when it is uttered.

    March 13, 2006 — 17:02
  • Christian,
    I think you’re objection is begging the question, here’s why. Your proposition (P) is a tensed proposition. Namely, that it is raining in Paris NOW. Suppose we expresses the proposition tenselessly by relativising it to a particular time:
    (P2) It is raining in Paris at 3:12pm, on March the 13th 2006
    Now we grant that God knows (P2) iff (P2) is true.
    But here it does not follow that (P2) is true at sometimes and false at others. in fact the shoe is on just the other foot; if (P2) is ever true or ever false it is eternally so.
    So if we take away the tensed language, i.e., the “now” we, not surprisingly, excoriate your agrument for God existing in time.
    Seems to me like you can’t embed a tense operator in (P), such as “it is now the case that” without arriving at your conclusion. But it is just possible that God knows things tenselessly.

    March 13, 2006 — 17:14
  • Christian Lee

    I totally agree that that strategy is available and that if there are tenseless truth-conditions for tensed propositions, my argument above does not work. I think your view is plausible, but requires a view of tense I think is mistaken.
    And my point is that the view you are suggesting seems required to make sense out of God being atemporal. That is not question begging, it is just drawing out the consequences of the view, consequences that A-theorists like myself will take to show the view that God is atemporal to be problematic.
    I think it is plausible to say that there is one proposition (P) that is expressed by tokens of the sentence type “It is now raining in Paris” and that the sentence doesn’t express propositions at different times. In the end that might be incorrect, but I think there are strong considerations for both views that need to be evaluated carefully.

    March 13, 2006 — 20:44
  • Christian, I understand that you find it plausible. You’re an A-theorist. But I don’t think I should find it plausible, because I think A-theories are decisively refuted both by our best physics and by our best philosophy of language (particularly with respect to truthmakers). I don’t think propositions could have tenses in their logical structure. If there’s a way to have that with a B-theory, then I’d be open to it hearing it, but I haven’t encountered it, and it doesn’t seem natural to me at all.

    March 13, 2006 — 22:23
  • Christian Lee

    I myself don’t know of any decisive refutations. I am a bit knew to philosophy of time. However, something you said suggest we might be talking past one another. So, you said “I don’t think propositions could have tenses in their logical structure.”
    The A-Theory I prefer is one which analyzes past and, maybe, future tense into relations of earlier than and later than the present. So, there is only one “real” tense and B-relations are analyzed in terms of it. And that property, presentness, is supposed to be picked out by ‘now’ which, as far as I can tell, doesn’t raise any interesting problems of logical form.
    I’m inclined to think B-theories have significant problems too.

    March 13, 2006 — 23:03
  • Given all the attention middle knowledge gets, it’s nice to see someone at least touch on physical premotion, even if only in passing. Strictly speaking, physical premotion (such as Garrigou-Lagrange accepts) is not supposed to be a form of theological determinism, any more than middle knowledge is supposed to be. Banezians like G-L claim that the divine predetermination is not necessitating (hence they would tend to deny theological determinism), although it is infallible. That is, one could will otherwise, but, given that God has willed that you’ll will a given way, it’s certain (but not necessary) that you will, because God’s will is sort of like an infallible persuasion.
    Of course, how one could coherently believe in infallible predetermination while disbelieving in determinism is quite the mystery.
    I like your truthmaker response to the knowledge objection, and think it very plausible. I wonder, however, if the objector might come back with some argument like the following: “What is needed for an account of divine knowledge is not merely an account showing that what God believes is true, but an account showing that the relation between object and belief is such that it is not merely happenstance that the belief is true. The true belief must be knowledge. The most natural way to make this relation ‘non-happenstantial’ is by causation. If God causes the object, given omnipotence, that would explain matters in such a way that you could see why it isn’t just happenstance that God’s belief is true. Likewise, if the object by existing causes God to know in some way, that would also give you a way to see that God’s belief’s being true isn’t just happenstance. So what else could this relation be? The mere fact that q (if A had not happened, God would not have believed p) is true doesn’t, on its own, help, because for all we know this could be pure happenstance. So we have a sort of grounding objection for this; what we need to know is not the truthmaker for p, but the truthmaker for q.”

    March 14, 2006 — 0:16
  • Heath White

    I think this is a decent response to the knowledge objection. Brandon’s “grounding objection” can _I think_ be answered by saying that since God exists simultaneously with every point in time, his knowledge can be viewed as a kind of perception of every action.
    I agree with Jeremy that we might just call “causation” “efficient causation” and call “truthmaking” “formal causation” and then the notion of causation wouldn’t be problematic.
    Christian raises a good point about tensed beliefs which is not entirely refuted by the B-theorists I think. Put it this way. It is entirely possible for me to have beliefs of the form, “It is raining now” without having any beliefs of the form, “It is raining at 12:00 noon EST.” Perhaps God has all (and only) the true beliefs of the latter, B-theorist form. Still, then, there are some true beliefs God lacks, namely ones of the tensed form. Whether this is problematic depends on whether you think omniscience requires knowing all true propositions or having all true beliefs. I suppose the B-theorist would just say (actually, the local B-theorists can tell us what they would say) that God’s lacking true beliefs of the form, “It is raining now” is no more problematic than God’s lacking true beliefs with other indexical elements, e.g. “It is raining here” or “I am left-handed.”

    March 14, 2006 — 8:38
  • Tom

    Hi friends-
    Alan Rhoda (who is on vacation right now I think) is a fellow poster here. You may find some of our ideas argued for here (http://www.alanrhoda.net/papers/opentheism.pdf) applicable to the issues Kevin raises.

    March 14, 2006 — 9:11
  • Christian, the best critique of A-theories is in the chapter in Ted Sider’s Four-Dimensionalism on time. It pretty much captures all the previous critiques and advances several new arguments. In all the stuff I’ve seen since then, I haven’t seen anything that really succeeds in resisting his arguments. I say this as someone who wanted to hold a moving spotlight view going into graduate school.
    I’m actually unaware of any arguments against B-theories besides what simply seems like hand-waving (e.g. the “thank goodness it’s over” argument and similar confusions). Dean Zimmerman thinks presentism has got to be knowable a priori (and you can picture him pounding his fist in good Moorean fashion when he says this), but I never got him to present any arguments for it. But there are at least two very strong considerations against it.
    Brandon, I’m unclear on how direct awareness of something by an omniscient and infallible being can count as happenstance. Isn’t God’s essential omniscience and infallibility enough to ground the truth of q?
    Heath, I wasn’t think in terms of formal causation but I suppose that does capture it better than efficient causation if you associate efficient causation with motion as everyone through the beginning of the early modern period did. But the fact is that we don’t anymore. Once we had causation at a distance, we expanded efficient causation to include things like fields and forces other than corpuscularian pushing. Now we have inklings of backward causation with electrons, including the idea that something you check after the fact causes a cat to have been dead or alive in the past, at least according to one interpetation of quantum mechanics. The concept of causation that had been efficient causation is not longer just motion, so I’m not sure we need to resist saying something causes something in God efficiently out of the worry that it’s going to cause a change in God. Timeless causation seems to me to be a viable option even within what’s generally taken to be efficient causation. Maybe that just means efficient causation has expanded to include formal causation, but that’s not how I was thinking of it. I’d be happy to say it’s that, though, if that makes the Thomists among us happier.
    As for the B-theory point, most B-theorists do think of tensed statements as indexicals. You could list of all the non-tensed facts, with what happens before and after other things. Set out the whole timeline like that. B-theorists then say that those are all the facts about time. A-theorists say there’s one further fact — what time it is now and therefore which things are past or future. B-theorists think there’s no such extra fact. So there’s nothing for God not to know if God doesn’t know what time it is, because there’s no such thing as what time it is an any non-time-indexed sense. Once you index to a time, you can say what time it is, because that just means you’re using an indexical the same way we can say something is here when we’re in a spatial location. But most theists say there’s no fact God doesn’t know if God doesn’t where here is. There’s no here for God. So too a timeless God, if the B-theory is correct, does not know what time it is for the very reason that there’s no fact (apart from time-indexing for temporal beings) about what time it is.

    March 14, 2006 — 9:45
  • Tom, Alan’s paper assumes presentism, which is what I’m denying, so it doesn’t really have any bearing on what I’m saying. I acknowledge that those who want to retain an A-theory will have trouble working out how God can be atemporal and yet know what time it is, but I think that’s just one further reason for divine atemporalists to reject A-theories. There are two attempts I know of to retain A-theories while saying God is atemporal (really three, but one is only pseudo-atemporal). I think what I want to say on this will take up more than I should put in a comment, so I’ll probably just put together a post on it myself.

    March 14, 2006 — 10:06
  • Tom

    My bad. I thought the debate was about the possibility of a coherent semantics given the A-theory.
    As for presentist-theists making sense of divine atemporality and omniscience, I agree it’s more than difficult. It looks positively impossible to me. I agree that divine atemporalists who want to retain omniscience have to reject the A-theory of time. But one could also be an A-theorist and maintain omniscience by rejecting divine timelessness.
    I haven’t read Ted Sider. Hope to.

    March 14, 2006 — 10:37
  • Kevin Timpe

    Let me try and comment on some of the other comments. Thanks to all for their thoughts.
    Jeremy and Heath,
    I’m certainly willing to allow that there are other kinds of causation besides efficient causation. So maybe I need to make it clearer that the kind of causation in the Knowledge Objection is efficient causation (as I think its proponents think). It’s not clear to me, however, that truthmaking is exactly the same as formal causation (but I’m willing to be convinced).
    So if physical premotion is “infallible persuasion,” how is this not a form of at least conditional necessity? And if it is, then since it comes from outside of the agent, how is the agent supposed to have libertarian free will if she is “infallibly persuaded”? And on Molinism: given the standard account (as in Flint) that the agent isn’t the truthmaker for the relevant counterfactuals of creaturely freedom, I think that there is a good consequence-style argument for the conclusion that if there were true CCFs (which I don’t think there are), then the agent wouldn’t be responsible for what she does.
    Also, in your ‘grounding objection’, you seem to be asking for an account of how God is in epistemic contact with truths, even if He’s omniscient. And I’ll be honest–I don’t have a full account. Since all the examples we do have are those involving causation and truthmaking, it isn’t surprising that those cases we are familiar with don’t help us get clear on God’s cognition. I think that angelic cognition might help here, but I’m not sure that’s any clearer than omniscience. But I don’t think that it merely happenstance that God believes what is true. I think it is necessarily true that God believes all and only true propositions. So I think I’m with Jeremy here.
    Two further comments. I’m wondering if the suggestions that efficient causation can be atemporal is really going to help, primarily because any sort of efficient causation ‘working’ on God seems to be problematic for divine simplicity, even if not for eternity per se. But, at the same time, it’s not clear to me that my account isn’t going to face the same problem, given that there does appear to be ‘truthmaking potency’ in God, and it isn’t clear that this is compatible with divine simplicity. Second, I realize that there is a real question if an atemporal God could now ‘now-indexicals’. And I don’t think I have anything new to say on this that hasn’t already been said in the literature. But if an atemporal God can’t have such knowledge, and ifthere are true propositions of this form (which depends on a lot of issues in time and tense that I don’t want to get into in my paper), then it does seem to follow that this would give a good reason for rejecting divine eternity. But I haven’t yet been convinced that both of these antecedents are true (though this might reflect the cogency of the arguments such much as my own psychology).

    March 14, 2006 — 10:52
  • Tom

    I was just reading a review of Sider’s book by Katherin Hawley (Unversity of St. Andrews) that’s forthcoming in Nous. In reviewing ch. 2 (Sider’s defense of the B-theory of time), she comments: “…the main challenger to the B-theory is presentism, the view that only the present is real and tensed facts are irreducible to tenseless ones. Sider challenges presentists with arguments about cross-time spatial relations, about truth-makers for claims about past and future, and about the special theory of relativity. The STR can seem to show that there is no one time which is the present, since whether some event across the Atlantic is present (simultaneous with my typing this sentence) depends upon the frame of reference from which we compare my typing with that transAtlantic event. With presentism rejected, the B-theory is presupposed throughout the rest of the book.��?
    Is this an accurate summary of Sider’s argument against presentism (viz., cross-time spatial relations, the grounding objection, and STR)?

    March 14, 2006 — 11:04
  • Christian Lee

    Thanks Jeremy for the tip, I’m going to read Sider today.
    As far objections to the B-Theory, I was thinking of (a) the ‘Thank goodness that’s over’ objection, and (b) the view that the only plausible B-theory requires indexes to dates/times and that this index requires a property of presentness. I think utterance-reflexive and utterrer-reflexive views cannot handle subjunctives like “If there were no people, it would still be sunny now”. This pushes us to a date theory, but dates need to be analyzed and I am hesitantly inclined to think that this cannot be done without a special present.
    I’m no presentist though.

    March 14, 2006 — 11:29
  • Tom, I believe those are the main arguments, though his presentation of them is much more detailed and involves a few different versions of them that I hadn’t seen elsewhere. So I wouldn’t assume that you’re familiar with what he says just because you’re familiar with the general claims of each argument type.
    As for the A-theory and timelessness, wait for my post. I’ve begun working on it, but my family doesn’t want to cooperate. Brian Leftow and Bill Craig have some interesting ways to try to get out of the argument. I’m less than satisfied with Craig’s view, but I think Leftow has something to say (even if I don’t like his fuller view.

    March 14, 2006 — 11:35
  • Christian,
    I don’t understand argument B. How does a notion of simultaneity required for time-indexing require presentness in the robust A-sense? I don’t see how indexicals are a problem in counterfactuals unless you hold to a Lewisian semantics of counterfactuals with counterparts. If indexicals work by pointing at a time de re, then doesn’t that explain the truth of such counterfactuals, because the same time de re is the one you’re talking about in the counterfactual condition of there being no people.
    Argument A is not an objection to a B-theory. The B-theory agrees with the A-theory that events taking place later than other events are in fact later than them, and that’s all you need to make sense of “thank goodness it’s over”. All the expression means is that you’re glad that you’re not experiencing the thing you earlier experienced. That makes perfect sense on a B-theory. The only view that doesn’t allow this is the view that our current utterances take place in the past, when the bad thing is still going on. Since no one holds that view, the argument doesn’t impress me very much.

    March 14, 2006 — 11:42
  • Kevin,
    I’ve never fully understood the reasoning behind physical premotion (I’ve never fully understood the reasoning behind its major historical rival, middle knowledge, either). But Banezians will often insist that even in the case of human beings it is possible (albeit only under certain circumstances) for a human being to influence another human being without necessitation and without coercion, but in such a way that it is certain what the influenced human being will do; so a fortiori God must be able to do it even more. Another common line of thought is that, given divine omnipotence, we should never rule out something as within God’s power unless we can show rigorously that it involves a contradiction; and they deny that it is a contradiction for God to predetermine a free action infallibly but in such a way that it is still libertarian-free (usually pointing to purported cases in which it happens with human beings and to theological cases like the impeccability of saints in heaven). The Molinists (with great plausibility, I think) denied this claim, holding that God must be a cooperator, not a predeterminer, in free actions, if we are to make any sense of their being free. (Both groups held that they were faithfully following the position of Thomas Aquinas.) On conditional necessity: I think Banezians would simply deny that there’s any conditional necessity; the conditional is certain, which is why God can infer the free action infallibly from its premotion, but it is not certain because of any necessity, but simply because God is necessarily omniscient and omnipotent.
    I don’t think one can answer the objection I suggested simply by appealing to omniscience and omnipotence, any more than Molinists and Banezians can answer similar objections just by appealing to omniscience and omnipotence (however much they’ve tried to do so). The perception/awareness suggestion made in several of the comments is promising but it seems that either we must hold that God’s knowledge is caused by free actions; or we must hold that God’s knowledge causes free actions; or we must have an account of perception/awareness that is not causal at all. I’m not really sure what such an account of perception/awareness would be.

    March 14, 2006 — 12:15
  • Christian Lee

    Take the conditional: (C) If it there were no people ever, then here in Boulder it would still be sunny now.
    I think (C) is true. What makes it true is that the closest world to this world where there are never people is a world where it is now sunny here in Boulder.
    There is an assumption that the conditional is true and I am happy to make it. According to token-utterance reflexive views, (C) is analyzed as:
    The closest world to this world where there are never people is a world where: The state of affairs of It being sunny here in Boulder is simultaneous with this utterance.
    But, this is wrong because there are no people or utterances in the world under consideration. So, these reflexive views do not work.
    Hence, the move to dates and demonstrating times. I think this is the view you like, and I like it too, because it avoids this particular worry. It also avoids worries raised by sentences like “There was a time when dinosaurs were present and no utterers were present.”
    Now, we can think de re about times, point to one mentally and saying, ‘this time’ and that seems right to me. This fact motivates the view I like and the view you seem to like. But, there are other facts I would like to explain, for instance, why we cannot think de re about future or past times. That is, I think we can directly think about this time, where ‘this’ picks out what I would call ‘the present’, but we cannot think directly about the first moment of 2007. More generally, we can only think directly about some times, not others, and I think there is something in virtue of which this is true. Here is a good explanation I think: We can only think directly about those times that are present.
    But if the fact that a time is present is to explain why we can think de re about it, then we have a good reason to think there is some property that satisfies this explanatory role. But, this is just “the present” on my view, the cutting edge of all causal processes.

    March 14, 2006 — 12:32
  • Christian, it sounds like you’re working with a B-theory account that reductively analyzes “X is present” with “X is simultaneous with this utterance”. I don’t think most B-theorists today have such a reduction. The truth conditions for statements involving tensed terms aren’t as simple as that.
    In the case of ‘now’ in the sentence in question, there’s actually a pretty easy answer to the truth conditions. (C) says “If it there were no people ever, then here in Boulder it would still be sunny now.” The truth conditions for that sentence consist of the location ‘here in Boulder’ refers to, the time ‘now’ refers to, and its being sunny all coinciding in the spacetime manifold of the possible world that’s nearest to the actual world with no people.
    This is pretty much of a piece with the analogy between space and time that B-theorists insist on. If space is like time, then ‘now’ functions with time the same way ‘here’ functions with space. Therefore, if B-theories are going to have problems with counterfactuals with now-statements, then both A-theories and B-theories will have the same sort of problem with here-statements unless they adopt the hereist view, according to which there’s something special about here merely because it’s where I am, in addition to all the facts about where things are and how they relate to each other spatially. No one holds that sort of view, though, so you have to be able to dodge the argument with ‘here’. B-theorists just do the same sort of thing with ‘now’. I think this is going to be true of most any semantic argument against B-theories.

    March 14, 2006 — 13:09
  • Kevin Timpe

    I agree with you that Banezians want to say the sort of thing that you attribute to them. I just think that what they are saying is incorrect and/or incoherent. The kinds of cases you point to all seem to involve character formation; but the incompatibilist doesn’t think that conditional necessity by one’s own character undermines free will so long as one has libertarian freedom in the formation of said character. But setting of our character by something outside of ourselves would undermine said freedom. (On a related note, I think that the libertarian can give an account of why the redeemed in heaven will not be able to sin despite having libertarian free will along exactly these lines.)
    Second, regarding ‘epistemic contact.’ When I wrote that, I had Stump’s chapter on God’s knowledge in her new Aquinas book in mind. Consider the following passage from page 184:
    On Aquinas’s account of God’s knowledge, then, for God, too, direct and immediate cognition of things outside the mind involves at least three elements: (a) God’s being in epistemic contact with everything he cognizes, (b) God’s possessing a concept or intelligible form of what he cognizes, and (c) God’s applying that concept or form to what he is in epistemic contact with.
    I was trying to get at the first of these three issues in my post, since this seems to be the issue at hand. At times, Aquinas uses the following kinds of language to illustrate what he means by God’s having epistemic contact: simple awareness, intellectual observation, view, vision, gaze, etc…. Of course, these need to be taken with a grain of analogy. God’s epistemic contact is like ours, except it doesn’t invovle physical faculties or causation (helpful, isn’t it!).
    But then Stump goes on to say that noone has an adequately worked out account of epistemic contact, and so the fact that Aquinas doesn’t just shows that what he’s said is incomplete, not that it isn’t helpful (or true):
    “We might find Aquinas’s talk of God’s ‘seeing’ mysterious…: how does God accomplish this ‘seeing’? As far as I can see, Aquinas provides no further help in analyzing God’s epistemic contact than to hold that God applies his ideas to what he cognized and that God atemporally ‘sees’ thing other than himself. In this respect, then, Aquinas’s account is just incomplete. It is worth noticing, however, that currently standard accounts of human cognition are also incomplete in analogous respects, contrary to what one might suppose. Although it is clear taht concepts or mental representations have to be applied to what the cognizer is in epistemic contact with, no one has more than a rudimentary idea of what such application consists in or of what has gone wrong in agnosia patients who are no longer capable of it. Furthermore, although it is true that what is cognized acts causally on the cognizer’s senses, for that causal connection to count as epistemic contact at all, the sensory data produced in that way must undergo some processing by the central nervous system. Causal connection between some object and say, an eye in a vat would not constitute epistemic contact. But sensory input by itself underdetermines the result of the central nervous system’s processing. How is the result of that processing related to the thing cognized, then? Or, to put it another way, how is it that the result of the processing constituted epistemic contact with the extramental things that generated the sensory input? At the moment, at any rate, nobody knows. The incompleteness of Aquinas’s account of God’s knowledge looks less surprising when we recognize that contemporary accounts of human knowledge are incomplete in the same way.”
    Of course, this doesn’t mean that Aquinas’s incomplete account is right, or that the incompleteness isn’t a problem at all. But I do think it goes to show just how tricky an issue ‘epistemic contact’ is all the way around.

    March 14, 2006 — 13:20
  • Christian Lee

    Jeremy, I’m not working with the B-Theory account you suggest. Half of the point of my last post was to say exactly that, that token reflexive view cannot give truth conditions for statements about worlds without utterers. No bid deal I say, we should use dates and times, just like you suggest. On this view, ‘now’ refers to a time. Then, I proposed a problem for that view. I said it cannot explain why we can have de re thoughts about some times and not others. However, a view which acknowledges ‘the present’ as selecting a special property can.
    Likewise, ‘here’ will refer to the location of the utterance and ‘I’ will refer to the agent of the utterance. And I also think that either (a) there is a special property of ‘hereness’ and ‘I-ness’ or (b) there is not, and this is exactly why space and time are different.
    Now, although I am less interested in theological implications of the view, I’d add that the B-Theory you suggest seems just as problematic for an atemporal view of God as the A-Theory I am suggesting. On the view that ‘now’ is indexical and picks out ‘this time’ and where God can believe it is now 6:00 in Boulder on Earth, God has a de re belief about a time. I suggest that that is sufficient condition for a being to be in time. Conversely, suppose it is not a sufficient condition, then why are we to think that ‘we’ are in time? I think the answer is straightforward, we should think we are in time because we can have indexical beliefs about times.

    March 14, 2006 — 13:32
  • Kevin,
    We’re agreed on physical premotion; I don’t think it’s ultimately coherent either, although I think it’s as interesting and clever an attempt as middle knowledge. Physical premotion is always intended to be distinguished from moral premotion (e.g., character formation), which almost everyone agrees God can do without violating free will; but it doesn’t seem the Banezian can find any analogous case stronger than moral premotion.
    I’d forgotten that Aquinas-as-interpreted-by-Stump doesn’t have an account of epistemic contact; I don’t think this is quite the right way to interpret Aquinas (I think both the Banezians and the Molinists are more correct in seeing Aquinas as saying that there is some sort of causal epistemic contact) but I take your point about the difficulty of providing a solid account of epistemic contact. I just worry that it gets dangerously close to being a defense that protects everyone, Banezian, Molinist, and other, from everyone else (since everyone can, and does, appeal to divine omniscience and omnipotence as a reason why their view is coherent and tenable).

    March 14, 2006 — 14:15
  • Divine Atemporality and Tensed Facts

    In the comments on Kevin’s post Divine Eternity and Libertarian Free Will, several commenters have raised the problem of how an atemporal God can know what time it is. There are several directions someone can go with this question, and I wanted to comm…

    March 14, 2006 — 14:40
  • Kevin, doesn’t the doctrine of continuous creation help with the problem of how God knows things in time? After all, if God is continously sustaining the existence of all things across all time, then God should have direct causal contact with everything that happens, even if it doesn’t mean God is causing any event or God’s knowledge is being caused by those evens. Maybe I just haven’t fully grasped what question you’re looking into, but this seems to me to be good as at least part of the answer.
    Christian, somehow I didn’t see the second half of your post. I just looked back and read all that for the first time. I’m not sure what happened. It was probably that I got halfway through reading your comment and got interrupted then forgot there was more.
    I don’t think your second point is a problem for a B-theorist. Why couldn’t a B-theorist think about the current time in a de re manner? After all, it’s the only time I have contact with. The distinction between my connection with this time and my connection with all other times seems to me to be a B-fact, so B-theorists and A-theorists alike can explain why I can refer de re to this time but not to others.
    As for what distinguishes between things in time and things not in time, I think the most obvious way to do that is in terms of whether the thing in question experiences succession. It sounds fairly ad hoc to me to define being in time in terms of something that doesn’t immediately have to do with being in time but knowing something about time, expecially when it conveniently forces someone in on e direction on this question.
    But I don’t think B=theories have that consequence anyway. For one thing, why couldn’t God refer to a time de re in the same way we refer de re to various places in our immediate vicinity without being at all of them, at least if eternity means being in a parallel time dimension that has relations something like simulaneity with all times. God is in the vicinity of every moment in time in a way enough like our being in the vicinity of lots of points in space that we’re not at. I think that should be enough for de re reference.
    I’ve got the post on A-theories and atemporality up now, for those only checking this post.

    March 14, 2006 — 15:30
  • Christian Lee

    Jeremy, I do think a B-Theorist can think about the current time de re. When she thinks ‘this time’ she ipso facto thinks about the current time, the only time she can possibly think about de re. Yet, she cannot think de re about some time t’ that is either later than or earlier than the current time, t. Now, how exactly does a B-Theorist explain this? Is she going to say that I can think de re about t because t is not earlier than or later than the time I can think about? No. Is she going to say I cannot think de re about t’ because t’ is not “the current” time? No. Well, I suggest there is no B-Theorist explanation for the fact that I can think de re about the current time, but I cannot think de re about some past time.
    I have an explanation as an A-Theorist: The times I can think about de re instantiate presentness.
    Second, I wasn’t suggesting that we define ‘being in time’ in terms of ‘being directly aware of some time’. I am suggesting that being directly aware of a time gives us a sufficient reason to think, justifies our belief that, we are in time. But, the question is then, if God is directly aware of time in the sense that she can think de re about the current time, and think ‘this time’, then why should we not also think that God is in time?
    And again, it is unclear to me that space is like time. In fact, it is unclear to me that we can think de re about places at all, instead of objects that occupy places, from whence it is inferred that there is a placed occupied by that object thought about de re. The belief about places would be a general proposition, not singular.
    But suppose that’s wrong, then I suggest we can think de re about those locations that we are at, and only those that we are at, assuming that ‘being at a location’ doesn’t entail having a part at that location. The locations we could think de re about would be those we are perceptually aware of. But again, the analogy between space and time breaks down. We are not perceptually aware of the past or the future. Again, I would like to explain this by referring to the fact that the past and future are not present, and that that is why we are not perceptually aware of them.

    March 14, 2006 — 16:15
  • Kevin Timpe

    I share your hesitation about the inablity to give an adequate notion of epistemic contact. So I’d like to see some of the smart epistemology folks that are reading this say something that will make clear the notion of epistemic contact. If preferable, I’d also like it by friday! 🙂

    March 14, 2006 — 17:36
  • Is she going to say I cannot think de re about t’ because t’ is not “the current” time? No.
    Actually, the correct answer is yes. She’s going to say she can’t think about it because it’s not the current time. Then she’ll say what a B-theorist says about what it means to say something is the current time. Ultimately the reason she can’t think about it de re is because she doesn’t have direct contact with that time.
    Ultimately, I’m less sure than you that anyone can think about any time de re to begin with, A-theory or not. The main problem is that our thoughts don’t take place at a time but in an interval. Are we thinking about exactly the interval that it takes for us to form the thought? I’m not sure we can do the kind of individuation that we ordinarily take ourselves to be doing. The best way to capture what’s going on may have to be a very complicated set of truth conditions, and I don’t see how those would favor either a B-theory or an A-theory.
    being directly aware of a time gives us a sufficient reason to think, justifies our belief that, we are in time
    As I said in the other thread, there’s an easy response to this. Whatever you’re using the expression ‘being in time’ to mean, it’s not the same thing atemporalists are denying about God. Atemporalists deny that God experiences things in succession, not that God is directly aware of times. Atemporalists insist that God is directly aware of all times at once. If you want to use the words ‘being in time’ to describe that, then atemporalists just need to come up with another English expression to describe the view they have in mind, perhaps just saying that God doesn’t experience things in succession but experiences all times directly all at once.
    I have to say that if we can only think de re about objects in places but not about places, then I see little motivation for thinking we can think de re about times, even apart from my worry above about thoughts taking time to occur. Why can’t I just say the same thing about times? We can’t think about times de re but just about things in time. More carefully, we can’t think about spacetime locations de re but can think about what occupies spacetime regions de re and then refer demonstratively to the regions they occupy via our de re thought of the thing in that region.

    March 15, 2006 — 8:43
  • Christian

    “Ultimately the reason she can’t think about it de re is because she doesn’t have direct contact with that time.”
    That just shifts the explanation backwards one step. First, it is just false, because she will have direct contact with that time. So, she will be able to think de re about it. But we need to ask: Why can she not think de re about some time that is not the current time? Again, I don’t see how a B-Theorist explains this. “The current” time in her mouth is equivalent to “This time” and so when explain why she can think about it de re, and not later times de re, her explanation will be: She can think de re about this t at t because t is this t.
    That’s not an explanation. Suppose she moves to a token-reflexive view. The times one can think de re about are times simulateous with the thought. This may be tenable in the end, but I would like to look for a deeper explanation, a property that has explanatory power.

    March 15, 2006 — 11:55
  • Christian Lee

    “Atemporalists deny that God experiences things in succession, not that God is directly aware of times.”
    That’s fine. Then we are talking past one another. Then the whole point about ‘now’ beliefs needs to be modified. I think the argument will be:
    p1. If God has the ‘now’ belief that this time is present and God is atemporal, then God’s “now belief” picks out every time.
    p2. If God’s “now belief” picks out every time, then God has a necessarily false belief.
    p3. If God has a necessarily false belief, then God is not omniscient.
    c. If God has the ‘now’ belief that this time is present and God is atemporal, then God is not omniscient.
    defense of p2: necessarily, the property of being present is had by only one time at a time. The a-theory is true.
    defense of p3: omniscience requires having only true beliefs.
    defense of p1: you wrote: “Atemporalists insist that God is directly aware of all times at once.”

    March 15, 2006 — 13:40