Multiple Attributes Disproofs of the Existence of God
(2) Omniscience and Immutability
(3) Omniscience, Eternity, and Time
2. Omniscience and Immutability
by Norman Kretzmann
3. Omniscience, Eternity, and Time
by Anthony Kenny
Here we come back to the problem of whether an immutable God knows what time it is. God must know that it is now t1 and not t2, and at a later time God must know that it is now t2 and not t1, which seems to cause His knowledge to be variable and God to be mutable.
One interesting but incorrect, in my view, solution is to posit that God does change but only in such a way so as to preserve His perfection of omniscience. That is, God’s knowledge of the current time or of present-tense propositions changes with time or the truth-value of those propositions but only in order to keep Him all-knowing. God knows at t1 that it is then t1, and He knows at some later time t2 that it is then not t1, but this changing knowledge merely safeguards His omniscience. Thus, although at any given time the content of God’s knowledge is different from the content of His knowledge at any other time, He still knows everything there is to know at all times. (That may be the reason for William Lane Craig’s suggestion that logically prior to creation God exists in “undifferentiated time,” and after creation God is in time.) Nor does God appear to lose His pure actuality on this theory, since at all times His mind remains “adjusted” to all truths, changing as they are.
This solution, however, is to be rejected, because it unbecomingly anthropomorphizes God.
Kenny’s paper begins with a critique of the doctrine of divine timelessness. According to Aquinas, he says, God can know future contingents “because God does not see future contingent facts as being future but as being present; future contingents are present to God. It is, St. Thomas says, nearer the truth to say that if God knows a thing, then it is than to say that if he knows it, then it will be.” (The Impossibility of God, 210) Now this is correct, to a degree. What Aquinas really says is that
all things that are in time are present to God from eternity, not only because He has the types of things present within Him, as some say; but because His glance is carried from eternity over all things as they are in their presentiality. Hence it is manifest that contingent things are infallibly known by God, inasmuch as they are subject to the divine sight in their presentiality; yet they are future contingent things in relation to their own causes. (ST, I, 14, 13)
To see future things “in their presentiality” means to see them as if they were present, already having come to be. It does not mean that God thinks that future events really are present. (The important question is how does God know these future events. We can’t surely be content with saying that God sees the future as if it were present and that it just magically happens to be exactly like the real future. So, my answer is that God knows future contingents by perfectly predicting them.) Kenny continues that “on St. Thomas’s view, my typing of this paper is simultaneous with the whole of eternity. Again, on this view, the great fire of Rome is simultaneous with the whole of eternity. Therefore, while I type these very words, Nero fiddles heartlessly on.” (211) There are two ways of getting out of this difficulty. First is to argue that only the present moment, the now is simultaneous with the whole of eternity. The fire of Rome was simultaneous with God’s eternal present, but is no longer such. In fact, on some theories the fire of Rome no longer even exists. The second is to point out that the simultaneity in question is merely metaphorical. Perhaps what this term means with respect to the time-relationship between God and the world is that for God the presentiality of a past event looks as vivid as the presentiality of a future event. But nothing about this admittedly awesome characteristic entails an annulment of the difference between tenses for us creatures.
Back to time. Kenny points out that Aquinas thinks that God knows propositions (”enunciable things”), but
not after the manner of enunciable things, as if in His intellect there were composition or division of enunciations; for He knows each thing by simple intelligence, by understanding the essence of each thing; as if we by the very fact that we understand what man is, were to understand all that can be predicated of man. This, however, does not happen in our intellect, which discourses from one thing to another, forasmuch as the intelligible species represents one thing in such a way as not to represent another. Hence when we understand what man is, we do not forthwith understand other things which belong to him, but we understand them one by one, according to a certain succession. On this account the things we understand as separated, we must reduce to one by way of composition or division, by forming an enunciation. Now the species of the divine intellect, which is God’s essence, suffices to represent all things. (ST, I, 14, 14)
Our author attributes to Aquinas, correctly, it seems, the view that God does not express His knowledge in propositions. Moreover, just as two different sentences can express a single proposition (e.g., “It is raining” in English and “Дождит” in Russian mean the exact same thing, viz., that it is raining), so two different propositions can express the same item of knowledge. An example is, “Today is Friday” (uttered on Friday) and “Yesterday was Friday” (uttered on Saturday). “God’s knowledge is not expressed in propositions, and so he can know the same item of knowledge permanently and unchangingly. It is only because we are temporal changing beings that we have to express the one item of knowledge first in one proposition and then in another.” (217) Kenny is not satisfied with that solution for a number of reasons, but it is surely an interesting attempt to escape the problem.
Hector-Neri Castañeda’s approach is different. He first points out that the two “now”s in the statement
(4a) First X knows that it is now t1 and not t2, and then X knows that it is now t2 and not t1.
express indexical references by the speaker and not by X. That is, if I utter (4a), then by “now” I mean my own present and not X’s present. Reformulating (4a) in order to get rid of such indexicals yields
(4b) At t1 X knows [tenselessly] that it is [tenselessly] then t1, but not t2, and at t2, later than t1, X knows that it is then t2, but not t1.
Castañeda comments: “semantically the difference is enormous: (a) ‘now’ does, while ‘then’ does not, express an indexical reference by the speaker; (b) ‘then’ does, while ‘now’ does not, attribute to X an indexical reference to time t1 in the first and to time t2 in the second conjunct; (c) whereas sentence (4a) cannot be used by Kretzmann or anybody else to make exactly the same statement at times other than t1 and t2, sentence (4b) can be used repeatedly at any time by anybody to make exactly the same statement on each occasion of its utterance.”
Note that (4b) is not equivalent to
(4c) At t1 X knew that it was t1 at t1, but not t2, and at t2 he knew that it was t2 at t2, but not t1.
This is because X does not need to know what time it is at either t1 or t2 in order for (4c) to be true. It is always obviously t1 at t1, etc.
What have we accomplished? We have rephrased (4a) into an equivalent statement (4b), such that X can potentially know all four propositions in which we are interested (”it is thent1 t1” ; “it is thent1 not t2” ; “it is thent2 t2” ; “it is thent2 not t1” ) at both t1 and t2. To show how this can be true, our author uses a trick. Given the principle
(P) If a sentence of the form ‘X knows that a person Y knows that …’ formulates a true statement, then the person X knows that statement formulated by the clause filling the blank ‘…’,
we can transform (4b) into
(4d) Time t2 is later than t1, and at t1 X knows both (1) that it is thent1 t1, but not t2, and (2) that somebody knows (or would know) at t2 that it is (would be) thent2 t2, but not t1.
Why do we need a “somebody” to replace X? Because X is in time, namely, at t1, and while he is at t1, he cannot be at t2 to know that it will then be t2. But if somebody else knows it at t2, and X knows (at t1) that that person knows it, then he in effect knows (at t1) that at t2 it is (or would be) then t2, etc. All four relevant propositions are thereby known by X at t1. Clever, isn’t it? However, I don’t think that this rather elaborate procedure is necessary. For
(4e) Time t2 is later than t1, and at t1 X knows both (1) that it is thent1 t1, but not t2, and (2) that if X were at t2, then it would be thent2 t2, but not t1
is equally effective. X can conclude from the fact that he knows that the counterfactual in (2) is true at t1 that the second pair of propositions is true as much as the first pair is.
Thus, if God can know all four propositions at any time, then Kretzmann’s inference from God’s knowing what time it is to His being mutable is invalid. But we are not done yet, for with God we can rephrase (4d) in yet another way. Let
(4f) be: Time t2 is later than t1, and at t1 God knows both (1) that it is thent1 t1, but not t2, and (2) that, because He is omniscient, God knows at t2 that it is thent2 t2, but not t1.
God knows at t1 that He knows at t2 that p. Hence God knows at t1 that p. QED.
Conclusion. With enough ingenuity, solutions to questions such as “Does God know what time it is?” which preserve both His omniscience and immutability can be found even if one does not accept the B theory of time.
References:
Castañeda, Hector-Neri. “Omniscience and Indexical Reference,” The Journal of Philosophy, Vol. 64, No. 7. (Apr. 13, 1967), pp. 203-210.