The Classical Ockhamist Response to the Dilemma of Freedom and Foreknowledge

Previously I argued that there is a powerful argument against the claim that libertarian free action (defined in the last post) is compatible with God's infallible foreknowledge (or belief) of those actions, roughly, with God's omniscience (S is omniscient = df S knows every true proposition that is logically possible to know and S believes no false proposition). I will repeat the bare bones of the argument for convenience:

The TI Argument

Following William Hasker, I will use the incompatibilist argument offered by Linda Zagzebski, though names and actions will be different. With the contemporary emphasis on “going green,” it seems best to retire poor Jones and his ever-running lawnmower. Hasker opted for one “Cuthbert” and his iguana. I find it unnecessarily cruel to name people Cuthbert, and so will modify Hasker’s modification of Zagzebski’s argument with Al Dente and his plate of Fettuccini Alfredo.

Let three moments of time be ordered such that t1 < t2 < t3

1. Suppose that God infallibly believes at time t1 that Al dente will eat Fettuccini Alfredo at t3. (premise)

2. The proposition God believes at t1 that Al dente will eat Fettuccini Alfredo at t3is accidentally necessary at t2. (from the principle of the necessity of the past)

3. If a proposition p is accidentally necessary at t and p strictly implies q, then q is accidentally necessary at t. (transfer of necessity principle)

4. God believes at t1 that Al Dente will eat Fettuccini Alfredo at t3 entails Al Dente will eat Fettuccini Alfredo at t3. (from the definition of infallibility)

5. So the proposition Al Dente will eat Fettuccini Alfredo at t3 is accidentally necessary at t2. (from 2-4)

6. If the proposition Al Dente will eat Fettuccini Alfredo at t3 is accidentally necessary at t2, it is true at t2 that Al Dente cannot do otherwise than to eat Fettuccini Alfredo at t3. (premise)

7. If when Al Dente does act he cannot do otherwise, he does not do it freely. (principle of alternative possibilities)

8. Therefore, Al Dente does not act at t3 freely. (from 5-7)

This argument gives us a powerful formal argument against the compatibility of God’s foreknowledge of the (libertarian) free actions of human beings. I will refer to the above argument as the TI argument (the argument for Theological Incompatibilism).


The Ockhamist solution

The Ockhamist view denies the accidental necessity of God’s past beliefs about future free actions, and so denies premise (2) of the argument. The Ockhamist denies the accidental necessity of God’s past beliefs by making use of the hard/soft fact distinction. Though this distinction is notoriously difficult to explicate, it can be explained adequately enough to understand the gist.

A “hard fact” about the past is a fact that is entirely about the past.¹ For example, take the fact about the collapse of the Twin Towers on 9/11. That fact is entirely about the past. On the other hand, a “soft fact” is not entirely about the past. God’s belief in September 2009, that Al Dente will eat Fettuccini Alfredo for dinner on December 25, 2009, expresses a fact that is not entirely about the past. Part of the belief refers to what will happen in the future and its truth is capable of being affected by as of yet to occur free choices. Thus, if the hard/soft fact distinction can be explicated, and if God’s past belief is indeed a soft fact, then God’s past beliefs about the future free actions of humans is not accidentally necessary, and therefore premise (2) is false.

Another popular Ockhamist response, given by Alvin Plantinga, is to argue that we do have power to change the past, but this power is not causal power it is “counterfactual power.” It is “counterfactual” since Plantinga admits that, if God knows Al Dente will eat Fettuccini Alfredo for dinner, then Al will eat Fettuccini Alfredo.² But, Plantinga argues, Al has the power to act such that if Al so acted, then God would have had a different belief. So, even though God may have in fact believed a certain proposition P, Al Dente still has the power to do something such that if he were to do it, God would not have believed P. The explication just given is why this Ockhamist solution is dubbed “counterfactual power over the past.” This argument looks like it denies the accidental necessity of God’s past belief, and so premise (2) of the TI argument denied.

In this post I will not deal with Plantinga's Ockhamist answer but only with what I'm calling the "classical Ockhamist" answer, viz., the claim that God's past belief is a soft fact about the past.

Some problems with the Ockhamist solution

Recall the claim made above regarding God’s belief about a paradigm case of free action: God infallibly believed on September 1, 2009, that Alvin Dente would eat a plate of Fettuccini Alfredo for dinner on December 25, 2009. We can put this more rigorously as:

(G) God believed at t1 that Al Dente would eat Fettuccini Alfredo at t3.

We then asked whether, given (G), Al Dente was free to refrain from eating at t3. A reason for giving a negative answer to this question is based on the principle of the fixity of the past (PFP, hereafter).³ Our argument has been that:

(G) is a fact about the past relative to t2, so it is not within Al Dente’s power to bring it about that ~ (G); i.e., (G) is outside of the control of Al’s power.⁴ However, (G) entails that Al Dente will eat at t3. Whatever is entailed by something outside of someone, S’s, control is itself outside of S’s control. Therefore, whether Al Dente can refrain or do other than eat at t3 is outside of his control.⁵

Recall that the Ockhamist denies the accidental necessity of God’s past beliefs, and so denies premise (2) of the TI argument. The Ockhamist denies that God’s past beliefs about the future free actions of agents are accidentally necessary by claiming that God’s past beliefs in this area are not hard facts; rather, they are soft facts. This means, Ockhamists tell us, that God’s past beliefs are not subject to PFP.

To see why the Ockhamist makes this claim, and why the move does not work, it will serve us to look at some putative examples of hard facts and soft facts vis-à-vis (G). Here are three paradigm cases of hard facts, cases all Ockhamists would count as hard facts:

(1) In 1492, Columbus sailed the ocean blue.

(2) Jones uttered the sentence, “Al Dente will eat Fettuccini Alfredo on December 25, 2009,” on September 1, 2009.

(3) Jones believed that, “Al Dente will eat Fettuccini Alfredo on December 25, 2009,” on September 1, 2009.

These facts about the past are contrasted with another type of fact, facts called soft facts. Paradigm cases of these facts might be examples like:

(4) In 1492, Columbus sailed the ocean blue, 517 years before Al Dente at Fettuccini Alfredo on December 25, 2009.

(5) Jones uttered the true sentence, “Al Dente will eat Fettuccini Alfredo on December 25, 2009,” on September 1, 2009,

(6) Jones correctly believed that, “Al Dente will eat Fettuccini Alfredo on December 25, 2009,” on September 1, 2009,

Ockhamists want to say that (G) is similar to (4) – (6), and so (G) should be classified as a soft fact about the past. But this way lies trouble. First, there is no accepted explication of the hard/soft fact distinction. This makes progress difficult, to say the least. But, despite this hurdle, the explication on offer must at least do two things. As Widerker points out,⁶ the Ockhamist must (i) make explicit the property P that (G) and (4) – (6) have in common; and (ii), they must show that it is in virtue of having P that makes them soft facts. But the various explications have had trouble doing just that. In other words, the property soft facts are said to have do not show that it is because of this property or feature that facts like (4) – (6) are not subject to PFP.

We can illuminate this point by briefly looking at some features soft facts are said to have that hard facts do not have and which make them not subject to PFP.⁷ One view is that soft facts entail facts about the future. This is of course true of facts like (4) – (6). But is it because of this property that (4) – (6) are not subject to PFP? No; and here is why: Many hard facts (facts subject to PFP) entail facts about the future. Here are two:

(7) Walter Payton scores his first touchdown at t1 entails the future fact, relative to t2, that Walter Payton does not score his first touchdown at t3.

(8) God delivered the Israelites from Egypt at t1 entails that God exists at t2 – tn (this assumes that God is eternal).

Another view is that a soft fact entails the obtaining of a future state of affairs. Again, (4) – (6) share this feature (the future state of affairs that obtains is Al Dente eating his pasta dinner). But certainly it is not because of this property that (4) – (6) are not subject to PFP. Again, some ostensible hard facts (facts subject to PFP) entail future facts. Here is a plausible assumption for the theist to make: Necessarily, if God decrees that X will occur at t3, then X will occur at t3. Therefore, if God decrees at t1 that Al Dente will eat Fettuccini Alfredo at t3, then Al will eat Fettuccini Alfredo at t3. Certainly, God’s decree at t1 is a hard fact about the past relative to t2, yet God’s decree at t1 entails the future state of affairs at t3.⁸ The basic problem these accounts face, as Widerker points out, is that there certainly seem to be cases where one temporally distinct event E1 is necessarily connected, either metaphysically or logically, to another temporally distinct event E2 that obtains later than E1. Yet, in these cases, E1 is considered, on most analyses, to be a hard fact about the past, and thus subject to PFP.

The Ockhamist, not surprisingly, still wants to call (G) a soft fact; and they beg our patience as they work on explicating the hard/soft fact distinction. After all, they tell us, (4) – (6) certainly look like soft facts in that they are not subject to the fixity of the past. At this juncture, we should make more explicit what is meant by the past being fixed. Hasker claims that a hard fact, one subject to PFP, is a fact that is really about the past and also not within anyone’s power to make false.⁹ Widerker concurs, offering the following account of PFP:

(PFP) If a given event occurs at time t, then no one has it within his power at a time later than t to bring it about that the event did not occur at t.

Given this view of PFP, we can see why there is the intuition that (4) – (6) are not subject to PFP, and are thus soft facts. Given this view of PFP, we see that although we do not have the power to bring about¹⁰ the nonoccurrence of the past event, we do have the power (on LFW) to bring it about that an event does or does not exemplify the contingent proposition it refers to, e.g., Al Dente eats Fettuccini Alfredo on December 25, 2009. Applying this to one of our putative soft facts:

(6) Jones correctly believed that, “Al Dente will eat Fettuccini Alfredo on December 25, 2009,” on September 1, 2009,

we can see that, on LFW, Al Dente can bring it about that he refrains from eating Fettuccini Alfredo on December 25, 2009. In doing so he does not cause the event of Jones’ believing to not occur, he merely and harmlessly causes the non-obtaining of the contingent property expressed in (6).¹¹ Furthermore, on this analysis of PFP, we see why (6)’s counterpart, (3), is subject to PFP.

Given our work up to this point we are now in a position to show why (G) is disanlagous to (4) – (6) in at least one relevant area. Take our definition of Omniscience:

O: S is omniscient = df S knows every true proposition that is logically possible to know and S believes no false proposition.

O entails that God Infallibly Believes (GIB) every true proposition, thus:

GIB: S infallibly believes = df S believes every true proposition that is logically possible to believe and S believes no false proposition.

Now, take (G):

(G) God believed at t1 that Al Dente would eat Fettuccini Alfredo at t3.

Given GIB and (G), we can see that if Al Dente refrained from eating the plate of Fettuccini Alfredo on December 25, 2009, then God would have held a different belief than the one he in fact held. That is, the event described in (G) would have been different, not simply the exemplification of the contingent proposition referred to. But this violates PFP, and thus it attributes to Al Dente the causal power over the past. Since no one has causal power over the past, a claim Ockhamists do not wish to deny, then Al Dente cannot refrain from eating his plate of pasta, and therefore, Al Dente does not act freely when he so eats.

Of course, the Ockhamist is free to present an account of the hard/soft fact distinction where facts like (3) do not express hard facts, but this seems unlikely. Certainly, as many have pointed out: the past beliefs of people seem to be as good an example as any of something that is fully past, and it would be an ad hoc move, invoked to save a theory, to claim that the past beliefs of a divine person are not fully past. It certainly seems that my belief at 6:00 am this morning is, at 6:01, something completely in the past, and thus unalterable. Yet, there is another way out. The Ockhamist may affirm that we do not have causal power over the past, but we do have another kind of power over the past: counterfactual power. I will take up this objection in a later post, Lord willing.


____________________________

¹ One view is that the fact’s grammar points to the past (using past tense) while its content is not entirely about the past (cf. David Hunt, “The Simple Foreknowledge View,” in Divine Foreknowledge: Four Views, 77).

² Of course, since Al Dente and his dinner is my illustration, Plantinga does not refer to that example specifically.

³ One reason to think that the past is accidentally necessary is that it is fixed and lies outside our control.

⁴ The kind of power and control spoken of here is assumed to be the kind desired by libertarianism, i.e., the power to refrain from doing some action, or to do otherwise.

⁵ Widerker (2002) claims that this argument is given by Jonathan Edwards. Plantinga also presents this argument as a “powerful” expression of the incompatibilist’s worry (Ockham’s Way Out, 181). This argument is also expressed in Hasker (1998) and Linville (1995).

⁶ David Widerker, “Troubles With Ockhamism,” The Journal of Philosophy, v. 87, 9 (Sept. 1990), 465.

⁷ I’m using Widerker as a rough guide, ibid, 465-470.

⁸ Of course, libertarians would not say Al Dente ate freely, but that point is irrelevant to my example. Similar points could be made in light of God promising X. Suppose God promises at t1 to allow into heaven all those who trust in Jesus Christ, and Jones trusts in Jesus Christ at t3. Given the plausible assumption that “If God promises that X, then X,” this fact entails that Jones will be in heaven at t4, where t4 > t3.

⁹ William Hasker, God, Time, and Foreknowledge, 82, though he goes on to offer a more salient and rigorous account, pp. 88-90.

¹⁰ S “brings about” Y iff: There exists an X such that S causes X to be the case and (X&H) => Y and ~ (H=>Y), where ‘H’ represents the history of the world prior to it coming to be the case that X (Hasker, “A New Anti-Molinist Argument,” Religious Studies, 35 (1999) 291.)

¹¹ Though agents do not always have this power, see Widerker, Troubles, 471.