Page 12

causal, between the antecedent and the consequent. And, of course, there is no
connection, no relevance, between the antecedent and the consequent of (F).

Now take the counterfactual conditional--not the counterfactual of freedom, but the

counterfactual simpliciter:

(G) If I were to offer my wife the choice between liver and onions or ice cream, then

she would choose ice cream.

I believe, and I think I know, that (G) is true. I might even say to you, "If you knew my
wife, then you would know that she would choose ice cream in this situation." Yet notice
the antecedent of this last sentence very carefully: "If you knew my wife..." What am I
saying? Am I not saying that if you knew, like I know, my wife's character, her beliefs,
her desires, habits, etc., then you would know what she would choose?

Why do we think that counterfactual conditionals like (G) have truth-value, and that

we can and do know them? Isn't it because we think (or assume) that there is a
connection, probably a causal connection, between the antecedent and the consequent? I
think so.

But now let us suppose that my wife has libertarian freedom. Let us suppose that,

given any set of antecedent circumstances, such as my offering her the choice of liver and
onions or ice cream, she is free in that circumstance to choose one or the other, to choose
ice cream or refrain from ice cream. It seems to me that once we make this assumption,
once we interject libertarianism into the situation, things have to change. They have to
change because the connection we might otherwise assume between the antecedent and
the consequent has been severed. What grounded the truth of (G) without this
assumption, namely, my wife's character, can no longer serve that function. Assuming
that the agents who are the subjects of counterfactuals have libertarian freedom
obliterates any connection, any relevance between the antecedent and the consequent of
those propositions. And if there is no such connection, then it seems that CFs are
analogous to proposition (F): "If the moon is made of green cheese, then Alvin Plantinga
is an android." If we think that the latter is absurd and ought not to be accepted as true, if
we think, in other words, that (F) is ungrounded, then we ought to think that CFs are
ungrounded as well.

The real culprit here, then--the basis of the GO--is libertarianism, or perhaps more

generally, indeterminism. As Peter Van Inwagen has said, the GO "depends on no other
features of free acts other than the fact that they are undetermined (a consequence of
incompatibilism)."

Now it is not my intent in this paper to critique indeterminism. It is

my point here only that indeterminism has implications--implications for counterfactuals
concerning free creatures that God is alleged to know prior to his creative decision, and

Peter Van Inwagen, "Against Middle Knowledge." This is an unpublished paper which can found

at www.soci.niu.edu/phildept/speakers/VanInwagen_AgainstMiddleKnowledge.html.