I had recently thought to finally make an effort to complete a post that has been in the draft phases for around 7 years now in which I revisited necessitarianism, a position I formerly held (example). There was already much to say on the subject even before Amy Karofsky published an interesting - if flawed - book providing a rare defense of necessitarianism (link). With that publication, I thought to use it as a foil for some reflections I've meant to publish for a while now.
Was Augustine a Necessitarian?
One argument Karofsky makes is that Augustine was a necessitarianism:
Augustine also believes that the future is necessitated. Combining Christianity with aspects of Plato's metaphysical, epistemological, and ethical theories, he maintains that there is an ultimate, necessary basis that is a fully good and all-knowing God. Because God is omniscient, God knows what will occur in the future, and because God's intellect and will are unchanging, so, too, is the way that the world will unfold. God even knows what a person will choose to do. However, Augustine explains, such divine foreknowledge does not cause or force the person to choose in the way that they do; the person still chooses of their own will. (pg. 10)
...it is, of course, possible to live by the theory of necessitarianism - Parmenides, the Stoics, Augustine, Spinoza, and I are all cases in point. (A Case for Necessitarianism, pg. 158)
In reality, to suggest Augustine believed the future is "necessitated" or was a "necessitarian" is misleading, to say the least. The simplest refutation is that Augustine argued our will is not necessitated: "We sin not by necessity but by the will" (link).
Now, the future is indeed foreknown by God. In fact, Karofsky is correct that Augustine believes God's knowledge to be eternal (The Confessions, Book XI, link). But this does not, to Augustine, imply necessitarianism. When Evodius asks, in On the Free Choice of the Will, "How is it that future events God foreknows do not happen necessarily?" Augustine's answer is, in part, that "God foreknows all the things of which He is the author and yet is not the author of all the things He foresees. He is not the evil author of these things..."
It may be inconsistent for Augustine to hold that God's "eternal" knowledge is [seemingly] contingent on future events like human free will - events and free will that is not necessitated - but that is indeed what it appears that Augustine believed, and nothing in The Confessions indicates otherwise.
Perhaps one might argue that the above only reflects Augustine's earlier thought. On the Free Choice of the Will was written at least a decade before The City of God, the other work by Augustine which Karofsky cites. But it turns out that Karofsky's own citation (book 5, chapter 10) is precisely what rules Augustine out as a necessitarian. Augustine writes:
Wherefore, neither is that necessity to be feared, for dread of which the Stoics labored to make such distinctions among the causes of things as should enable them to rescue certain things from the dominion of necessity, and to subject others to it. Among those things which they wished not to be subject to necessity they placed our wills, knowing that they would not be free if subjected to necessity. For if that is to be called our necessity which is not in our power, but even though we be unwilling effects what it can effect — as, for instance, the necessity of death — it is manifest that our wills by which we live up-rightly or wickedly are not under such a necessity; for we do many things which, if we were not willing, we should certainly not do. This is primarily true of the act of willing itself — for if we will, it is; if we will not, it is not — for we should not will if we were unwilling. But if we define necessity to be that according to which we say that it is necessary that anything be of such or such a nature, or be done in such and such a manner, I know not why we should have any dread of that necessity taking away the freedom of our will. For we do not put the life of God or the foreknowledge of God under necessity if we should say that it is necessary that God should live forever, and foreknow all things; as neither is His power diminished when we say that He cannot die or fall into error — for this is in such a way impossible to Him, that if it were possible for Him, He would be of less power. But assuredly He is rightly called omnipotent, though He can neither die nor fall into error. For He is called omnipotent on account of His doing what He wills, not on account of His suffering what He wills not; for if that should befall Him, He would by no means be omnipotent. Wherefore, He cannot do some things for the very reason that He is omnipotent. So also, when we say that it is necessary that, when we will, we will by free choice, in so saying we both affirm what is true beyond doubt, and do not still subject our wills thereby to a necessity which destroys liberty. (link)
Later in The City of God, a concrete example of prelapsarian Adam illustrates exactly what Augustine means by "liberty" and "free will," and his statements, such as the following, are unintelligible within the framework of Karofsky's necessitarianism: "the first freedom of will which man received when he was created upright consisted in an ability not to sin, but also in an ability to sin" (link).
How Karofsky can be aware of such a work yet classify Augustine as a necessitarian is a mystery, for this is exactly what Karofsky denies is possible to a necessitarian: "a necessitarian will reject any account that takes free will to mean that a person can sometimes choose one of (sic) more than one possible alternative beliefs... Instead, a necessitarian believes... nothing could have been otherwise" (A Case for Necessitarianism, pg. 153).
To drive the point home, Augustine not only believed that man is - or, at least, was (prior to the first sin) free - he believed that God is free. In The Enchiridion, Augustine says, "both baptism and death, were submitted to by Him, not through a pitiable necessity, but of His own free pity for us" (link). And again, "The omnipotent God, then, whether in mercy He pities whom He will, or in judgment hardens whom He will, is never unjust in what He does, never does anything except of His own free-will, and never wills anything that He does not perform." While God's grace is necessary for our salvation, salvation itself not necessitated but is rather a "free gift."
What is "Theistic Contingentarianism"?
Karofsky defines contingentarianism as the contradictory of necessitarianism. The former (contingentarianism) is the view that "at least some thing about the universe could have been otherwise in some way or other" (A Case for Necessitarianism, pg. 1). The latter view (necessitarianism), as mentioned above, is that "nothing could have been otherwise." Insofar as many theists defend divine freedom in accepting that God could have refrained from creating, theistic contingentarianism is a species of contingentarianism.
Christianity and Theistic Contingentarianism
What motivates theistic contingentarianism? There's probably more than one answer to this question, so I'll answer for myself one reason why I now disbelieve necessitarianism. If necessitarianism is true, creation is necessitated. What I eventually have come to conclude is that necessitarianism is incompatible with the biblical doctrine of creation and/or God.
Christians believe in a Creator-creature distinction. Compare this to Karofsky's monism as expressed in pages 137-144. She admits that she believes, for example that "the essence of one thing contains the essence of every other." Any attempt to develop a theistic necessitarianism along such lines would be pantheistic, anti-biblical, and a non-starter for Christians.
While the above makes a case for why a Christian might be motivated to accept theistic contingentarianism, Karofsky argues that theistic contingentarianism in fact fails. As such, the book implicitly attacks Christianity.
One such argument it fails is as follows:
In general, it seems that any theory that attempts to account for contingency by appeal to entities that are in some way necessary and in some way contingent will fail. Indeed, the other leading theory that posits a basis for modalities that is partly contingent and partly necessary also fails, and does so for pretty much the same reasons that dispositionalists account fail. That theory is what I will call contingentarian theism.As I explained in the Introduction, a contingentarian-theist typically maintains that although God is a necessary being, God has free will, and, as such, God could have chosen to create a different world with different inhabitants. God's necessary nature provides for the necessity of laws, essences, and eternal truths, while God's free will provides for the contingency of the existence of the entitites in the world and some of their properties. However, there are many problems with the contingentarian theistic account of contingency, and many of those problems are similar to those that arise for dispositionalist accounts.Both the dispositionalist and the contingentarian-theist attribute a contingent feature to an otherwise necessary basis. The dispositionalist attributes potency to the essential nature of dispositional properties and the contingentarian-theist attributes free will to the necessary nature of God. As I showed earlier, the contingent feature of dispositions not only fails to account for contingency but also prevents the dispositionalist from providing an adequate account of necessity. A similar problem arises in contingentarian theism and can be expressed as a Euthyphro-type dilemma that might go like this: When God chooses to create from among possible alternative, either God makes the choice for some reason, or God makes the choice for no reason. If God chooses to create this world for some reason (say that it is the best), that reason determines God's choice, in which case, the choice could not have been made differently, and the outcome is necessary and not contingent. However, if God chooses for no reason (God just happens to choose this world), then God's choice is arbitrary and any choice can follow, in which case anything is possible, and nothing is necessary. So, either nothing is contingent, or nothing is necessary.The problem is not that difficult to see. Both the dispositionalist and the contingentarian-theist recognize that a necessary basis is needed in order to have necessary natural laws and necessary eternal truths. Both also recognize that if the basis is entirely necessary, there can be no contingency. In a desire to avoid necessitarianism, they attribute some contingent feature to that basis. But then that basis is such that it could have been otherwise in at least some way, in which case it is not necessary and cannot provide an adequate basis for the necessity of certain laws and truths. In essence, one cannot have one's cake and eat it, too.Finally, the further condition that was derived from Shalkowski's argument works to show that both dispositionalism and theistic contingentarianism fail because, according to that condition, only what is necessary can be a member of basis. (Karofsky, A Case For Necessitarianism, pgs. 76-77)
Paragraphs 1 and 2 set the context for the argument. Paragraphs 3 spells out an argument to which I intend to reply below, as it is similar to the following arguments I made when I was a necessitarian (with some differences, on which, again, see below):
Why did God decree to maximally manifest His glory? I don't see any other possible conclusion than something similar to this or what I said to my friend. In fact, to say there are multiple possible worlds is simply to say God's instantiation of this possible world was not necessary. On this supposition, can there be a reason God instantiated this possible world? Would not such a reason imply the necessity of the instantiation? If not, then is not the alleged reason an arbitrary one? (link)
I then elaborated on this argument here:
My current response would be that nothing which occurs is unnecessary, for that would imply, as Clarke wishes to avoid, that God’s will is arbitrary. Sure, on the assumption that there are multiple possible worlds, God may have a reason for instantiating some particular possible world, but as on Clarke’s view such a reason would not be necessary, it would still be arbitrary. I may as well ask what God’s reason was for choosing His reason for creating this possible world over against any other reason capable of being chosen which might have led to the instantiation of another possible world, and there could be no answer because the reason itself was chosen arbitrarily. (link)
And once more, I make the same essential argument here:
If I ask what the basis or reason is that, given that world A (in which God creates) and world B (in which God never creates) are both agreeable to His nature, God chose to instantiate world A rather than world B, the answer “it was agreeable to His nature” doesn’t suffice. For world B would also have been agreeable to His nature. The arbitrarity objection stands. (link)
To try to put the argument as forcefully as I can think to, one further way to state it is to suppose God was free to create or not create as He has. On this supposition, further suppose we say God has some reason[s] w to create this world. Reason[s] w would not necessitate the creation of this world, so conceivably, God could have refrained from creating this world for reasons x.
Now, any reason[s] y which we might posit as an explanation for why God chose reason[s] w (on which account He created this world) rather than reason[s] x (on which account He might have refrained from creating this world) would be reason[s] subject to the same point ad infinitum.
That is, reason[s] y would not necessitate choice of reason[s] w (on which account He created this world), so conceivably, God could have had chosen reason[s] x (on which account He might have refrained from creating this world) for reason[s] z.
Reason[s] w do not explain why God did not refrain from creating this world. Reason[s] y do not explain why God did not choose reason[s] x, etc. In short, no appeal to "reasons" can serve as a full-stop explanation of why God chose to create this world rather than refrain from creating this world. On theistic contingentarianism, any attempt to work out why this world was created rather than not would generate an infinite regress. So what Karofsky argues is similar to what I too have entertained.
So stated: should a theistic-contingentarian regard this objection as problematic?
I now argue that it is not problematic.
Was this creation necessitated? Did God have to create at all? If divine sufficiency is true and pantheism is false, I think not.
This does beg the question as to whether God's choice to create this world is "arbitrary." In a sense, the answer to this must be yes. In another sense, no.
When I was a child, one of the first philosophically interesting things I read was in Animorphs: "...can you decide to do nothing? That's a decision too." (K. A. Applegate, The Message). My dad also listened to Rush quite frequently, and their following refrain in the song Freewill communicates the same point: "If you choose not to decide, you still have made a choice."
That is, God either could have chosen to create - as He did - or chosen not to create. But a choice simpliciter had to be made. A choice was necessary even though one specified choice was not. Accommodating our language to the timelessness of God's eternal choice is not easy, but however one wants to phrase it, the reality of having to make a timeless choice itself was not arbitrary.
Neither was God's choice arbitrary in the following sense - God could have had [and indeed does have] reasons for creating this world "so that the manifold wisdom of God might now be made known through the church to the rulers and the authorities in the heavenly places." One may ask for what reason He wanted to make that known. I believe the answer to all such teleological questioning is that all things exist for the glory of God. Although such reasons for His choice are no more necessitated than the choice to create itself, they are still reasons.
This departs from Karofsky: in her third paragraph, when she says that if necessitarianism were false, then "any choice can follow," "that anything is possible," or that "nothing is necessary." If I understand her correctly, I don't follow her reasoning here and disagree with her. That God, being free, is not limited to one choice does not imply "any choice can follow" in the sense, for example, of "any choice [which we could imagine]." We have epistemic limitations, as Karofsky herself acknowledges. These limitations may prevent us from understanding why some choices which we might imagine God could have made might not, in fact, "follow" or be consistent with the nature of God.
Finally, then, in what way must we admit that God's choice is "arbitrary"? God's free choice is the final, metaphysical (causal) explanation of how this world came to be. If we attempt to ask why or for what reason this world was chosen rather than not so created in the sense of seeking a deeper, more fundamental metaphysical explanation for God's choice, I don't think we find one. While it could have "pleased" God to refrain from creating this world, He was "pleased" to freely choose to create this world. He did not have to choose to create this world, and the same can be said for any "reasons" God may have had in freely choosing to create this world. Again, however, this is not a problem, for God had to choose, and He [freely] has.
Moving on, in reply to Karofsky's fourth and fifth paragraphs on pages 76-77 of her book (cited above), God is indeed the necessary basis for the contingentarian-theist. But I don't think this statement of hers follows: "if the basis is entirely necessary, there can be no contingency." A thorough reply would require a much longer post than I intend here to provide, as it would require a full-fledged exposition of one's metaphysical views to form a complete response to Karofsky. I will, however, offer a few reflections.
Initially, I didn't understand on what grounds she thinks a necessary being would have to [timelessly] act in a way that is necessitated rather than merely in a manner consistent with the nature of said being. In a different context, she makes a similar assertion: "the divine essence includes the divine will and thus includes everything that God chooses" (pg. 12). By initial appearances, she conflates God's will with the exercise thereof.
Additionally, given her first argument against Aquinas on page 11, I am not sure to what extent Karofsky is familiar with distinctions between God's natural and free knowledge. Her second and third arguments against Aquinas also lead me to wonder how much exposure she has to theistic replies to Euthyphro's dilemma.
An alternative explanation for pages 11-12 might be that Karofsky is only attempting to internally critique the philosophers about whom she is writing (Aquinas, Leibniz). Even if that is the case, her case against theistic contingentarianism on page 76 seemed lacking, and given that I disagreed with her exposition of Augustine, if I knew more about Aristotle and/or Leibniz, I could imagine finding myself disagreeing with her historical analysis of them too.
In general, it was my impression, then, that Karofsky's arguments against theistic contingentarianism here were laden with hidden, unsubstantiated metaphysical assumptions. I thought she might regard certain, controversial views of God - e.g. absolute divine simplicity - as standard for all Christians.
As it turns out, she lays out the metaphysical background for her criticisms a little later on (pgs. 81-82, 92ff., 137ff.). I'll plan to address these in a separate post, but I will touch on my disagreement with them below as a precursor of further thoughts.
Is Non-Theistic Necessitarianism Epistemically Viable?
While I might at some other time discuss Karofsky's ethical and metaphysical views in more detail, for now, I will make one criticism - a crucial one, in my mind - of Karofsky's necessitarianism. Is it epistemically viable?
Earlier, I mentioned Karofsky's monism to be of such a nature that "the essence of one thing contains the essence of every other." Correspondingly, and in the same context, Karofsky also concedes that on her view, "the meaning of one term includes the meaning of all terms" (pg. 143).
These and other such similar statements she makes means she believes herself and fellow necessitarians to be committed to the doctrine of internal relations. This would be epistemically problematic, as Gordon Clark pointed out (before, ironically, he too changed his mind to necessitarianism - without addressing any criticisms of it that I mention in this post - cf. his book The Trinity, pgs. 111-119):
Reenactment of a thought is possible, nonetheless, because it can be separate from this immediacy without alteration. Not only so, it can be separated from other thoughts without alteration. Thus history becomes possible.
This self-identity of the act of thought has been denied by two extreme views. The first view is that of idealism, the theory of internal relations, the notion that everything is what it is because of its context. This makes history impossible. To know any one thing it would be necessary to know its context; i.e. to know the whole universe. Knowledge would thus be restricted to the explicit consciousness of the omniscient Absolutes; and Collingwood, though he may be Beckett, does not claim to be the Absolute...
What follows if it is true that psychological analysis presupposes a “complete knowledge of the psychological possibilities of life”? It would follow, would it not, that historical analysis also presupposes a complete knowledge of historical possibilities. In short, it would be impossible to know anything without knowing everything.
Such a Platonic or Hegelian requirement of omniscience is a serious philosophical problem. It is not to be dismissed thoughtlessly. The meticulous scholar, J. H. Hexter, in his Reappraisals of History, castigates historical relativism as a fad and insists on the “rudimentary distinction” between knowing something and knowing everything. But he omits all philosophic justification for this distinction.
Undoubtedly this distinction must be maintained, if a human being is able to know anything at all. Make omniscience the prerequisite of partial knowledge, and partial knowledge vanishes. But Bultmann, like Hexter, offers no help: less help, in fact, for Bultmann lets the requirements of omniscience stand. (Historiography, Secular and Religious, pgs. 225-226, 334)
Given that Karofsky admits epistemic limitations throughout her book, these and similar arguments (more of which can be read here) pose serious difficulties to her. She does not and indeed cannot know the meaning of all terms; therefore, on her own grounds, she cannot know the meaning of even one term. Her brand of necessitarianism entails epistemic skepticism.
Thus, I think the problem of partial knowledge illustrates why a theistic contingentarian metaphysic is more defensible than Karosky's: it is more cohesive with a sound epistemology (link, link).