The Mechanics of Divine Foreknowledge and Providence

A Time-Ordering Account

T. Ryan Byerly

• Publisher: Bloomsbury Academic
• Year: 2014
• Print ISBN:
  9781501318269
• eBook ISBN:
  9781623566869
• Edition: 1
• Topic: Theology & Religion
• Subtopic: Free Will & Determinism in Philosophy

How exactly could God achieve infallible foreknowledge of every future event, including the free actions of human persons? How could God exercise careful providence over these same events? Byerly offers a novel response to these important questions by contending that God exercises providence and achieves foreknowledge by ordering the times.

The first part of the book defends the importance of the above questions. After characterizing the contemporary freedom-foreknowledge debate, Byerly argues that it has focused too narrowly on a certain argument for theological fatalism, which attempts to show that the existence of infallible divine foreknowledge poses a unique threat to the existence of creaturely libertarian freedom. Byerly contends, however, that bare existence of infallible divine foreknowledge cannot threaten freedom in this way; at most, the mechanics whereby this foreknowledge is achieved might so threaten human freedom.

In the second part of the book, Byerly develops a model for understanding the mechanics whereby infallible foreknowledge is achieved that would not threaten creaturely libertarian freedom. According to the model, God infallibly foreknows every future event because God has placed the times that constitute the history of the world in primitive earlier-than relations to one another. After defending the consistency of this model of the mechanics of divine foreknowledge with creaturely libertarian freedom, the author applies it to divine providence more generally. A novel defense of concurrentism is the result.

Part One

From the Existence of Divine
Foreknowledge to Its Mechanics

1

The Foreknowledge Argument

The Introduction to this text began with two questions—one about the relationship between human freedom and divine foreknowledge and one about the relationship between human freedom and divine providence. In Part I, my focus is on the question concerning freedom and foreknowledge: “If God knew long ago with perfection what I will do each day for the rest of my life, then how could it be that what I do each day for the rest of my life is genuinely up to me?”

Discussion of this question among contemporary philosophers has tended to focus on a certain argument that attempts to show that an infallible God’s having long ago possessed exhaustive foreknowledge implies that no human being performs any action she performs freely. For ease of reference, I will call this argument the foreknowledge argument. In this chapter, I present a careful version of the foreknowledge argument, compare it with two similar fatalistic arguments, and offer an up-to-date discussion of responses to it which reject one of its premises or inferences. My goal will be to establish criteria for judging a response to the foreknowledge argument to be successful and to cast significant doubt on whether any response to the argument which rejects one of its premises or inferences satisfies these criteria. One might conclude from this that the foreknowledge argument is a success. But I think that judgment is premature. Rather, my hope is to pique the reader’s interest in the prospects for developing a successful response to the argument which does not involve rejecting one of the argument’s key premises or inferences.

1. The foreknowledge argument

The aim of the present section is to present and explain a leading version of the foreknowledge argument, one which has been called by some authors the best version of this argument.¹ As we will see, the foreknowledge argument makes salient a puzzle which is generated by two motivations which many religious believers in the west have shared—a motivation on the one hand to maintain a commitment to divine cognitive perfection and a motivation on the other hand to uphold a certain kind of freedom for human beings.

We can start with the following informal presentation of the foreknowledge argument, which we will later compare to informal versions of parallel arguments for causal fatalism and logical fatalism:

There’s nothing I can now do about what anyone, including God, believed long ago. But, then, if God believed long ago that I will do certain things each day for the rest of my life, and God cannot be mistaken, then there’s nothing I can now do about these things, either. So, if God believed long ago that I will do certain things each day for the rest of my life and God cannot be mistaken, then whether I do those things isn’t up to me.

This informal line of reasoning is likely to strike a chord with any reader who has given much thought to the question of whether infallible divine foreknowledge and human freedom are compatible. And, it is not far off from the most carefully developed contemporary statements of the foreknowledge argument.

The informal version of the foreknowledge argument just presented can be precisified into a formal version which makes its key logical features and commitments explicit. The formal version of the foreknowledge argument I will present is a conditional proof. Assuming that an infallible God believed some claim in the distant past concerning an arbitrarily selected human being’s future action, the argument attempts to show that this action could not be done freely. Thus, infallible divine foreknowledge and human free action are incompatible.

We can get the formal version of the foreknowledge argument started by adopting two suppositions which we will assume for the purpose of providing the conditional proof. The first supposition concerns a divine forebelief:

(1) God believed at t₁ that Elizabeth would sing a love sonnet to John at t₁₀₀.

Let t₁ refer to a time long, long before time t₁₀₀, even before Elizabeth or John came into being. Stipulating that t₁ is earlier than t₁₀₀ is what makes the foreknowledge argument an argument about foreknowledge—or, more precisely, forebelief. As indicated already, the selection of Elizabeth and her action of singing a love sonnet is entirely arbitrary.

The second supposition we need concerns divine infallibility. What we will focus on is a requirement for divine infallibility which says that God can’t be wrong in his beliefs.² It is tempting to explain this idea as follows: necessarily, if God believes something, it is so. The motivations for claiming that God meets this requirement for infallibility typically derive from strong commitments concerning divine cognitive perfection. For example, classical theism holds that it is central to divine cognitive perfection that God is omniscient. This is to say, roughly, that for every proposition p, if p, then God knows that p. Further, it isn’t just that God simply happens to know every proposition that is the case but rather that he couldn’t have failed to know each proposition that is the case. Thus, necessarily, for every proposition p, if p, then God knows p. However, plausibly, any person who knows a proposition p both believes p and doesn’t believe not-p. So, given the classical theist’s view that God couldn’t fail to know each proposition which is the case, it follows that, necessarily, for any proposition p, if p, then God believes p and God doesn’t believe not-p. And from this claim the requirement of divine infallibility stated earlier can be proven: namely, it is necessarily the case that if God believes p, then p.³

The claim that, necessarily, if God believes p then p is not quite the claim that we will use for our argument, however. Instead, we will use a claim which focuses both more narrowly on divine beliefs about the actions of human agents at times and more specifically on divine beliefs held at times. Informally, the claim we will focus on says: necessarily, if God believes at any time that some human agent performs an action at some time, then that agent does perform that action at that time. Formally, we can express our second supposition as follows:

(2)

_L ∀t, t′, S, A (God believes at t that S does A at t′ → S does A at t′)

In (2), “

_L” indicates logical necessity, “→” indicates the material “if-then” conditional, and “∀” indicates the universal quantifier. Thus, in English, (2) says: “it is logically necessary that, for any times t and t′, any human agent S and any action A, if God believes at t that S does A at t′, then S does A at t′.”

There are several further notes I should offer to help explain the meaning of (2) and to explain why (2), rather than some similar claim about infallibility, is being put to work in the foreknowledge argument. First, I use

_L and speak of logical necessity to distinguish the necessity here from other kinds of necessities. I use the language of “logical necessity” and “logical possibility” to indicate that a thing is necessary or possible in an absolutely unqualified way, in contrast to things which are possible or necessary subject to certain qualifications. An example of a claim which is necessarily the case in this unqualified way is the claim all green things are colored. It is impossible that this claim be false, not just subject to certain qualifications, but absolutely. It is always and under every condition impossible for something to be green and non-colored, not just impossible relative to certain times or certain conditions. (2) claims that it is necessary in this unqualified way—necessary at all times and under all conditions—that if God believes at any time a claim about what an agent does at some time, then this agent does what God believes he or she does at that time. The necessity involved here is the same sort of necessity involved in the classical theist’s claim that God is necessarily such that for any proposition p, if p, then God knows p.

Second, as highlighted earlier, (2) concerns divine beliefs held at times rather than simply divine beliefs. The reason for focusing attention in (2) on divine beliefs held at times is that (1) likewise focuses on a divine belief held at a time. Prima facie, someone who is willing to suppose, with (1), that God holds beliefs at times and who is also attracted to the view that God can’t go wrong in his beliefs, should be happy to endorse supposition (2).

Finally, it is noteworthy that (2) focuses only on divine beliefs about what agents do at times rather than divine beliefs about anything whatsoever. For instance, we might have imagined that instead of (2), the following claim could have been used in the foreknowledge argument: necessarily, if God believes any proposition p at any time t, then p. Apart from the fact that, following (1), the foreknowledge argument is itself already focused only on a belief about what an agent does at a time, it turns out that there are also theoretical advantages to restricting (2) to divine beliefs about what agents do, or at least what happens, at times. For, there are objections which would face (2) if it concerned divine beliefs about just anything, objections that are overcome given the restrictions found in (2) as it is stated. To see this, one needs only to observe that there are some philosophers who have been attracted to the view that certain propositions change their truth-values.⁴ For example, it may be true at one time that Socrates is bent and not true at some later time that Socrates is bent. Other propositions, such as Socrates is bent at time t more plausibly, do not change their truth-values. If Socrates is bent at t, then at every time it is true that Socrates is bent at t. A person who thought that some, though not all, propositions can change their truth-values would be unlikely to explain divine infallibility by claiming that necessarily, for every proposition p, if God believes p at any time, then p. Such a person may nonetheless be quite happy saying that necessarily if God believes at any time that an agent S does an action A at t, then S does A at t. Thus, restricting the objects of divine beliefs in (2) to propositions concerning what agents do at times has a significant theoretical advantage over a similar claim about divine infallibility concerning divine beliefs about just anything.

Claims (1) and (2) are the two suppositions of the foreknowledge argument. The remainder of the argument attempts to show that suppositions (1) and (2) imply that Elizabeth’s singing her love sonnet to John at t₁₀₀ is not something she does freely. In order to do so, the argument will need to make several additional claims about the nature of the past, necessity, and freedom.

Let us begin with a key claim about the past, a claim sometimes called the Principle of the Necessity of the Past or the Principle of the Fixity of the Past:

(3) ∀t, t′, x [(x obtains at t and t < t′)→ (

_A at t′ that x obtains at t)]

Here “∀” is used as before, while “<” is used to indicate “is earlier than” and “

_A” is used to indicate “it is accidentally necessary that”. Thus, in English, (3) claims: “For all times t and t′ and each state of affairs x, if x obtains at t and t is earlier than t′, then it is accidentally necessary at t′ that x obtained at t.” A bit more intuitively and less cumbersomely, (3) affirms that every state of affairs that obtained in the past is now necessary—accidentally necessary.⁵

The key to understanding claim (3) and its plausibility is to understand the notion of accidental necessity,

_A. Accidental necessity, in contrast to logical necessity, is a sort of necessity subject to certain conditions—chiefly temporal conditions. Its distinguishing mark is that it is a necessity relativized to times. Nothing is accidentally necessary unqualifiedly; what is accidentally necessary is necessary only at some time or other.^{6, 7}

Is there anything that is necessary at some times but not at others, possible at some times but not at others? Intuitive examples suggest an affirmative answer. For example, given that Lincoln was shot, it is not now possible that Lincoln has not been shot. Lincoln’s not having been shot is now impossible. Of course, there may have been a time, say, on the night when Lincoln was making his way to the theater, when it was possible for Lincoln not to be shot. It seems to follow then that there is a state of affairs, that of Lincoln’s not being shot, which was at one time possible, but is no longer possible. Conversely, Lincoln’s being shot is an example of something which has to be at some times—i.e., it is necessary at some times—such as the present, but didn’t have to be at other times. Thus, there is some sort of necessity which is accidental—a necessity relative to times.

Examples of past states of affairs are very useful in illustrating the concept of accidental necessity. And this may be for very good reason. For, it may be that every past state of affairs is accidentally necessary. Everything that has happened in the past now cannot fail to have happened. But this does not seem to be the case with the future. At least certain future happenings seem as if they now could happen and now could not happen. It seems as if it is now possible for them to happen and now possible for them not to happen. That the past is accidentally necessary may then be part of what sets i...

Table of contents

• Cover
• Halftitle Page
• Title Page
• Contents 
• Acknowledgments
• Introduction
• Part One: From the Existence of Divine Foreknowledge to Its Mechanics
• Part Two: A Time-ordering Account of Foreknowledge and Providence
• Bibliography
• Index
• Imprint