Running head: FOREKNOWLEDGE 1

Arminianism and Molinism on Divine Foreknowledge

Nathan Justice

A Senior Thesis submitted in partial fulfillment

of the requirements for graduation
in the Honors Program
Liberty University
Spring 2017
FOREKNOWLEDGE 2

Acceptance of Senior Honors Thesis

This Senior Honors Thesis is accepted in partial

fulfillment of the requirements for graduation from the
Honors Program of Liberty University.

______________________________
Gaylen Leverett, Ph.D.
Thesis Chair

______________________________
Joshua Chatraw, Ph.D.
Committee Member

______________________________
David Beck, Ph.D.
Committee Member

______________________________
Marilyn Gadomski, Ph.D.
Honors Assistant Director

______________________________
Date
FOREKNOWLEDGE 3

Table of Contents
I. Arminianism .......................................................................................................... 5
A. Arminian Definition of Divine Foreknowledge .............................................................5
B. Biblical Evidence for Arminianism ...............................................................................6
C. Theological Evidence for Arminianism ........................................................................9
D. Philosophical Evidence for Arminianism......................................................................9
1. The reality of future contingents...................................................................................... 9
2. Robert Kane’s event-causal libertarian freedom. ........................................................ 11
E. Critiques of Arminianism ........................................................................................... 14
1. Biblical critiques. ............................................................................................................. 14
2. The compatibility of divine foreknowledge and human action. .................................. 17
3. Critiques of libertarian freedom. ................................................................................... 26
4. Simple foreknowledge and providential control........................................................... 36
II. Molinism ............................................................................................................ 41
A. The Molinist Definition of Divine Foreknowledge ...................................................... 41
B. Biblical Evidence for Molinism ................................................................................... 44
C. Theological Evidence for Molinism ............................................................................ 45
D. Philosophical Evidence for Molinism ......................................................................... 47
E. Critiques of Molinism ................................................................................................. 48
1. Biblical critiques. ............................................................................................................. 48
2. Theological critiques. ...................................................................................................... 49
3. Semantic and metaphysical objections. ......................................................................... 52
4. Critiques of the conjunction between Molinism and libertarian freedom ................. 56
III. The Author’s View ............................................................................................ 60
IV. Conclusion......................................................................................................... 63
Bibliography ........................................................................................................... 64
FOREKNOWLEDGE 4

Abstract

Evidence is examined concerning the coherence of divine foreknowledge as

defined by Arminianism and Molinism. Arminianism argues that God has complete and

infallible knowledge of the future, and attempts to simultaneously maintain a strong view

of libertarian freedom. Molinism agrees with the Arminian stance on foreknowledge and

human freedom, but argues that middle knowledge must also be posited for God to have

strong providential control over His creation. It is argued that Molinism better accounts for

the biblical data and provides a more coherent theological and philosophical position, since

Arminianism cannot provide a strong theory of providential control. Subsequently, a

defense is given for Reformed Molinism, which combines Molinism with Robert Kane’s

event-causal version of libertarian freedom.

FOREKNOWLEDGE 5

Arminianism and Molinism on Divine Foreknowledge

Throughout church history, Christians have commonly held that God has complete

knowledge of the future, commonly known as divine foreknowledge. This naturally

developed from the doctrine of omniscience: if God knows all true propositions, and there

are true propositions about the future, then God must have divine foreknowledge. However,

skeptics concerning divine foreknowledge have challenged this line of reasoning by

arguing that God’s omniscience does not include complete knowledge of the future.

Consequently, theologians and philosophers have attempted to defend the traditional

understanding of divine foreknowledge against new objections. In this study, these

defenses of divine foreknowledge will be examined, with the goal of finding which view

is the most biblically consistent and theologically/philosophically coherent. Specifically,

Arminianism and Molinism will be thoroughly scrutinized. As a result of this investigation,

Reformed Molinism seems to be the most theologically/philosophically coherent position.

I. Arminianism

A. Arminian Definition of Divine Foreknowledge

Arminianism argues for the traditional understanding of divine foreknowledge,

which says that God has “complete and infallible knowledge of the future.”1 This definition

of divine foreknowledge is also called “simple foreknowledge”, because Molinists accept

God’s complete and infallible knowledge of the future while also affirming God’s middle

knowledge. In defending simple foreknowledge, Arminians represent the most common

view of divine foreknowledge throughout church history as well as the most common

1. David Hunt, “The Simple-Foreknowledge View,” in Divine Foreknowledge: Four Views, edited
by J.K. Beilby and P.R. Eddy (Downers Grove, IL: InterVarsity Press, 2001), 65.
FOREKNOWLEDGE 6

interpretation of the biblical data. In distinction from Calvinism, which holds that God

foreknows because He foreordains, Arminius argued that

God foreknows future things through the infinity of his essence, and through the
pre-eminent perfection of his understanding and prescience, not as he willed or
decreed that they should necessarily be done, though he would not foreknow them
except as they were future, and they would not be future unless God had decreed
either to perform or to permit them.2

Though the content of what God knows is agreed upon by Arminians and Calvinists, how

God knows the future has been fiercely debated since the time of Arminius and Calvin.

B. Biblical Evidence for Arminianism

Throughout the Old Testament, the prophets and kings consistently declare God’s

knowledge of and mastery over the future. In Psalm 139, David proclaims that “before

there is a word on my tongue, Behold, O Lord, You know it all … And in Your book were

all written / The days that were ordained for me, When as yet there was not one of them.”3

This psalm not only attests to God’s foreknowledge of David’s words, but also His control

over David’s entire future. Similarly, 1 and 2 Kings have an entire promise-fulfillment

motif, with many of these promises involving the free choices of human beings. 4 By

2. Jacob Arminius, “Certain Articles To Be Diligently Examined And Weighed Because Some
Controversy Has Arisen Concerning Them Among Even Those Who Profess The Reformed Religion,” in
The Works of Jacob Arminius, vol. 2, translated by James Nichols (Buffalo, NY: Derby, Orton, and Mulligan,
1853), accessed January 21, 2017. http://wesley.nnu.edu/arminianism/the-works-of-james-arminius/volume-
2/controversial-articles-of-faith/.

3. Verses 4, 16.

4. Roy provides the following examples: 1 Kings 11:34-37 cf. 1 Kings 12:20 and 2 Kings 8:19; 1
Kings 13:2 cf. 2 Kings 21:26, 22:8-13, 23:3; 1 Kings 16:2-4 cf. 16:11-12; 1 Kings 17:13-16; 1 Kings 21:19
cf. 22:29-38; 1 Kings 21:23 cf. 2 Kings 9:6-10, 35-36, 10:17; 2 Kings 7:1-2, 16-20; 2 Kings 20:17-18 cf.
24:12-14; 1 Kings 13:5; Josh. 6:26 cf. 1 Kings 16:34; 2 Kings 1:16-17; 2 Kings 2:21-22; 2 Kings 4:43-44; 2
Kings 10:30 cf. 15:12; 1 Kings 14:5-6; 1 Kings 20:22, 26; 1 Kings 22:22, 28-40; 2 Kings 3:18-19, 24-25; 2
Kings 8:13, 15; 2 Kings 19:32-36. See S.C. Roy, How Much Does God Foreknow? A Comprehensive Biblical
Study (Downers Grove, IL: IVP Academic, 2006).
FOREKNOWLEDGE 7

emphasizing the fulfillment of God’s promises, the author of 1 and 2 Kings is arguing for

God’s covenantal faithfulness during the tumultuous years of infighting and bloodshed.5

Outside these two books, predictive prophecy is a prominent feature of the Old Testament,

whether it addresses certain actions of the Israelites, future judgment from Yahweh, or the

Messiah’s arrival. Since this genre is so prevalent, it would be futile to try to cite every

example of it.6 Consequently, the primary example will be Isaiah 40-48, which explains

why God uses predictive prophecy so frequently.

In Isaiah 40-48, God puts Israel and its idols on trial. As the prosecutor, God argues

that Israel has turned away from Him, the true God, to serve false deities. In order to

distinguish Himself as the true and living deity, God challenges the idols to predict the

future: “Let them bring forth and declare to us what is going to take place; As for the former

events, declare what they were, that we may consider them and know their outcome. Or

announce to us what is coming; Declare the things that are going to come afterward, that

we may know that you are gods.”7 God is grounding His own claim to deity on His ability

to proclaim the future; to prove Himself worthy of allegiance, God says that He will bring

forth Cyrus to rule as king and that He will deliver Israel from foreign captivity.8

5. Roy argues that “These numerous and widely varied examples from 1-2 Kings convincingly
demonstrate that the prophets of Israel are indeed predictors of the future. They can do so as spokespersons
of Yahweh, precisely because Yahweh does in fact know both the near and distant future. And he knows all
of the future, including the future free decisions of human beings. Thus to be true to the overwhelming
teaching of 1-2 Kings, we must affirm that the foreknowledge of Yahweh is truly exhaustive.” See How Much
Does God Foreknow, 43.

6. Roy records 2,323 predictive prophecies in both the Old and New Testaments in How Much Does
God Foreknow.

7. Isaiah 41:22-23.

8. Cf. 41:25-28; 42:8-9; 43:9-12; 44:7-8; 44:26-45:6, 20-21; 46:9-11; 48:3-11.

FOREKNOWLEDGE 8

Importantly, this prophecy involves hundreds of human choices, including the monumental

task of appointing Cyrus as king. Since God is staking His position as God of Israel on His

own ability to prophesy the future, it would be ludicrous to claim that God lacks exhaustive

foreknowledge of all events, including human choices.

Jesus makes a similar claim in the Gospel of John during the Passover meal.

Throughout His ministry, Jesus made several predictions about His future death at the

hands of the scribes and Pharisees, since he was trying to explain why His mission as

Messiah would defy contemporary Jewish expectations.9 Subsequently, Jesus gave the

following test for His own messianic claims: “From now on I am telling you before it

comes to pass, so that when it does occur, you may believe that I am He.”10 In other words,

Jesus states that if His predictions come true, then the disciples will know that He really

was the Messiah. He then predicted the betrayals of both Peter and Judas, which involved

the free decisions of both men. Just like God in Isaiah 40-48, it makes no sense for Jesus

to rest His messianic claim on His own ability to foreknow events if He lacked divine

foreknowledge.

Finally, several passages explain how God’s plan of salvation was decided before

the foundation of the world.11 This plan involves thousands of free human choices, if not

millions. To say that God lacks exhaustive divine foreknowledge simply ignores passages

which clearly describe God foreknowing people. Once one combines the predictive

9. Cf. John 2:19-22; Matt. 16:21; 17:22-23; 20:17-19; Mark 8:31; 9:31; 10:33-34; Luke 9:22, 44;
18:31-33.

10. John 13:19.

11. Cf. Rom. 8:29-30; 1 Pet. 1:2, 18-20; Eph. 1:3-14; 3:11; 1 Cor. 2:7; 2 Tim. 1:9-10; Rev. 13:8;
17:8.
FOREKNOWLEDGE 9

prophecies; God and Jesus both staking their identity claims on their ability to know the

future; and the other passages describing God’s foreknowing certain people and actions; it

becomes exceedingly difficult, if not impossible, to deny exhaustive divine foreknowledge.

C. Theological Evidence for Arminianism

Arminians argue that God’s foreknowledge makes the best sense of prayer, divine

guidance, and hope in God’s ultimate triumph over evil. In each of these areas, exhaustive

divine foreknowledge provides greater assurance of God’s promises. First, God knows the

outcomes of human prayers; He knows what trials we will face and which blessings will

come to fruition. This naturally leads to more trustworthy guidance from God. Since God

knows the consequences of human actions, we can know that the weightiest decisions we

make are understood and accounted for by God. Moreover, God knows the logistics of His

own victory over evil since He has foreknown it eternally. Through His exhaustive

knowledge of future events, God “can know how to order and rule our own lives and indeed

all of history to ensure that his purposes of grace will be ultimately victorious.”12 The

security and confidence of God’s people rests in His complete and infallible knowledge of

the future.

D. Philosophical Evidence for Arminianism

1. The reality of future contingents. Additionally, a positive philosophical case

can be made for divine foreknowledge. This case begins by defining God’s omniscience as

His knowledge of all true propositions. Since this definition is readily accepted by every

Christian position, the focus of the debate shifts to the truth value of future-tense

12. Roy, How Much Does God Foreknow, 277.

FOREKNOWLEDGE 10

statements. Peter Geach argued that “you can no more name an as yet non-existent object

than you can christen a baby not yet conceived, or ring a bell not yet cast … such a name

is not a name of an object until it has an object to name. And so existential quantification

cannot be applied to objects that do not yet exist.”13 Others follow this same line of

reasoning in an argument from truth against divine foreknowledge.14

However, why must truth only correspond to what presently exists? William Lane

Craig argues that for future-tense statements to be true, the present-tense version of the

same statement must also be true at some point in time. For instance, if someone said on

March 1, 2016 that “Donald Trump will win the presidential election,” that statement is

true because Donald Trump won the presidential election. It does not matter if the event

exists at the time the statement is made; it still corresponds to the fact that Donald Trump

won the presidential election. Second, the facts that constitute past-tense or present-tense

statements also constitute future-tense statements. It seems rather arbitrary to deny the truth

of a fact based on the tense in which it is stated. Third, this position results in a reductio ad

absurdum: if truth only corresponds to presently-existing events, and past events no longer

exist in the present, then to deny the truth of future-tense statements would also require the

denial of past-tense statements, which severely limits human knowledge. Given these three

arguments, it is reasonable to hold that future-tense statements or propositions can be true.

Therefore, God’s omniscience must include exhaustive foreknowledge.15

13. Peter Geach, Providence and Evil (Cambridge: Cambridge University Press, 1977), 54.

14. The argument from truth says that 1) Future events have not yet occurred. 2) A true statement
corresponds to what has occurred. 3) Hence, no statements about the future can be true (since the future has
not yet occurred). See Norman Geisler, The Battle for God: Responding to the Challenge of Neotheism
(Grand Rapids, MI: Kregel, 2001), 59.
FOREKNOWLEDGE 11

Beyond this argument for divine foreknowledge, Arminians also provide several

arguments for libertarian freedom, which is another key aspect of their position. There are

many different versions of libertarian freedom, although every variation can fit into three

broad categories: non-causalist or simple indeterminist, agent-causal, and event-causal or

causal indeterminist.16 Arminians could fit into any of these three categories; yet this paper

will focus on Robert Kane’s version of event-causal libertarian freedom, which has been

one of the most prominent libertarian views of the past 30 years.

2. Robert Kane’s event-causal libertarian freedom. For Kane, how one defines

a person’s “will” is vitally important to understanding human freedom. At the basic level

is a person’s appetitive will, which deals with the person’s desires and preferences; these

act as the motivation for a person’s actions. For instance, a person’s desire to eat an apple

is the reason why he or she buys an apple from the store. Second, there is the rational will,

which controls one’s choices, decisions, and intentions; for example, a person exercises

her rationality when she chooses to buy an apple rather than a pear. 17 Third, there is the

15. Craig develops these arguments in The Only Wise God: The Compatibility of Divine
Foreknowledge and Human Freedom (Wipf & Stock Publishing, 2000), 56-61.

16. “Agent-causal theories postulate ‘a sui generis form of [nonevent] causation’ by an agent or
substance that is not reducible to causation by states or events of any kind involving the agent, physical or
mental. Noncausalist or simple indeterminist theories insist that free choices or actions are uncaused events,
which are nonetheless explicable in terms of an agent’s reasons or purposes. Causal indeterminist or event-
causal theories maintain that agents cause their ‘free actions via [their] reasons for doing so, but
indeterministically.’” See Robert Kane, “Introduction: The Contours of Contemporary Free-Will Debates
(Part 2)”, in The Oxford Handbook of Free Will, 2nd edition, edited by Robert Kane (New York, NY: Oxford
University Press, 2011), 20.

17. “Choices and decisions are to be understood as mental acts whose function is to settle
uncertainty, terminate deliberation, and issue in intentions. Intentions are, in turn, states of mind that persist
through time and guide action by way of providing plans for organizing both immediate and future action.”
See Michael McKenna, “Compatibilist Ultimacy: Resisting the Threat of Kane’s U Condition*”, in
Libertarian Free Will: Contemporary Debates, edited by David Palmer (New York: Oxford University Press,
2014), 73.
FOREKNOWLEDGE 12

striving will, which is the efforts one makes to carry out one’s intentions. For libertarian

freedom to exist, Kane holds that there must be some indeterminacy between a person’s

appetitive will and rational will:

… it would be no threat to a libertarian conception of free will if the process giving

rise to desires and preferences was a deterministic one. Nor would it threaten
libertarian freedom if the relation between intentions once formed and overt actions
was also deterministic. What would, however, undermine libertarian freedom is if
the relation between an agent’s reasons once acquired and her subsequent choices
and decisions was deterministic. Settling uncertainty about what to do, for Kane,
given the reasons one has, is where the incompatibilists will locate distinctly
libertarian freedom.18

Additionally, Kane classifies three different kinds of free acts. First, there are Plain Free

Acts (PFAs), which are actions done voluntarily, on purpose, and without coercion. Kane

believes that these actions can occur in both deterministic and indeterministic situations,

thereby accommodating to both compatibilist and libertarian conceptions of free will.

Second, there are Freely Willed Acts (FWAs), which are done voluntarily, on purpose, and

without coercion, but also are performed out of an intention which the person is ultimately

responsible for. This means that a person could be motivated in a deterministic fashion,

thereby being unable to act otherwise, and still be able to perform an FWA.19

18. Ibid., 74. I recognize that the word “indeterminacy” has a specific meaning in quantum physics,
and this scientific use of the word has implications on the free will debate. However, Kane uses
“indeterminacy” to refer to a lack of causal necessity, and I will be using the word in the same fashion
throughout this essay. For those interested in the scientific side of the debate, I recommend reading Part II,
“Physics, Determinism, and Indeterminism,” of The Oxford Handbook of Free Will, 2nd edition, edited by
Robert Kane (New York: Oxford University Press, 2011).

19. “In the cases of many FWAs, probably most, all that is required is that the pertinent
psychological ingredients in the proximal etiology of her acts be ones for which she is personally responsible.
These ingredients have to be hers, in the sense that she helped form them. These motives, purposes, and
character traits might themselves be secured within her overall psychic constitution so that she ‘just is’ the
way she is at the time roughly coincident with the action. In this sense, she might, at that time, be unable to
do otherwise, and so fail to satisfy an AP [alternate possibility] condition within the context of that time
frame. Indeed, her performing of these FWAs thus might arise from locally deterministic causes that are
internal to her own motivational psychology. But they will still count as FWAs so long as the proximal will-
FOREKNOWLEDGE 13

The third kind of free act, Self-Forming Acts (SFAs), are most important in Kane’s

analysis, since they are the only kind of act where indeterminacy must occur. SFAs are the

actions that permanently define one’s character; they shape the course of one’s personal

history. Now these kinds of actions are not so powerful as to drastically change lives

because of one decision (though that may occur in extreme circumstances); lives are

changed “by repeated activity, in piecemeal fashion, through a series of SFAs over the

course of a life. Hence, the freedom of our agency, when we act of our own free wills is,

at least sometimes, an achievement that represents the long-term shaping of our dearest

conceptions of our own characters.”20 Thus, Kane holds that many decisions may arise

deterministically and persons can still be held responsible; but for persons to be held

responsible, one’s character must have been solidified over time through a succession of

SFAs.

These SFAs enable a person to be ultimately responsible, where ultimate

responsibility (UR) for an action requires that

an agent must be responsible for anything that is a sufficient cause or motive for
the action’s occurring. If, for example, a choice issues from, and can be sufficiently
explained by, an agent’s character and motives (together with background
conditions), then to be ultimately responsible for the choice, the agent must be at
least in part responsible by virtue of choices or actions voluntarily performed in the
past for having the character or motives he or she now has.21

constituting psychic ingredients themselves are the products of other FWAs wherein, at earlier times in her
history, the agent acted in such a way as to then directly shape her character, her motives, and purposes by
choosing or deciding in conditions of uncertainty involving torn decisions.” See McKenna, “Compatibilist
Ultimacy,” 76.

20. Ibid.

21. Robert Kane, “Libertarianism”, in John Martin Fischer et al., Four Views on Free Will (Oxford,
UK: Blackwell Publishing, 2007), 14.
FOREKNOWLEDGE 14

Yet for UR to be attributed to an agent, SFAs must also be will-setting actions, where

“agents make choices or decisions between two or more competing options and do not

settle on which of the options they want more, all things considered, until the moment of

choice or decision itself.”22 Additionally, for there to be two or more competing options,

Kane thinks that there must be more than one way a person could act voluntarily,

intentionally, and rationally. He calls this the plurality conditions for free will.23

One can summarize Kane’s analysis into a three-step argument, which McKenna

labels the Ultimacy Argument (UA):

1. A person acts of her own free will only if she is the ultimate source of her act.
2. If determinism is true, no one is the ultimate source of her acts.
3. Therefore, if determinism is true, no one acts of her own free will.24

Thus, Kane’s analysis (as summarized by McKenna) necessitates that any compatibilist

account of free will must account for a person’s ultimate responsibility. If it cannot do so,

then it seems that compatibilism cannot support moral responsibility.25

E. Critiques of Arminianism

1. Biblical critiques. Biblically, one main point of contention between skeptics

about divine foreknowledge and traditional Arminians is predictive prophecy. In the

22. Ibid., 20.

23. Ibid.

24. Michael McKenna, “Compatibilist Ultimacy: Resisting the Threat of Kane’s U Condition*”, in
Libertarian Free Will: Contemporary Debates, edited by David Palmer (New York: Oxford University Press,
2014), 80.

25. Compatibilism, as a general view of human freedom, holds that “even if every act we perform
is caused by something outside ourselves (such as natural causes or God), we are still free, for we can still
act according to our character and desires.” See John Frame, The Doctrine of God (Phillipsburg, NJ: P & R
Publishing, 2002), 136. In Kane’s terminology, compatibilism holds that even if a person is not the ultimate
source of her act, she is still free since she acts according to her desires.
FOREKNOWLEDGE 15

section on Biblical Evidence for Arminianism, it was noted that predictive prophecy is a

prominent feature of the Old Testament and is difficult for skeptics to properly explain. In

response, they say that every predictive prophecy in scripture fits into one of three

categories: prophecy can express God’s intention to do something in the future irrespective

of creaturely decision; it can express God’s knowledge of what will happen due to

necessary conditions currently existing in the world; or it may be a conditional prophecy

about what will happen if something obtains.26 However, each of these options proves

problematic for the skeptic.

If the skeptic’s analysis is correct, then the first option, God’s intention to do

something irrespective of creaturely choices, must account for very few prophecies in

scripture since they accept libertarian freedom. If libertarian freedom is extremely valuable

to God, then it is unlikely that He would run roughshod over it to accomplish His purposes;

and since many of the prophecies in scripture involve the free choices of human beings,

these skeptics must hold that most prophecies fit into the other two categories. Yet the

second option, God’s knowledge of what follows from necessary conditions, removes any

theological significance from these predictions. Humans can know what necessarily

follows from certain actions; consequently, these predictions would not be especially

compelling or express God’s uniqueness. Subsequently, Arminians and Molinists accept

the reality of conditional prophecies in scripture; they just view these prophecies as

restating God’s moral standard rather than predicting the future. However, if skeptics are

26. This schema was developed in Richard Rice, “Biblical Support for a New Perspective,” in The
Openness of God: A Biblical Challenge to the Traditional Understanding of God (Downers Grove, IL:
Intervarsity Press, 1994), 51-52, and has been widely accepted by others.
FOREKNOWLEDGE 16

saying that God’s conditional prophecies are statements of His knowledge of what would

occur given certain circumstances, then this is an assertion of God’s middle knowledge,

which skeptics regularly deny. Thus, these three categories cannot account for every

predictive prophecy in scripture, and the skeptic’s analysis fails.27

Even if the skeptics’s explanations did account for every predictive prophecy, they

still fail to account for Isaiah 40-48 and John 13:19. Greg Boyd interprets Isaiah 46:9-11

and 48:3 by appealing to God’s intention to bring about His own purposes: “The Lord is

not appealing to information about the future he happens to possess; instead, he is appealing

to his own intentions about the future. He foreknows that certain things are going to take

place because he knows his own purpose and intention to bring these events about.”28 Yet

this interpretation ignores the earlier chapters where God explicitly states that His criterion

for true deity is knowledge of the future (e.g. Isaiah 41:22-28; 42:8-9). Additionally, none

of the skeptics previously cited have explained John 13:19, where Jesus grounds His

messianic purpose in His own ability to predict the future. Given the paltry explanation of

these two passages by skeptics, one can only agree with Bruce Ware’s analysis:

Since God himself declares the criterion by which the question of his deity is to be
evaluated and established, and since that criterion is the possession of a knowledge
of the future that can be declared and its truthfulness verified (or falsified) by the
unfolding of future events, how utterly impertinent and presumptuous to deny of
God divine foreknowledge and so deny the very basis by which God himself has
declared that his claim to deity shall be vindicated and made known.29

27. These arguments are made in Craig, The Only Wise God, 43-44; and Ware, God’s Lesser
Glory, 132-134.

28. Gregory Boyd, God of the Possible: A Biblical Introduction to the Open View of God (Grand
Rapids, MI: Baker Book House, 2000), 30.

29. Bruce Ware, God’s Lesser Glory: The Diminished God of Open Theism (Wheaton, IL:
Crossway Books, 2000), 104.
FOREKNOWLEDGE 17

The other passages given in defense of exhaustive divine foreknowledge are poorly

interpreted by skeptics as well. Boyd tries to argue that God predetermines some things

without predetermining everything, even though Psalm 139:16 says that God ordained

every day for David before he lived one of them. In response to texts declaring God’s plan

for salvation to be through Christ before the foundation of the world (Eph. 1:3-14; 3:11; 1

Cor. 2:7; 2 Tim. 1:9-10; Rev. 13:8; 17:8), Boyd and other skeptics argue that God

predetermined that Jesus would die, but did not predetermine for specific people to kill

Him.30 Yet if God decided before He created the world that Christ would be crucified, then

He did so irrespective of creaturely decisions. Consequently, He had to plan for persons to

crucify Jesus. Considering the number of people involved in Jesus’s betrayal and

crucifixion, it seems unlikely that God could orchestrate Jesus’s crucifixion based on his

knowledge of the past, present, and probability calculations concerning the future. Thus, it

seems much more plausible that God has exhaustive foreknowledge.

2. The compatibility of divine foreknowledge and human action. Although the

biblical evidence seems to conclusively favor divine foreknowledge, Arminians face

tougher challenges by simultaneously holding to exhaustive divine foreknowledge and

libertarian freedom. The problem may be shown through the following argument for

theological fatalism:

1. God’s being omniscient necessarily implies that if Jones mows his lawn on
Saturday afternoon, then God believed at an earlier time that Jones would mow his
lawn on Saturday afternoon.
2. Necessarily, all of God’s beliefs are true.
3. No one has the power to make a contradiction true.

30. Boyd argues that the Bible “never suggests that the individuals who participated in this event
were predestined to do so or foreknown as doing so. It was certain that Jesus would be crucified, but it was
not certain from eternity that Pilate, Herod, or Caiaphas would play the roles they played in the crucifixion.
They participated in Christ’s death of their own free wills.” See God of the Possible, 45.
FOREKNOWLEDGE 18

4. No one has the power to erase someone’s past beliefs, that is, to bring it about that
something believed in the past by someone was not believed in the past by that
person.
5. No one has the power to erase someone’s existence in the past, that is, to bring it
about that someone who existed in the past did not exist in the past.
6. So if God believed that Jones would mow his lawn on Saturday afternoon, Jones
can refrain from mowing his lawn only if one of the following alternatives is true:
a. Jones has the power to make God’s belief false;
b. Jones has the power to erase God’s past belief; or
c. Jones has the power to erase God’s past existence.
7. But alternative (a) is impossible (this follows from steps 2 and 3).
8. And alternative (b) is impossible (this follows from step 4).
9. And alternative (c) is impossible (this follows from step 5).
10. Therefore, if God believes that Jones will mow his lawn on Saturday afternoon,
Jones does not have the power to refrain from mowing his lawn on Saturday
afternoon; that is to say, Jones is not free.31

This argument is essentially a more nuanced version of the argument against exhaustive

divine foreknowledge; subsequently, traditional Arminians must show how divine

foreknowledge and libertarian freedom can coexist without infringing on each other.

The first attempt to reconcile foreknowledge and libertarian freedom is the

Boethian view. Named after Saint Boethius, this view argues that since God exists outside

of time, He knows all things at once, and there is no sequential time in His thought. In other

words, “He simply knows in one eternal Now.”32 Because God exists outside of time, His

knowledge is not temporally prior to Jones’s action, which means that the argument cannot

get off the ground. Just like my watching Jones mow his lawn from across the street does

not affect his freedom to continue or stop mowing, God’s eternal knowledge of Jones’s

31. In The Only Wise God, William Lane Craig takes a whole chapter (“The Argument for
Theological Fatalism”) to present this syllogism. However, Craig cites Nelson Pike as the original formulator
of the argument. For Pike’s version, see his article “Divine Omniscience and Voluntary Action,” The
Philosophical Review 74, no. 1 (Jan. 1965): 27-46, accessed January 21, 2017,
http://www.jstor.org/stable/2183529.

32. Geisler, The Battle for God, 62.

FOREKNOWLEDGE 19

lawn mowing does not coerce or force Jones to perform that action. Therefore, the Boethian

concludes that God can foreknow human actions without eliminating human freedom.

Since its initial statement in Boethius’s The Consolation of Philosophy, the

Boethian view has been popular among laypersons as a go-to response to the sovereignty-

freedom debate, but it has received harsh criticism in the academic community. The

simplest critique is to recognize that even if God’s timeless knowledge does not coerce a

person into performing an action, “the timeless realm is as ontologically determinate and

fixed as the past … we have no more reason to think that we can do anything about God’s

timeless knowledge than about God’s past knowledge. If there is no use crying over spilt

milk, there is no use crying over timelessly spilling milk either.”33 Even though it’s true

that God’s timeless knowledge does not coerce a person into acting a certain way, it does

not rebut the argument for theological fatalism. In order to circumvent the force of the

argument, one must add something to God’s timeless knowledge to maintain the

conjunction of foreknowledge and human freedom.34

Due to the failure of the Boethian view, many philosophers have tried to rebut the

argument for theological fatalism by adopting two different variants of Ockhamism. Both

versions rely on the distinction between soft and hard facts. Soft facts are facts whose

grammar indicates the past but whose content belongs at least partly in the future; hard

33. Linda Zagzebski, “Recent Work on Divine Foreknowledge and Free Will,” in The Oxford
Handbook of Free Will, 1st edition, edited by Robert Kane (New York, NY: Oxford University Press, 2002),
52.

34. Molina also gives a more complex critique in Disputation 48, sections 9-11 (102-103 of
Freddoso’s translation) and Disputation 49, sections 15-16 (122-124 of Freddoso’s translation). The reader
can decide if Molina’s objection is cogent or not.
FOREKNOWLEDGE 20

facts are facts whose grammar and content is entirely focused on the past.35 In the first

version of Ockhamism, it is argued that God’s beliefs about the future are soft facts, and

that this maintains libertarian freedom. For instance, Ockhamists will argue that God

believes that Adam will sin is a soft fact by

unpacking what it is for God to hold a belief about the future and then showing that
this fact is constituted not just by what is now the case but also (in part) by what
will be the case later … if God’s belief is constituted as the belief that Adam will
sin only retroactively, once Adam actually sins, then nothing about God’s prior
belief would appear to be inconsistent with either Adam’s freedom to sin or his
freedom not to sin.36

To understand how this argument works, it will be necessary to give further nuance to the

argument against the conjunction of divine foreknowledge and human freedom.

The tension between divine foreknowledge and human freedom centers around the

idea of accidental necessity, also known as temporal necessity.37 It works as follows: God

is omniscient, so He knows all true propositions, including propositions about the future.

This means that, as an example, God knows whether a person will have a glass of water

rather than a glass of milk at lunchtime before he makes that choice. Yet God’s

foreknowledge is also infallible, meaning that His knowledge of the future cannot be

wrong. Consequently, if God knows that a person will choose to have a glass of water

before he chooses to do so, and His knowledge is infallible, then he cannot choose other

than what God already knows he will choose. Otherwise, if he chose to have the glass of

35. Hunt, “The Simple-Foreknowledge View,” 84.

36. Ibid.

37. Both terms are used interchangeably, so I will use “temporal necessity” because it gives the
reader a clearer understanding of the idea being conveyed.
FOREKNOWLEDGE 21

milk, then God’s foreknowledge would be wrong and He would not be omniscient. Thus,

his choosing the glass of water is now temporally necessary.

To maintain the coexistence of libertarian freedom and divine foreknowledge,

Ockhamists argue that soft and hard facts are two kinds of facts that cannot overlap; a

proposition can only be a soft fact or only be a hard fact. Additionally, they introduce three

different principles: the Fixity of the Past Principle, which states that no one can do

anything during a present or future time period to render false a truth strictly about a past

time period; the Mixed Conjunction Principle, which states that the conjunction of one

proposition strictly about a time period and another not strictly about that time period is

also not strictly about that time period; and the Equivalence Principle, which states that if

two propositions are equivalent, then one of them is strictly about a given time period just

in case the other is also strictly about that time period. Given these three principles, the

Ockhamist may argue that

Since "God believed during t [a point in time] that X would do A during T [a later
point in time]" and "God existed during t and X will do A during T" are equivalent,
the equivalence principle forces a choice between both being strictly about t on the
one hand and neither being strictly about t on the other. The mixed conjunction
principle guarantees that "God existed during t and X will do A during T" is not
strictly about t. So, "God believed during t that X would do A during T" must also
fail to be strictly about t.

Because the Fixity of the Past Principle only applies to truths strictly about a past time

period, and “God believed during t that X would do A during T” is not strictly about the

past, then God’s belief about the future could be falsified by human actions, meaning that

God’s belief is not temporally necessary and libertarian freedom is maintained. This kind
FOREKNOWLEDGE 22

of response has been heavily critiqued in the scholarly literature, but there are other views

which seem more promising.38

Although the Boethian and Ockhamist views seem inadequate to rebut the argument

for theological fatalism, Luis de Molina proposed a unique response to it that adds a fourth

alternative to premise 6 (see page 18): d) if Jones were to act in a different way than what

God foreknows, then God would have believed that instead of what he believes now.

Molina says that “even if (i) the conditional is necessary (because … these two things

cannot both obtain, namely, that God foreknows something to be future and that the thing

does not turn out that way), and even if (ii) the antecedent is [temporally necessary],

nonetheless the consequent can be purely contingent.”39 In essence, Molina argues that the

foreknown event is the ground of God’s foreknowledge; if the ground of His

foreknowledge were different than it is now, then God’s foreknowledge would be different

than it is now:

Pike is certainly correct that God’s infallibility prevents his holding a false belief.
But that same infallibility guarantees that if Jones were to refrain, God would have
held a different belief. Since God cannot be mistaken or fooled, he would have
foreknown from eternity if Jones were going to refrain. God cannot hold a false
belief. Therefore, whatever Jones will do, God foreknows it. If Jones were to act
differently, God’s true belief would not have been false; rather his belief would
have been different.40

38. The three definitions and the quote were from D.E. Brant, “On Plantinga’s Way Out,” Faith and
Philosophy 14, no. 3 (July 1997): 338-340, accessed June 27, 2016, DOI:10.5840/faithphil199714327. In the
original article, Brant used t’ to symbolizes “a later point in time”, but I substituted that with T to make the
difference clearer, since the quotation marks made it hard to see the apostrophe. Also, I cannot discuss the
objections to Ockhamism due to page constraints. However, I recommend that the reader start by reading
Brant’s article, which gives a thorough critique of Alvin Plantinga’s version of Ockhamism.

39. See Disputation 52, section 34 of On Divine Foreknowledge (page 189 of Freddoso’s
translation).

40. Craig, The Only Wise God, 71. Craig puts it another way on page 74: “God’s foreknowledge is
chronologically prior to Jones’s mowing the lawn, but Jones’s mowing the lawn is logically prior to God’s
FOREKNOWLEDGE 23

Because the event itself grounds God’s foreknowledge, the event itself cannot be

temporally necessary; thus, human action is still free.

Just like the Boethian and Ockhamist views, the Molinist rebuttal does not escape

criticism. David Hunt argues that this strategy does not avoid accidental necessity:

Does it make any difference that God’s belief depends on what I do and that it
would have been different if I were to act differently? No, for he has already held
this belief, and my action must be consistent with this fact about the past … God’s
believing that I will press the button does not leave me any alternatives to pressing
the button, since he can’t hold that belief and I fail to press the button.41

However, it seems that Hunt’s criticism misses the point of Molina’s argument. What

follows from the argument for theological fatalism is that the action will happen, not that

the event will necessarily happen: “From God’s foreknowledge of x we can be absolutely

sure that x will happen. But it does not have to occur; it is possible for it to fail to happen.

What is impossible is a situation in which God foreknows x and x fails to happen, for this

would be a logical contradiction.”42 Therefore, it seems that Molina’s rebuttal to the

argument for theological fatalism is coherent.

Nevertheless, if one does not find Molina’s rebuttal convincing, there is one more

strategy available to Arminians. Hunt labels it the Augustinian view, as he argues that

Augustine has been misinterpreted by those who espouse compatibilism regarding

determinism and human freedom. The traditional interpretation of Augustine held that

Augustine was a soft determinist who accepted that human actions are causally necessary.

foreknowledge. Jones’s mowing is the ground; God’s foreknowledge is its logical consequent; Jones’s
mowing is the reason why God foreknows that Jones will mow the lawn.”

41. Hunt, “The Simple-Foreknowledge View,” 86.

42. Craig, The Only Wise God, 73.

FOREKNOWLEDGE 24

By contrast, Hunt argues that Augustine rejected both causal necessity and the power of

contrary choice:

Temporal necessity is determined by the temporal order; but what is relevant to free
agency, Augustine maintains, is the causal/explanatory order. The two orders
normally coincide: what is prior in the one order is prior in the other. In cases of
divine foreknowledge, however, the two orders diverge, and what is temporally
closed (because infallibly foreknown) may remain causally/explanatorily open; as
Augustine notes in The City of God, "a man does not therefore sin because God
foreknew that he would sin" (V.lO). This is enough for Augustine to regard W as
free despite the fact that God's foreknowledge of W renders it unavoidably
necessary.43

The result is a new form of libertarian freedom: even if a person does not have the power

of contrary choice, she can freely choose as long as there are no causal constraints on her

freedom.

Some Arminians may have a problem with rejecting the power of contrary choice,

otherwise known as the principle of alternate possibilities (PAP), because it has been at the

core of the Arminian position for many years. Boyd echoes this sentiment when he says

that “Person X is free regarding an action A if and only if it’s genuinely possible for him

to do ~A.”44 Yet for Hunt and others, PAP seems intuitively false. Henry Frankfurt

developed supposed counterexamples to PAP which denied the power to do otherwise and

yet seemingly maintained morally responsibility. Counterexamples to PAP usually involve

a situation where a mechanism is placed inside a person’s head without her knowledge,

and if she were to form the intention of performing a specific action, that mechanism would

43. David P. Hunt, “On Augustine’s Way Out,” Faith and Philosophy 16, no.1 (January 1999): 10,
accessed June 2, 2016, DOI: 10.5840/faithphil19991612

44. Gregory Boyd, “An Open-Theist Response,” in Divine Foreknowledge: Four Views, edited by
J.K. Beilby and P.R. Eddy (Downers Grove, IL: InterVarsity Press, 2001), 108.
FOREKNOWLEDGE 25

activate and prevent her from performing the action. In the counterexample, the person

chooses a different action, and the mechanism does not activate. Consequently, it is argued

that even though she could not have done otherwise (the mechanism would have activated

and prevented her from doing so), she is still responsible for the choice she makes since

she was causally uninhibited.

For traditional libertarians, the invocation of a mechanism in someone’s head seems

implausible and uncompelling. Consequently, there has been considerable debate over

Frankfurtian counterexamples to PAP.45 In his contribution to the overall discussion, Hunt

has argued that divine foreknowledge provides a better counterexample to PAP than

Frankfurt’s argument:

To whatever extent it is clear in the Frankfurt counterexamples that an action can

be unavoidable without this jeopardizing its libertarian freedom, it is at least as clear
(if not more so) in divine foreknowledge cases that the foreknown action can be
unavoidable yet libertarianly free. The same intuitions that support Frankfurt's
argument, when brought to bear on divine foreknowledge of human actions,
provide direct support for the claim that these actions can be libertarianly free
despite their inevitability. There is no need to seek indirect support for this
judgment via a consideration of Frankfurt counterexamples—indeed, doing so can
only muddy the waters by making freedom in the face of divine foreknowledge
appear on a par with, and to require support from, freedom in the face of
counterfactual intervention.46

However, the discussion of Frankfurtian counterexamples and PAP, as well as the

discussion of the tension between divine foreknowledge and human freedom, is only

45. For a nice representation of the overall discussion, see J.M. Fischer, “Responsibility and
Control,” The Journal of Philosophy 79, no. 1 (January 1982): 24-40; David Widerker, “Libertarian Freedom
and the Avoidability of Decisions,” Faith and Philosophy 12, no. 1 (January 1995): 113-118; and J.M.
Fischer, “Libertarianism and Avoidability: A Reply to Widerker,” Faith and Philosophy 12, no. 1 (January
1995): 119-125.

46. David Hunt, “Frankfurt Counterexamples: Some Comments on the Widerker-Fischer Debate,”
Faith and Philosophy 13, no. 3 (July 1996): 400, accessed May 29, 2016, DOI: 10.5840/faithphil199613331.
FOREKNOWLEDGE 26

necessary if libertarian freedom itself is intelligible. Consequently, it is necessary to

consider critiques to libertarian freedom to see if this core tenet of Arminianism is even

required for moral responsibility.

3. Critiques of libertarian freedom. To begin, there are objections raised

specifically against Robert Kane’s event-causal view of libertarian freedom. First, Michael

McKenna argues that Kane’s UR condition is not necessary for moral responsibility by

emphasizing the contextual nature of ultimacy:

We speak, for instance, of the original or ultimate source of Perrier drinking water
as being a spring somewhere in the south of France, and we regard the refrigerator
or the bottling plant as, by contrast with the spring, a mediated source for the
sparkling water in our glass. In ordinary contexts, it would never dawn on us to
think that whether the water in our glass really originated in France turned on
whether determinism was true or not.47

Since the notion of ultimacy is context-sensitive, McKenna emphasizes the need to

determine the context of ultimacy in discussions of moral responsibility. In his view, “the

proper context for settling whether a person ultimately formed herself, or pertinent features

of her character or will, via an SFA, is located in the domain of ordinary folk psychological

discourse.”48 Even if libertarians disagree with contextualizing ultimacy in the domain of

folk psychological discourse, McKenna thinks that it’s important to realize that there are

many contexts in which ultimacy does not depend on whether determinism is true or false.49

In response, Kane partially agrees with McKenna’s analysis. He accepts that the

notion of ultimacy is contextual, and that the source of Perrier drinking water does not

47. McKenna, “Compatibilist Ultimacy,” 85.

48. Ibid., 86.

49. Ibid.
FOREKNOWLEDGE 27

depend on whether determinism is true. However, he insists that in the context of free will

and moral responsibility, it really does matter whether the person is responsible for their

own character in some fashion or whether it is entirely due to deterministic factors:

… if determinism is true, wherever you stop, there would always be conditions in

the past such that once they had occurred (given the laws), it was settled that a
person’s later will (character, motives, and purposes) would be exactly as it is now,
though nothing whatever the person voluntarily and intentionally did or omitted
played any role in producing or bringing about those conditions.50

Although McKenna made some valuable points about contextualizing ultimacy, his point

fails to challenge and/or defeat Kane’s understanding of ultimate responsibility.

Another objection to Kane’s view is the issue of competing desires and intentions.

In Freedom of the Will, Jonathan Edwards harshly critiqued the idea that a person is in a

state of complete indifference whenever she decides to perform an action, and that her

desires and intentions do not decisively lean in one way or the other. He insisted that “to

suppose the Will to act at all in a state of perfect indifference, is to assert that the mind

chooses without choosing. To say that when it is indifferent, it can do as it pleases, is to

say that it can follow its pleasure, when it has no pleasure to follow.” 51 Edwards thought

that the confusion resulted from the focus of the indifference. He held that the Arminian is

thinking of indifference towards the objects he is choosing (each cake looks equally

appetizing), whereas Edwards was speaking of indifference in the choice itself.

Subsequently, he said that “the determination is not about the objects, unless indirectly and

50. Robert Kane, “New Arguments in Debates on Libertarian Free Will: Responses to
Contributors,” in Libertarian Free Will: Contemporary Debates, edited by David Palmer (New York:
Oxford University Press, 2014), 183.

51. Jonathan Edwards, Freedom of the Will, 1754, reprint (Grand Rapids, MI: Christian Classics
Ethereal Library, 1999), 48.
FOREKNOWLEDGE 28

improperly, but about the actions, which it chooses for other reasons than any preference

of the objects, and for reasons not taken at all from the objects.”52 In other words, one may

be indifferent about the cake he chooses, but he is not indifferent about the choice to act

itself.

In response, Kane would likely agree with Edwards, in that persons do not choose

out of a state of indifference. In all situations, a person is inclined to perform an action. In

this sense, both Edwards and Kane disagree with noncausal libertarians, who were

Edwards’s primary intellectual opponent. However, the uniqueness of Kane’s view is that

it focuses on the importance of circumstances where persons have the desire to complete

two different efforts:

These [are] circumstances in which (i) we are deliberating between competing

options; (ii) we intend to choose one or the other but cannot choose both; (iii) we
have powerful motives for wanting to choose each of the options for different and
competing reasons; (iv) there is a consequent resistance in our will to either choice,
so that (v) if either choice is to have a chance of being made, effort will have to be
made to overcome the temptation to make the other choice; and (vi) we want to
give each choice a fighting chance of being made because the motives for each
choice are important to us. The motives for each choice define in part what sort of
person we are; and we would be taking them lightly if we did not make an effort in
their behalf. These conditions, I contend, are the conditions of ‘self-forming’
actions (SFAs).53

Kane readily accepts that persons always have intentions for performing actions, but he

insists that there must not be any external or internal causal coercion involved in SFAs.

A third and final objection to Kane’s event-causal view of libertarian freedom takes

the form of an infinite regress argument: if SFAs are required for free and morally

52. Edwards, Freedom of the Will, 50.

53. Kane, “New Arguments,” 198.

FOREKNOWLEDGE 29

responsible actions, then must these be preceded by other SFAs? If so, then it seems like

those SFAs would need to be preceded by other SFAs, and on goes the infinite regress.

Subsequently, Kane recognizes this infinite regress and argues that SFAs need not be

preceded by more SFAs. Instead, an agent must simply have teleological guidance control

(TGC). This kind of control is exhibited in persons when they tend “through feedback loops

and error correction to converge on a goal in the face of perturbations … the kind of control

(TGC) agents exercise over the efforts or volitional streams that may lead to SFAs—in

contrast to the [plural voluntary control] they exercise over the SFAs themselves—is a

compatibilist kind of control.”54 However, TGC is also compatible with indeterminism,

meaning that persons are responsible irrespective of the necessity involved in the action.

Kane recognizes that the resulting view is quite remarkable when he calls the reader to

Note where we thus arrive: multiple parallel goal-directed cognitive processes

(volitional streams, as I also call them) simultaneously exercised by the agent, over
each of which the agent has ‘one-way’ or singular voluntary control (TGC), when
occurring together, make possible ‘more-than-one-way’ or plural voluntary control
(PVC), since the agent might succeed in attaining the goal of either process at a
given time voluntarily, on purpose and for reasons. Stated differently, two cognitive
processes over each of which an agent has what Fischer calls guidance control,
exercised simultaneously and in parallel, give rise to what he calls regulative
control, the power at a time to bring about a choice by attempting to bring it about
and the power at that same time to bring about an alternative choice by attempting
to bring it about.55

Beyond the critiques of Kane’s specific view, there are several critiques of

libertarian freedom in general that arise from both philosophical and theological issues.

54. Kane, “New Arguments,” 199. Plural voluntary control is when agents “are able to bring about
whichever of the options they will, when they will to do so, for the reasons they will to do so, on purpose,
rather than accidentally or by mistake, without being coerced or compelled in doing so or wiling to do so, or
otherwise controlled in doing or willing to do so by any other agents or mechanisms.” From Robert Kane,
“Libertarianism”, 30.

55. Ibid., 200.

FOREKNOWLEDGE 30

The most prominent critique of libertarian freedom is the luck/chance objection, which

argues that indeterminism makes all human actions arbitrary or the result of luck or chance.

Before answering this objection, Kane starts by explaining five key aspects of his position.

First, he holds that indeterminism need not be involved in every act a person makes of her

own free will; indeterminism is only necessary for SFAs. Second, he holds that SFAs occur

when one encounters two difficult choices that she must choose between, which involves

some internal tension that Kane attributes to indeterminacy in a person’s neural processes.56

Third, Kane labels this indeterminism in the neural processes parallel processing.

To explain this idea, Kane posits a woman who encounters a mugging on her way to work.

She has two competing motivations: help the victim, or continue her commute and avoid

getting involved. Given the complexity of the idea, it will be useful to quote Kane here at

length:

… imagine that two competing neural networks are involved, each influencing the
other and representing her conflicting motivations. (These are complex networks
of interconnected neurons in the brain circulating impulses in feedback loops that
are generally involved in higher-level cognitive processing). The input of one of
these networks consists in the woman’s desires and motives for stopping to help
the victim. If the network reaches a certain activation threshold (the simultaneous
firing of a complex set of ‘output’ neurons), that would represent her choice to help.
For the competing network, the inputs are her ambitious motives for going on to
her meeting; and its reaching an activation threshold represents the choice to go on.
Now imagine further that the two networks are connected so that the indeterminism
that is an obstacle to her making one of the choices is present because of her
simultaneous conflicting desire to make the other choice—the indeterminism thus
arising from a tension-creating conflict in the will, as noted. Under such
circumstances, when either of the pathways reaches an activation threshold which
amounts to choice, it would be like your solving the mathematical problem by
overcoming the indeterministic background noise generated by the presence of the
other pathway. And just as when you solved the mathematical problem despite the

56. In Kane’s presentation of libertarian freedom, he always discusses two competing choices or
motives. Of course, it seems possible that there could be more than two competing motives, but in all his
examples, Kane always uses two motives/choices.
FOREKNOWLEDGE 31

presence of this indeterminism, one could say you did it and are responsible for it,
so one can say this as well, I would argue, in the present case, whichever one is
chosen. The network through which she succeeds in reaching a choice threshold
will have succeeded despite the indeterminism present because of the existence of
the competing networks.57

Fourth, he holds that the indeterminism involved in the choice is an ingredient in

the larger teleological activities of the agent, where the indeterminism functions as an

interference of the cognitive processes which could result in either the failure or success of

the goal. Fifth, he holds that this indeterminism, when functioning as an obstacle to the

teleological activity of the agent, does not exclude responsibility if the activity reaches its

intended goal. Assuming these five steps, Kane argues that the luck objection does not

apply to his version of libertarian freedom.58

The first version of the luck objection argues that the agent did not bring about the

action, but rather that it “just happened.” Kane responds by stating that “the agent brings

about the choice that is made by engaging in a goal-directed process of trying or attempting

or making an effort to bring about that particular choice for the reasons favoring it, and by

succeeding in attaining this goal, whichever choice is made.”59 The second version argues

that the agent did not have control over the choice, but Kane says that the agent has the

power to make the event be or not be, as well as the power to make the competing choice

be, because “either of the efforts in which the agent is engaged might succeed in attaining

57. Robert Kane, “Rethinking Free Will: New Perspectives on an Ancient Problem,” in The
Oxford Handbook of Free Will, 2nd edition, edited by Robert Kane (New York, NY: Oxford University
Press, 2011), 388-389.

58. The five aspects of Kane’s position are given in “New Arguments in Debates on Libertarian
Free Will,” 193-195. The third step is the same as described in the quote from The Oxford Handbook of Free
Will, but the quote gave more detail than Kane’s description in “New Arguments in Debates on Libertarian
Free Will,” so I used the quote instead.

59. Kane, “New Arguments in Debates on Libertarian Free Will,” 195.

FOREKNOWLEDGE 32

its goal despite the chance of failure,” with the chance of failure coming from the

indeterminacy in the cognitive processes.

The third, fourth, and fifth versions of the objection stipulate that the choice was

not made intentionally/voluntarily/for reasons, but Kane responds by pointing out that the

choice would have resulted from a goal-oriented cognitive process aimed at making that

choice; and since the alternative was also available to the agent, it would have been made

voluntarily and for reasons given by the agent. Finally, the sixth version of the objection

questions the agent’s moral responsibility for the action. Yet Kane holds that agents in

SFAs have plural voluntary control, which is “the power to act in more than one way and

to be able to do so, either way, voluntarily, intentionally, and rationally,” and this seems

satisfactory for the agent to be considered morally responsible for that choice.60 Therefore,

since all six versions of the luck objection have been sufficiently accounted for by Kane’s

analysis, it seems that the luck objection fails to challenge libertarian freedom.

The underlying factor behind these different versions of the luck objection is the

idea that “if the occurrence of a choice depends on the occurrence of some undetermined

or chance events (e.g., quantum events) in the brain over which the agent lacks control,

then whether or not the choice occurs would appear to be the matter of luck or chance,”

which removes responsibility.61 In other words, the objector thinks that one needs to micro-

manage each neuron in order to be responsible for the action. However, this would be a

misunderstanding of cognitive processes. To perform purposive actions, what one needs is

60. Ibid., 196.

61. Kane, “Rethinking Free Will,” 393.

FOREKNOWLEDGE 33

macro-control of processes involving many neurons—processes that may succeed

in achieving their goals despite the interfering or hindering effects of some
recalcitrant neurons. We do not micro-manage our actions by controlling each
individual neuron or muscle that might be involved. But that does not prevent us
from macro-managing our purposive activities (whether they be mental activities
such as practical reasoning, or physical activities, such as arm-swingings) and being
responsible when those purposive activities attain their goals.62

Thus, instead of the indeterminacy being the cause of the action, the reasons/motives of the

person and his efforts to overcome the temptation to perform the other choice are the cause,

and the indeterminacy acts as a causally relevant hindrance or interference for the

achievement of the outcome.63

In addition to the luck objection, Jonathan Edwards critiqued non-causal

libertarians on their understanding of causation. In his time, Arminians argued that any

form of necessity eliminated human freedom. Edwards retaliated by arguing that this

conception of causation could never work. First, saying that an event has a cause of its

existence and yet is not connected to that cause is inconsistent, for it directly denies the

definition of contingency. An effect that is contingent to its cause is necessarily dependent

upon its cause for its existence; to say otherwise is unintelligible. Second, if the effect was

not necessarily connected to the cause, then it would have no cause or reason for its

existence, which is directly opposing the reasoning previously considered. Third, to

suppose that events may have causes which they are not directly connected to is to say that

62. Ibid., 395.

63. As an example of a causally relevant hindrance or interference, Kane uses the example of a
vaccine: “A vaccination may hinder or lower the probability that I will get a certain disease, so it is causally
relevant to the outcome. But if I get the disease despite it, the vaccination is not the cause of my getting the
disease, though it was causally relevant, because its role was to hinder that effect. The causes of my getting
the disease, by contrast, are those causally relevant factors (e.g., the infecting virus) that significantly
enhanced the probability of its occurrence.” See ibid., 394.
FOREKNOWLEDGE 34

an effect’s cause is not its cause, which is an explicit contradiction. Thus, Edwards found

it “clearly manifest, that every effect has a necessary connection with its cause, or with that

which is the true ground and reason of its existence.”64

Kane would agree with Edwards’s analysis, since he finds non-causal

libertarianism to be unintelligible as well. Subsequently, he would likely argue that his own

event-causal view is not susceptible to such critique, and he could argue so for two reasons.

First, Kane accepts that most actions are examples of teleological guidance control, and

since this is compatible with determinism, he would not disagree with Edwards. Second,

Kane could argue that self-forming actions are connected to their cause, but that cause is

hindered due to the competing cognitive processes. In other words, Kane could recognize

the necessary connection between cause and effect without giving up indeterminism.

Although the philosophical objections to Kane’s event-causal libertarian freedom

seem ineffective, John Frame provides theological objections to libertarian freedom in

general. One group of criticisms points out that the Bible never explicitly teaches, grounds

moral responsibility in, places any positive value on, or judges anyone’s conduct by

reference to libertarian freedom. In this case, Frame is absolutely right; libertarian freedom

is a philosophical concept that is never explicitly proposed in scripture. Likewise, one could

argue that compatibilist freedom is not explicitly proposed either. Frame argues that the

“integrity of human personality is not possible in a libertarian construction, for on that view

the will must always be independent of the heart and all of our other faculties,” but it seems

64. These arguments come from Edwards, Freedom of the Will, 59-61. The quote specifically comes
from page 61.
FOREKNOWLEDGE 35

that this critique does not hold for Kane’s event-causal view, since the heart and all other

faculties are necessary for the cognitive processes we undergo when making a decision. 65

Additionally, Frame argues that since people won’t sin in heaven, and since God

cannot act against His own holy character, then libertarian freedom must be false.66 Yet

Kane’s view does not rest upon the power of contrary choice; instead, it maintains that a

person must be ultimately responsible for the action, which requires that the action itself

must be made intentionally and without coercion, or that the action deterministically

follows from a prior SFA. In this sense of ultimate responsibility, God is a unique case;

since His character is perfectly good and immutable, He has never required any “self-

forming” or “will-setting.” Consequently, God has never performed a SFA, but He could

still perform FWAs (see page 12) and be ultimately responsible for those choices.

Similarly, people in heaven will have formed their wills through a succession of SFAs

throughout their lifetime. As Christians live on earth, their desires and motives will become

more in tune with God’s perfectly good desires. Consequently, it is plausible that they

would perform FWAs in heaven and, due to their will being permanently set to follow God,

not sin. Thus, Frame’s objections fail to apply to Kane’s event-causal view.

Frame’s final category of objections to libertarian freedom also questions the

explanatory power of simple foreknowledge. These objections focus upon God’s

providential control of the world: how does God ensure that His plan will come to pass if

humans have libertarian freedom? Scripture attests to God’s control over all creation,

65. John Frame, The Doctrine of God, 142.

66. The argument from the lack of sin in heaven comes from page 140 of The Doctrine of God,
while the argument from God’s character comes from page 142.
FOREKNOWLEDGE 36

including the many sins committed by people throughout history. Arminius himself

declared that God’s providence “preserves, regulates, governs and directs all things and

that nothing in the world happens fortuitously or by chance.”67 Given this strong view of

God’s control over creation, one must ask if simple foreknowledge is sufficient to “govern

and direct all things.”

4. Simple foreknowledge and providential control. Skeptics concerning divine

foreknowledge have argued that simple foreknowledge has the same amount of

providential control as present knowledge; thus, simple foreknowledge is insufficient for

the strong view of divine providence stated above. The argument starts by defining simple

foreknowledge as complete and infallible knowledge of the future. What follows from this

definition proves problematic:

since foreknowledge can be utilized in a providentially beneficial manner only if

there is a time at which what is foreknown can influence a divine decision that is
itself not also already foreknown, there can exist no conceivable context in which
[simple foreknowledge] would enable God to make providentially beneficial
decisions that he would not be able to make without this knowledge.68

In other words, foreknowledge would only be helpful if God has not made up His mind

about how to act; but because God foreknows both the future events in the actual world

and His own response to those events, foreknowledge does not give him any advantage.

67. Jacob Arminius, “A Declaration of the Sentiments of Arminius”, in The Works of Jacob
Arminius, vol. 1, translated by James Nichols (Buffalo, NY: Derby, Orton, and Mulligan, 1853), accessed
June 11, 2016, http://wesley.nnu.edu/arminianism/the-works-of-james-arminius/volume-1/arminius-
theological-sentiments/.

68. David Basinger, The Case for Freewill Theism: A Philosophical Assessment (Downers Grove,
IL: InterVarsity Press, 1996) 55.
FOREKNOWLEDGE 37

This could also result in a violation of the Doxastic Principle, which states that “it is

impossible to hold the belief that p while deciding to bring it about that p.”69

David Hunt seeks to avoid this problem by narrowing the scope of the Doxastic

Principle. He does so by delineating between accessed and unaccessed knowledge. In

human experience, accessed knowledge is what one consciously thinks of at a moment in

time; unaccessed knowledge is what one holds in his subconscious and accesses when

needed. Hunt thinks that God’s knowledge could work in the same way, in that God would

“possess complete unaccessed foreknowledge and simply access whatever would

maximize providential control without stultification.”70 Tomis Kapitan agrees with this

analysis, arguing that the Doxastic Principle would only be plausible if revised to account

for accessed and unaccessed knowledge: “It is impossible for a rational, self-reflective

agent to consciously entertain the belief that he/she will perform an action A while, at the

same time, deliberating about whether to perform A.”71

Nevertheless, Kapitan also recognizes that another principle undergirds the

Doxastic Principle, one that involves the notion of doxastic openness: “If at t1 S decides to

do action A, then at some time t2 appropriately prior to t1, S believes that it is yet open for

him/her to do A.”72 Doxastic openness implies that God forms intentions and deliberates

between different options, but if God foreknows what He will do in the future, then this

69. David Hunt, “Divine Providence and Simple Foreknowledge,” Faith and Philosophy 10, no. 3
(June 1993): 399, accessed May 29, 2016, DOI: 10.5840/faithphil19931036.

70. Ibid., 409.

71. Tomis Kapitan, “Providence, Foreknowledge, and Decisions Procedures,” Faith and Philosophy
10, no. 3 (July 1993): 416, accessed May 29, 2016, DOI: 10.5840/faithphil19991612.

72. Ibid., 415.

FOREKNOWLEDGE 38

deliberation and intention-formation cannot occur, leaving God providentially impotent.

Subsequently, Hunt suggests that there is a difference between propositional and practical

beliefs which may avoid this doxastic problem: “what one comes to believe as a result of

foreknowledge is a propositional belief about what will happen, whereas what one comes

to believe as a result of deciding is a practical belief about what to do … even if the

propositional belief is acquired first, it may still be necessary to go through the actual

process of decision-making in order to achieve the practical belief.”73

For this distinction between propositional and practical beliefs to work, Hunt posits

that before God chose to create the world, He thought through possible sets of conditions

He would encounter. As He thought through the possible situations, He decided that if

some event X obtains, then it would be best to respond with action Y. Hunt thinks that God

did this for every possible instance of providential control that He might perform, and that

some of these instances might require God to act before the temporal occurrence of X.

Once He has decided what to do for every possible instance of providential control, God

accesses His simple foreknowledge and determines which of these possible sets of

conditions will actually occur in creation. Because God knows which conditions will

obtain, He subsequently acts to accomplish the plans He formulated before His

foreknowledge was accessed. If this conception of God’s mental activity is correct, then

“simple foreknowledge allows God to undertake providential interventions that would not

73. Hunt, “Divine Providence and Simple Foreknowledge,” 410.

FOREKNOWLEDGE 39

otherwise have been feasible, and only complete foreknowledge can guarantee that no

conditional decision with a true antecedent will go unimplemented.”74

Although Hunt argues that this model bypasses the doxastic problem at hand, he

fails to see the fatal flaw in his argument. Even if God did this planning without accessing

His foreknowledge, once He decides to access it, then He consciously knows that He will

perform a certain action while also deliberating about whether to perform that action, which

violates the revised Doxastic Principle. Thus, the doxastic problem remains. Yet David

Basinger wonders if Hunt’s argument could be salvaged by replacing God’s complete and

infallible foreknowledge with “a justified belief about what will quite probably come

about.”75 If God only believed that something would probably happen, then He might still

be doxastically open per Kapitan’s definition.

David Basinger answers his own question by examining God’s desires. Since God

is a perfectly good being, it seems that He would be committed to doing as much as He

possibly could to accomplish His desired ends, since His own ends would also be perfectly

good. If God is committed to achieving His own ends, then He must consider the relevant

data to understand what is necessary to meet His own goals. Otherwise,

it is not only possible that he will make decisions that fail to actualize his desired
ends to the extent they could have been actualized if he had considered all of the
relevant data available, it is possible that his decisions will in fact significantly
hinder the actualization of his creative goals. Thus, if we assume that God is a
perfectly good being, we must conclude, I believe, that he would never formulate
conditional decisions before creation that did not contain in the antecedent

74. The entire argument, which ends with this quote, is found in “Divine Providence and Simple
Foreknowledge,” 413. Note: This sequence is an explication of the logical process in God’s mind; to interpret
this as a temporal sequence would be a misinterpretation of Hunt’s position.

75. Basinger, The Case for Freewill Theism, 126.

FOREKNOWLEDGE 40

conditions all the relevant data to which he would have access when these decisions
were implemented.76

Once one agrees that God formulated these conditional decisions and consulted the relevant

data in order to do so, then it follows that God must have knowledge of His own actions

and the consequences of those actions as part of the relevant data being considered. This is

eminently plausible, as God could commit Himself to an action and then discover that this

decision would not accomplish His own ends. To ensure that His own perfectly good goals

would be met, God must have complete knowledge of His own future actions and their

consequences. Yet if He requires complete knowledge of His own actions and their

consequences, then He cannot form conditional plans based on probability, which loops

directly back into the Doxastic Problem. Thus, if simple foreknowledge is adopted, then

God cannot act providentially in His own creation.77

The resulting image of God is defined by impotence, which clearly contrasts God’s

depiction in scripture: “The mind of man plans his way, but the Lord directs his steps.”78

To properly account for the biblical text, one’s view of divine foreknowledge must allow

God to providential guide the world according to His perfectly good will. Moreover, one’s

view of divine foreknowledge must support a coherent view of human freedom and moral

responsibility. The next view under consideration—Molinism—attempts to balance both

76. Ibid., 127-128.

77. This has been a paraphrase of Basinger’s argument. If one wishes to read the entire
formulation in Basinger’s own words, consult his appendix, “Simple Foreknowledge and Providential
Control,” in The Case for Freewill Theism.

78. Proverbs 16:9 NASB. There are other passages that describe God’s providential control, but
this verse seemed most straightforward.
FOREKNOWLEDGE 41

aspects by positing divine middle knowledge. It remains to be seen whether Molinism

adequately accounts for divine foreknowledge or whether it falls flat.

II. Molinism

A. The Molinist Definition of Divine Foreknowledge

To properly understand the Molinist definition of divine foreknowledge, one must

first understand middle knowledge; and to understand middle knowledge, one needs to

understand the several compartments of God’s knowledge. Consequently, it will be best to

start by defining God’s natural knowledge, which is the first compartment. As has been

discussed previously, Christians hold that God is, by nature, omniscient. Thus, natural

knowledge refers to what God must know by nature of being omniscient. This category

includes knowledge of all logically possible worlds as well as knowledge of all

metaphysically necessary truths.79 This natural knowledge is also prevolitional, meaning

that God had this knowledge prior to any creative act.80

Another compartment of God’s knowledge is free knowledge. When God decided

to create a world, He did so out of His own divine will; He was not forced to create by

anyone/anything else. By choosing to create a world, He foreknows what exists in the

actual world. Since this knowledge results from His creative act, and God freely chose to

create the world, it is labeled free or postvolitional knowledge. In other words, God could

79. Freddoso defines metaphysical necessity as a state of affairs obtaining “at every moment in
every possible world.” God’s natural knowledge includes perfect comprehension of the natures of all possible
creatures, all of the active and causal powers these creatures might exercise, and all of the different space-
time arrangements of creatures so that He knows how they will interact with each other. From his
“Introduction,” 11-12.

80. Ibid.
FOREKNOWLEDGE 42

have decided to not create a world and thereby not have free knowledge while still having

natural knowledge.81

In between God’s natural and free knowledge is His middle knowledge, which is

unique to Molinism.82 Through his middle knowledge, God knows conditional future

contingents, otherwise known as counterfactuals of freedom.83 This particular kind of

proposition indicates what effects “would in fact be produced, remotely or immediately, by

indeterministic secondary causes under a condition or hypothesis that specifies a possible

spatio-temporal arrangement of secondary causes.”84 In other words, counterfactuals of

freedom indicate what God could create in any possible world given the secondary causes

included in each possible world. Additionally, middle knowledge shares characteristics of

both natural and free knowledge. Middle knowledge is prevolitional like natural

knowledge, meaning that God has no control over which counterfactuals of freedom are

true and which are false; yet middle knowledge is also contingent like free knowledge, in

81. Thomas Flint refers to four different logical moments, where “in the third logical moment, God
decides upon a particular creative act of will—he decides which beings to create in which circumstances.
From this divine decision and the knowledge which precedes it flow not only all contingent creaturely events
ultimately precipitated by God’s creative action, but also (and immediately) the fourth logical moment, in
which God knows all the contingent truths under his control—that is, in which he has free knowledge.” See
his discussion in Divine Providence: The Molinist Account (Ithaca, N.Y.: Cornell Univ. Press, 1998), 43.

82. Other positions, such as Calvinism, hold that God knows what persons would do when placed
in certain circumstances. However, they would not separate this from God’s natural knowledge; Molinism is
unique in its separation of natural and middle knowledge.

83. In his own writing, Freddoso refers to them as conditional future contingents because it makes
sense within Molina’s own terminology. Later Molinist writings refer to them as counterfactuals of freedom,
and William Lane Craig sometimes calls them subjunctive conditionals.

84. Freddoso, “Introduction,” 21.

FOREKNOWLEDGE 43

the sense that counterfactuals of freedom that are true could have been false depending

upon the possible world in question.85

By introducing the concept of divine middle knowledge, Molina provided a way to

ground God’s complete and infallible foreknowledge in His own decree without sacrificing

the libertarian freedom of free creatures. The logical process flows as follows: natural

knowledge enables God to know infallibly how Peter could act in any logically possible

world. Middle knowledge enables God to know infallibly how Peter would act given any

possible combination of secondary causes and circumstances. God then decides to actualize

one of these worlds, with the actual world containing the situation where Peter denied

Christ three times. God’s free knowledge is His knowledge of the actualization of this

situation. In other words,

given that God knows infallibly by His middle knowledge that Peter would in fact
freely deny Christ if placed in those circumstances, it follows that by decreeing that
Peter will be in those circumstances God ensures that it is true from eternity that
Peter will freely sin—long before Peter actually brings about the effect in question.
So God freely knows that Peter will sin because it is already true by divine decree
that Peter will sin, just as God freely knows that a given necessary effect will obtain
because it is already true by divine decree that it will obtain. And it is in this sense—
and this sense alone—that God knows by His free knowledge that S will obtain
because S will in fact obtain.86

85. Ibid., 47. Freddoso makes the distinction between the three kinds of knowledge clearer by using
Adam’s sin in the garden as an example on page 47: “By His natural knowledge God knows that it is
metaphysically possible but not metaphysically necessary that Adam will sin if placed in the garden; by His
free knowledge he knows that Adam will in fact be placed in the garden and will in fact sin. What He knows
by His middle knowledge, on the other hand, is something stronger than the former but weaker than the latter,
namely, that Adam will sin on the condition that he be placed in the garden. So God has middle knowledge
… if and only if He has comprehensive prevolitional knowledge of conditional future contingents.”

86. The paragraph is paraphrased and the quotation is taken from ibid., 54. Flint also importantly
notes that “Free knowledge, then, is neither causally nor explanatorily prior to the true future contingent
foreknown. Indeed, in the sense that the truth of a proposition must be thought of as prior to the fact that
someone knows that truth, it seems evident that the true future contingent should be seen as explanatorily
prior to God’s knowledge of it.” Cf. Divine Providence, 45.
FOREKNOWLEDGE 44

B. Biblical Evidence for Molinism

As far as biblical evidence for divine foreknowledge goes, Molinists and Arminians

agree that the Bible clearly teaches God’s complete and infallible knowledge of the future.

Rather, the issue is whether scripture teaches God’s knowledge of counterfactuals of

freedom. Molina and other Molinists cite two passages that seem to suggest God’s middle

knowledge: 1 Samuel 23:6-13 and Matthew 11:20-24. In 1 Samuel 23, David is told that

Saul and his large army are marching toward Keilah to capture David and his men. When

David hears this report, he asks God whether Saul will come to Keilah to capture him, and

God says that Saul will march on the city. David then asks if the inhabitants of Keilah will

turn him and his men over to Saul, and God says that they will. Consequently, David and

his men leave before Saul arrives and they avoid getting captured.

Now this result raises some questions. Why did God answer affirmatively to both

of David’s questions if David did not actually get captured? Rather than denying God’s

foreknowledge, it makes more sense to posit that God knew the following counterfactual:

“If David remains in Keilah until Saul arrives, the men of Keilah would turn David over to

Saul.” This situation only makes sense if God knows counterfactuals of freedom. Likewise,

Jesus displays knowledge of counterfactuals when He judges Chorazin and Bethsaida:

“Woe to you Chorazin! Woe to you, Bethsaida! For if the miracles had occurred in Tyre

and Sidon which occurred in you, they would have repented long ago in sackcloth and

ashes.”87 Without knowledge of counterfactuals, Jesus could not honestly make this

statement.

87. Matt. 11:21. Other passages that are less cited but also display God’s counterfactual knowledge:
Jeremiah 38:17-18 and John 15:22-24. The passages cited in this section were referenced by William Lane
FOREKNOWLEDGE 45

Additionally, Molinists hold to a strong account of divine providence. However, it

would take too much space to fully explicate the biblical case for strong providential

control, so the reader must examine the evidence at his or her leisure.88

C. Theological Evidence for Molinism

Admittedly, the Molinist position cannot be directly proven from scripture, since

the concept of middle knowledge requires counterfactuals to be known prevolitionally, and

this is not mentioned by any passage of scripture. In other words, middle knowledge is a

philosophically derived concept rather than something that is derived from scripture.

Nevertheless, the Molinist position proves to be very compelling when one sees its

implications for Christian doctrine.

First, Molinism provides a coherent view of divine providence. Since God knows

what any person would in fact do in any set of circumstances, He may order the world to

meet His goals and purposes without compromising human freedom and the resultant

moral responsibility. The world is such that God will certainly have the ultimate victory

over evil because He chooses what circumstances to actualize, and this enables Him to

ensure that He will be glorified and exalted by what occurs in creation.89

Craig in his chapter, “The Middle-Knowledge View,” in Divine Foreknowledge: Four Views, edited by J.K.
Beilby and P.R. Eddy (Downers Grove, IL: InterVarsity Press, 2001), 123-124.

88. For a book-length discussion of the relevant biblical passages, see D.A. Carson, Divine
Sovereignty and Human Responsibility: Biblical Perspectives in Tension (Atlanta, GA: John Knox Press,
1994). For more coverage of the New Testament data, see chapter four, “God’s Control: Its Efficacy and
Universality,” in Frame, The Doctrine of God.

89. Craig even argues that God could providentially guide the world to have an optimal balance of
saved and unsaved persons, where “those who never in fact hear the gospel are persons who would not
respond to it if they did hear it. God brings the gospel to all those who he knows will respond to it if they
hear it. Thus, the motivation for the missionary enterprise is to be God’s ambassadors in bringing the gospel
to those whom God has arranged to freely receive it when they hear it. No one who would respond if he heard
it will be lost.” Cf. The Only Wise God, 150-151.
FOREKNOWLEDGE 46

Second, Molinism provides a coherent picture of prophecy. The Molinist could

provide a model of prophecy that involves God’s middle knowledge, thereby ensuring the

certainty of the event while also ensuring the freedom of the person performing the

prophesied action. Flint argues that

we need to think of God as prevolitionally recognizing via middle knowledge that,

if he willed to put Peter in C*, Peter would deny Jesus, and God could, if he so
willed, prophesy that denial. On the basis of this (and other) prevolitional
knowledge, God could then perform a single act of will that would have among its
results both Peter’s being in C* and Jesus’s prophesying the denial; and only
logically posterior to this single act of will would God have foreknowledge.90

Therefore, God can have complete certainty about the actual occurrence of the action

prophesied due to His providential guidance via middle knowledge while maintaining the

freedom of the person being prophesied.

Beyond these two theological topics, there have been several essays arguing for

Molinist understandings of the perseverance of the saints, biblical inspiration, infallibility,

and theistic evolution.91 In sum, the Molinist doctrine of middle knowledge has some wide-

reaching explanatory power that Open Theism and Arminianism cannot match.

90. Flint, Divine Providence, 209.

91. I will not be discussing these articles in this essay, but I encourage the reader to explore these
on his or her own time. For perseverance of the saints, see William Lane Craig, “‘Lest Anyone Should Fall’:
A Middle Knowledge Perspective on Perseverance and Apostolic Warnings,” International Journal for
Philosophy of Religion 29 (1991): 65-74. For biblical inspiration, see William Lane Craig, “‘Men Moved by
the Holy Spirit Spoke from God’ (1 Peter 1:2): A Middle Knowledge Perspective on Biblical Inspiration,”
Philosophia Cristi 1 (1999): 45-82. For infallibility, see Thomas Flint, “Middle Knowledge and the Doctrine
of Infallibility,” in Philosophy of Religion, vol. 5 of Philosophical Perspectives, edited by Jas. E. Tomberlin
(Atascadero, CA: Ridgeway, 1991), 373-393. For theistic evolution, see Del Ratzsch, “Design, Chance and
Theistic Evolution,” in Mere Creation, edited by William Dembski (Downers Grove, IL: Intervarsity Press,
1998), 289-312.
FOREKNOWLEDGE 47

D. Philosophical Evidence for Molinism

Much of the philosophical evidence for Molinism has been explained in the

previous section on Philosophical Evidence for Arminianism. Thus, if one wants evidence

for Molinism’s conjunction of divine foreknowledge and human freedom or for its support

for libertarian freedom, turn back to that section. This section will give an argument for

middle knowledge that was presented by William Lane Craig and explain its premises. The

argument runs as follows:

1. If there are true counterfactuals of creaturely freedom, then God knows these truths.
2. There are true counterfactuals of freedom.
3. If God knows true counterfactuals of creaturely freedom, God knows them either
logically prior to the divine creative decree or only logically posterior to the divine
creative decree.
4. Counterfactuals of creaturely freedom cannot be known only logically posterior to
the divine creative decree.
5. Therefore, God knows true counterfactuals of freedom (from 1 and 2).
6. Therefore, God knows true counterfactuals of creaturely freedom either logically
prior to the divine creative decree or only logically posterior to the divine creative
decree (from 3 and 5).
7. God knows true counterfactuals of creaturely freedom logically prior to the divine
creative decree (from 4 and 6).92

The first premise is a logical implication of divine omniscience, which was

previously defined as God’s knowledge of all true propositions. If counterfactuals of

freedom are true propositions and God does not know them, then God is not omniscient.

The second premise is the most controversial, and objections to the premise will be handled

later. For now, it will be assumed that it is true. The third premise states the logical

alternatives: either God knows counterfactuals logically before or after the divine creative

decree.

92. Craig, “The Middle-Knowledge View,” 136-137.

FOREKNOWLEDGE 48

The fourth premise is also somewhat controversial, but there is plausible support

for it. If God only knew counterfactuals of freedom after His creative decree, then He

would be deciding what each person would do in every actual circumstance rather than

having His middle knowledge be based upon what a person would do in every possible set

of circumstances. In other words, “if God knows counterfactual truths about us only

posterior to this decree, then there really are no counterfactuals of creaturely freedom. If

there are such counterfactuals, they must be true logically prior to the divine decree.”93

Given these four premises, then the rest of the argument logically follows; and since

premise 7 is the definition of middle knowledge, it follows that Molinism is true.

E. Critiques of Molinism

1. Biblical critiques. The first objections to Molinism are focused on the biblical

evidence for God’s knowledge of counterfactuals of freedom. As was explained previously,

there are only three or four passages of scripture that can plausibly be taken as evidence

for divine counterfactual knowledge, but even these passages are disputed. Objectors argue

that these passages are only rhetorical, and that God and/or Jesus is using the subjunctive

mood to emphasize the point he is making. However, the Molinist can rightly point out that

the burden of proof is on the skeptic to show that this language is only rhetorical; since no

one has successfully made that case, it seems reasonable to recognize the biblical evidence

for God’s counterfactual knowledge.94

93. Ibid., 143.

94. This objection and the response to it is provided by Freddoso in his “Introduction,” 62-63.
FOREKNOWLEDGE 49

2. Theological critiques. In addition to the biblical objections, Thomists from

Molina’s period raised three theological objections to Molinism that need to be addressed.

The first objection focuses on the asymmetry thesis, which states that God relates to good

and evil in two different ways. Normally the asymmetry thesis is understood as the

intention/permission distinction, which argues that God intends good events to occur but

only permits evil events to occur. Based on this distinction, Thomists argued that “God’s

total causal contribution to free actions is nondetermining and hence that the very same

causal influence on God’s part may result indifferently in either a good act or an evil act.”95

Without this intention/permission distinction, God could be considered the author of both

good and evil actions.

In response, it is important to distinguish between God’s concurrence and God’s

grace. Every traditional Christian position has believed that “God’s general causal

influence is required in order for secondary causes to bring about any effects whatever

anywhere in the created world.”96 This cooperation with secondary causes is termed God’s

general concurrence. For Molina, God concurrently upheld secondary causes by directly

acting on the effect itself, in contrast to the Bañezian position that God acted directly on

the secondary agent.97 Furthermore, Molina argued that God’s concurrence is intrinsically

95. Freddoso, “Introduction,” 64.

96. Ibid., 17.

97. The Bañezians were followers of the Dominican Domingo Bañez, who Freddoso describes as
“Molina’s chief theological nemesis … who taught at Salamanca from 1577 to 1600 and was active in the
official ecclesiastical inquiry into Molina’s writings” (134, note 13). They were Thomists who also accepted
the reality of counterfactuals of freedom, but they argued that counterfactuals are located after the God’s
creative decree, so that God decides which counterfactuals are true and which are false. For a broader
discussion of the Bañezian position, see Freddoso, “Introduction”, 36-42.
FOREKNOWLEDGE 50

neutral and is only rendered efficacious by the secondary causes, whereas Bañezians

believed that God’s concurrence is intrinsically efficacious so that they could keep the

intention/permission distinction intact.98

Rather than grounding the intention/permission distinction in God’s concurrence,

Molina grounds it in God’s grace. Molina thinks that grace is a particular cause in contrast

to the general causation of divine concurrence. He thinks that God’s grace inclines a person

toward the good without causally necessitating their choice of the good. In other words,

God may causally enable a person to choose the good and that person could still choose to

make the evil choice. Consequently, the Molinist may still uphold the asymmetry thesis:

when a good act is elicited, it is perfectly proper to say that it was elicited because
of God’s grace acting as a cause that inclined the agent internally toward such an
act. When an evil act is elicited, it is perfectly proper to say that God’s particular
causal influence on the subject has been impeded, so that His only effective causal
contribution to the act has been His general concurrence, which is intrinsically
neutral in the strong sense and hence does not determine in any way the specific
moral character of the act.99

A second Thomist objection invokes the principle of predilection, which states that

no one would be better than another person unless he were loved and helped more by God.

This objection argues that Molinism violates this principle, since two persons could be

placed in the same circumstances and they would act differently, with the result that the

one person may boast in his acceptance and use of God’s grace. Molinists may counter

argue that this principle does still hold, since “it is God who gratuitously singles Peter out

by arranging things in such a way that Peter will freely act well. By the same token, God

98. This explanation of concurrence, as well as the difference between Molinist and Bañezian
conceptions of concurrence, are found in ibid., 17-18.

99. This paragraph and the quotation may be found in Freddoso, “Introduction,” 64-65.
FOREKNOWLEDGE 51

permits Judas to sin by allowing him to be so situated that, as God knows via middle

knowledge, he will freely sin.” In other words, God is still giving more grace to some

people instead of other people, which meets the criteria of the principle of predilection.100

The third theological objection involves petitionary prayer. The reader might recall

a time in his or her life where a prayer was offered for a friend involving a request for that

person’s salvation, a plea for that person to change a destructive habit, or a number of other

things. If one accepts Molinism, then one must reconcile the conjunction of strong

providential control and libertarian freedom. Since libertarian freedom maintains that one

agent cannot causally determine another agent’s actions, Bañezians argued that if God

cannot directly cause us to act in certain ways, then He could not effectively answer

prayers, which means that Molinism diminishes the importance of prayer.101

To answer this objection, two distinctions are in order. First, it is arguable that some

aspects of conversion are the result of God’s direct causation (e.g. removing a heart of

stone and replacing it with a heart of flesh, as described by Ezekiel 36:26) while others

involve the will of the person converted (e.g. choosing to sin no more). In this way,

Molinists could accept God’s direct causal influence while still holding to libertarian

freedom. Second, one needs to distinguish between strong and weak actualization. When

God strongly actualizes something, He directly causes it to occur; when He weakly

actualizes something, He brings about the states of affairs which result from His direct

causation of something else. Accordingly, whatever happens in conversion and

100. This objection and the quotation that responds to it come from Freddoso, “Introduction,” 66.

101. Flint mentions this objection in Divine Providence, 110.

FOREKNOWLEDGE 52

sanctification that does not involve human free will may be the result of God’s strong

actualization, and whatever happens in conversion and sanctification that does involve

human free will may be the result of God’s weak actualization. Thus, Molinism provides a

plausible account of God’s interaction with humanity without diminishing the importance

of prayer.

3. Semantic and metaphysical objections. Although the biblical and theological

objections to Molinism prove unsuccessful, many philosophical objections to Molinism

have been raised ever since Plantinga reintroduced it in The Nature of Necessity. The most

prominent objection to Molinism has been the “grounding” objection, which argues that

counterfactuals of freedom lack metaphysical grounds and therefore cannot be true or false,

meaning that God cannot know them. Robert Adams was the first modern philosopher to

raise this objection in his 1977 article, “Middle Knowledge and the Problem of Evil.” To

his mind, the only plausible metaphysical grounding of counterfactuals are the intentions,

desires, and character of the persons mentioned in the counterfactual. However, traditional

conceptions of libertarian freedom hold that a person may act against their character; thus,

the only plausible counterfactuals would say that it is probable agent S would do action A,

which fails to meet the Molinist criteria for counterfactuals that S would in fact do A.102

While Adams’s analysis initially seems plausible, there are two problems that seem

to undermine it. First, Wierenga points out that the proposition “Saul may act out of

character” can be understood in two different ways: it either indicates an ability that Saul

has (Saul has it within his power to act out of character), or it indicates a logical possibility

102. Robert M. Adams, “Middle Knowledge and the Problem of Evil,” American Philosophical
Quarterly 14, no. 2 (April 1977): 109-117, accessed June 25, 2016, http://www.jstor.org/stable/20009657.
FOREKNOWLEDGE 53

(it might be that David stays in the city and Saul acts out of character, deciding not to

besiege the city). It seems that Adams argument is that since either of these interpretations

is plausible, it cannot be said that Saul’s intentions or character entails the certainty of the

counterfactual. Yet this potential argument fails due to Adams’s own understanding of

metaphysical grounding:

Perhaps we can avoid settling that issue, however, if we recall that Adams is willing
to allow that a causally sufficient condition can ground a proposition. Since a
causally sufficient condition for the truth of a proposition need not entail the
proposition, it seems clear that Adams does not require that a ground of a
proposition entail the proposition. Hence, the fact that [Saul’s character and
intentions] does not entail [the claim of the counterfactual] does not seem to show
that it does not ground it.103

If Adams himself thinks that lack of entailment does not necessarily result in lack of

metaphysical grounding, then Molinist counterfactuals have metaphysical grounding as

well even if they lack entailment.

Second, Freddoso argues that Molinist counterfactuals have metaphysical

grounding like the grounding of absolute future contingents. A realist about the absolute

future holds that a future-tense proposition has proper metaphysical grounds now if “there

will be at some future time adequate metaphysical grounds for the truth of its present-tense

counterpart p.” In other words, all that is necessary to metaphysically ground a future-tense

proposition is that “some agent has caused or will cause the corresponding present-tense

propositions to be true.” Subsequently, Freddoso argues that Molinist counterfactuals could

be grounded in a similar way: “it seems reasonable to claim that there are now adequate

metaphysical grounds for the truth of a conditional future contingent Ft (p) on H just in

103. The argument of this paragraph and the quote may be found in Edward Wierenga, The Nature
of God: An Inquiry into Divine Attributes (Ithaca: Cornell University Press, 1989), 141-143.
FOREKNOWLEDGE 54

case there would be adequate metaphysical grounds at t for the truth of the present-tense

proposition p on the condition that H should obtain at t.” Unless one wants to deny

knowledge of the future, then one should accept the grounding of Molinist counterfactuals

as well.104

Based on Freddoso’s argument, a formula for metaphysical grounding can now be

created: “‘It X the case (Y) that z’ is now grounded iff it X the case (Y) that ‘z is now

grounded.’” For instance, if one wants to know whether the statement “It might be the case

that Donald Trump wins the presidential election in 2020” is grounded, then

we need to look where the temporal or modal operator X tells us to look (into the
past of our world, the future of our world, some other world), look precisely where
the specifier Y tells us to look (three years ago, two months from now, the nearest
world in which I offer Gore the bribe), and see if the present-tense proposition z is
grounded there by the causal activity of some agent. If it is, the proposition in
question is grounded; if it isn’t, the proposition is ungrounded.105

This formula requires a proposition’s grounding to be determined by its proximity to the

actual world; but what if a possible world where a person refrains from an action is equally

proximal to the actual world as another possible world where the same person performs the

same action? If the propositions “tie” in this way, then a false proposition has the same

grounding as a true proposition. Since false propositions are not grounded, it supposedly

follows that true propositions are ungrounded as well.106

104. The quotations in this paragraph are found in Freddoso, “Introduction,” 72-73. To read the
whole argument, read pages 69-75.

105. Both quotes are from Flint, Divine Providence, 134. The Trump example is my own.

106. Ibid., 135. This same objection is also raised by David Hunt in “A Simple-Foreknowledge
Response” in Divine Foreknowledge: Four Views, edited by J.K. Beilby and P.R. Eddy (Downers Grove,
IL: InterVarsity Press, 2001), 153-154.
FOREKNOWLEDGE 55

Subsequently, both Plantinga and Flint have responded by appealing to primitive

properties that apply to certain possible worlds rather than others:

A Z-less world [where Z stands for an action in a possible world] may be no less
similar to the actual world than is a Z-ful one if only fundamentally non-conditional
propositions are considered. But if, as Plantinga contends, ‘one feature determining
the similarity of worlds is whether they share their counterfactuals,’ then the Z-ful
world may well be more similar to the actual world due to the fact that, both in it
and in the actual world, (c  z) is true, whereas the same counterfactual is false in
the relevant Z-less worlds.107

Some have worried that such a response contradicts the standard semantics for

counterfactuals, but Mares and Perszyk have shown that it does not, since these primitive

properties “can help determine which closeness relations are appropriate for the assessment

of [counterfactuals of creaturely freedom]. The semantics is still used to give a

compositional theory of truth for counterfactuals.”108 Thus, this “primitive facts” view

seems to adequately resolve the “tie” argument.

Nevertheless, the rebuttal of the “tie” argument results in the introduction of a

metaphysical issue, commonly termed the “priority” problem. It essentially argues that God

cannot know counterfactuals of freedom prior to His act of creation, rendering them useless

for providential action in the world. I do not have the space to cover this objection here,

but Plantinga gave a plausible response that seems to defuse the objection.109

107. Flint, Divine Providence, 135. Note that (c  z) would be false in Z-less worlds because the
agent would choose to perform the action rather than refrain; he has the ability to refrain from Z, but chooses
to perform Z instead, meaning that the Z-ful world contains the true counterfactual.

108. Edwin Mares and Ken Perszyk, “Molinist Counterfactuals,” in Molinism: The Contemporary
Debate, edited by Kenneth Perszyk (Oxford: Oxford University Press, 2011), 110.

109. For formulations of the priority problem, see Hugh J. McCann, “The Free Will Defense,” in
Molinism: The Contemporary Debate, edited by Kenneth Perszyk (Oxford: Oxford University Press, 2011),
259, and Adams, “Middle Knowledge and the Problem of Evil,” 113-114. Plantinga’s reply may be found in
“Replies to My Colleagues,” in Alvin Plantinga, edited by James Tomberlin and Peter van Inwagen
(Dordrecht: D. Reidel, 1985), 376.
FOREKNOWLEDGE 56

4. Critiques of the conjunction between Molinism and libertarian freedom. Up

until now, the objections examined have focused either on semantic issues or metaphysical

grounding issues pertaining to counterfactuals of freedom. Thus far, it seems like those

issues have been sufficiently rebutted by Molinists. Yet there is one category of objections

not yet examined which focuses on Molinism’s adherence to libertarian freedom.

Beginning with William Hasker in 1986, a slew of arguments and counterarguments have

been written debating whether counterfactuals of freedom eliminate a person’s free will.110

To keep things concise, only Hasker’s most recent version of the argument will be

explained.

To begin, Hasker defines the crucial notion of “bringing about” something:

“A brings it about that Y iff [if and only if]: For some X, A causes it to be the case that X,

and (X & H) ⇒ Y, and ~ (H ⇒ Y), where ‘H’ represents the history of the world prior to its

coming to be the case that X and ‘⇒’ stands for broadly logical or metaphysical

necessity.”111 Using this definition, and assuming that “□→” symbolizes the connective in

a counterfactual, he builds the following argument:

110. If the reader wishes to follow the paper trail, one may do so as follows: William Hasker, “A
Refutation of Middle Knowledge,” Noûs 20, no. 4 (December 1986): 545-557; Freddoso, “Introduction,” 75-
78; Robert Adams, “An Anti‐Molinist Argument”, in James E. Tomberlin, ed., Philosophical Perspectives 5
(1991): 343–53; William Lane Craig, “Robert Adams’s New Anti-Molinist Argument,” Philosophy and
Phenomenological Research 54 (1994): 857-861; William Hasker, “Middle Knowledge: A Refutation
Revisited,” Faith and Philosophy 12, no. 2 (April 1995): 223-236; William Hasker, “Explanatory Priority:
Transitive and Unequivocal: A Reply to William Craig”, Philosophy and Phenomenological Research 57
(1997): 389–93; William Lane Craig, “On Hasker’s Defense of Anti-Molinism,” Faith and Philosophy 15,
no. 2 (April 1998): 236-240; Flint, Divine Providence, chapters 6 and 7; William Hasker, “A New Anti-
Molinist Argument,” Religious Studies 35 (1999): 291-297; and William Hasker, “Anti-Molinism is
Undefeated!” Faith and Philosophy 17, no. 1 (January 2000): 126-131.

111. This definition is originally found in Hasker, “A New Anti-Molinist Argument,” 291, but I
found it in William Hasker, “The (Non-) Existence of Molinist Counterfactuals,” in Molinism: The
Contemporary Debate, edited by Kenneth Perszyk (Oxford: Oxford University Press, 2011), 31.
FOREKNOWLEDGE 57

1. Agent A is in circumstances C, the counterfactual of freedom ‘C □→ Z’ is true of

her, and she freely chooses to do Z. (Molinist premise)
2. A is in C, and it is in A's power to refrain from doing Z. (From premise 1 and
definition of libertarian freedom)
3. It is in A's power to bring it about that: A is in C, and A refrains from doing Z. (From
premise 2)
4. If it is in A's power to bring it about that P, and ‘P’ entails ‘Q’ and ‘Q’ is false, then
it is in A's power to bring it about that Q. (Power Entailment Principle)
5. (A is in C and refrains from doing Z) ⇒ (C □→ ~ Z). (Molinist premise)
6. If it is in A's power to bring it about that A is in C and refrains from doing Z, and
‘(C □→ ~ Z)’ is false, then it is in A's power to bring it about that (C □→ ~ Z).
(From premises 4 and 5)
7. It is in A's power to bring it about that (C □→ ~ Z). (From premises 1, 3, 6)
8. It is not in an agent’s power to bring about the truth of the counterfactuals of
freedom about her.
9. It is not in A’s power to bring it about that (C □→ ~ Z). (From premise 8)112

The crucial step in this argument is the move from premise 7 to premise 8. To

support this move, Hasker claims that counterfactuals of world-actualization are an

important part of God’s decision to create a world, and therefore are a key part of a world’s

history. But if these counterfactuals are part of a world’s history, then “‘H’ does entail ‘C

□→ X’, where ‘C □→ X’ is a true counterfactual of freedom. But if this is so, then we

created free agents do not bring about the truth of counterfactuals of freedom about us;

there is no possible world in which we do this.”113 Consequently, Hasker thinks that

premise 8 is adequately supported, resulting in the impossibility of libertarian freedom.

Thomas Flint disagrees with Hasker’s conclusion, arguing that his objection to

Molinism fails even if one assumes that counterfactuals of freedom are included in a

world’s history, such that “(BA*) A brings it about that Y iff: For some X, A causes it to be

112. Premises 1-7 are on pages 31-32 of Hasker, “The (Non-) Existence of Molinist
Counterfactuals”; premises 8 and 9 are on page 34 of the same chapter.

113. Ibid., 33.

FOREKNOWLEDGE 58

the case that X, and (X & H) ⇒ Y, and ~ (H ⇒ Y), where ‘H’ represents the history of the

world (including all true counterfactuals of creaturely freedom) prior to its coming to be

the case that X.”114 The fatal flaw of Hasker’s argument arises from the assumption that “if

it's not possible that A brings it about that X, then it's not possible that A has the power to

bring it about that X.”115 Using this assumption, Flint revises premises 6 and 7 of the

argument to say that (6*) In no possible world does A bring it about that (C □→ ~ Z), and

(7*) In no possible world does A have the power to bring it about that (C □→ ~ Z). This

revision accurately captures Hasker’s assumption while maintaining the flow of the

original argument, yet it also results in a contradiction; although 6* is true given the revised

definition of “bringing about”, 7* is not true. This is because “(7*) amounts to the claim

that (7) is true in no possible world. And as we've already seen, the Molinist will quite

reasonably say that, given (BA*), (7) is in fact true—true in the actual world—and hence

true in some possible worlds.”116

If (6*) is true and (7*) is false, then Flint has derived a counterexample to Hasker’s

assumption, thereby giving a reason to reject Hasker’s move from premise 8 to premise 9.

In the end, Flint thinks that arguments like Hasker’s—those that try to invoke notions of a

world’s history or numerous definitions of “bringing about”—miss the crucial issue of the

whole debate. In Flint’s mind, the whole debate revolves around the Molina’s view that

God’s middle knowledge depends upon what persons would do in actual circumstances:

114. Thomas Flint, “Whence and Whither the Molinist Debate: A Reply to Hasker,” in Molinism:
The Contemporary Debate, edited by Kenneth Perszyk (Oxford: Oxford University Press, 2011), 40-41.

115. Ibid., 41.

116. The revised versions of premises 6 and 7, as well as the quote, are from ibid., 42.
FOREKNOWLEDGE 59

“Even though (C □→ Z) is true and A is in C, A still is able to do ~ Z. Had she done ~ Z,

(C □→ Z) wouldn't have been true. Instead, (C □→ ~ Z) would have been true. So A has

the power to do something (~ Z), such that, had she done it, a counterfactual that is true—

namely, (C □→ Z)—would have been false.”117 Molinists and anti-Molinists have argued

fiercely over the plausibility of this position, but Flint thinks that revising the same

arguments over and over again will move the discussion nowhere.

In sum, Molinism has been the most biblically consistent and philosophically

coherent position thus far. It has seemingly answered all objections, and provided a

balanced way of accepting both strong providential control over creation and libertarian

freedom. Nevertheless, it seems that some theological aspects of Molinism should be

corrected. Most well-known Molinists, such as William Lane Craig, ardently hold to an

Arminian view of soteriology, where God offers prevenient grace to all people so that they

may choose to follow Him.118 By contrast, the author holds that the Reformed view makes

better sense of the biblical data, and will therefore argue that Reformed Molinism provides

the best option for maintaining a biblical soteriology and a coherent view of human

responsibility.119

117. Flint, “Whence and Whither the Molinist Debate,” 43.

118. “Arminian-Wesleyan theologians concur with the Augustinian-Calvinist position that humans
are unable to believe in Christ without the prior drawing of the Holy Spirit; however, they deny that this
drawing is the same as regeneration. They instead maintain that this drawing is prevenient grace, or prior
ability-supplying grace, that the Spirit gives to all humanity (John 12:32).” See Kirk R. MacGregor,
“Regeneration,” in The Lexham Bible Dictionary, edited by John D. Barry et al. (Bellingham, WA: Lexham
Press, 2016).

119. Admittedly, the Arminian-Calvinist debate over soteriology is one of the most contentious
areas of Christian theology, so simply stating that the Reformed view better accords with the biblical data is
questionable, to say the very least. To back this claim is far beyond the scope of this paper, so the reader
should investigate the varying positions in due time.
FOREKNOWLEDGE 60

III. The Author’s View

Calvin argued that the Fall resulted in the corruption of humanity’s reason and heart

(i.e. intentions and character). This is central to the Christian religion, so it seems very

unwise to deny this claim. He then argued that this corruption results in humanity’s

inability to choose salvation in Christ and inability to even desire the good. This seems

correct; humans have lost what Frame calls moral freedom.120 Yet the entailment between

one’s loss of moral freedom and one’s loss of libertarian freedom is unclear.

Of course, a person’s moral freedom is what should be truly desired, since it allows

people to seek after God and pursue Him wholeheartedly. Calvin thought that

Man, since he was corrupted by the fall, sins not forced or unwilling, but
voluntarily, by a most forward bias of the mind; not by violent compulsion, or
external force, but by the movement of his own passion; and yet such is the
depravity of his nature, that he cannot move and act except in the direction of
evil.121

Yet it is far from clear how the natural disposition of the human heart necessarily leads to

a compatibilist account of human freedom. If a person (who was not empowered by God’s

prevenient grace) performed an SFA, could they not choose between two conflicting

reasons, both of which are evil? Is there any reason why this could not be the case, unless

a person presupposed a compatibilist view of human freedom?

On the other hand, is this a possibility on the Molinist scheme? Molina himself held

to a strong view of libertarian freedom, where “at [time] t [person] P contributes causally

to [state of affairs] S only if (i) at t P contributes causally to S and (ii) P’s contributing

120. Frame, The Doctrine of God, 135-136.

121. John Calvin, Institutes of the Christian Religion 2.3.5.

FOREKNOWLEDGE 61

causally to S does not obtain at t by a necessity of nature and (iii) the total causal activity

at t of causes other than P is compossible with P’s not causally contributing at t to S.”122

Kane’s analysis could readily accept this definition as well; the only difference would be

that the power of contrary choice is not necessary for every decision to be morally culpable,

since PFAs and FWAs do not require this power.

By adopting Kane’s event-causal view, a nuanced view of human freedom begins

to emerge. One may hold that many, if not all, desires and intentions arise from a corrupted

will, so that one is naturally inclined towards wrongdoing. Next, one can grant that many

decisions are the result of teleological guidance control, where one path is clearly more

appealing than the other and one follows his desires without thought. These would be

common actions which one (usually) puts little thought to: what food to eat, what music

one enjoys listening to, and what clothes one chooses to wear. However, Christians also

believe that God intervenes and imparts His prevenient grace, which inclines a person

toward the good. It is at these moments, when God’s grace clashes with carnal nature, that

self-forming acts occur.123

This clash of grace and human will is most visible at the moment of conversion.

Through His middle knowledge, God knows that if a person is given enough grace and

placed in the right circumstances, then that person would decide to accept Christ as his or

her Lord and Savior. If God elects an individual to salvation, then His particular grace

122. Freddoso, “Introduction,” 27.

123. Kane thinks of these clashes in primarily physical terms, where neural processes are clashing
in the brain. While I would accept this clashing in the brain, I also hold that humans experience spiritual
conflict, where “the spirit” and “the flesh” conflict with one another. Although this conflict readily
accommodates to substance dualism, it could apply to a neo-Aristotelian metaphysic as well.
FOREKNOWLEDGE 62

begins to change the person’s heart and he or she is drawn to Him. At the moment of

conversion itself, God changes the person’s heart apart from any act of the person’s own

will; since no one would ever decide to follow God without His help, God must change the

spiritual orientation of a person’s soul. With the changed heart, the person now has two

desirable options before him or her: choose God and follow His perfect plan, or reject

God’s call and live a life of rebellion. Because God has placed the person in the set of

circumstances where he would in fact choose Him, the person overcomes the

indeterminacy in his neural processes and surrenders his life to Christ.

After conversion, there will be more opportunities for self-forming actions as part

of the process of sanctification. The flesh and the spirit will continue to do battle, resulting

in difficult choices. Nevertheless, God’s middle knowledge allows Him to place persons

in the right circumstances so that they will choose Him over the desires of the flesh. Thus,

He can guarantee that person’s perseverance throughout the rest of his life on earth.

Now this is a basic version of the author’s proposed model; to give it the proper

nuance would go well beyond the scope of this paper. Still, this initial form captures both

the biblical data and lived experience. In the life of a Christian, there is a tug-of-war

between the desires of the flesh and the desires of the spirit that is readily incorporated into

Kane’s view. Furthermore, it merges well with Molinism so that God still has strong

providential control over His creation. In sum, Reformed Molinism expands Kane’s view

by incorporating it into a Molinist framework and adding the conflict between flesh and

Spirit to the already-existing clash between conflicting motivations or intentions,

combining the best aspects of Molinism and Reformed soteriology.

FOREKNOWLEDGE 63

IV. Conclusion

At their core, Arminianism and Molinism agree about the proper definition of

divine foreknowledge. Both camps hold that the biblical evidence supports God’s

completely and infallibly knowing the future; the difference arises when one addresses the

implications of this definition. Arminians hold that divine foreknowledge alone is

sufficient to explain God’s providential control over creation and the coherent conjunction

of God’s sovereignty and human responsibility. However, it was shown that divine

foreknowledge alone is incapable of supporting a strong theory of providential control,

since it eliminates God’s ability to deliberate between options. By contrast, Molinism

establishes a basis for strong providential control through its affirmation of divine middle

knowledge, and it coherently maintains God’s sovereignty and human responsibility.

Additionally, Reformed Molinism conjoins Robert Kane’s event-causal libertarianism with

Molinism and Reformed soteriology to produce the most biblically consistent and

philosophically/theologically coherent view available.

FOREKNOWLEDGE 64

Bibliography

Adams, R. M. “Middle Knowledge and the Problem of Evil.” American Philosophical

Quarterly 14, no. 2 (April 1977): 109-117. Accessed June 25, 2016.
http://www.jstor.org/stable/20009657.

Arminius, Jacob. “Certain Articles To Be Diligently Examined And Weighed Because

Some Controversy Has Arisen Concerning Them Among Even Those Who
Profess The Reformed Religion.” In The Works of Jacob Arminius, Vol. 2.
Translated by James Nichols. Buffalo, NY: Derby, Orton, and Mulligan, 1853.
Accessed January 21, 2017. http://wesley.nnu.edu/arminianism/the-works-of-
james-arminius/volume-2/controversial-articles-of-faith/.

_____. "A Declaration of the Sentiments of Arminius." In The Works of Jacob Arminius,
Vol. 1. Translated by James Nichols. Buffalo, NY: Derby, Orton, and Mulligan,
1853. Accessed June 11, 2016. http://wesley.nnu.edu/arminianism/the-works-of-
james-arminius/volume-1/arminius-theological-sentiments/.

Barry, John D. et al., eds. The Lexham Bible Dictionary. Bellingham, WA: Lexham
Press, 2016.

Basinger, David. The Case for Freewill Theism: A Philosophical Assessment. Downers
Grove, IL: InterVarsity Press, 1996.

Boyd, G. A. God of the Possible: A Biblical Introduction to the Open View of God. Grand
Rapids: Baker Book House, 2000.

_____. “An Open-Theist Response.” In Divine Foreknowledge: Four Views. Edited by

J.K. Beilby and P.R. Eddy. Downers Grove, IL: InterVarsity Press, 2001.

Brant, D. E. “On Plantinga’s Way Out.” Faith and Philosophy 14, no. 3 (July 1997): 334-
352. Accessed June 27, 2016. DOI:10.5840/faithphil199714327.

Calvin, John. Institutes of the Christian Religion. Translated by Henry Beveridge.

Edinburgh: The Calvin Translation Society, 1845.

Carson, D.A. Divine Sovereignty and Human Responsibility: Biblical Perspectives in

Tension. Atlanta: John Knox Press, 1994.

Craig, William L. “The Middle-Knowledge View.” In Divine Foreknowledge: Four

Views. Edited by J.K. Beilby and P.R. Eddy. Downers Grove, IL: InterVarsity
Press, 2001.
FOREKNOWLEDGE 65

_____. “‘No Other Name’: A Middle Knowledge Perspective on the Exclusivity of

Salvation Through Christ.” Faith and Philosophy 6 (1989): 172-188. Accessed
February 5, 2017. DOI: 10.5840/faithphil19896212.

_____. “On Hasker’s Defense of Anti-Molinism.” Faith and Philosophy 15, no. 2 (April
1998): 236-240. Accessed June 24, 2016. DOI: 10.5840/faithphil199815220.

_____. The Only Wise God: The Compatibility of Divine Foreknowledge and Human
Freedom. Wipf & Stock Publishing, 2000.

Edwards, J. Freedom of the Will. 1754. Reprint, Grand Rapids, MI: Christian Classics
Ethereal Library, 1999.

Flint, Thomas P. Divine Providence: The Molinist Account. Ithaca, N.Y.: Cornell Univ.
Press, 1998.

_____. “Whence and Whither the Molinist Debate: A Reply to Hasker.” In Molinism:
The Contemporary Debate. Edited by Kenneth Perszyk. Oxford: Oxford
University Press, 2011.

Frame, J.M. The Doctrine of God. Phillipsburg, NJ: P & R Publishing, 2002.

Freddoso, Alfred J. “Introduction.” In On Divine Foreknowledge: Part IV of the

Concordia. Edited and translated by Alfred J. Freddoso. Ithaca: Cornell
University Press, 1988.

Geach, P. T. Providence and Evil. Cambridge: Cambridge University Press, 1977.

Geisler, Norman L. The Battle for God: Responding to the Challenge of Neotheism.
Grand Rapids, MI: Kregel, 2001.

Hasker, W. “Anti-Molinism is Undefeated!” Faith and Philosophy 17, no. 1 (January

2000): 126-131. Accessed June 24, 2016. DOI: 10.5840/faithphil20001715.

_____. “Middle Knowledge: A Refutation Revisited.” Faith and Philosophy 12, no. 2
(April 1995): 223-236. Accessed June 24, 2016. DOI:
10.5840/faithphil199512216.

_____. “A Refutation of Middle Knowledge.” Noûs 20, no. 4 (December 1986): 545-557.
Accessed June 20, 2016. http://www.jstor.org/stable/2214984.

Hunt, D. P. “Divine Providence and Simple Foreknowledge.” Faith and Philosophy 10,
no. 3 (June 1993): 396-416. Accessed May 29, 2016. DOI:
10.5840/faithphil19931036.
FOREKNOWLEDGE 66

_____. “Frankfurt Counterexamples: Some Comments on the Widerker-Fischer Debate.”

Faith and Philosophy 13, no. 3 (July 1996): 395-401. Accessed May 29, 2016.
DOI: 10.5840/faithphil199613331.

_____. “On Augustine’s Way Out.” Faith and Philosophy 16, no.1 (January 1999): 3-26.
Accessed June 2, 2016. DOI: 10.5840/faithphil19991612.

_____. “A Simple-Foreknowledge Response.” In Divine Foreknowledge: Four Views.

Edited by J.K. Beilby and P.R. Eddy. Downers Grove, IL: InterVarsity Press,
2001.

_____. “The Simple-Foreknowledge View.” In Divine Foreknowledge: Four Views.

Edited by J.K. Beilby and P.R. Eddy. Downers Grove, IL: InterVarsity Press,
2001.

Kane, Robert. “Introduction: The Contours of Contemporary Free-Will Debates (Part 2)”.
In The Oxford Handbook of Free Will. 2nd edition. Edited by Robert Kane. New
York, NY: Oxford University Press, 2011.

_____. “Libertarianism.” In John Martin Fischer et al. Four Views on Free Will. Oxford,
UK: Blackwell Publishing, 2007.

_____. “New Arguments in Debates on Libertarian Free Will: Responses to

Contributors.” In Libertarian Free Will: Contemporary Debates. Edited by David
Palmer. New York: Oxford University Press, 2014.

Kapitan, T. “Providence, Foreknowledge, and Decisions Procedures.” Faith and

Philosophy 10, no. 3 (July 1993): 415-420. Accessed May 29, 2016. DOI:
10.5840/faithphil19991612.

Mares, Edwin and Ken Perszyk. “Molinist Counterfactuals.” In Molinism: The

Contemporary Debate. Edited by Kenneth Perszyk. Oxford: Oxford University
Press, 2011.

McKenna, Michael. “Compatibilist Ultimacy: Resisting the Threat of Kane’s U

Condition*.” In Libertarian Free Will: Contemporary Debates. Edited by David
Palmer. New York: Oxford University Press, 2014.

Molina, Luis de. On Divine Foreknowledge: Part IV of the Concordia. Edited and
translated by Alfred J. Freddoso. Ithaca: Cornell University Press, 1988.

Pike, Nelson. “Divine Omniscience and Voluntary Action.” The Philosophical Review
74, no. 1 (Jan. 1965): 27-46. Accessed January 21, 2017.
http://www.jstor.org/stable/2183529.
FOREKNOWLEDGE 67

Plantinga, Alvin. “Replies to My Colleagues.” In Alvin Plantinga. Edited by James

Tomberlin and Peter Van Inwagen. Dordrecht: D. Reidel, 1985.

Rice, Richard. “Biblical Support for a New Perspective.” In Pinnock et al. The Openness
of God: A Biblical Challenge to the Traditional Understanding of God. Downers
Grove, IL: Intervarsity Press, 1994.

Roy, S.C. How Much Does God Foreknow? A Comprehensive Biblical Study. Downers
Grove, IL: IVP Academic, 2006.

Ware, Bruce A. God's Lesser Glory: The Diminished God of Open Theism. Wheaton, IL:
Crossway Books, 2000.

Wierenga, E. R. The Nature of God: An Inquiry into Divine Attributes. Ithaca: Cornell
University Press, 1989.

Zagzebski, Linda. “Recent Work on Divine Foreknowledge and Free Will.” In The
Oxford Handbook of Free Will. 1st edition. Edited by Robert Kane. New York,
NY: Oxford University Press, 2002.