The Timetable and the Driver: Divine Foreknowledge and Libertarian Freedom in the Christian and Islamic Traditions

Presented by Claude for Zia H. Shah, MD

God’s infallible foreknowledge is compatible with libertarian human freedom, because knowing an event, even infallibly and from eternity, is not what causes it; the free agent’s own act is what makes the knowledge true.

The argument for theological fatalism depends on a hidden slide from “necessarily, if God knows p, then p” to “if God knows p, then p is necessary.” Boethius and Aquinas (eternity), Ockham (soft facts), the modern “dependence” response (Origen, Plantinga, Merricks) and Molina (middle knowledge) each block that slide at a different premise. Taken together they make a strong cumulative case for compatibility.
The Islamic tradition independently reached each of these moves. Averroes and al-Ghazali distinguish eternal from originated knowledge, as Boethius did. Ibn ‘Arabi’s dictum that “knowledge follows the known” anticipates the Ockhamist and dependence solutions. Maturidi teaching on created capacity (istita’a) and real choice (ikhtiyar) supports libertarian agency. The Quran itself states counterfactuals of freedom (6:28; 8:23), as Molinism requires.
Swiss trains are a good illustration. According to SBB’s January 2026 figures as reported by Le News, “94.1% of Swiss passenger trains running on the federal network arrived on time” in 2025, yet the timetable that “knew” this drove none of them. Divine foreknowledge is the perfect record of what free agents actually do, not the engine that makes them do it.

Abstract

This essay argues that human free will, understood in the strong libertarian sense, can coexist with God’s exhaustive and infallible foreknowledge. It frames the problem with Nelson Pike’s 1965 argument and Linda Zagzebski’s formulation of theological fatalism in the Stanford Encyclopedia of Philosophy. It then shows that the fatalist argument rests on a modal confusion and on contestable premises about the necessity of the past and the transfer of necessity. Four classical Western responses are reconstructed and assessed: Augustine’s and Boethius’s (and Aquinas’s) appeal to divine eternity, Ockham’s distinction between hard and soft facts, the modern “dependence” solution, and Luis de Molina’s doctrine of middle knowledge. The main objections are weighed: Zagzebski’s “timeless fatalism,” the grounding objection of Robert Adams and William Hasker, and the open-theist alternative. The essay then shows that the Islamic tradition anticipates, parallels, and sometimes improves on these solutions. Its sources include the Quran’s own pairing of a divine “Book” with insistent calls to moral striving, the hadith literature on qadar, the debates among the Qadariyya, Jabriyya, Mu’tazila, Ash’ariyya and Maturidiyya, Averroes’s distinction between eternal and originated knowledge, Ibn ‘Arabi’s principle that “knowledge follows the known,” and Muhammad Iqbal’s dynamic theism. A central section develops an extended analogy: Swiss trains arrive on time not because the timetable causes them to, but because drivers bring them in. Five further analogies are tested for their strengths and limits. The essay concludes that foreknowledge is best understood as the perfect record of freedom rather than its rival, and that the Preserved Tablet is a ledger of what free agents do, not a script that compels them.

1. Introduction: An Old Worry with a Sharp Edge

Every believer in an omniscient God eventually meets the same worry, usually late at night. If God already knows what I will choose tomorrow, and God cannot be wrong, then how can I choose otherwise? And if I cannot choose otherwise, in what sense is my choice mine? The worry is ancient. It troubled Augustine’s interlocutor Evodius, archive Substack Boethius in his prison cell in Pavia, and the early Muslim community, which divided over qadar within a century of the Prophet’s death. Nelson Pike gave it its modern edge in “Divine Omniscience and Voluntary Action” (Philosophical Review, 1965). As the Stanford Encyclopedia of Philosophy notes, Pike “gets the credit for clearly and forcefully presenting the dilemma in a way that produced an enormous body of work by both compatibilists and incompatibilists.” stanford

I write as a physician who has spent a professional lifetime watching people choose: to quit smoking or not, to take the medication or not, to forgive or not. Clinical medicine assumes, without ever stating it, that patients are agents. Our counsel, consent forms, and moral praise and blame all presuppose that the person before us could have done otherwise. I also write as a Muslim who reads the Quran’s declaration that “God does not change the condition of a people until they change what is in themselves” (13:11) as a charter of human responsibility. The Quran’s equally firm declaration that God holds “the keys of the unseen” (6:59) is not a repeal of that charter. My thesis is that the two declarations are compatible, and that both the Christian-philosophical and the Islamic traditions have given us the tools to show it.

The essay proceeds as follows. Section 2 states the fatalist argument precisely and identifies its weak joints. Sections 3 to 6 reconstruct the major Western responses: Augustine and Boethius, Aquinas, Ockham and the dependence solution, and Molinism. Section 7 weighs the objections and explains why I reject open theism. Sections 8 to 11 present the Islamic material and show its correspondences with the Western solutions. Section 12 is the special section on the Swiss trains and their companions. A thematic epilogue closes the essay.

2. The Argument for Theological Fatalism, and the Modal Fallacy at Its Heart

Linda Zagzebski’s entry “Foreknowledge and Free Will” in the Stanford Encyclopedia of Philosophy (first published 2004, substantively revised 2026) states the “Basic Argument for Theological Fatalism” with admirable clarity. Let T be the proposition that you will answer the telephone at 9 a.m. tomorrow:

Yesterday God infallibly believed T.
If E occurred in the past, it is now-necessary that E occurred then. (Principle of the Necessity of the Past)
So it is now-necessary that yesterday God believed T.
Necessarily, if yesterday God believed T, then T. (Definition of infallibility)
If p is now-necessary, and necessarily (p → q), then q is now-necessary. (Transfer of Necessity)
So it is now-necessary that T.
If it is now-necessary that T, you cannot do otherwise than answer.
So you cannot do otherwise.
If you cannot do otherwise, you do not act freely. (Principle of Alternate Possibilities)
So you do not answer freely.

The SEP observes that there is “a consensus that this argument or something close to it is valid,” and that compatibilists must therefore reject premise (1), (2), (5) or (9). The entry maps the classical positions onto these premises. Boethius and Aquinas deny (1) as stated, because God does not believe anything “yesterday.” Ockham denies (2) in its full generality. Molina (and perhaps Scotus) may deny (5). Augustine, on some readings, is best understood as denying (9).

Before turning to those answers, we should expose the cruder version of the argument that circulates in popular discussion, because it commits an outright fallacy. The popular version runs: “Necessarily, if God knows that I will do X, then I will do X; God knows that I will do X; therefore, necessarily, I will do X.” This trades on a scope ambiguity. The true premise is necessitas consequentiae, the necessity of the consequence: □(Kp → p). It is impossible for God to know p and for p to be false. The conclusion the fatalist wants is necessitas consequentis, the necessity of the consequent: Kp → □p. The first does not entail the second. “Necessarily, if I know you are sitting, you are sitting” is true. “If I know you are sitting, you are necessarily sitting” is false, since you could have stood. Aquinas made exactly this point when he wrote that the proposition “Everything known by God must necessarily be” must be distinguished. Taken de re, of the thing, it “is divided and false.” Taken de dicto, of the saying, it is true (Summa Theologiae I, q.14, a.13, ad 3).

The careful fatalist knows this, which is why Zagzebski’s argument does not commit the crude fallacy. It replaces simple necessity with temporal necessity, the “now-necessity” of the past, and adds a transfer principle. The real debate is therefore about premises (1), (2) and (5): whether God’s knowledge is “in the past,” whether everything in the past is fixed in the relevant sense, and whether fixity transfers across entailment. The great figures of both traditions have been working on exactly these joints.

3. Augustine and Boethius: Foreknowledge of a Will Is Foreknowledge of a Will

Augustine. In Book III of On Free Choice of the Will (De libero arbitrio), Evodius puts the problem plainly: he wonders “how God can have foreknowledge of everything in the future, and yet we do not sin by necessity.” Augustine’s answer has two parts. First, he observes that foreknowledge as such is not coercive. If you foreknew that someone would sin, “your foreknowledge would not force him to sin,” and in the same way that memory of past events does not compel them to have happened, God’s foreknowledge does not compel future events. Second, and more profoundly, he argues that what God foreknows is our will, and a will is by definition something in our power: “our will would not be a will if it were not in our power. And since it is in our power, we are free with respect to it.” Hence: “Since God foreknows our will, the very will that he foreknows will exist,” and God “foreknows our sins in such a way that our wills remain free and are within our power.”

This is a deceptively simple move. God’s foreknowledge is accurate, so it must represent its object as it actually is. If the object is a free volition, then God foreknows a free volition, and accurate foreknowledge cannot turn the free into the unfree without ceasing to be accurate. Augustine repeats the argument in City of God V.9–10 against Cicero, who had denied foreknowledge to save freedom. Augustine refuses the trade: “we embrace both.” Our wills are among the causes God foreknows, and “our wills also have just so much power as God willed and foreknew that they should have.”

Boethius. The most influential solution was written by a condemned man. In Book V of The Consolation of Philosophy (c. 524), Lady Philosophy first sets out the dilemma with full force (V.3). She then lays down an epistemological principle: “everything which is known is known not according to its own nature but according to the nature of those comprehending it” (V.6). The kind of knowing depends on the kind of knower. God’s nature is eternal, so we must ask what eternity is. Her answer became the standard definition for the next millennium:

“Eternity, then, is the complete, simultaneous and perfect possession of everlasting life” (interminabilis vitae tota simul et perfecta possessio), V.6, trans. V. E. Watts.

A temporal being, even one that never began and never ends, “proceeds in the present from the past into the future.” It has already lost yesterday and does not yet possess tomorrow. God, by contrast, possesses the whole of His life at once. It follows that God does not, strictly speaking, fore-know anything. His knowledge is better called providentia, a looking-forth from above, than praevidentia, a seeing-before. All events, past, present and future, are present to Him in a single eternal “now.”

Boethius then gives his famous illustration. When we see at the same moment a man walking on the ground and the sun rising in the sky, our seeing imposes no necessity on either. The sun rises by natural necessity and the man walks by his own will, but our simultaneous seeing of both changes neither. “So also the divine gaze looking down on all things does not disturb the quality of things which are present to it at its own level, but future with respect to time.” This is where Boethius introduces his distinction between two necessities. Simple necessity belongs to things by their nature, as all men are mortal. Conditional necessity is the necessity that “if you know that someone is walking, he must be walking.” The second is no constraint on the walker. A free act seen by God is conditionally necessary, given that God sees it, and simply contingent in itself.

What Boethius achieves is to remove the “fore” from foreknowledge. If God has no “yesterday,” premise (1) of the fatalist argument is false as stated, and premise (2), the necessity of the past, never gets a grip.

4. Aquinas: The Watcher on the Height and the Cause of Causes

Thomas Aquinas adopted and sharpened the Boethian view. In Summa Theologiae I, q.14, a.13, he asks whether God’s knowledge is of future contingent things and answers that it is. He explains how in a sentence that has become classic:

“All things that are in time are present to God from eternity, not only because He has the types of things present within Him, as some say; but because His glance is carried from eternity over all things as they are in their presentiality. Hence it is manifest that contingent things are infallibly known by God, inasmuch as they are subject to the divine sight in their presentiality; yet they are future contingent things in relation to their own causes.”

The last clause is the key. An event can be present to God and still future and contingent relative to its proximate causes in time. Aquinas then offers his image: “he who goes along the road, does not see those who come after him; whereas he who sees the whole road from a height, sees at once all travelling by the way.” The traveller on the road cannot see who is coming round the bend. The watcher on the height sees them all, and his seeing does not push any of them along. In the Summa contra Gentiles (I.66) he adds the image of the centre of a circle, equally present to every point on the circumference.

Aquinas adds something Boethius did not develop: a metaphysics of primary and secondary causation. God is the first cause of all being, including the being of free acts. But God causes according to the mode of the secondary cause. He causes necessary things necessarily and contingent things contingently, and He moves the will as a will, that is, voluntarily (ST I, q.19, a.8; I, q.83, a.1, ad 3). Divine causality is not one more push among created pushes. It is the giving of being to the whole order of causes, including the self-determining causes we call persons. On this view human beings are not puppets, because the puppeteer’s strings would be a created cause competing with the will. God is not a competitor on that level at all.

Libertarians, myself included, must handle this Thomist doctrine with care. Read in a strongly “premotion” sense, as the later Báñezian Thomists read it, it threatens to make God the sufficient determiner of every choice. That reading is precisely what Molina rose to resist. But the epistemic part of Aquinas, the claim that God knows contingents as present without making them necessary, stands independently, and it is that part I endorse.

5. Ockham, Soft Facts, and the Dependence Solution

William of Ockham (c. 1287–1347) was dissatisfied with the eternity solution, partly because he thought Scripture and ordinary speech require that God knew things before they happened. In his Tractatus de praedestinatione et de praescientia Dei respectu futurorum contingentium (translated by Marilyn McCord Adams and Norman Kretzmann as Predestination, God’s Foreknowledge, and Future Contingents, 1969), he attacks premise (2) of the argument instead. stanford

Ockham grants that the past has a special necessity, which he calls necessitas per accidens or accidental necessity. But he denies that every truth expressed in the past tense has it. Some propositions are “about the past” in their wording but “about the future” in their substance. “It was true yesterday that Socrates would sit today” is past-tensed in form, but its truth depends on what Socrates does today. Contemporary philosophers call these soft facts about the past, as distinct from hard facts such as “Socrates sat yesterday,” which are fully fixed. God’s past belief that you will answer the phone tomorrow, Ockham argues, is a soft fact. It is “about” tomorrow, and tomorrow is still in your hands.

Marilyn McCord Adams revived this approach in 1967 (“Is the Existence of God a ‘Hard’ Fact?”). As the SEP candidly notes, her first criterion proved too generous: it made even “Jack sat yesterday” soft, because it entails that Jack will not sit tomorrow for the first time. A decade of refinements followed, from Freddoso, Hoffman and Rosenkrantz, Kvanvig, Wierenga, Craig and others. Alvin Plantinga’s “On Ockham’s Way Out” (Faith and Philosophy, 1986) cut through the thicket. stanford We need not define softness in general, he argued; paradigm cases suffice. T itself, “you will answer the phone tomorrow,” is plainly not a hard fact about the past. If God is essentially omniscient and necessarily existent, T and “God believed T yesterday” entail each other. Assuming hardness is closed under logical equivalence, God’s past belief cannot be hard either.

Plantinga then gave the view its most useful form by defining accidental necessity in terms of counterfactual power. The relevant claim is not that I can change the past, which is incoherent. It is this: “It was within Jones’ power at t2 to do something such that if he did it, God would not have held the belief he in fact held at t1.” There is one actual past, but had Jones chosen differently, God would always have believed differently. And this counterfactual dependence “need not be a causal dependency,” as the SEP stresses. Jones does not reach back in time and push God’s mind about. God’s mind simply tracks Jones.

That brings us to what the SEP now treats as a distinct position, the dependence solution, associated recently with Trenton Merricks (“Truth and Freedom,” 2009). Its ancient source is Origen of Alexandria, who wrote in his Commentary on Romans: “it will not be because God knows that an event will occur that it happens; but, because something is going to take place it is known by God before it happens.” stanford The order of explanation runs from the event to the knowledge, not from the knowledge to the event. God believes T because T. T is not true because God believes it. The fatalist’s argument treats God’s belief as a fixed fact from which my act must be derived. The dependence solution reverses the arrow and insists that my act is the ground and God’s belief the grounded.

The SEP is fair about the costs. The deepest worry is that God’s past beliefs “seem to be as good a candidate for something that is strictly past as almost anything we can think of,” and that Ockhamism can look ad hoc. I think the worry is weaker than it appears, for a reason the SEP itself supplies: whether premise (2) applies may depend on “special features of the divine mind.” That is not special pleading. A belief that is infallible by nature is precisely one whose content is fixed by its object, which makes it the paradigm of a dependent fact. We shall see in Section 10 that the Islamic tradition said exactly this eight centuries ago.

6. Molinism: Middle Knowledge and Providence Without Puppetry

The Jesuit Luis de Molina (1535–1600) wanted more than compatibility. He wanted providence, a God who does not merely watch free creatures but governs the world through them without violating them. His Concordia liberi arbitrii cum gratiae donis, divina praescientia, providentia, praedestinatione et reprobatione (Lisbon, 1588), “The Harmony of Free Choice with the Gifts of Grace, Divine Foreknowledge, Providence, Predestination and Reprobation,” proposed that God’s knowledge has three logical, not temporal, “moments.”

Natural knowledge (scientia naturalis): God’s knowledge of all necessary truths and all possibilities, including every possible creature in every possible circumstance and everything it could do. This knowledge is prior to God’s will and independent of it.
Middle knowledge (scientia media): God’s knowledge of the counterfactuals of creaturely freedom, that is, what any possible free creature would freely do in any possible set of circumstances. These truths are contingent, because the creature could have chosen otherwise, so they are not part of natural knowledge. Yet they are prior to God’s creative decree and not up to God, so they are not part of free knowledge either. Hence “middle.”
Free knowledge (scientia libera): after God decrees to actualize one of the “feasible” worlds that middle knowledge discloses, He knows exhaustively everything that will actually happen.

Providence works through the second moment. Knowing what each free creature would do in each circumstance, God can choose which circumstances to actualize, and thereby bring about His purposes through free choices He neither causes nor compels. The biblical warrant usually cited includes David at Keilah (1 Samuel 23:10–13), where God tells David what the men of Keilah would do if he stayed, and Jesus’s remark that Tyre and Sidon would have repented had they seen His miracles (Matthew 11:21).

Modern philosophy of religion revived Molinism in force. Alvin Plantinga’s The Nature of Necessity (1974) and his Free Will Defense rely on counterfactuals of freedom: it is possible that every feasible world containing free creatures contains some moral evil, a condition Plantinga called “transworld depravity.” William Lane Craig’s The Only Wise God (1987) presented middle knowledge as the reconciliation of foreknowledge and freedom for a wide readership. Thomas P. Flint’s Divine Providence: The Molinist Account (1998) is the most rigorous defence. Kenneth Keathley’s Salvation and Sovereignty: A Molinist Approach (2010) applies the model to soteriology. google Building on Timothy George, it replaces the Calvinist TULIP with ROSES: Radical depravity, Overcoming grace, Sovereign election, Eternal life, Singular redemption. Keathley calls Molinism “the only game in town for anyone who wishes to affirm a high view of God’s sovereignty while holding to a genuine definition of human choice, freedom, and responsibility.” thegospelcoalition

The grounding objection. The best-known critique was pressed by Robert Merrihew Adams (“Middle Knowledge and the Problem of Evil,” 1977) and developed by William Hasker in God, Time, and Knowledge (1989). What makes it true, before any creature exists, that Peter would deny Christ in circumstances C? Not Peter’s actual choice, which has not occurred and in many worlds never does. Not Peter’s character, since that would determine him and defeat libertarian freedom. Not God’s will, since that would make the counterfactual God’s decree. If nothing grounds these counterfactuals, Adams and Hasker argue, there is nothing for God to know. Hasker adds an argument that if the counterfactuals are prior to and independent of my choice, I cannot bring them about, and so I am not free with respect to the choices they entail.

Molinists reply in two ways. Some deny that counterfactuals need truthmakers beyond themselves, just as many deny that past-tense truths need presently existing truthmakers. Others ground them in the creatures’ own possible free self-determination. My own view is that the grounding objection is real but not decisive, and that it is less a problem for the compatibility thesis than for Molina’s ambition to explain providence. Even if middle knowledge failed, Boethian eternity and the dependence solution would still secure compatibility. Molinism is an optional, though attractive, upper storey on a sound ground floor.

7. Objections Weighed, and Why Open Theism Is Not the Answer

Zagzebski’s “timeless fatalism.” The strongest objection to the Boethian solution comes from a philosopher sympathetic to it. Zagzebski argues (1991, ch. 2; and in the SEP) that the fatalist argument can be rebuilt with “the necessity of the timeless realm” in place of the necessity of the past: (1t) God timelessly knows T; (2t) if E is in the timeless realm, it is now-necessary that E; (3t) so it is now-necessary that T. “The timeless realm is as much out of our reach as the past.” Eternity, on this view, only relocates the problem.

Defenders have three replies, and I find the combination persuasive. First, as Zagzebski herself concedes, “the intuition of the necessity of the timeless realm is probably weaker than the intuition of the necessity of the past.” We have robust intuitions about spilled milk and none about eternity. Second, Katherin Rogers argues that the eternal “now” is more like the present than the past. What is present is not fixed against the agent who is now acting; it is the very arena of her acting. If my free act is present to God as I perform it, God’s seeing it is no more a constraint than a spectator’s seeing a runner is a constraint on the runner. Third, the Boethian view combines naturally with the dependence solution. Even a timeless knowing is a knowing of its object, and its content is grounded in what the object is. “Out of reach” in the sense of unalterable is not the same as “independent of me” in the sense of ungrounded by my act.

Prophetic objects. Peter van Inwagen worries that a timeless God might inscribe a prophecy on a stone tablet in time, a “Freedom-denying Prophetic Object.” This is a serious point for both traditions, since the Quran speaks of a divine Book. I return to it in Section 10. My answer is that a revealed prophecy of a free act is itself a soft fact, grounded in the act it reports.

Open theism. A family of thinkers, including Richard Swinburne and William Hasker in his “limited foreknowledge” phase, along with open theists such as Gregory Boyd and Alan Rhoda, concludes that God does not infallibly know future free acts, because either there are no truths to know or they cannot be known. stanford I reject this view, for three reasons. Logically, as the SEP observes, denying future contingent truth has costs: it threatens bivalence and leaves puzzles about the asymmetry between past and future truths. Religiously, it shrinks God’s knowledge to fit our theory, and no reading of “with Him are the keys of the unseen; none knows them except Him” (6:59) makes room for a God surprised by tomorrow. Dialectically, it is unnecessary. If foreknowledge does not entail necessity, there is no need to amputate foreknowledge to save freedom. Open theism concedes the fatalist’s inference, and that inference is the very thing in question.

8. The Quranic Data: A Book, a Measure, and a Charter of Responsibility

The Quran does not present divine knowledge and human freedom as rival claims to be balanced. It states both with full force, often within a few verses of each other.

Divine knowledge and the Book.

وَعِندَهُۥ مَفَاتِحُ ٱلْغَيْبِ لَا يَعْلَمُهَآ إِلَّا هُوَ ۚ وَيَعْلَمُ مَا فِى ٱلْبَرِّ وَٱلْبَحْرِ ۚ وَمَا تَسْقُطُ مِن وَرَقَةٍ إِلَّا يَعْلَمُهَا وَلَا حَبَّةٍ فِى ظُلُمَـٰتِ ٱلْأَرْضِ وَلَا رَطْبٍ وَلَا يَابِسٍ إِلَّا فِى كِتَـٰبٍ مُّبِينٍ

“And with Him are the keys of the unseen; none knows them except Him. And He knows what is on the land and in the sea. Not a leaf falls but that He knows it. And no grain is there within the darknesses of the earth and no moist or dry [thing] but that it is [written] in a clear record.” (Surah al-An’am 6:59, Sahih International)

مَآ أَصَابَ مِن مُّصِيبَةٍ فِى ٱلْأَرْضِ وَلَا فِىٓ أَنفُسِكُمْ إِلَّا فِى كِتَـٰبٍ مِّن قَبْلِ أَن نَّبْرَأَهَآ ۚ إِنَّ ذَٰلِكَ عَلَى ٱللَّهِ يَسِيرٌ

“No misfortune can happen, either in the earth or in yourselves, that was not set down in writing before We brought it into being—that is easy for God.” (Surah al-Hadid 57:22, Abdel Haleem)

إِنَّا كُلَّ شَىْءٍ خَلَقْنَـٰهُ بِقَدَرٍ

“Indeed, all things We created with predestination [bi-qadar, by measure].” (Surah al-Qamar 54:49, Sahih International)

قُل لَّن يُصِيبَنَآ إِلَّا مَا كَتَبَ ٱللَّهُ لَنَا هُوَ مَوْلَىٰنَا ۚ وَعَلَى ٱللَّهِ فَلْيَتَوَكَّلِ ٱلْمُؤْمِنُونَ

“Say, ‘Never will we be struck except by what Allah has decreed for us; He is our protector.’ And upon Allah let the believers rely.” (Surah al-Tawbah 9:51, Sahih International)

Notice what 57:22 and 9:51 are about: misfortunes, what “befalls” or “strikes” us. They are verses of consolation, meant, as 57:23 continues, “so you need not grieve for what you miss or gloat over what you gain.” They are not verses about the authorship of moral choices.

Human choice and responsibility.

فَمَن شَآءَ فَلْيُؤْمِن وَمَن شَآءَ فَلْيَكْفُرْ

“…so whoever wills—let him believe; and whoever wills—let him disbelieve.” (Surah al-Kahf 18:29, Sahih International)

إِنَّا هَدَيْنَـٰهُ ٱلسَّبِيلَ إِمَّا شَاكِرًا وَإِمَّا كَفُورًا

“Indeed, We guided him to the way, be he grateful or be he ungrateful.” (Surah al-Insan 76:3, Sahih International)

وَنَفْسٍ وَمَا سَوَّىٰهَا ۝ فَأَلْهَمَهَا فُجُورَهَا وَتَقْوَىٰهَا ۝ قَدْ أَفْلَحَ مَن زَكَّىٰهَا ۝ وَقَدْ خَابَ مَن دَسَّىٰهَا

“And [by] the soul and He who proportioned it, and inspired it [with discernment of] its wickedness and its righteousness, he has succeeded who purifies it, and he has failed who instills it [with corruption].” (Surah al-Shams 91:7–10, Sahih International)

إِنَّ ٱللَّهَ لَا يُغَيِّرُ مَا بِقَوْمٍ حَتَّىٰ يُغَيِّرُوا۟ مَا بِأَنفُسِهِمْ

“Indeed, Allah will not change the condition of a people until they change what is in themselves.” (Surah al-Ra’d 13:11, Sahih International)

لَا يُكَلِّفُ ٱللَّهُ نَفْسًا إِلَّا وُسْعَهَا ۚ لَهَا مَا كَسَبَتْ وَعَلَيْهَا مَا ٱكْتَسَبَتْ

“Allah does not charge a soul except [with that within] its capacity. It will have [the consequence of] what [good] it has gained, and it will bear [the consequence of] what [evil] it has earned.” (Surah al-Baqarah 2:286, Sahih International)

أَلَّا تَزِرُ وَازِرَةٌ وِزْرَ أُخْرَىٰ ۝ وَأَن لَّيْسَ لِلْإِنسَـٰنِ إِلَّا مَا سَعَىٰ

“That no bearer of burdens will bear the burden of another, and that there is not for man except that [good] for which he strives.” (Surah al-Najm 53:38–39, Sahih International)

مَّنْ عَمِلَ صَـٰلِحًا فَلِنَفْسِهِۦ ۖ وَمَنْ أَسَآءَ فَعَلَيْهَا ۗ وَمَا رَبُّكَ بِظَلَّـٰمٍ لِّلْعَبِيدِ

“Whoever does righteousness—it is for his [own] soul; and whoever does evil [does so] against it. And your Lord is not ever unjust to [His] servants.” (Surah Fussilat 41:46, Sahih International)

فَمَن يَعْمَلْ مِثْقَالَ ذَرَّةٍ خَيْرًا يَرَهُۥ ۝ وَمَن يَعْمَلْ مِثْقَالَ ذَرَّةٍ شَرًّا يَرَهُۥ

“So whoever does an atom’s weight of good will see it, and whoever does an atom’s weight of evil will see it.” (Surah al-Zalzalah 99:7–8, Sahih International)

مَّآ أَصَابَكَ مِنْ حَسَنَةٍ فَمِنَ ٱللَّهِ ۖ وَمَآ أَصَابَكَ مِن سَيِّئَةٍ فَمِن نَّفْسِكَ

“What comes to you of good is from Allah, but what comes to you of evil, [O man], is from yourself.” (Surah al-Nisa’ 4:79, Sahih International)

Two features of this cluster matter for the argument. The vocabulary of kasb and iktisab, earning and acquiring (2:286), is the Quran’s own, and it later became the technical term of Ash’ari theology. And the verse “your Lord is not ever unjust to [His] servants” (41:46) ties responsibility to justice. God’s names al-‘Adl (the Just) and al-Hakam (the Judge) would be hollow if the judged had no real say in what they did. Reward and punishment presuppose real choice. A God who scripted the sin and then punished the sinner would be zallam, unjust, which is exactly what 41:46 denies.

Divine transcendence of time.

هُوَ ٱلْأَوَّلُ وَٱلْـَٔاخِرُ وَٱلظَّـٰهِرُ وَٱلْبَاطِنُ ۖ وَهُوَ بِكُلِّ شَىْءٍ عَلِيمٌ

“He is the First and the Last, the Ascendant and the Intimate, and He is, of all things, Knowing.” (Surah al-Hadid 57:3, Sahih International)

وَإِنَّ يَوْمًا عِندَ رَبِّكَ كَأَلْفِ سَنَةٍ مِّمَّا تَعُدُّونَ

“And indeed, a day with your Lord is like a thousand years of those which you count.” (Surah al-Hajj 22:47, Sahih International)

تَعْرُجُ ٱلْمَلَـٰٓئِكَةُ وَٱلرُّوحُ إِلَيْهِ فِى يَوْمٍ كَانَ مِقْدَارُهُۥ خَمْسِينَ أَلْفَ سَنَةٍ

“The angels and the Spirit will ascend to Him during a Day the extent of which is fifty thousand years.” (Surah al-Ma’arij 70:4, Sahih International)

ٱللَّهُ لَآ إِلَـٰهَ إِلَّا هُوَ ٱلْحَىُّ ٱلْقَيُّومُ ۚ لَا تَأْخُذُهُۥ سِنَةٌ وَلَا نَوْمٌ … يَعْلَمُ مَا بَيْنَ أَيْدِيهِمْ وَمَا خَلْفَهُمْ ۖ وَلَا يُحِيطُونَ بِشَىْءٍ مِّنْ عِلْمِهِۦٓ إِلَّا بِمَا شَآءَ

“Allah—there is no deity except Him, the Ever-Living, the Sustainer of [all] existence. Neither drowsiness overtakes Him nor sleep… He knows what is [presently] before them and what will be after them, and they encompass not a thing of His knowledge except for what He wills.” (Surah al-Baqarah 2:255, Ayat al-Kursi, Sahih International)

The variable ratio between “a day with your Lord” and human years (one thousand in 22:47, fifty thousand in 70:4) is a pointed way of saying that divine time is not measured by our clocks. Read with 57:3, where God is at once “the First and the Last,” the verses come close to Boethius’s tota simul. God is not further along our timeline than we are; He embraces both its ends. Ayat al-Kursi adds that God knows “what is before them and what is behind them,” the same image of the watcher on the height that Aquinas would use.

Quranic counterfactuals of freedom. Most strikingly for the Molinist comparison, the Quran states what free agents would do in circumstances that never obtain:

بَلْ بَدَا لَهُم مَّا كَانُوا۟ يُخْفُونَ مِن قَبْلُ ۖ وَلَوْ رُدُّوا۟ لَعَادُوا۟ لِمَا نُهُوا۟ عَنْهُ وَإِنَّهُمْ لَكَـٰذِبُونَ

“But what they concealed before has [now] appeared to them. And even if they were returned, they would return to that which they were forbidden; and indeed, they are liars.” (Surah al-An’am 6:28, Sahih International)

وَلَوْ عَلِمَ ٱللَّهُ فِيهِمْ خَيْرًا لَّأَسْمَعَهُمْ ۖ وَلَوْ أَسْمَعَهُمْ لَتَوَلَّوا۟ وَّهُم مُّعْرِضُونَ

“Had Allah known any good in them, He would have made them hear. And if He had made them hear, they would [still] have turned away, while they were refusing.” (Surah al-Anfal 8:23, Sahih International)

These are exactly what Molina called counterfactuals of creaturely freedom: “if they were returned [a circumstance that will not obtain], they would [freely] return to what was forbidden.” The classical exegetes saw their hypothetical character. Al-Tabari’s report from Ibn Zayd glosses 8:23 as “if He made them hear, after knowing that there is no good in them, they would not benefit,” and later commentators describe the clause as stated “by way of supposition and hypothesis” (‘ala sabil al-fard wa-l-taqdir). The Quran thus attributes to God knowledge of what free creatures would do in non-actual circumstances. That is the datum Molinism theorizes, centuries before Molina. The Quran does not supply Molina’s theory of logical priority to the creative decree. But it does settle the question the grounding objection raises against Molinists: whether there are such truths for God to know. For a Muslim, there are, because God states some of them.

9. The Hadith on Qadar: Belief in the Decree as a Spur to Action

The qadar tradition in hadith is often read as fatalist. Read carefully, it consistently refuses the fatalist inference.

The Hadith of Jibril. In the famous narration recorded by Muslim (Sahih Muslim, Kitab al-Iman, no. 8), ‘Umar ibn al-Khattab reports that a stranger, later identified as the angel Jibril, asked the Prophet about faith. The Prophet answered with six articles, the last being “that you affirm your faith… in the Divine Decree about good and evil” (wa tu’mina bi-l-qadari khayrihi wa sharrihi). The chain itself is instructive. ‘Umar’s son ‘Abdullah narrated it to rebut Ma’bad al-Juhani of Basra, who was reportedly “the first man who discussed qadr” by denying it. Belief in qadar is an article of faith, but what it affirms is God’s comprehensive knowledge and measuring. It does not affirm the unreality of human agency.

“Work, for everyone is eased.” In Sahih al-Bukhari (no. 4949; cf. 6596, 7551) and Muslim, ‘Ali reports that the Prophet, at a funeral, scratched the ground with a stick and said that “there is none of you but has his place written for him, either in Paradise or in the Hell-Fire.” The companions asked the fatalist’s question exactly: a-fa-la nattakil? “Shall we not depend on this and give up work?” The Prophet answered: “No! I’malu fa-kullun muyassarun li-ma khuliqa lahu. Work, for everyone is eased toward what he was created for.” He then recited 92:5–10: “As for him who gives and fears God and affirms the best, We shall ease him toward ease…”

The structure of the reply is decisive. The Prophet does not deny the record. He denies the inference from the record to passivity. The record is fulfilled through the work, not in spite of it, and the Quranic verses he recites make the giving and the God-consciousness the human side of the “easing.” The fatalist treats the Book as a cause that makes striving pointless. The Prophet treats striving as the very content the Book records.

“We flee from the decree of God to the decree of God.” In Sahih al-Bukhari (no. 5729) and Sahih Muslim (no. 2219), as also in Malik’s Muwatta’, ‘Umar sets out for Syria and learns at Sargh of an outbreak of plague. After consultation he decides to turn back. Abu ‘Ubayda objects: “Is it flight from the decree of Allah?” ‘Umar replies: “Better that someone other than you had said it, Abu ‘Ubayda. Yes. We flee from the decree of Allah to the decree of Allah (nafirru min qadari-llahi ila qadari-llah).” He then gives the analogy of camels in a valley with a fertile slope and a barren one: whichever you graze them on, you graze them by the decree of God. For ‘Umar, qadar is not one path that cancels choice. It is the whole valley, every slope of which is within God’s knowledge and measure, and the shepherd must still choose where to lead his flock. As a physician I find this the most modern text in the tradition. It is a public-health decision, made by deliberation, justified by a revealed principle of quarantine, and explicitly declared to be within the decree rather than against it.

“You stole by God’s decree; we cut your hand by God’s decree.” A widely circulated report has a thief brought before ‘Umar plead that God had decreed his theft, to which ‘Umar replied: “You stole by the decree of God, and we cut your hand by the decree of God.” I must flag that I have not been able to locate this report in the canonical collections. The hadith-verification service HadithAnswers.com reached the same result in February 2020: “I have not come across this statement of Sayyiduna ‘Umar in any primary Hadith source. Certain non primary sources like the books on Gharibul Hadith have cited this … without a chain. I could therefore not verify the authenticity.” (Its Arabic reference may concern a different saying of ‘Umar about thieves.) So it should be treated as an edifying athar of uncertain chain, not as established history. Its logic is nonetheless impeccable, and it matches the plague narration: if the decree excuses the crime, it equally “excuses” the punishment. The fatalist excuse refutes itself, because it makes moral discourse, including the thief’s own plea, unintelligible.

10. The Theological Schools: From Qadariyya and Jabriyya to Kasb and Istita’a

Qadariyya and Jabriyya. The earliest polemic set the Qadariyya, those who affirmed human power (qadar in the sense of capacity) over their acts, often to the point of denying God’s prior decree, against the Jabriyya, the compulsionists associated with Jahm ibn Safwan, for whom human acts are attributed to us only metaphorically, as motion is attributed to a falling stone. Mainstream Sunni theology defined itself against both, just as Western orthodoxy rejected both Pelagianism and fatalism.

The Mu’tazila. The Mu’tazila made ‘adl, divine justice, one of their five principles and drew the conclusion that human beings create or originate (yakhluqun, yuhdithun) their own acts by a power God gives them. Otherwise, they argued, God would be the author of the very evils He punishes. This is a recognisably libertarian position: real power to do or not do, located in the agent. Its weakness, in the eyes of its critics, was that it seemed to set up a second creator beside God. Its strength was moral. It took 41:46, “your Lord is not ever unjust to His servants,” with full seriousness.

The Ash’ari doctrine of kasb. Abu al-Hasan al-Ash’ari (d. 935/6), a former Mu’tazili, sought a middle way using the Quran’s own word kasb, “acquisition.” ibnarabisociety +2 God creates (khalq) every act, including human acts, but the human being acquires it. The usual account is that God creates in the agent, simultaneously with the act, a temporary created power (qudra hadītha) that has no causal efficacy, and the act’s being “accompanied” by this power and by the agent’s will is what makes it his for moral purposes. Al-Baqillani tried to give the created power some role in determining the act’s moral mode (its being obedience or disobedience). Al-Juwayni, in his late al-‘Aqida al-Nizamiyya, moved further toward granting the created power real effect. Al-Ghazali, in al-Iqtisad fi al-I’tiqad and the Ihya’, defended kasb as the mean between jabr and Mu’tazili “creation.”

In fairness I must say plainly what Ash’ari critics, from the Maturidis to Ibn Taymiyya, have said: kasb in its classical form is very hard to distinguish from determinism. If God alone creates both the act and the power, and the power is causally idle, then “acquisition” names a correlation, not an agency. The History of Muslim Philosophy edited by M. M. Sharif summarizes al-Ash’ari’s position as one on which acquisition “means only a general coincidence of the divine power and human actions,” and remarks that al-Ash’ari “did not favour the idea of the freedom of choice.” ilmgate al-islam The Ash’aris were sometimes ironically called “the middle jabr” (al-jabr al-mutawassit). ibnarabisociety My own sympathy for al-Ghazali, whose occasionalism I have studied elsewhere, does not extend to pretending this difficulty away. Occasionalism can coexist with libertarian freedom only if the choice (the volition) is genuinely the agent’s, even if the physical execution is God’s customary concomitant. That is a modification the later Ash’ari tradition gestured toward but did not fully make.

The Maturidi position. Abu Mansur al-Maturidi of Samarqand (d. 944), a contemporary of al-Ash’ari, offers what I consider the most promising classical Sunni foundation for libertarian agency. encyclopedia Wikipedia In his Kitab al-Tawhid he holds that God creates human acts according to the human being’s own choice. The human contribution is real choice (ikhtiyar) and acquisition (iktisab), made possible by a God-given capacity (istita’a) that is “valid for opposite acts,” that is, a two-way power to do right or wrong. According to the Sharif volume, al-Maturidi grounds human obligation (taklif) “not in man’s possessing the power to create an action, but [in] the freedom to choose (ikhtiyar) and the freedom to acquire an action (iktisab).” He also appeals to the consciousness of freedom, holding that its denial “will lead to the denial of all human knowledge.” Most important for our purposes, al-Maturidi holds that the eternal divine decree “is not inconsistent with human freedom, nor does it imply any compulsion on the part of man, because it is an eternal record based on foreknowledge.”

That last formula is the Islamic statement of the dependence solution in five words: an eternal record based on foreknowledge. The decree, as it bears on human acts, is God’s knowing-and-recording of what we will freely choose, not a script that produces the choice. A two-way capacity that is God-given but genuinely the agent’s is, in contemporary terms, libertarian agency within a theistic metaphysics of dependence.

Ibn Taymiyya and Ibn al-Qayyim. The Hanbali reformers rejected both Mu’tazili “self-creation” and Ash’ari occasionalism. Ibn Taymiyya (d. 1328) set out the “levels” (maratib) of qadar: God’s knowledge, His writing, His will, and His creation. He insisted, against the Ash’aris, that created causes, including the human will, have real efficacy that God places in them. His student Ibn Qayyim al-Jawziyya devoted a large treatise, Shifa’ al-‘Alil fi Masa’il al-Qada’ wa-l-Qadar wa-l-Hikma wa-l-Ta’lil, to showing that the decree and real human causation, and indeed divine wisdom and purposiveness, are compatible. Their position is not libertarian in the Mu’tazili sense, since God’s will remains the sufficient ground of all that occurs. But their refusal to make created causes idle is a significant ally against fatalism.

Ibn Sina, al-Ghazali and Averroes on divine knowledge. The philosophers approached the question from the side of knowledge rather than power. Avicenna (Ibn Sina) held that God, being immutable, knows particulars only “in a universal way” (‘ala nahw kulli). He knows the eclipse as an item in the necessary causal order, not as happening now, since knowledge of the “now” would introduce change into God. Al-Ghazali, in the thirteenth discussion of Tahafut al-Falasifa, made this one of three counts on which he charged the philosophers with unbelief, because it seemed to deny that “not an atom’s weight… escapes Him” (34:3). His counter-argument is the important one for us. Change in the object does not require change in the knowledge. God’s single, eternal knowledge of an eclipse encompasses its being future, present and past, and what changes is only the relation (idafa) between the knowledge and the temporally located event. This is very close to Boethius: one eternal act of knowing, many temporal relations to it.

Averroes (Ibn Rushd, d. 1198) answered al-Ghazali in the Tahafut al-Tahafut, the Fasl al-Maqal (1179) and its appendix on divine knowledge (the Damima), and the Kashf ‘an Manahij al-Adilla (1179). His decisive move is that divine knowledge and human knowledge share only a name. In the Damima he writes that “if the name of ‘knowledge’ is predicated of both originated and eternal knowledge, it is predicated by sheer homonymy.” Eternal knowledge does not undergo the change that originated knowledge does when its object comes into being. “The mistake in this matter has arisen simply from making an analogy between the eternal Knowledge and originated knowledge.” God does not know beings “with originated knowledge but with eternal Knowledge.” In the Kashf, Averroes presents the Quranic teaching on qada’ wa qadar as a middle path between those who make human beings wholly compelled and those who make them independent creators. Acts are ours, but they are possible only through causes outside us that God has ordered.

The parallel with Boethius and Aquinas is close: knowledge takes its mode from the knower, and eternal knowledge does not stand in a “before” relation to its objects. But honesty requires a caution. Averroes also says, in the same text, that “the eternal Knowledge is the cause of beings, not their effect as originated knowledge is.” Some scholars, notably Catarina Belo, therefore read his theory of qadar as “both deterministic and predestinarian.” The libertarian can accept Averroes’s eternity point without his causal claim, or can read “cause” as God’s creative knowledge of the natures and capacities of things, including the two-way capacity of rational agents, rather than of each particular choice. On either reading, Averroes’s main contribution, the refusal to model God’s knowledge on ours, is exactly what defeats premise (1) of the fatalist argument.

11. Ibn ‘Arabi: “Knowledge Follows the Known”

The most striking Islamic anticipation of the Ockhamist and dependence solutions comes from Muhyi al-Din Ibn ‘Arabi (1165–1240). His treatment of the “mystery of destiny” (sirr al-qadar) rests, as Bakri Aladdin shows in a study for the Muhyiddin Ibn Arabi Society, on two linked ideas: “the notion that ‘Knowledge is subject to the known’, and… the concept of the ‘established entities’ (al-a’yan al-thabita).” ibnarabisociety

The a’yan thabita are the “fixed” or “established” entities: the realities of all possible things as they are eternally known to God, “whether or not they exist in the cosmos.” Ibn ‘Arabi distinguishes their “thingness of fixity” (shay’iyyat al-thubut), their being as known, from their “thingness of existence,” which they receive when God says “Be!” Each fixed entity, in Aladdin’s paraphrase, “includes its own independent personality and its own specific destiny.” The crucial principle, which Ibn ‘Arabi claims with characteristic pride to have discovered, is that divine knowledge follows these realities:

“Knowledge is subject to the known, and what is known is subject to knowledge, so understand! This is a great and subtle matter, which to my knowledge no one has mentioned before, or if they have, it has not come to my attention.” (al-Futuhat al-Makkiyya, as quoted by Aladdin)

Aladdin notes that the principle that “knowledge is subject to the known” (al-‘ilm tabi’ li-l-ma’lum) had in fact already been discussed by the Mu’tazila and the Ash’aris. It was a kalam commonplace that knowledge, unlike power, has no causal effect on its object. Ibn ‘Arabi’s contribution was to make it the key to qadar. In the chapter on ‘Uzayr in the Fusus al-Hikam, he defines qada’ as God’s “judgement over things… according to His Knowledge of them and in them,” and qadar as “the appointment in time of what the things are in their ‘ayn and nothing more.” He then draws the remarkable conclusion that “the judgement of qada’ over things only occurs from these things themselves,” so that “to God belongs the decisive proof” (cf. Quran 6:149). God judges according to what things are, and what they are is the content of His knowledge, which follows them.

The correspondence with Western thought is close and, I think, illuminating:

With Origen and the dependence solution: “not because God knows that an event will occur does it happen; but because it is going to take place, it is known by God.” Ibn ‘Arabi’s al-‘ilm tabi’ li-l-ma’lum states the same explanatory asymmetry.
With Ockham’s soft facts: God’s eternal knowledge of what I will do is “about” me. Its content is fixed by my ‘ayn, my reality, not imposed on it. In Plantinga’s terms, had my reality been otherwise, God’s knowledge would always have been otherwise.
With Molina’s middle knowledge, partly: God’s knowledge of the fixed entities includes their possible states in all circumstances, which resembles natural plus middle knowledge.

The Sufi dictum that “knowledge follows the known; it does not cause it” therefore says what Origen said and what Merricks says: the dependence runs from the act to the knowledge. A caution is again in order. Ibn ‘Arabi’s ontology, in which the entities “have never smelt the breath of existence” and the only true existence is God’s, raises its own questions about whether the fixed entity’s “choice” is a choice in the libertarian sense. Some readers find in it a deeper determinism, in which one is simply what one eternally is. I do not claim Ibn ‘Arabi as a libertarian. I claim his epistemic principle, that knowledge follows its object, as a decisive answer to the fatalist’s inference, whatever one makes of his metaphysics.

The Preserved Tablet and the Pen. This principle also gives us the right reading of the Quran’s “Book” (kitab), the Preserved Tablet (al-Lawh al-Mahfuz, 85:22), and the Pen (al-Qalam, 68:1) that, according to a report from Ibn ‘Abbas cited by al-Qurtubi on 57:22, was commanded at its creation to “write,” so that “it wrote what will be until the Day of Resurrection.” A record is an epistemic object. A ship’s log records the voyage and does not steer the ship. If the Tablet is the inscription of God’s knowledge, and God’s knowledge follows the known, then the Tablet follows the known. It is the perfect ledger of what free agents will do, not the program that makes them do it. Al-Maturidi’s phrase, “an eternal record based on foreknowledge,” says exactly this. Van Inwagen’s “Freedom-denying Prophetic Object” is defused in the same way. A tablet inscribed with “Zayd will freely give charity on Friday” is true because Zayd will freely give. Had he been going to withhold, the tablet would always have read differently. It is a soft fact in stone.

Iqbal: the creative ego and the open possibility. Muhammad Iqbal’s The Reconstruction of Religious Thought in Islam (lectures of 1928–29; published 1930, enlarged 1934) is a modern Muslim interlocutor whom a libertarian must both welcome and correct. In Lecture III, “The Conception of God and the Meaning of Prayer,” Iqbal considers the view of divine knowledge as “a single indivisible act of perception which makes God immediately aware of the entire sweep of history, regarded as an order of specific events, in an eternal ‘now’.” He attributes it to Dawwani, ‘Iraqi and Josiah Royce. He grants that “there is an element of truth in this conception” but worries that a closed, finished history leaves no room for divine creativity or human freedom. His alternative: “The future certainly pre-exists in the organic whole of God’s creative life, but it pre-exists as an open possibility, not as a fixed order of events with definite outlines.” In Lecture IV, “The Human Ego—His Freedom and Immortality,” he insists that the ego is a free, creative centre capable of “consciously participating in the creative life of his Maker.”

Iqbal’s passion for human creativity is entirely right, and his insistence that divine knowledge is “a living creative activity” rather than a passive mirror is a useful corrective to static images. But insofar as he denies that God knows the future as actual, he drifts toward the open theism I rejected in Section 7. My reply to Iqbal is the reply of this whole essay: he need not pay that price. If knowledge follows the known, then God’s knowing the outcome of my creative act does not make it any less mine or any less creative. The future is “open” in the only sense freedom requires, namely open to me, its author. Interestingly, Iqbal himself, in Lecture II, describes divine time as free from “the quality of passage” and not admitting “divisibility, sequence, and change.” That is a Boethian thought which, if followed through, supports the compatibility thesis rather than undermining it.

The Ahmadiyya contribution. Within the modern Ahmadiyya tradition, Hazrat Mirza Bashir-ud-Din Mahmud Ahmad delivered a celebrated address on Taqdir-e-Ilahi (Divine Decree) at the Jalsa Salana in Masjid Nur, Qadian, in 1919. Hazrat Mirza Masroor Ahmad, the community’s current head, described it in his Friday Sermon of 19 February 2021 as “a masterpiece of Islamic theology,” and summarized it as “stating that belief in divine decree and the existence of God Almighty are inseparable,” while carefully distinguishing the domains of decree from the domain of human effort. This is a theme common to the Ahmadiyya commentarial tradition, which has generally read the Quran’s taqdir as God’s establishment of laws and measures within which human beings genuinely choose. Mirza Tahir Ahmad’s Revelation, Rationality, Knowledge and Truth (1998) belongs to the same modern apologetic current. It insists on the consonance of revelation with reason and with the scientific picture of a lawful universe, and that consonance is the background against which a non-fatalist reading of qadar is natural. I note these works as evidence that the compatibilist reading is not a Western import into Islam but a living position within modern Muslim thought.

12. The Swiss Timetable and the Train Driver: Foreknowledge Without Causation

12.1 The central example

Swiss trains are famous for arriving on time, and the reputation is earned. The Swiss Federal Railways (SBB) reported in January 2026, as covered by Le News, that “94.1% of Swiss passenger trains running on the federal network arrived on time” in 2025, a record and “an improvement of 0.9 percentage points on the previous year” (93.2 per cent in 2024). SBB’s standard is strict: a train counts as punctual only if it arrives “with less than 3 minutes” of delay at its stops, measured against the published timetable, and trains cancelled at short notice count as late. sbb Connection punctuality, the rate at which planned transfers actually work, reached 98.6 per cent in 2025 according to RailMarket’s report on SBB’s figures, “slightly below the 98.7% recorded in 2024.” Since 1982 the Swiss system has run on a Taktfahrplan, a clock-face timetable. sbb Under the Bahn 2000 concept, trains reach the main hubs just before the full or half hour and leave just after, so that connections interlock across the whole country. admin (A technical note: since 2020 SBB’s headline measure has been train punctuality rather than the older passenger-weighted “customer punctuality,” which is why I speak of trains rather than passengers.)

Now consider the timetable. On the morning of a given Tuesday, the timetable “says” that the 08:02 from Bern will arrive in Zürich at 08:58. A traveller who consults it “knows,” in an ordinary fallible sense, when the train will arrive. In most cases the knowledge is borne out. Ask, then: what brings the train in at 08:58? Not the timetable. A printed sheet, a digital display, or a database entry has no tractive force. The train arrives because a driver, using judgement, skill and attention, applies the throttle and the brakes. She reads the signals, responds to the conditions, and brings the train into the platform at the appointed minute. The timetable corresponds to the arrival. It does not produce it. If the driver had been late, the timetable would have been wrong. The correctness of the timetable depends on the driver, not the driver on the timetable.

This is precisely the structure of the dependence solution and of Ibn ‘Arabi’s principle that knowledge follows the known. The timetable’s accuracy is grounded in the driver’s performance. If we imagined a perfect timetable, one that was never wrong, it would still be the driver who brought every train in. The timetable would be perfect because it perfectly corresponded to what drivers in fact did, not because it pulled the trains along. The fatalist reasons: “The timetable says 08:58; the timetable is infallible; therefore the driver must arrive at 08:58 and has no say in the matter.” That is the modal fallacy in railway uniform. What is necessary is only the conditional: if the infallible timetable says 08:58, the train arrives at 08:58. The arriving is still the driver’s doing. Her braking at the right moment is not made right by the timetable. The timetable is made right by her braking.

The example has a further, Molinist, layer that is worth drawing out. The Swiss timetable is not a passive guess. Its designers built it on deep knowledge of what drivers would do in various circumstances: line speeds, gradients, dwell times, and how crews respond to a given signal pattern. They chose to “actualize” a timetable, a set of circumstances, in which competent, free professionals would bring trains into hubs at interlocking times. That is how providence works on the Molinist model. God does not drive the train. He arranges the network, the circumstances, in which free drivers, freely driving, bring about the purposes He intends. The Taktfahrplan is a human image of a world arranged so that free acts interlock into a coherent whole.

The example’s limits must also be stated. A human timetable is fallible (5.9 per cent of SBB trains in 2025, “only 1 in 17,” arrived late, down from 6.8 per cent in 2024, according to SBB’s figures as reported by Le News) and prescriptive: it is also an instruction to the driver, which creates a normative pressure divine knowledge as such does not. And the timetable precedes the arrival in time, whereas on the Boethian view God’s knowledge is not temporally prior at all. These limits cut in favour of compatibility rather than against it. If even a prescriptive, temporally prior document does not cause the punctual arrival, still less does a purely cognitive, eternal knowing cause a free act. And the 94.1 per cent, being an empirical record, is in fact the best image of what God’s knowledge is: a perfect record of what free agents actually did.

12.2 Five companion analogies

(1) The historian and the past. A historian knows that Caesar crossed the Rubicon in 49 BC. Her knowledge is certain and cannot now be falsified. Yet nobody supposes that her knowledge caused Caesar to cross, or that Caesar crossed unfreely because historians would later know it. Augustine used this very analogy: just as “when you remember past events you do not compel them to have happened,” God’s foreknowledge does not compel. Strength: it shows plainly that the direction of knowledge (later knower, earlier event) is irrelevant to causation. Knowledge rests on the event, not vice versa. Limit: the historian’s knowledge comes after the event and is caused by its traces, while the fatalist’s worry is about knowledge that comes before. The analogy therefore needs Boethius’s eternity, or Ockham’s softness, to transfer it fully.

(2) The recorded match watched later. Suppose a football final is played live on Saturday, and I watch the recording on Sunday without knowing the score. A friend who watched it live knows that the striker will shoot left in the 89th minute. From my vantage point he “foreknows” it. Nothing in his knowledge makes the striker’s choice unfree; the striker chose left on Saturday, freely, under pressure, and could have chosen right. Strength: this is the best image of Boethian eternity. To God, our “future” is like a match already fully present to Him, and His seeing it is no more constraining than my friend’s having watched it. It also shows how one event can be “future” relative to one viewer (me) and “present or past” relative to another (my friend). Limit: the recording implies that the match is over. Careless use can suggest that the future is already settled in a way that bypasses present agency. The answer is that it is “settled” only by the free choices that constitute it, just as the match result was settled by the players’ play.

(3) The parent who knows the child. A mother knows that, offered a choice between the chocolate cake and the steamed broccoli, her six-year-old will choose the cake. Her knowledge is well founded and reliably correct. The child still chooses freely, and indeed the mother knows because she knows the child’s free character. Strength: it is intuitive, personal, and shows that knowledge can track character without manufacturing it. It also echoes the Quranic counterfactuals (6:28; 8:23), where God’s knowledge of what people would do reflects what is in them (“had God known any good in them”). Limit: parental knowledge is fallible and inferential. Children sometimes surprise us, and that sliver of possible surprise is part of why we feel the child to be free. If the mother’s knowledge were infallible and based on character alone, the libertarian would worry that character determines choice. The analogy illustrates non-causation but not infallibility.

(4) The chess grandmaster. A grandmaster playing a club player can often foresee the opponent’s next several moves and arrange the board so that the opponent’s free replies walk into a mating net. Strength: this is the best image of Molinist providence. The grandmaster does not move the opponent’s pieces. He chooses the circumstances (his own moves), knowing how the opponent would respond, and thereby achieves his purpose through the opponent’s freedom. Limit: the grandmaster’s foresight is probabilistic, and his purpose is adversarial. God’s middle knowledge, on the Molinist view, is certain, and His purposes are benevolent: He arranges circumstances in which free creatures can flourish, not in which they are trapped. The analogy also brings in the grounding objection. The grandmaster’s foresight is grounded in the opponent’s patterns of play, and Molinists must say what grounds God’s foresight of a libertarian creature.

(5) The mathematician and the student. A professor knows the answer to the problem she sets her class. A student works it out, makes choices about method, errs, corrects herself, and arrives at the answer by her own effort. The professor’s prior knowledge of the answer did nothing to bring the student there. Strength: it separates knowledge of outcome from causation of process, and the dignity of the student’s achievement is untouched by the teacher’s foreknowledge. It also captures the hadith “work, for everyone is eased toward what he was created for”: the answer is known, and still the work must be done. Limit: the answer to a mathematics problem is necessary, while the outcome of a free act is contingent. The professor knows what the answer is, not what the student will write. The analogy therefore illustrates non-causation of process but not foreknowledge of contingent choice.

A foil: the weather forecaster. The weather forecast is often offered as an analogy, and it is instructive mainly as a disanalogy. The forecaster knows tomorrow’s rain by extrapolating present causes, and the knowledge is fallible because the system is chaotic. If God’s knowledge of our acts were like this, knowledge from present causes, then it would either be fallible or, if infallible, would presuppose deterministic causes. That is exactly the knowledge the libertarian cannot attribute to God. The SEP makes the same point: knowledge of “what is going to happen” on the basis of present tendencies yields at most a “mutable future” and not infallibility. Eternity and dependence, not prediction, are the right models of divine knowledge.

What the analogies show together. No single analogy captures all the features of divine knowledge: infallibility, priority (or eternity), and non-causation. Each captures one or two. The train and the historian show that correctness depends on the act. The recorded match shows how eternity can make foreknowledge into a kind of seeing. The grandmaster shows how providence can work through freedom. The parent and the mathematician show that knowing and doing are different relations. Not one of them involves the knowledge causing the act. That is the cumulative point. The fatalist needs knowledge to be a cause, and nothing in our experience of knowledge, human or by extension divine, suggests that it is.

13. The Correspondence Mapped

We can now set out the correspondences the essay has been tracing:

Western solution	Core idea	Islamic parallel	Assessment
Boethius / Aquinas (eternity)	God knows all as present in one eternal “now”; knowledge takes the mode of the knower	Quran 57:3; 22:47; 70:4; al-Ghazali’s single eternal knowledge with changing relations (idafa); Averroes’s eternal vs. originated knowledge “by sheer homonymy”	Close parallel; Averroes adds the apophatic warning against modelling God’s knowledge on ours
Ockham (soft facts)	God’s “past” beliefs about future acts are not fixed independently of those acts	Maturidi: the decree is “an eternal record based on foreknowledge”	Parallel in substance, though without Ockham’s tense logic
Dependence solution (Origen, Plantinga, Merricks)	Knowledge depends on the known, not vice versa	Ibn ‘Arabi and kalam: al-‘ilm tabi’ li-l-ma’lum, “knowledge follows the known”	Strong anticipation; arguably the Islamic tradition’s clearest formulation
Molina (middle knowledge)	God knows counterfactuals of freedom prior to the creative decree	Quran 6:28; 8:23 state counterfactuals of freedom; Ibn ‘Arabi’s a’yan thabita include possible states	The Quran supplies the data; Molina supplies the theory of logical priority
Libertarian agency	Two-way power located in the agent	Mu’tazili qudra; Maturidi istita’a “valid for opposite acts” and ikhtiyar	Maturidi synthesis is the most promising Sunni basis

Where does the Islamic tradition improve on the Western solutions? In three places, I suggest. First, the Quranic counterfactuals give a scriptural existence proof of the truths Molinism needs, which blunts the grounding objection’s force for a believer. Second, the kalam principle that knowledge follows the known is stated as a general epistemic law, not as an ad hoc exception for God, which answers the SEP’s charge that Ockhamism is “ad hoc.” Third, the Prophetic praxis tradition (“work, for everyone is eased,” “we flee from the decree of God to the decree of God”) turns the speculative debate into a lived ethic. Belief in the decree becomes a reason for effort, deliberation and responsibility rather than resignation.

Epilogue: The Ledger and the Hand That Writes

There is an old Arabic image of the qalam, the Pen, that “wrote what will be.” We tend to imagine it as an author’s pen, composing a plot that the characters must then act out. But there is another kind of pen: the recorder’s pen, the scribe’s, the clerk’s who keeps the ledger of a busy port. Everything that happens in the port is written down, completely and without error, and yet not a single ship sails because of the ledger. The ships sail because their captains choose to sail them.

The Quran, I have argued, asks us to imagine the Pen in the second way when it speaks of our deeds. “Whoever does an atom’s weight of good will see it.” The seeing at the end presupposes the doing in the middle, and the doing is ours. The Book is complete, and that completeness is a glory of God’s knowledge. But a complete record of free acts is not a cage around them. It is a tribute to their reality. Only something that really happens can be truly recorded.

Philosophers have sometimes made the problem of foreknowledge sound like a contest between God and the human being for possession of the future, as if every truth God knows must be subtracted from what we can do. Boethius, in his cell, saw through this. The divine gaze, he said, “does not disturb the quality of things.” Ibn ‘Arabi, in Konya, said the same in another idiom: “knowledge is subject to the known.” ‘Umar, at Sargh, said it in the language of a statesman: we flee from the decree of God to the decree of God. And a Swiss train driver, easing into Zürich at 08:58, says it without words. The timetable was right because she was skilled, and her skill did not become less her own because the timetable was right.

A physician learns to respect two things at once: the regularities of the body, which are as predictable as any timetable, and the stubborn, irreducible freedom of the patient, who may or may not take the medicine. Faith asks for the same double respect at a higher level: for a God whose knowledge leaves nothing out, and for a human self whose choices are really its own. The tradition, Christian and Muslim alike, has given us the conceptual means to hold both. What remains is the work, and as the Prophet told his companions at the graveside when they asked whether they should simply rely on what was written, the work is still ours to do.

Bibliography

Scripture and hadith

The Qur’an. Arabic text: standard Uthmani text as presented at quran.com/tanzil.net. Translations: Sahih International (Jeddah: Abul-Qasim/Al-Muntada al-Islami, 1997); M. A. S. Abdel Haleem, The Qur’an: A New Translation (Oxford: Oxford University Press, 2004).
al-Bukhari, Sahih al-Bukhari, nos. 4949, 5729, 6596, 7551 (numbering as at sunnah.com).
Muslim ibn al-Hajjaj, Sahih Muslim, Kitab al-Iman no. 8 (Hadith of Jibril); no. 2219 (plague narration).
Malik ibn Anas, al-Muwatta’, Kitab al-Jami’, bab on the plague.

Western primary sources

Augustine, On Free Choice of the Will (De libero arbitrio), Book III, trans. Thomas Williams (Indianapolis: Hackett, 1993).
Augustine, The City of God (De civitate Dei), Book V, chs. 9–10.
Boethius, The Consolation of Philosophy, Book V, trans. V. E. Watts (Harmondsworth: Penguin, 1969); Loeb ed., trans. S. J. Tester (Cambridge, MA: Harvard University Press, 1973).
Thomas Aquinas, Summa Theologiae I, q.14, a.13; q.19, a.8; q.83, a.1; Summa contra Gentiles I.66.
William of Ockham, Predestination, God’s Foreknowledge, and Future Contingents, trans. Marilyn McCord Adams and Norman Kretzmann (New York: Appleton-Century-Crofts, 1969; 2nd ed. Indianapolis: Hackett, 1983).
Luis de Molina, Concordia liberi arbitrii cum gratiae donis, divina praescientia, providentia, praedestinatione et reprobatione (Lisbon, 1588); Part IV trans. Alfred J. Freddoso as On Divine Foreknowledge (Ithaca: Cornell University Press, 1988).
Origen, Commentary on the Epistle to the Romans.

Western secondary sources

Adams, Marilyn McCord. “Is the Existence of God a ‘Hard’ Fact?” Philosophical Review 76 (1967).
Adams, Robert Merrihew. “Middle Knowledge and the Problem of Evil.” American Philosophical Quarterly 14 (1977).
Craig, William Lane. The Only Wise God: The Compatibility of Divine Foreknowledge and Human Freedom. Grand Rapids: Baker, 1987.
Flint, Thomas P. Divine Providence: The Molinist Account. Ithaca: Cornell University Press, 1998.
Hasker, William. God, Time, and Knowledge. Ithaca: Cornell University Press, 1989.
Keathley, Kenneth. Salvation and Sovereignty: A Molinist Approach. Nashville: B&H Academic, 2010.
Merricks, Trenton. “Truth and Freedom.” Philosophical Review 118 (2009).
Pike, Nelson. “Divine Omniscience and Voluntary Action.” Philosophical Review 74 (1965).
Plantinga, Alvin. The Nature of Necessity. Oxford: Clarendon Press, 1974.
Plantinga, Alvin. “On Ockham’s Way Out.” Faith and Philosophy 3 (1986).
Rogers, Katherin. “The Necessity of the Present and Anselm’s Eternalist Response to the Problem of Theological Fatalism.” Religious Studies 43 (2007).
Stump, Eleonore, and Norman Kretzmann. “Eternity.” Journal of Philosophy 78 (1981).
van Inwagen, Peter. “What Does an Omniscient Being Know about the Future?” Oxford Studies in Philosophy of Religion 1 (2008).
Zagzebski, Linda. The Dilemma of Freedom and Foreknowledge. New York: Oxford University Press, 1991.
Zagzebski, Linda. “Foreknowledge and Free Will.” Stanford Encyclopedia of Philosophy (first published 2004; substantive revision 2026).

Islamic primary sources

al-Ash’ari, Abu al-Hasan. Kitab al-Luma’ and Maqalat al-Islamiyyin.
al-Ghazali, Abu Hamid. Tahafut al-Falasifa (The Incoherence of the Philosophers), trans. Michael E. Marmura (Provo: Brigham Young University Press, 1997/2000); al-Iqtisad fi al-I’tiqad; Ihya’ ‘Ulum al-Din, Kitab al-Tawhid wa-l-Tawakkul.
Ibn ‘Arabi, Muhyi al-Din. al-Futuhat al-Makkiyya; Fusus al-Hikam, trans. R. W. J. Austin as The Bezels of Wisdom (New York: Paulist Press, 1980).
Ibn Qayyim al-Jawziyya. Shifa’ al-‘Alil fi Masa’il al-Qada’ wa-l-Qadar wa-l-Hikma wa-l-Ta’lil.
Ibn Rushd (Averroes). Fasl al-Maqal with the Damima (appendix on divine knowledge), 1179; al-Kashf ‘an Manahij al-Adilla, 1179, trans. Ibrahim Y. Najjar as Faith and Reason in Islam (Oxford: Oneworld, 2001); Tahafut al-Tahafut, trans. Simon van den Bergh (London: Luzac, 1954).
Ibn Sina (Avicenna). al-Shifa’, al-Ilahiyyat VIII.6.
Ibn Taymiyya. al-‘Aqida al-Wasitiyya; treatises on qadar in Majmu’ al-Fatawa.
al-Juwayni, Imam al-Haramayn. al-‘Aqida al-Nizamiyya.
al-Maturidi, Abu Mansur. Kitab al-Tawhid.
al-Qurtubi. al-Jami’ li-Ahkam al-Qur’an, on 57:22.
al-Tabari. Jami’ al-Bayan, on 8:23.

Modern Islamic and secondary sources

Aladdin, Bakri. “The Mystery of Destiny (sirr al-qadar) in Ibn ‘Arabi and al-Qunawi.” Muhyiddin Ibn Arabi Society.
Ahmad, Mirza Bashir-ud-Din Mahmud. Taqdir-e-Ilahi (Divine Decree), address at Jalsa Salana, Qadian, 1919.
Ahmad, Mirza Tahir. Revelation, Rationality, Knowledge and Truth. Tilford: Islam International Publications, 1998.
Belo, Catarina. “Ibn Rushd on God’s Decree and Determination (al-qada’ wa-l-qadar).” Al-Qantara 27 (2006).
Iqbal, Muhammad. The Reconstruction of Religious Thought in Islam. Lahore, 1930; Oxford: Oxford University Press, 1934. Lectures II–IV.
Sharif, M. M., ed. A History of Muslim Philosophy, vol. 1. Wiesbaden: Otto Harrassowitz, 1963. Chapter on Maturidism.
Shah, Zia H. Essays at thequran.love.
Swiss Federal Railways (SBB). Punctuality statistics and press release on 2025 results (January 2026).