# Foreknowledge Is Causation Objection Defeater ## Intro The objection: God's foresight is not a passive stare. He is not a spectator who happens to catch the game early. He is the one who made the players, the field, and the rules, and He knew the score before the whistle. So His knowing is His causing. Talk of "He only foresees, He does not determine" is a distinction with no difference, and it is used to keep God's hands clean while He wrote the whole thing. The reply is not a slogan about roosters and sunrises, though the slogan is true. It is a claim about the direction of explanation. Beliefs answer to facts; facts do not answer to beliefs. When God believes that Judas betrays, the reason the belief is true is Judas. Reverse that and you have not described knowledge at all, you have described decree, and decree is a further claim the objector still has to argue for. There is one place where the objection has real teeth, and it should be handed to the objector rather than hidden: Aquinas himself wrote that God's knowledge is the cause of things. That sentence is quotable and it looks like a concession from the strongest possible witness. It is not, and the paragraph it sits in says why, in the same article. Working through that text carefully is the most useful thing this page does. ## Cheatsheet **30-second reply.** Knowing and causing are different relations, and the difference is the direction of explanation. My belief that it rained is true because it rained; the rain is not true because of my belief. Now say what makes God's belief about Judas true. It is Judas. If you want to say God *decreed* Judas rather than *knew* Judas, that is a different claim, and you have not argued for it, you have assumed it by naming knowledge "causation." And be careful, because the equation eats everything: a doctor who reliably predicts a prognosis has now caused the disease, and God's knowledge of His own eternal nature has now caused His nature. The equation does not survive its own generality. **Fast facts.** - The move has a classical name: *post hoc ergo propter hoc*, treating temporal priority as explanatory priority. - Aquinas does say "the knowledge of God is the cause of things", *Summa Theologiae* I q.14 a.8. In the same article he says it causes only *as joined to will*. Quote both halves before the objector quotes the first. - [Acts 2:23](/codex/acts-2-23/) uses foreknowledge and human culpability in one sentence about the same event. - Greek *proginōskō* in [Romans 8:29](/codex/romans-8-29/) is contested between an epistemic reading (God knew in advance) and a relational one (God fore-loved, set covenant regard upon), and *both* readings defeat this objection. - Molinism's feasibility constraint means even a God who selects a world from among the ones available is not thereby the author of what free creatures do in it. **Counter-moves.** 1. **Ask for the truthmaker.** "What makes God's belief about Judas true?" Every answer either concedes the distinction or converts the objection into naked determinism. 2. **Hand them the Aquinas quotation yourself**, then read the rest of the article. Pre-empting their best evidence is worth more than surviving it. 3. **Run the generality test.** Apply the equation to any ordinary reliable knower and watch it produce absurdities. 4. **Separate the three objections that arrive together**: knowing-is-causing, creating-while-knowing, and the fixity-of-the-past argument. Only the third is hard, and it is handled at [Theological Fatalism Objection Defeater](/codex/theological-fatalism-objection-defeater/). 5. **Refuse "distinction without a difference" as a verdict.** It is a conclusion, not an argument. Make them show the two relations have the same logical properties, and they cannot, because one is symmetric under counterfactual variation and the other is not. **Concessions to grant.** - God is **not** a passive observer. He is Creator, and creation plus knowledge does raise a genuine question about responsibility. That question is real, and it is not this one. - On classical theism, God's knowledge is **not** derived from observing the world the way ours is. Say so before they do. - Some Christians, consistent Reformed determinists, **do** hold that God determines all things. This objection is not a refutation of Christianity; at most it is an argument for one Christian position over another. Naming that dissolves half its force. - Aquinas's formula is **a real datum**, not a misquotation. Grant its authenticity and dispute its meaning. **Closing line.** You have not shown that knowing is causing. You have renamed causing as knowing, and then discovered causation inside it. If you want the conclusion that God determines every act, argue for the decree directly. It is a serious position with serious defenders. It is just not something you get for free from the word "knew." ## In full The objection asserts that divine foreknowledge is not distinct from divine determination: since God is Creator rather than observer, His knowledge of what creatures will do can only be knowledge of what He has arranged, so the epistemic-versus-causal distinction is a verbal screen for a single relation. This defeater grants the premise that God is not a spectator, and denies that the conclusion follows. Knowledge and causation differ in the direction of explanation, in their counterfactual behavior, and in their logical properties; the classical formula that appears to identify them (Aquinas, *Summa Theologiae* I q.14 a.8) explicitly conditions the identification on the joining of will, which is the theist's point rather than the objector's. Where the objection is reformulated as an argument from *creating while knowing*, it becomes a different and better argument, which is answered by the feasibility constraint on middle knowledge or by the primary-and-secondary causation framework. Where it is reformulated as an argument from the fixity of the past, it becomes theological fatalism, which is a genuinely hard argument handled separately at [Theological Fatalism Objection Defeater](/codex/theological-fatalism-objection-defeater/). The concept-level treatment is [Foreknowledge vs Causation](/codex/foreknowledge-vs-causation/). ## Argument structure | # | Premise | |---|---| | **P1** | Knowledge and causation are distinct relations, distinguished by the direction of explanation: beliefs answer to facts, facts do not answer to beliefs. | | **P2** | The two relations behave differently under counterfactual variation, which is a test the objection fails. | | **P3** | The classical formula that God's knowledge causes things is conditioned, in its own text, on the joining of divine will, so it does not identify knowledge with causation. | | **P4** | Reformulated as creating-while-knowing, the objection is a different argument, and it is answered by feasibility or by primary-and-secondary causation. | | **P5** | The equation proves too much: applied generally, it makes every reliable knower a cause and makes God's self-knowledge the cause of God. | | **C** | **Foreknowledge is not causation, and the inference from God knew to God determined is unsound.** | ## Form Defensive, built on a category distinction (P1, P2), a concession-audit of the objector's best evidence (P3), a redirect of the reformulated objection to where it is actually answered (P4), and a reductio (P5). The argument does not attempt to prove that God is *not* a determiner; a Reformed reader may hold that He is, on other grounds. It proves only that the determination does not follow from the knowing, which is what the objection claimed. ## Grammar and hermeneutics The objection is philosophical, but it regularly recruits one Greek word, and the exchange goes badly for anyone who has not looked at it. **The vocabulary.** *Proginōskō* (verb, "to know beforehand") and *prognōsis* (noun) occur in [Acts 2:23](/codex/acts-2-23/), [Romans 8:29](/codex/romans-8-29/), [Romans 11:2](/codex/romans-11-2/), [1 Peter 1:2](/codex/1-peter-1-2/), and [1 Peter 1:20](/codex/1-peter-1-20/). The prefix *pro-* is temporal or logical priority; the stem is *ginōskō*, the ordinary verb for knowing, which in biblical Greek inherits the Hebrew *yada'* range and can carry relational and covenantal weight (see [G1097 - ginosko](/codex/g1097-ginosko/) and [H3045 - yada](/codex/h3045-yada/)). **The two readings.** 1. **Epistemic.** God knew in advance what would occur. Standard in Arminian and Molinist exegesis, and the plain sense of the word outside theology. 2. **Relational or elective.** God fore-*loved*, set covenantal regard on persons in advance, as in [Amos 3:2](/codex/amos-3-2/), "You only have I known among all the families of the earth", where knowing plainly is not information. Standard in Reformed exegesis, and grammatically defensible: the object of *proginōskō* in [Romans 8:29](/codex/romans-8-29/) is *hous*, "whom", persons rather than propositions about persons, which is a real point in its favor. **Why the objection loses on either reading.** This is the useful observation, and it is worth stating explicitly rather than picking a side. On reading 1, *proginōskō* is straightforwardly epistemic and the objection needs an argument to convert knowing into causing, which is what the rest of this page denies it. On reading 2, the word is not about foreknowledge at all but about elective love, so it cannot be evidence that God's *foreknowledge* determines anything; the determination, if any, would come from election, which is a distinct doctrine argued from distinct texts. The objector needs *proginōskō* to be simultaneously epistemic (so that it is foreknowledge) and causal (so that it determines), and no reading of the word delivers both. **The clause structure of the strongest text.** [Acts 2:23](/codex/acts-2-23/) is one sentence with two coordinate causes: *tē hōrismenē boulē kai prognōsei tou theou*, by the determined plan and foreknowledge of God, and *dia cheiros anomōn*, through the hand of lawless men. Peter uses *boulē* (settled purpose) and *prognōsis* (foreknowledge) as two things, joined by *kai*, not as a hendiadys where one word explains the other. Then he assigns the killing to human agents with a preposition of instrumentality and holds the audience guilty for it in the next breath. If the inspired writer had thought the foreknowledge absorbed the human agency, the sentence would not be built this way, and the sermon would not end with a call to repent ([Acts 2:38](/codex/acts-2-38/)). **What not to claim.** Do not assert that *proginōskō* "cannot" mean elective love; it can, and Reformed exegetes have good grammatical grounds. Do not assert that the *pro-* prefix settles the question of causality; prefixes mark priority, not agency. Overreaching on the Greek is the fastest way to lose a debate that the philosophy already wins. --- ## P1, Knowledge and causation differ in the direction of explanation ### Affirmative case 1. **Truthmaking runs one way.** A belief is true because of what it is about. My belief that it rained is true because it rained; the rain does not obtain because of my belief. Whatever else divine knowledge is, if it is knowledge then its content answers to reality, and the objection needs to deny this rather than assume its opposite. 2. **The fallacy has a name and a shape.** Treating what comes earlier as therefore explanatory is *post hoc ergo propter hoc*. The rooster crows before dawn and does not cause it; the barometer falls before the storm and does not cause it. The objection is that pattern with the temporal gap enlarged. 3. **The distinction is not invented for this debate.** Epistemology and metaphysics are different disciplines because knowing and being are different relations. No one accepts the equation anywhere else in philosophy, and adopting it only for God is special pleading in the objector's direction. ### Anticipated objections 1. *"God is not a barometer. He is the Creator, so His knowledge cannot be a reading taken off an independent world."* 2. *"You are describing how human knowledge works and projecting it onto God."* ### Rebuttals 1. Granted in full, and it is the strongest thing the objector says. But the conclusion still does not follow. Creating a world with genuinely free creatures and knowing what they do in it are two acts, and the second is not the first performed again. What creation gives the objector is a question about responsibility for the arrangement, which is answered at P4, not a proof that knowledge just is causation. 2. The direction-of-explanation point is not a human peculiarity; it is what makes a state a *knowing* rather than a *doing*. Classical theism agrees that God does not know by observation, and it still denies that His knowledge of contingent creaturely acts is identical to His causing them. Aquinas holds both, which is the subject of P3. --- ## P2, The two relations behave differently under counterfactual variation ### Affirmative case 1. **The counterfactual test.** Had Judas not betrayed, God would not have believed that he would. Had God not believed it, Judas might still have done it or not; nothing about the act is settled by the belief's absence. The dependency is asymmetric, and asymmetric dependence is exactly what distinguishes a tracking relation from a producing one. 2. **Causation supports intervention; knowledge does not.** If A causes B, then manipulating A is a route to changing B. Nobody thinks that changing God's belief is a route to changing what a creature does, and the objector does not think so either, which is a tacit admission that the relation is not causal. 3. **Reliability is not production.** A perfectly reliable predictor is still a predictor. Raise reliability to infallibility and you have raised it to certainty, not converted it into agency, because certainty is a property of the knower's relation to the fact and not a force exerted on the fact. ### Anticipated objections 1. *"Counterfactuals about God are unintelligible. There is no world where God believes falsely."* 2. *"You are smuggling in libertarian freedom. On determinism the counterfactual test fails."* ### Rebuttals 1. The counterfactual is not "God believes falsely", which is indeed impossible, but "had the act differed, the belief would have differed." That is not only intelligible, it is entailed by infallibility, since an infallible knower's beliefs covary perfectly with the facts. Perfect covariance is the strongest possible version of the tracking relation, so the objection's own premise supplies the point. 2. On determinism the counterfactual test does fail, but notice what that means: the objector then needs determinism as an independent premise, and the objection has become an argument *from* determinism rather than an argument *for* it. That is a legitimate debate ([Hard Determinism](/codex/hard-determinism/), [Compatibilism](/codex/compatibilism/)) and it is not this one. --- ## P3, The Thomistic formula does not identify knowledge with causation ### Affirmative case 1. **What Aquinas actually wrote.** *Summa Theologiae* I q.14 a.8 is titled "whether the knowledge of God is the cause of things", and he answers yes. This is the objector's best possible citation, and it should be produced by the theist first. 2. **The condition inside the same article.** Aquinas immediately restricts it: God's knowledge is the cause of things *insofar as His will is joined to it*, and knowledge without the assent of will is not causal, which he calls the difference between speculative and practical knowledge. The causal work is done by will, not by knowledge; knowledge supplies the form of what is willed. Remove the will and the article's own answer is no. 3. **Primary and secondary causation.** In the same *Summa*, God as first cause acts through genuine created causes rather than instead of them (I q.105 a.5): fire really heats, and the will really chooses, precisely as instruments through which God works. On this framework, saying God causes an act is not saying the creature did not. Eleonore Stump's *Aquinas* (2003) develops the double-agency reading. 4. **The formula covers God's knowledge of creatures as creatable, not His knowledge of free acts as free.** Aquinas's concern in q.14 a.8 is that God's knowledge is architectural, the way an artisan's plan is the cause of the artifact. Applying it to contingent free acts is an extension that Aquinas himself handles differently, in q.14 a.13, where the eternal-present account does the work. ### Anticipated objections 1. *"Aquinas is a Christian determinist, so quoting him against determinism is dishonest."* 2. *"Primary and secondary causation is a distinction without a difference. If the first cause settles the outcome, the secondary cause is decorative."* 3. *"Molinists reject the Thomistic framework anyway, so you cannot use both."* ### Rebuttals 1. Whether Aquinas is a determinist is genuinely disputed, and the honest position is that his account is not the modern determinist thesis and does not map cleanly onto it. But the point here does not require Aquinas to be a libertarian. It requires only that the sentence the objector quotes does not say what they need, and the article's own restriction settles that regardless of Aquinas's wider commitments. 2. This is the best of the three objections and deserves a real answer rather than a formula. The distinction earns its keep if created causes are genuinely causes, that is, if they contribute the specific character of their effects and if the world would have been different had they acted otherwise. Whether Thomism can secure that for free acts is a live intramural dispute; what is not disputable is that the framework is not the claim "God's knowing is God's causing", which is the claim under review. 3. Correct, and this defeater does not need both to be true. P3 shows that the objector's flagship citation fails on its own terms. P4 gives a Molinist route and a Thomist route separately, so a reader may take whichever they hold. --- ## P4, Creating-while-knowing is a different argument, and it has answers ### Affirmative case 1. **State the better objection honestly.** The stronger version is not "knowing is causing" but "God chose to actualize *this* world knowing every act in it, so He bears authorship of those acts." That argument is worth engaging, and it is not defeated by the epistemic-causal distinction. 2. **The Molinist answer: feasibility.** God can actualize only worlds that are *feasible*, and which worlds are feasible depends on the counterfactuals of creaturely freedom, which are not up to God. Selecting from a constrained menu is not authoring the menu. Plantinga's transworld-depravity argument is the sharpest form: it may be that no feasible world contains free creatures who never sin. See [Molinism](/codex/molinism/) and [Counterfactuals of Freedom](/codex/counterfactuals-of-freedom/). 3. **The Thomist answer: concurrence.** God's causality operates at a different level than creaturely causality, sustaining the creature's own real agency rather than overriding it. The creature's act is wholly from God as first cause and wholly the creature's as proper cause, which are not competing shares of one quantity. See [Divine Providence](/codex/divine-providence/). 4. **The biblical pattern assigns both.** [Genesis 50:20](/codex/genesis-50-20/), the brothers meant evil and God meant good, in one act. [Acts 2:23](/codex/acts-2-23/) and [Acts 4:27-28](/codex/acts-4-27-28/) do the same with the crucifixion. Scripture is comfortable assigning full divine purpose and full human culpability to a single event, which is the pattern any adequate account has to reproduce. ### Anticipated objections 1. *"Feasibility is a fiction. If the counterfactuals are true prior to creation, God still picked the world where I fall."* 2. *"Concurrence is word magic. Two total causes of one effect is not a coherent notion."* 3. *"Then God is at least reckless, creating knowing the harm."* ### Rebuttals 1. He picked the world; He did not pick what you would freely do in it, because that is the input to feasibility rather than an output of His choice. The residual complaint is not about knowledge or causation any more, it is the problem of evil in its creation-selection form, which is a real objection with its own treatment. Redirect rather than defend on the wrong ground. 2. It is coherent on any account where the two causes operate at different levels rather than dividing one workload, which is the entire point of the primary-secondary distinction. The author-and-character analogy is imperfect but exact on the relevant point: everything in the character's action is the author's doing and everything in it is also the character's, and neither reduces the other's contribution. Whether the analogy holds for real agents is contested, and it is fair to say so. 3. That is the problem of evil, not the foreknowledge objection, and it should be named as a change of topic. It is answered elsewhere: [Problem of Evil](/codex/problem-of-evil/), [Free Will Defense](/codex/free-will-defense/), [Soul-Making Theodicy](/codex/soul-making-theodicy/). Letting an exchange drift from foreknowledge to evil without noticing is how a winnable argument gets lost. --- ## P5, The equation proves too much ### Affirmative case 1. **It swallows ordinary knowledge.** A physician who reads a scan and knows with practical certainty how the disease will progress has, on the equation, caused the progression. A parent who knows exactly which child will take the last cookie has caused the taking. Every reliable knower becomes an agent of what they know, which nobody accepts. 2. **It swallows God's self-knowledge.** God knows His own nature perfectly and eternally. If knowledge is causation, God's knowledge causes God's nature, which contradicts aseity and makes the divine nature dependent on a divine act. See [Aseity](/codex/aseity/). 3. **It swallows the past.** God knows perfectly what happened yesterday. If knowing is causing, His present knowledge causes yesterday's events, which is backward causation. The objector will reject that, and the rejection concedes that knowledge of X is not causation of X in at least one direction. Once conceded for the past, the burden is on them to show why the future is different, and the only available answer, that the future is not yet fixed, is open theism rather than determinism. 4. **The equation is unstated in the argument.** No premise ever defends "knowledge is causation"; it is smuggled in by describing God's knowledge with causal verbs. An argument whose crucial step is a choice of vocabulary is not an argument. ### Anticipated objections 1. *"Human cases are disanalogous. The physician did not create the patient."* 2. *"Self-knowledge is different, God's nature is necessary."* 3. *"Backward causation is different because the past is fixed."* ### Rebuttals 1. Then the operative premise is *creation*, not *knowledge*, and the objection should be restated as the creation-selection argument at P4. That is progress, and it should be named as such: the objector has just abandoned the thesis that foreknowledge is causation. 2. Agreed, which is why it is a counterexample. A universal equation that requires an exemption clause for the most perfect case of knowledge is not a universal equation. The exemption is the concession. 3. Also agreed, and also a counterexample for the same reason. Notice how the objection now needs both an exemption for necessary truths and an exemption for the past, at which point the "equation" is a claim about exactly one case, future contingents, held for no stated reason beyond the conclusion it is meant to produce. --- ## Live-cite kit **Scripture.** - [Acts 2:23](/codex/acts-2-23/), the determined plan and foreknowledge of God, and the hands of godless men, in one sentence with two grammatical agents. - [Acts 4:27-28](/codex/acts-4-27-28/), the same structure with named human actors doing "whatever Your hand and Your purpose predestined to occur." - [Genesis 50:20](/codex/genesis-50-20/), you meant evil against me, God meant it for good; one act, two intentions, no dissolving of either. - [Romans 8:29](/codex/romans-8-29/), whom He foreknew He also predestined; the text the objection reaches for, and the one whose two readings both defeat it. - [Isaiah 46:9-10](/codex/isaiah-46-9-10/), declaring the end from the beginning; the foreknowledge claim at full strength, with no causal verb applied to human sin. - [1 Samuel 23:11-12](/codex/1-samuel-23-11-12/), God tells David what Saul *would* do, David leaves, and it does not happen. God's knowledge of an unrealized conditional caused nothing. **Scholarly.** - Thomas Aquinas, *Summa Theologiae* I q.14 a.8, with the *insofar as His will is joined to it* restriction read aloud, and I q.105 a.5 on primary and secondary causation. - Boethius, *Consolation of Philosophy* V.3-6, the classical statement that the mode of knowledge belongs to the knower and does not impose necessity on the known. - Augustine, *City of God* V.9-10, foreknowledge and voluntary will held together against Cicero. - Alvin Plantinga, *God, Freedom, and Evil* (Eerdmans, 1974), on feasible worlds. - William Lane Craig, *The Only Wise God* (Baker, 1987), the accessible full-length treatment. - Eleonore Stump, *Aquinas* (Routledge, 2003), on double agency. **Aphorism.** - "The forecaster does not make the weather, and the calendar does not make the year." - "Beliefs answer to facts. Reverse that and you are not describing knowledge, you are describing decree, and you still have to argue for it." - "You need one word to be epistemic enough to be foreknowledge and causal enough to determine. It cannot be both at once." ## Tactical notes - **Ask for the truthmaker in the first minute.** "What makes God's belief about Judas true?" This single question does more work than any speech, because every answer either grants the distinction or restates determinism as a premise. - **Give them the Aquinas quotation before they find it.** Producing the opponent's best evidence yourself, then reading the qualification in the same article, is the most persuasive thirty seconds available on this topic. - **Separate three objections that arrive as one.** Knowing-is-causing (this page), creating-while-knowing (P4, and beyond that the problem of evil), and the necessity-of-the-past argument ([Theological Fatalism Objection Defeater](/codex/theological-fatalism-objection-defeater/)). Objectors move between them without noticing; the person who tracks the difference controls the exchange. - **Do not overclaim on the Greek.** *Proginōskō* has a defensible relational reading. The argument does not need the epistemic reading to win, and insisting on it hands a prepared objector an easy correction. - **Concede that Reformed determinism exists.** The objection at most favors one Christian account over another. Saying so early removes its anti-Christian framing and costs nothing. - **Watch for the slide into the problem of evil.** It is the most common drift, and the honest move is to name it and offer to change topics rather than to answer a different objection under the old heading. ## See also - [Foreknowledge vs Causation](/codex/foreknowledge-vs-causation/), the concept hub for the epistemic-versus-causal distinction - [Theological Fatalism Objection Defeater](/codex/theological-fatalism-objection-defeater/), the harder version of the objection, which concedes this one - [Theological Fatalism](/codex/theological-fatalism/), the six response families to the fixity-of-the-past argument - [God Is Surprised Objection Defeater](/codex/god-is-surprised-objection-defeater/), the texts pressed against exhaustive foreknowledge - [Molinism](/codex/molinism/), middle knowledge and the feasibility constraint - [Counterfactuals of Freedom](/codex/counterfactuals-of-freedom/), possible worlds against feasible worlds - [Divine Providence](/codex/divine-providence/), concurrence and the primary-secondary framework - [Compatibilism](/codex/compatibilism/), the account that accepts determination and keeps responsibility - [Libertarian Free Will](/codex/libertarian-free-will/), the account the objection targets - [Hard Determinism](/codex/hard-determinism/), the position the objection needs as an independent premise - [Predestination](/codex/predestination/), the doctrine downstream - [Problem of Evil](/codex/problem-of-evil/), where the creation-selection version of the objection actually belongs - [Aseity](/codex/aseity/), the attribute the equation would violate - [Arguments](/codex/arguments/), the master index
## Common questions this page answers **Q: If God knew I would sin, did He make me sin?** No, and the reason is the direction of explanation. God's belief that you would sin is true because you sin; you do not sin because of His belief. Knowledge answers to its object. To get from His knowing to His causing you need a further claim, that God decrees every act, and that claim has to be argued from other grounds rather than extracted from the word "knew." **Q: Isn't the difference between knowing and causing just a word game?** No, and there is a test. Causation supports intervention: if A causes B, changing A is a route to changing B. Nobody thinks changing God's belief is a route to changing what a creature does. And the dependence is asymmetric: had the act been different, the belief would have been different, but not the reverse. Two relations with different logical properties are not one relation with two names. **Q: Didn't Aquinas say God's knowledge causes things?** He did, in *Summa Theologiae* I q.14 a.8, and it is the best line the objection has. In the same article he restricts it: God's knowledge is causal only insofar as His will is joined to it, which is his distinction between speculative and practical knowledge. The causal work belongs to the will. Quote the whole article and the citation stops supporting the objection. **Q: What does "foreknew" mean in [Romans 8:29](/codex/romans-8-29/)?** It is disputed. *Proginōskō* can mean knowing in advance, and it can mean setting covenantal regard on someone in advance, the way "you only have I known among all the families of the earth" ([Amos 3:2](/codex/amos-3-2/)) plainly does not mean God lacked information about other nations. The useful point is that the objection fails on either reading: on the first the word is epistemic and still needs an argument to become causal, and on the second it is not about foreknowledge at all, so it cannot show that foreknowledge determines anything. **Q: God is the Creator, not a spectator, so how can His knowledge be passive?** Grant the premise. God is not taking readings off an independent world, and on classical theism His knowledge is not derived from observation at all. What does not follow is that knowing and causing collapse into one act. Creating a world containing free creatures and knowing what those creatures do are two things, and the genuine question the premise raises is about responsibility for the arrangement, which is the creation-selection argument answered by feasibility or by concurrence, not the claim that foreknowledge is causation. **Q: Doesn't [Acts 2:23](/codex/acts-2-23/) say God planned the crucifixion, so the men who did it were not responsible?** The verse says the opposite, in one sentence. It attributes the event to the determined plan and foreknowledge of God, and the killing to the hands of godless men, joining both without softening either. Peter then tells that audience to repent ([Acts 2:38](/codex/acts-2-38/)). A writer who thought divine purpose absorbed human agency would not have built the sentence that way or ended the sermon that way.