Justified True Belief

Mapping the Landscape of Good Reasons for the Truth of Christianity

Christian Evidences

Evidence & Arguments for Christian Theism

Divinity of Christ

Biblical Evidence for High-Christology: Jesus is God

TBD

Old Testament Criticism

Evidence for the Reliability of the OT Bible

Common Objections

Objection Analyses to Christian Theism

World Religions

Critical Analyses of Non-Christian Religions

Philosophy

Phileō Sophia - to Love Wisdom.

Metaphysics: Free Will

Whether Our Choices Are Up to Us

God's Knowing What You Will Do Does Not Make You Do It

(●) God foreknows every choice his creatures will make, and his foreknowing takes nothing away from the one who chooses. + Tomorrow morning you will stand at the kettle and decide whether to phone your brother. Maybe you will. Maybe you will put it off one more week, the way you did last time. Whatever you settle on, two things Christians say about that moment sound, the first time anyone sets them side by side, as though they cannot both be true. The first is that God already knows which way it goes. Not guesses shrewdly, not reads you well. Knows, and has known since before there was a kettle or a brother or a you. The second is that the decision is yours. Nothing and nobody settles it except you, standing there with the water going off the boil. That is what it means for a choice to be free: nothing settles it but the one who makes it, for reasons of his own. Theological fatalism says the two cannot stand together. If God knew in advance what you would do, the fatalist argues, then you could not have done anything else, and a deed you could not have avoided was never really yours. That is a serious argument, and it has had serious friends. Its first half says that God's foreknowing a deed makes the deed unavoidable. Its second half says that an unavoidable deed was never really yours. Jonathan Edwards, the eighteenth-century American theologian, argued the first half at length: God's foreknowledge, he held, proves every choice necessary. He denied the second half, holding that a necessary choice is no less the chooser's own. Some Christians, known today as open theists, have found the argument so forceful that they gave up God's knowledge of future free choices rather than live with it. And its modern form, a carefully built argument, has occupied professional philosophers for sixty years and shows no sign of going away. The open theist's escape keeps human freedom by shrinking God. It costs far more than it saves, because a God who has to wait and see what happens is not the God who declares the end from the beginning. So the argument defends one claim: God's knowing what you will do does not make you do it. It is deductive, which means its premises work like a locked chain: if every link holds, the conclusion cannot slip out. The fatalist's argument, for its part, comes in two versions, and both start from something Christians affirm, that God knows every choice before it is made (P1). The popular one runs straight from God's being certain to your act's being unavoidable, and it goes wrong at a single step (P2). The serious one never takes that step. It leans instead on two other things, that God cannot be wrong and that nobody can change the past, and it needs a different answer (P3, P4). Who holds what, on theological fatalism: - Support: Jonathan Edwards argued its first half in 1754, and Nelson Pike gave it its modern form in 1965. - Oppose: Origen answered it in the third century, and Alvin Plantinga gave the answer this argument adopts. The argument that follows refutes it.

(P1) God knows every choice his creatures will ever make. + God's knowledge of the future leaves nothing out. It reaches the free acts of people as surely as it reaches the motions of the stars. (1) Declaring the future is the test Scripture uses to tell God from an idol. - Isaiah stages a trial and summons the idols to the stand. The challenge put to them has nothing to do with thunder or spectacle: "Tell us what is to come hereafter, that we may know that you are gods" (Isaiah 41:23). Knowing what has not happened yet is offered as the credential of deity itself. - God's own answer to that challenge is a claim to do the one thing the idols cannot. He is the one "declaring the end from the beginning and from ancient times things not yet done" (Isaiah 46:10). - He also says, in so many words, why he bothered to announce things ahead of time: "before they came to pass I announced them to you, lest you should say, My idol did them" (Isaiah 48:5). A prediction issued after the event proves nothing whatever about the predictor. The entire force of the claim is that it came first. (2) The knowledge reaches particular acts of particular people. - The claim goes further than the broad shape of history, and further than a report of what God has resolved to do himself. "Even before a word is on my tongue, behold, O Lord, you know it altogether" (Psalm 139:4). A word half-formed and about to leave a mouth is about as free and as small as a human act ever gets. - Jesus tells the disciples about the betrayal before it happens, and he says exactly why he is telling them: "I am telling you this now, before it takes place, that when it does take place you may believe that I am he" (John 13:19). The foretelling of one man's treachery is put forward as evidence that the speaker is who he says he is. - The prediction to Peter is specific enough that anyone could check it before sunrise: "before the rooster crows twice, you will deny me three times" (Mark 14:30). That is not a tendency or a temperament. It is a count and a deadline. - Cyrus, the Persian king, is named outright and told what he will do with Jerusalem (Isaiah 44:28-45:1). Who wrote those chapters, and when, is an old argument. Neither answer changes what the text is doing. • Most critical scholars assign them to the time of the exile, on several grounds at once: the audience addressed is already in Babylon, Jerusalem is spoken of as fallen and waiting to be rebuilt, and the vocabulary shifts. Old Testament scholars still defend a single author in print, and the argument between them belongs to their discipline. • On the traditional dating, the text names a man who has not yet been born. On the exilic dating, it still announces the fall of Babylon and the rebuilding of a ruined city before either had happened. Either way, announcing what is not yet done is offered as the credential of the God who speaks, which is what (P1) needs. (3) The verses that simply say God knows everything are not what carries the weight. - Two verses get quoted in this dispute more than all the others together, and neither one decides it. Hebrews says that "no creature is hidden from his sight" (Hebrews 4:13). John says that "he knows everything" (1 John 3:20). - Both verses are as broad as a sentence can be, and that is exactly the trouble. Neither one stops to say whether tomorrow's free acts are among the things there are to know, which is the very point in dispute. An open theist, one of the Christians who hold that the future is genuinely unsettled until it happens, can affirm both sentences without blinking. - The weight sits on the other texts, on the trial in Isaiah and the rooster in the courtyard. Those are not summaries of a doctrine. They are the argument God makes about himself, staged in a courtroom and tested against rivals who are invited to try the same thing. (4) A being who did not know the future would not be the greatest there could be. - The reasoning is old, and it is short. God is the greatest being there could possibly be, which is what Anselm meant when he called him that than which nothing greater can be conceived. A being who knows everything there is to know is greater than one who does not. So God knows everything there is to know. - Open theists have a reply ready for that. A free act nobody has performed yet is not sitting there waiting to be known, they say, so failing to know it is no more a defect in God than failing to draw a square circle is a defect in a draftsman. - That reply comes in two versions. On the first, there is as yet no truth at all about what you will freely do tomorrow. On the second, the truth is there, but nobody whatever, God included, could get at it in advance. - Both versions come at a steep price. Ordinary speech treats yesterday's "he will resign" as having been true or false all along. And when Jesus foretells one man's treachery at one supper table and offers it as a reason to believe that he is who he says he is, the foretelling only works if there was something there to get right in the first place (John 13:19). - The second version pays the steeper price of the two. A foretelling offered as a ground of faith needs more than a truth that was merely there. It needs a truth that was within reach of the one who spoke it, and that is exactly what the second version denies. (5) The texts that look like God learning speak of him in human terms. - These belong on the table before anyone else puts them there. God "regretted that he had made man on the earth" (Genesis 6:6). After the golden calf "the Lord relented from the disaster that he had spoken of bringing on his people" (Exodus 32:14). At Nineveh, "God relented of the disaster that he had said he would do to them, and he did not do it" (Jonah 3:10). - Read cold, with nothing else in view, those look like a God who did not see it coming. Pretending otherwise would be dishonest. - For the relenting texts, Jeremiah states the standing rule the stories are following. If God announces that he will pluck up a nation, "and if that nation, concerning which I have spoken, turns from its evil, I will relent of the disaster that I intended to do to it" (Jeremiah 18:7-8). The same policy runs in reverse for a promise (Jeremiah 18:9-10). So an announced judgment carries a condition whether or not the condition is spelled out, and a condition being met is not a surprise to the one who attached it. - The king of Nineveh gives the game away himself: "Who knows? God may turn and relent" (Jonah 3:9). The uncertainty in that story belongs to the Ninevites, who had no idea what was coming. It was never said to belong to God. - Genesis 6:6 has no announced condition anywhere near it, and it does not need one. The verse pairs the regret with grief of heart, and grief is not surprise. A father who saw from ten years off exactly how his son would turn out can still be cut open by watching it happen. - The same books say the opposite of what the surface suggests, sometimes on the same page. Samuel tells Saul that "the Glory of Israel will not lie or have regret, for he is not a man, that he should have regret" (1 Samuel 15:29), and Balaam tells Balak that "God is not man, that he should lie, or a son of man, that he should change his mind" (Numbers 23:19). Therefore, the future God knows includes the choices his creatures will make, down to the word before it is spoken.

(P2) A thing certainly known is not thereby a thing that had to happen. + Being certain that a thing will happen is not the same as its having to happen. The fatalist's argument, in the form most people meet, slides from the one to the other at a single step, and naming that step does more good than any amount of appeal to mystery. (1) Two sentences that look almost alike say quite different things. - Start in a kitchen. Your daughter is standing in the doorway and you are looking straight at her. Necessarily, if you see her standing in the doorway, she is standing in the doorway. Your seeing guarantees that much and not one thing more. It does not plant her feet. If she turned around and went back upstairs, you would simply be watching that instead. - Now put God where your eyes were, and tomorrow where the doorway was. Two sentences come out of that. They sound nearly identical, and only one of them is true. - True: necessarily, if God knows you will do something, you will do it. The word "necessarily" governs the whole of the if-then. Given the knowing, the doing follows, and there is no way to have the one without the other. - Not true: if God knows you will do something, then you will necessarily do it. Here "necessarily" has slid off the if-then and landed on the last part all by itself, where it says something else entirely: that the deed could not have gone otherwise. - Medieval theologians had names for the pair being confused, and Aquinas used them (Summa contra Gentiles I.67; Summa Theologiae I.14.13). • The necessity of the consequence is the airtightness of the inference, the fact that the second thing really does follow from the first. • The necessity of the consequent is the thing inferred being unable to have been otherwise. • One way to keep them straight is that the consequence is the whole if-then, and the consequent is only its second half. - The first necessity is true here. The second does not follow from it, and yet the fatalist's whole case needs it. - Anselm had drawn the same line before him, separating a necessity that forces a thing to happen from a necessity that merely follows it once it is happening (De concordia I). (2) Certainty describes the knower, not the event. - To be certain is to be without doubt, and that is a fact about a mind. For an event to be necessary is for it to have been unable to turn out any other way, and that is a fact about the world. Sliding from the one to the other is the engine of the whole mistake. - Think of it this way: a forecaster who has not been wrong in a thousand years says it will rain on Thursday, and on Thursday it rains. Her certainty did not fall on the fields. The rain could perfectly well have failed, and had it been going to fail, she would have forecast a dry Thursday instead, because that is what being that good at forecasting amounts to. - The fatalist moves from "God cannot be mistaken" to "the deed cannot be otherwise." One of those is a claim about a knower, and the other is a claim about an act. Nothing carries you from the first to the second. (3) The argument runs just as well with God left out of it. - Take God out of the picture and put in Smith, a neighbor who merely happens to be right about what you will do tomorrow. Necessarily, if Smith correctly believes you will do it, you will do it. Smith does correctly believe it. So, by the very same reasoning, you must. - Now take Smith out as well and leave the bare truth standing on its own. Necessarily, if it is true that you will do it, you will do it. It is true. So you must. - On that reasoning, every "I told you so" in history turns out to have been a sentence passed on the person told. The theological fatalist rejects that conclusion as firmly as anyone else does. - Be careful about what the substitution shows. It is a test of the inference, not a comparison between knowers. It would be no good at all to say that human foreknowledge compels nobody and therefore divine foreknowledge compels nobody, because a human forecaster can be wrong and God cannot. - What the substitution does show is narrower than that, and harder to escape. The inference uses nothing about the believer except that his belief is true. So whatever is wrong with it in Smith's hands is wrong with it in God's. - Pressed on this, the fatalist repairs his argument by putting back what the substitution left out. Now the claim is not merely that God believes. It is that his having believed is beyond anyone's reach, because the belief is lodged in a past nobody can get at and was held by a believer who could not have been mistaken. That is a different and a far better argument, and (P4) takes it on. (4) The popular version is an ancient puzzle. - Strip God out entirely and what is left is Greek logical fatalism, the fatalism you can run on a bare true statement. It is older than any Christian doctrine of foreknowledge, and adding God changes nothing in the reasoning except who finds it impressive. • Aristotle met it as the puzzle of tomorrow's sea battle: whether it is already true today that the fleets will fight (On Interpretation 9). • The Stoics, the Greek school that taught that everything happens by fate, were arguing over it before a word of the New Testament was written. - The rabbis held foreknowledge and freedom together without the least embarrassment. The Mishnah, the written record of their oral law, sets the two down side by side: everything is foreseen, and freedom of choice is given (Avot 3.15). (5) A constraint nobody can describe is not a constraint. - Notice how little the fatalist actually claims about the act itself. He does not say your choice was caused by anything. He grants that nothing pushed you, that no prior state settled it, that you weighed it up and decided. He says only that it was somehow fixed anyway. - Fixed by what? Not by a cause, since he has just conceded there was none. The answer usually given is that the constraint is logical rather than causal: what you do has to be consistent with a belief that was already held, and consistency really is a limit, even where nothing pushes. - Limits of that kind are lying around everywhere, and nobody calls them coercion. Think of a man who played basketball last winter. There is one thing he can no longer do, which is play basketball for the first time. That is impossible in the strictest sense there is, and yet it does not keep him off the court this evening. - Your case is the same. To say that your doing something else tomorrow would be inconsistent with a belief that was in fact held is a remark about which descriptions fit together. It is not a report of anything you are unable to do. - Clever arguments that something cannot happen are nothing new. Zeno, the ancient Greek who argued that motion is impossible, has one that says a flying arrow is at rest at every instant of its flight, and so never moves. His arguments are ingenious, they are genuinely hard to answer, and philosophers were still publishing solutions two thousand years later. Nobody ever concluded that the arrow was standing still. - Run the test yourself. Delete God's knowing from the picture, and the fatalist agrees the act is free. Put it back, and suddenly it is not. Nothing whatever about the act has changed in between. Nothing stands ready to step in if you waver, either, because knowing never steps in on anything. Knowledge that makes no difference to its object, and stands ready to make none, cannot be what makes the act unfree. It cannot even make the act unavoidable, which is all the first half of the fatalist's case asks of it. (6) The distinction does real work on a real case. - Luther called this very distinction "a play on words," and anybody who suspects he had a point deserves better in reply than a Latin phrase and a confident tone. - So here is the work it does. Suppose you will in fact keep an appointment tomorrow, and God knows it. Ask first what the fatalist needs in order to win: he needs it to be the case that you are unable to stay home. Now ask what he has actually shown, which is that if you stayed home, God would never have believed you were coming. - Those are not the same claim, and only the first of them is fatalism. The second is true and perfectly harmless, and (P3) and (P4) show why. Therefore, being certainly known by God is not a property that makes an act unavoidable.

(C1) Therefore, God's knowing a choice does not make it unavoidable. + God's knowing that you will do a thing does not make the thing unavoidable. That is what stands once the slide from "certainly known" to "had to happen" has been blocked. (1) The form is a general rule applied to a particular case. - The locked chain here has two links. If both hold, the conclusion cannot slip out; you either break a link or take what comes. - The general rule is (P2): being certainly known does not make a thing unavoidable. (P1) puts God's knowledge of your choices under that rule, since if any knowledge anywhere is certain, his is. Apply the rule to the case and the conclusion is simply sitting there: his knowing a choice does not make it unavoidable. - Neither premise has to be certain. Each has to be more believable than its denial, which is the standard a deductive argument is held to, and it is a real standard rather than a lenient one. A premise more likely false than true sinks the whole thing. - Denying each premise costs something quite different. Deny (P2), and the conclusion goes with it, at a price: every true guess anybody has ever made about the future turns out to have been a sentence passed on the person guessed about. - Deny (P1) instead and the conclusion survives with less work to do, because a God who does not know what you will say tomorrow was never going to threaten your saying it. That route costs a doctrine rather than refuting an argument, and it is the road open theism takes. (2) Knowledge is not a force. - In conversation the objection nearly always arrives with a picture attached. Foreknowledge gets imagined as a kind of advance pressure, a rail laid down ahead of the train, so that the knowing does something to the act. - It does nothing to the act, and (P2) is the reason. Nothing travels from the knower to the known. There is no rail. (3) The version philosophers press makes no such mistake. - The serious form of the argument grants every word of (P2). It agrees that God's knowing causes nothing at all, and it agrees that the two necessities are different things. - Then it makes a wholly separate move. God's believing happened in the past, and the past is over. Nobody can arrange for the Battle of Hastings to have been fought in 1067. - That argument commits no fallacy, and it is not easily dismissed. It is the real objection, and (P3) and (P4) answer it. Therefore, being certainly known by God does not make a choice unavoidable.

(P3) God knows a choice because it will be made, not the other way round. + Knowledge of what another person does conforms to what he does, wherever he is the one who settles it (D6, D15). That has been true of every knower there has ever been, and a knower who cannot possibly be wrong is no exception. (1) Knowledge of what others do takes its shape from what they do. - Your knowing that a friend is left-handed does not make him left-handed. He is left-handed, and that is why the belief you hold about him is a true one rather than a false one. - The same thing runs through every case anyone can think of. A true report depends on the event, and the event does not depend on the report. - The fatalist's picture turns that around. It stands the knowledge upstream of the act and lets the knowledge shape the act the way a mold shapes what is poured into it. Nothing in the idea of knowledge supports that, and everything in ordinary usage cuts against it. Foreknowledge follows an act the way a shadow follows a walker, and nobody has ever suspected the shadow of doing the walking. - Think of it this way: a wedding photograph is accurate because of what happened at the wedding. Nobody has ever supposed that the wedding happened because the photograph came out well. (2) God's knowing comes first in time and second in the order of reasons. - Two different orders are in play here, and each has a name. One thing is chronologically prior to another when it comes earlier in time, the way the wedding comes before the print. One thing is logically prior to another when it is the reason the other is as it is, the way the wedding is why the print shows what it shows. - So God's knowing is chronologically prior to the act, earlier in time, and logically posterior to it, which simply means later in the order of reasons. Once you have the two words, there is nothing strange about that pairing at all. - God's knowing does not depend on the act the way a photograph depends on a wedding, though. Light traveled from the wedding to the film, so there is a physical story to tell about the photograph. Nothing at all travels backward from a man's mowing his lawn on Saturday to a belief God held about it eighty years earlier. - The dependence is of another kind, the kind a true report has on the event it reports: the act is the measure God's knowledge has to match in order to be true (D15). (3) This answer goes back to the third century. - Origen gave it against Celsus, a pagan critic of the faith: a thing does not come about because God knows it; God knows it because it is going to come about (Contra Celsum II.20). - Augustine gave the same answer to the Roman statesman Cicero, who had reasoned that if the future were foreknown nothing would be in our power, and so gave up foreknowledge to keep freedom (City of God V.9-10). - John of Damascus fixed the formula the Eastern church has used ever since: God foreknows all things, but he does not predetermine all things (Exposition of the Orthodox Faith II.30). - Arminius, a Dutch theologian writing around 1600, put it in one sentence that could be carved over the whole dispute: "a thing does not come to pass because it has been foreknown or foretold; but it is foreknown and foretold because it is yet to come to pass." (4) A challenge from Aquinas denies the order itself. - God's knowledge, Aquinas says, is the cause of things, as a builder's knowledge is the cause of the house. On that picture the knowing would stand upstream of what it knows. - By his own account, though, it causes only when God's will is joined to it (D15). (5) The fatalist's hardest objection grants every word of this. - Admitting as much is no retreat, and it costs nothing. The serious critic does not claim that God's knowledge causes anything at all, and answering him as though he did would be answering a weaker man than the one standing there. - What he claims instead is that the knowledge is lodged in the past, eighty years back and out of anybody's reach, and that the past is closed for business. Answering that takes a further step, and (P4) is the step. Therefore, the direction of dependence runs from the act to the knowledge, and never from the knowledge to the act.

(P4) What God believed long ago about a choice depends on the choice itself. + A belief held long ago about tomorrow, by a believer who cannot be wrong, is exactly the kind of past fact that waits on tomorrow. That matters, because the strongest version of the objection never says that God's knowledge pushes anybody around. It says that God's belief is lodged in a past nobody can reach. (1) The objection at full strength: Jones and his lawn. - It went into print in 1965, and philosophers have been working on it ever since. Take Jones, a man who mows his lawn on Saturday afternoon. • Eighty years ago, God believed that Jones would mow his lawn on Saturday afternoon. • God is not merely accurate but essentially infallible. Being the God he is, he could not have held that belief and been wrong, and no creature could have made him wrong. • What happened eighty years ago is over and done with. Nobody now has any power over it at all. • So if Jones is genuinely able to refrain on Saturday, he must hold one of three powers: the power to make that eighty-year-old belief a false one, the power to bring it about that God held some other belief than the one he held, or the power to bring it about that the believer was never there. • The first is ruled out by what God is. The second and the third are powers over the past. Nobody has any of the three, so Jones was never able to refrain, and the mowing was not voluntary. - Nothing in that confuses the necessity of the consequence with the necessity of the consequent. It grants everything (P2) and (P3) establish, and it runs instead on two other things entirely: what God is, and the fixity of the past, the principle that nobody now has any power over what is over and done with. That is why the answer given in (P2) leaves it standing. - The last step is a further assumption of its own. It takes for granted that if Jones was never able to refrain, his mowing was not voluntary, and (C2) declines that step as well. (2) Facts about the past come in two kinds. - A hard fact about the past is one whose truth is complete in its own time: it was made true by how things stood then, and it waits on nothing still to come. That Socrates drank hemlock is a hard fact. Nothing that happens from here on has any bearing on it. - A soft fact about the past is one whose truth depends on what comes later. Take 1066. It is a fact about 1066 that the Battle of Hastings was fought that year, and that fact is hard as iron. It is also a fact about 1066 that it was the year of the last successful invasion of England, and that one is soft: whether it is true depends on whether another invasion ever succeeds, and if one does, it will turn out never to have been true. - From the outside the two look alike. Both are stated in the past tense and both are perfectly true. What separates them is which way they lean. A hard fact leans on nothing later. A soft fact leans forward, on what is still to come. - The labels are modern. The distinction behind them is not, because Ockham drew it in the fourteenth century, arguing that some past-tense statements are about the past only in their grammar. (3) God's believing something about a free act is a soft fact. - Start with what makes a past fact soft in the first place. It is soft when it depends on what is still to come, and God's ancient belief is exactly that. What he believed depends on what he knew, and what he knew depends on what will in fact happen, so the belief leans forward rather than standing there whatever the future does. - There is a second test, and it is the sharper of the two: ask what a fact about one moment guarantees about a later one. Picture a barometer on a kitchen wall, falling at noon. What does that noon reading guarantee about midnight? Nothing at all. It makes a storm likely, and likely is as far as it goes. A fact shut up inside its own moment behaves exactly like that reading. - Now set God's noon belief beside the barometer. His believing at noon that the storm will come does guarantee midnight, because the believer cannot be wrong. You cannot so much as say what he believed without saying what midnight holds. A fact that carries midnight around inside it was never a fact only about noon. (4) The usual case for calling the belief hard leaves out what it is about. - The objection needs God's ancient belief to be hard. That is a claim it owes an argument for, not something it may quietly help itself to. The case usually made is that believing is a state the believer was in at the time, and that how a thing stood at a moment is as fixed as the moment itself. - What that case leaves out is the content of the belief. A temperature can be described completely without mentioning anything outside the moment it was taken in. A belief cannot, because a belief is about something, and this one is about Saturday. - Take an ordinary case. "He believed on Monday that Friday's flight would land safely" reports a state of Monday, and in the same breath it reaches out to Friday. - Here the kind of believer makes all the difference. If he can be wrong, all that waits on Friday is whether his belief was true. If he cannot be wrong, then what he believed waits on Friday as well, since he believes only what will be so. That is what the barometer showed. - Which past facts are hard and which are soft is not something anyone can tell by looking. "She wore a blue coat that morning" and "she wore the coat she would later lose" are both about the same morning, and only the second of them waits on what happens next. (5) Jones needs no power over the past. - Jones cannot reach back and erase a belief. He cannot make a belief false. He cannot unmake the believer. All three really are impossible. The objection works only if those three exhaust his options, and that is the assumption it never stops to examine. - There is a fourth option, and it is an entirely ordinary one. Jones can do something other than what he will do, and if he were to do it, God would always have believed that instead. - Think of it this way: a woman took a job last year, and she could perfectly well have turned it down. Had she turned it down, God would never have believed she would take it. Nothing about the past has been reached back into. A different past would simply have been the one there was. - The whole difference is in the grammar. It is not "he can change what God believed." It is "were he to act otherwise, God would have believed otherwise." The first claims a power over the past. The second is a counterfactual, the philosopher's word for a statement about what would have been the case had things gone differently. - Counterfactuals point backward all the time, and nobody reaches back into anything. If the letter had arrived on Tuesday, it would have been posted on Monday. Nobody imagines that Tuesday's arrival went back and posted it. (6) The distinction was not invented to rescue foreknowledge. - It would be a fair complaint if the soft fact were a device built for this one job, and suspicion on that score is exactly the right instinct. - It is not such a device, because everybody needs the sorting, whatever they think about God. It was already true eighty years ago that you would get out of bed this morning at the hour you got out of bed, and that is a past-tense truth like any other. - If every past-tense truth were closed in the way the fatalist's argument needs, that plain fact about 1946 would have abolished your freedom at the alarm clock, with God left out of the story altogether. So the fatalist has to say why his principle does not reach that far, and the only answer available to him is that some truths about the past wait on the future. - The distinction earns its keep in places where God never comes up at all. Historians say that 1066 was the year of the last successful invasion of England. Commentators say that a match was a player's last. Nobody finds either remark mysterious, and nobody suspects it of having been invented for the occasion. Soft facts are ordinary furniture. Therefore, God's ancient belief about a free act is one of the facts that wait on the act, and not one of the facts that close it off.

(C2) Therefore, God's foreknowledge takes no choice out of the chooser's hands. + Nothing in God's foreknowledge takes a choice out of the hands of the one who makes it. A choice is free when nothing settles it but the one who makes it, for reasons of his own. God's knowing brings nothing about, so nothing in it settles what you do (P3). Nor does his knowing take away even the ordinary mark of a free choice, which is its usual sign rather than its definition: that you could have chosen otherwise, with everything before the choice just as it was (P4). (1) There are two roads from a fact to an unavoidable act. - Suppose some fact makes an act unavoidable. There are only two ways it could manage the trick. Either it is among the things that bring the act about, or it is itself beyond everybody's reach and guarantees the act, so that the act inherits that same fixity. - The first road is closed. Knowledge of what another person does conforms to what he does, so the knowing stands downstream of the act and is not among the things that bring it about (P3). - The second road is closed too. God's ancient belief is among the facts that depend on what will be done, so it is not beyond everybody's reach, and its pastness has no fixity to pass along (P4). - Think of the two roads as the only two ways a photograph could be blamed for a wedding. Either the camera arranged the marriage, or the picture was fixed in advance and the couple had to match it. Nobody believes either one, and the picture's being perfectly accurate does not supply a third. - Anybody who wants a third road has to name it. There is one serious candidate, and it travels in the opposite direction from the fatalist's. Rather than arguing from God's knowledge forward to the act's necessity, it argues that a genuinely open choice is not the sort of thing anyone could know in advance, so something must have settled the act before there was anything for God to know. That is a question about where foreknowledge comes from, not about what foreknowledge does (D5). (2) Nothing makes Jones mow the lawn. - From God's knowing that Jones will mow the lawn, it follows with certainty that Jones will mow the lawn. It does not follow that he must. - He can put the mower back in the shed and go golfing. He merely will not. And if he were to go golfing, God would have known that from eternity, and there would never have been a belief about mowing for anybody to worry over. - That ability is the ordinary mark of a free choice, and Jones has it even when the mark is worded at its strictest. "With everything before the choice just as it was" means holding fixed every hard fact: the weather, the lawn, the whole history of the world down to Saturday noon. It leaves out whatever itself waits on his choice, and God's ancient belief is exactly that (P4). Hold everything else where it was, and the shed door is still open. (3) The answer holds whichever way the open questions go. - How God knows a free act that has not happened yet is a different objection with a different answer (D5), and it is a good objection. - Whether God is timeless or everlasting in time is a question for the doctrine of God, and (P3) and (P4) run on either view. - Whether God knows what you will do because he decreed it, fixing from eternity that you would do exactly that, is the largest question in the neighborhood, and nothing from (P1) through (P4) answers it. Nor does it have to. They speak of the choice the fatalist has in view, one that nothing settles but the one who makes it. If God's decree settled a choice, the decree and not the knowledge would be what settled it (D6, D15). - A Christian who holds that God decrees every act and a Christian who holds that he decrees none can both say that foreknowledge as such takes nothing away. An answer that needs no verdict there is available to more people rather than fewer. (4) Freedom does not stand or fall on alternatives. - Suppose a fatalist somehow managed to show that a foreknown act could not have gone otherwise. He would still have to show that the man who acted was not its source, the one the act came from and the one who settled it. Nothing in the argument from foreknowledge so much as touches that question. - Picture a voter with an electrode that a neurosurgeon has secretly fitted in his brain, set to switch on and make him vote for one candidate if he were ever about to vote for the other. He votes for that candidate on his own, and the electrode never switches on. He could not have voted otherwise. The vote is still entirely his, because the electrode did nothing and nothing settled it but him. A leading defender of foreknowledge answers the fatalist with a case of just this kind. - This matters more than it looks at first glance. God cannot sin, and he is praised for his goodness. The saints in glory, the redeemed who are already home with him, cannot sin either, and they are freer than they have ever been. Any principle that made a missing alternative fatal to freedom would strip the freedom from both, and that is a price no Christian should be willing to pay. Therefore, God knows what you will do, and nothing in his knowing it takes the doing out of your hands.

On foreknowledge and freedom: Adams, Marilyn McCord. "Is the Existence of God a 'Hard' Fact?" Philosophical Review 76, no. 4 (1967): 492-503. Craig, William Lane. Divine Foreknowledge and Human Freedom: The Coherence of Theism I: Omniscience. Leiden: Brill, 1990. Craig, William Lane. The Only Wise God: The Compatibility of Divine Foreknowledge and Human Freedom. Grand Rapids: Baker, 1987. Craig, William Lane. Systematic Philosophical Theology, vol. IIa, On God: Attributes of God. Hoboken: Wiley-Blackwell, 2025. Finch, Alicia, and Ted A. Warfield. "The Mind Argument and Libertarianism." Mind 107, no. 427 (1998): 515-528. Fischer, John Martin, ed. God, Foreknowledge, and Freedom. Stanford: Stanford University Press, 1989. Freddoso, Alfred J. "Accidental Necessity and Logical Determinism." Journal of Philosophy 80, no. 5 (1983): 257-278. Hunt, David P. "On Augustine's Way Out." Faith and Philosophy 16, no. 1 (1999): 3-26. McKay, Thomas J., and David Johnson. "A Reconsideration of an Argument against Compatibilism." Philosophical Topics 24, no. 2 (1996): 113-122. O'Connor, Timothy, and Christopher Franklin. "Free Will." In The Stanford Encyclopedia of Philosophy, edited by Edward N. Zalta, 2022. Plantinga, Alvin. "On Ockham's Way Out." Faith and Philosophy 3, no. 3 (1986): 235-269. Zagzebski, Linda Trinkaus. The Dilemma of Freedom and Foreknowledge. New York: Oxford University Press, 1991. On freedom defined by its source: Craig, William Lane. "Free Will." Reasonable Faith, Question of the Week #320, June 3, 2013. Frankfurt, Harry G. "Alternate Possibilities and Moral Responsibility." Journal of Philosophy 66, no. 23 (1969): 829-839. On the doctrine of God's knowledge: Astrain, Antonio. "Congregatio de Auxiliis." In The Catholic Encyclopedia, vol. 4. New York: Robert Appleton, 1908. Beilby, James K., and Paul R. Eddy, eds. Divine Foreknowledge: Four Views. Downers Grove, IL: InterVarsity Press, 2001. Flint, Thomas P. Divine Providence: The Molinist Account. Ithaca, NY: Cornell University Press, 1998. Helm, Paul. Eternal God: A Study of God without Time. 2nd ed. Oxford: Oxford University Press, 2010. Helm, Paul. Reforming Free Will. Fearn: Christian Focus, 2020. Leftow, Brian. Time and Eternity. Ithaca, NY: Cornell University Press, 1991. Stratton, Timothy A. Human Freedom, Divine Knowledge, and Mere Molinism: A Biblical, Historical, Theological, and Philosophical Analysis. 2nd ed. Eugene, OR: Wipf and Stock, 2026. On the book of Isaiah: Motyer, J. Alec. The Prophecy of Isaiah: An Introduction and Commentary. Downers Grove, IL: InterVarsity Press, 1993. Oswalt, John N. The Book of Isaiah, Chapters 40-66. New International Commentary on the Old Testament. Grand Rapids: Eerdmans, 1998. Critic sources engaged: Bignon, Guillaume. Excusing Sinners and Blaming God: A Calvinist Assessment of Determinism, Moral Responsibility, and Divine Involvement in Evil. Eugene, OR: Pickwick, 2018. Boyd, Gregory A. God of the Possible: A Biblical Introduction to the Open View of God. Grand Rapids: Baker, 2000. Haack, Susan. "On a Theological Argument for Fatalism." Philosophical Quarterly 24, no. 95 (1974): 156-159. Hasker, William. God, Time, and Knowledge. Ithaca, NY: Cornell University Press, 1989. Pike, Nelson. "Divine Omniscience and Voluntary Action." Philosophical Review 74, no. 1 (1965): 27-46. Pike, Nelson. God and Timelessness. London: Routledge and Kegan Paul, 1970. Pinnock, Clark H., Richard Rice, John Sanders, William Hasker, and David Basinger. The Openness of God: A Biblical Challenge to the Traditional Understanding of God. Downers Grove, IL: InterVarsity Press, 1994. Prior, A. N. "The Formalities of Omniscience." Philosophy 37, no. 140 (1962): 114-129. Sanders, John. The God Who Risks: A Theology of Divine Providence. Downers Grove, IL: InterVarsity Press, 1998. Taylor, Richard. "Fatalism." Philosophical Review 71, no. 1 (1962): 56-66. van Inwagen, Peter. An Essay on Free Will. Oxford: Clarendon Press, 1983. van Inwagen, Peter. "Free Will Remains a Mystery." Philosophical Perspectives 14 (2000). Ancient and medieval sources: Anselm, De concordia I. Aquinas, Thomas, Summa contra Gentiles I.67. Aquinas, Thomas, Summa Theologiae I.14.8. Translated by the Fathers of the English Dominican Province. 2nd rev. ed. London: Burns Oates & Washbourne, 1920. Aquinas, Thomas, Summa Theologiae I.14.13. Aquinas, Thomas, Summa Theologiae I.83.1. Aristotle, On Interpretation 9. Augustine, City of God V.9-10. Augustine, On the Trinity XV.13.22. Translated by Arthur West Haddan. In Nicene and Post-Nicene Fathers, First Series, vol. 3, edited by Philip Schaff. Buffalo: Christian Literature Publishing, 1887. Boethius, The Consolation of Philosophy V.3-6. Cicero, On Fate 28-30. Duns Scotus, John, Lectura I, d. 39. John of Damascus, An Exact Exposition of the Orthodox Faith II.30. Mishnah, Avot 3.15. Ockham, William of, Predestination, God's Foreknowledge, and Future Contingents. Translated by Marilyn McCord Adams and Norman Kretzmann. 2nd ed. Indianapolis: Hackett, 1983. Origen, Contra Celsum II.20. Reformation and later: Arminius, Jacobus. The Works of James Arminius. Translated by James Nichols and William Nichols. London, 1825-1875. Edwards, Jonathan. Freedom of the Will. 1754. Part II, sections 11-12. Luther, Martin. The Bondage of the Will. 1525. Translated by J. I. Packer and O. R. Johnston. Westwood, NJ: Fleming H. Revell, 1957. Biblical citations: Genesis 6:6; 18:12-13, 21; 22:12; Exodus 32:14; Numbers 23:19; Deuteronomy 8:2; 1 Samuel 15:29; 2 Kings 20:6; Psalm 139:4; Isaiah 41:23; 44:28-45:1; 46:10; 48:5; Jeremiah 18:7-10; Jonah 3:9-10; Matthew 26:24; Mark 14:30; John 13:19; 19:11; Acts 2:23; Hebrews 4:13; 1 John 3:20.
+ The "soft fact" is a device invented to save the doctrine. God's believing something eighty years ago is a fact about eighty years ago, and saying you have power over it is saying you can change the past.
1. This is the objection that matters most. It comes from people who deserve a hearing. John Martin Fischer and Linda Zagzebski have both pressed it hard, and neither of them is a fatalist looking for trouble. Both simply think the Ockhamist reply falls apart on inspection. The reply takes its name from William of Ockham, the fourteenth-century English friar who argued that a past-tense truth can be about the past only in its grammar, and that is the distinction (P4) puts to work. 2. Nobody is claiming power over the past. Start by conceding everything that should be conceded. Jones cannot make God's past belief false. He cannot erase it. He cannot reach back and alter one second of 1946. If the reply needed any one of those, it would be finished on the spot. It needs none of them. What it says is an if-then statement about what would have been the case, and not a claim about what can be done to what was: had Jones acted otherwise, God would always have believed otherwise. That sentence asserts no power to change anything whatever. It describes a dependence. 3. A different past is not an altered past. Take a case with no theology anywhere in it. There is a letter in your drawer and you will post it tomorrow. That being so, it is already true today that the letter will be posted. Now suppose that tomorrow you burn it instead. You will not have changed today's truth about the letter. There will simply never have been that truth in the first place, and today will have contained a different truth all along. Nothing has been reached into. The past that would have been is a different past rather than an altered one, and the difference between those two ideas is exactly the difference between the objection and the reply. 4. Softness is not a special exemption carved out for God. The charge of convenience would stick if soft facts appeared nowhere else. They appear everywhere. "She wore the coat she would later lose" is a fact about a morning that waits on an afternoon. "That was his last game" waits on everything that follows. Whole classes of ordinary past-tense truths lean forward like this, and no one finds them mysterious. So what the objection has to show is not that softness is odd, but that God's belief in particular is hard. There is a real argument for that. 5. The best case for hardness turns on God's intrinsic state. The serious version does not merely assert that the past is fixed and leave it there. It offers a test for telling hard facts from soft ones, and the test is intrinsicness: a fact about a moment is hard when it concerns only how things stood in that moment, borrowing nothing at all from later ones. - That Socrates drank hemlock passes the test easily. - That 1066 was the year of the last successful invasion does not, since it borrows from every century since. - Believing, the objection goes on, is a state the believer was in at the time, as much a feature of that moment as his height. So God's believing eighty years ago is hard by the very test the Ockhamist needs. Grant the test. A belief still fails it, and for a reason peculiar to beliefs. A height can be given completely without mentioning anything outside the moment it was measured in. A belief cannot, because a belief is about something, and this one is about Saturday. Try describing God's state eighty years ago without mentioning Saturday anywhere, and what you will have described is some other belief, or else no belief at all. Whatever else God's ancient believing is, it is not shut up inside the moment it was held in. None of this makes every belief soft. A fallible believer's belief about Saturday is hard (D13): only its truth waits on Saturday, and the belief itself waits on nothing still to come. What a belief's being about Saturday shows is only that the intrinsicness test cannot be what decides the matter for God's belief. Nor does any of this end the dispute, and saying it did would be dishonest. What it does is put the burden back where it started. The objection cannot simply help itself to the fixity of the past, because the fixity of the past is the thing in question. What it owes is a test that sorts hard facts from soft ones and lands beliefs on the hard side, and the best test it has produced does not do that. 6. The critics' own verdict is narrower than the objection sounds. Zagzebski's conclusion is not that the Ockhamist answer is absurd. It is that every available answer to the dilemma pays somewhere, her own included, and that is an honest report of where the literature stands rather than a verdict handed to the fatalist. Of the bills on offer, the Ockhamist one looks like the smallest (D8). It gives up nothing about God, it needs no theory of time, and it asks only that a belief about the future be the kind of past fact that depends on the future. That is a judgment about costs rather than a proof, and it can be argued with. What it is not is a concession that the argument from the fixity of the past goes through. See also: Philosophy / Free Will: Counterfactuals of Freedom Can Be True: The Grounding Objection
+ Boethius settled this fifteen hundred years ago: God is outside time, so he does not foreknow anything, he sees it all at once. Why go through the soft facts at all?
1. The Boethian answer is real, not a dodge. It is one of the church's oldest replies to the problem. God's life, Boethius wrote, is the whole and perfect possession all at once of unending life, so every moment is present to him. He does not look ahead to your Thursday. He simply sees it, the way you see the room you are sitting in. Aquinas took the answer over, and its modern defenders are serious philosophers. The picture is genuinely helpful. If nothing is earlier or later for God, there is no ancient belief sitting in a closed past, and the argument from the fixity of the past (P4) loses its first step. 2. Timelessness by itself does not dispose of the argument. Here is the difficulty, and there is nothing obscure about it. What the fatalist's argument really runs on is not pastness at all. It runs on a fact being beyond everybody's reach, and pastness was only the most obvious example of such a fact. A timeless eternity is at least as far beyond reach as eighty years ago is, since nothing anybody does can get behind it or in front of it. Zagzebski built that parallel deliberately, and the dilemma duly comes back in a timeless form, with the fixity of eternity doing exactly the work the fixity of the past had been doing. The answer turns out to be the same one, and it transfers without strain. A fact is soft when it waits on what will happen, and nothing in that definition requires the fact to be a past one. The present truth that God timelessly knows your Thursday waits on your Thursday just as the ancient belief did. Defenders of the timeless view have a reply of their own here: eternity is not a past at all, they say, and a rule about the fixity of the past cannot simply be carried across to it. That exchange is still running, and it does not need to be decided either way. On either verdict the soft-fact answer finishes the job, which is why it was carrying the weight all along. 3. Neither premise needs a verdict on whether God is in time. Nothing in (P3) or (P4) requires one. If God is timeless, his knowledge is of an eternal present and the dependence still runs from act to knowledge. If he is everlasting in time, the same holds and the soft fact answers the fixity. That is a feature rather than an evasion. Whether God is timeless is a question for the doctrine of God, Christians of every communion are found on both sides of it, and an argument about human freedom that forced a choice there would be overreaching. See also: Philosophy / Free Will: Does God See All Times at Once? Eternity and the Boethian Answer
+ The distinction between the two necessities is a play on words. Luther said so, and he was right: wherever you park the word "necessarily," if God knows it then it is going to happen, and that is the end of your freedom.
1. Luther really did say it. He called the medieval theologians' distinction a play on words and thought it a piece of evasion, and anyone who has watched a philosopher slide a "necessarily" around a sentence will feel the force of the complaint. A distinction that cannot be cashed out for anything is a trick. A charge like that has to be met with a plain answer, not out-argued with a cleverer one. So the test is whether this particular distinction does any work, not whether it can be stated in Latin. 2. On a real case the two readings come apart. Suppose you will keep an appointment tomorrow and God knows it. - What everyone grants: you will keep it, and there is no way things could go in which God knows you will keep it and you do not. - What the fatalist needs: that you are unable to stay home. Those two come apart at once. From the first it follows that you will go. To get the second you need the further claim that you cannot stay home, and nothing whatever in the first supplies it. If you were to stay home, the situation would not be one in which God had known wrongly. It would be one in which God had always known you would stay home. 3. The charge convicts the objection, not the reply. Grant Luther his instinct that a merely verbal distinction proves nothing, and then apply that instinct evenhandedly. It is the fatalist's argument that depends on where a word sits. Move "necessarily" from the front of the if-then to the middle and his conclusion appears out of nowhere; move it back and the conclusion vanishes again. The reply is the one pointing that out. It is not the one relying on it. 4. Luther had already denied the freedom the distinction protects. He was not confused about any of this. On his view of the will there is nothing left for the distinction to save, so of course it looked idle to him. But that is a substantive position about human freedom rather than a point about logic, and it has to be argued on its own merits. Which means the appeal to Luther proves rather less than it seems to. It reports the verdict of a man who had already decided the larger question.
+ If God has known since before I was born what I will do, then it is all settled already. Nothing I do makes any difference, so I may as well stop trying.
1. This is the objection people actually live with. It usually arrives after a diagnosis, or a wasted decade, or a prayer that went unanswered. It is not a debating move, and it deserves a straight answer rather than a clever one. The thought behind it is that if the film has already been shot, watching it attentively changes nothing. The answer is not clever. It is that the objection runs two different things together under the one word "settled," and only one of them would make trying pointless. 2. Known in advance is not the same as settled without you. What is true is that there is a fact of the matter about what you will do, and that God knows it. That makes the outcome certain. But certainty settles nothing, because to settle a choice is to be what makes it go the way it goes, which is more than making its outcome certain. What would make trying pointless is something else: an outcome settled without you, so that it would come out the same whatever you did. Turn it around for a moment. What God foreknows about tomorrow is what you will in fact decide, weighing and hesitating and changing your mind and all. Your deliberating is not bypassed by his knowing it; your deliberating is part of what he knows. Take the objection seriously and notice what it quietly asks of you: to picture your own weighing up as a formality, a going through of motions whose outcome was fixed somewhere else. There is nothing in foreknowledge that makes it one. 3. The same reasoning would retire every effort anyone has ever made. That is how you can tell it is broken. The argument makes no use at all of the fact that the thing foreknown is a choice, so it runs just as well on anything else. - It is already settled whether the crop will come in, so why plant. - It is already settled whether the patient will recover, so why operate. - It is already settled whether the brakes will hold, so why have them checked. Nobody lives that way, the person making the objection included, and the reason is that in each of those cases the outcome runs through the effort. Remove the effort and you have not uncovered an outcome that was waiting underneath it all along. You have described a different outcome altogether. 4. Neither Scripture nor the church has ever reasoned this way. The same Bible that has God declaring the end from the beginning also has Paul working harder than any of the other apostles, Daniel praying three times a day with the windows open, and a whole city of Ninevites fasting in sackcloth on the bare chance that God might relent. Not one of those people was ever told that his effort was decorative. The mistake is old enough to have a classical name. Ancient philosophers called it the lazy argument, and it went like this: if you are fated to recover, calling the doctor is pointless, and if you are fated not to recover, calling him is pointless again. Chrysippus the Stoic answered that the recovery and the calling are fated together, which is the same observation in older words. The outcome runs through the effort (Cicero, On Fate 28-30). Fatalism, the belief that what will be will be no matter what anybody does, is not a Christian doctrine, and the church has been refusing it since the Fathers, its earliest teachers, first argued with the Stoics.
+ The real problem is not whether foreknowledge rules out freedom; it is that there is nothing yet for the knowledge to rest on. A free choice that has not been made yet leaves God nothing to go on.
1. This is a better objection. Credit where it belongs. A. N. Prior pressed it only after granting the Ockhamist reply on its own ground. He allowed that a past truth about the future can genuinely depend on what is still to come. Then he asked the awkward question: what could a past truth of that kind possibly be known on the basis of, given that the events it leans on have not happened yet? Without some present ground for the belief, he thought, the alleged knowledge is no better than a lucky guess that happened to come off. That is not fatalism, and it does not commit the mistake (P2) identifies. It asks how foreknowledge is possible at all, rather than whether foreknowledge and freedom can both stand. 2. The same objection is pressed from the Reformed side. The Reformed side, the Protestant tradition that runs back to John Calvin, presses the version you are likelier to meet in a church hall, and it runs this way. A choice that nothing before it settles, one that could go either way right up to the moment it is made, is not the kind of thing anybody could know infallibly in advance. So if God does know it infallibly, something must have settled it beforehand, and the only candidate for that job is God. This is a serious argument, and it is not the fatalist's argument. It runs from the possibility of the knowledge back to the settledness of the act, rather than from the knowledge forward to the act's necessity. The complaint is not about time at all, and its defenders show as much by what they decline to use. Its ablest modern advocate holds that God is timeless, which puts the Boethian escape within easy reach, and he does not take it. What he wants to know is what the knowledge rests on. 3. The objection assumes that God knows by looking. It takes for granted that there has to be a present scene for God to inspect, and that if the scene is not there yet then neither is the knowledge. Call that the perceptualist picture, the picture of knowing as a kind of seeing. It is what most people import without noticing when they say that God "sees" the future, and it sits behind almost every modern denial of foreknowledge. There is no reason God's knowledge should work that way, and there is good reason to think it does not. Nobody has ever asked through what window God sees that two and two make four. 4. (P3) and (P4) say nothing about how God knows. (P3) says God knows a choice because it will be made. (P4) says his ancient belief depends on the act. Neither one says a word about how the knowing gets accomplished, and neither one needs to. So this objection, even if it succeeded completely, would leave both premises standing and strike instead at omniscience, God's knowing everything there is to know, from a different side. Those are two different complaints. Keeping them apart is the difference between answering the man and talking past him. 5. How God knows a free act has answers of its own. The short form of the answer is that God does not acquire his knowledge by any process at all, because a perfect being knows every truth there is simply in virtue of what he is. The fuller form adds an account of the how, given in terms of what God knows every creature would freely do in any circumstance he might be placed in. Philosophers call that middle knowledge (D12). Both of those belong to the question of how foreknowledge is possible, and neither is needed to answer the fatalist. See also: Philosophy / Free Will: God Does Not Look Ahead: How Foreknowledge Is Possible Philosophy / Free Will: God Knows What You Would Freely Do, and Plans Around It
+ You have the answer backwards: God knows what you will do because he decreed that you would do it. Foreknowledge is no puzzle at all once you see that it rests on his will.
1. The decree reading solves the problem completely. Start by granting that this is not a muddle. If God has decreed everything that comes to pass, settling from eternity that each thing will happen exactly as it does, then his knowledge of the future is knowledge of his own intentions. No question arises about what the knowledge rests on, and none arises about the fixity of the past. The whole puzzle simply dissolves. It is also the answer of a large part of the church, held by people who have thought about it longer and harder than most of us ever will, and it will not do to treat it as a failure of nerve. 2. It buys the solution with the freedom it was meant to protect. If God's knowing that Jones will mow rests on a decree that settles the mowing itself, then what makes the mowing go the way it goes is the decree, not Jones. The decree is the mowing's source. Be exact about where the damage lies. It lies in the settling: a decree that settles the choice itself leaves Jones nothing to settle. It does not lie in a shut door, with no other way left open to Jones, because a resolve can shut a door without settling the choice. God could resolve to place a man only where he would freely choose one way, as (D10) supposes of Judas, and the choosing would still be the man's. That damage is not some incidental cost to be absorbed. It is the very conclusion the fatalist wanted, arrived at by a friendlier road. Foreknowledge did not take Jones's freedom. The decree did. 3. The decree relocates the question instead of answering it. Whether God governs by determining each choice, and whether a decreed sin leaves its sinner responsible, is an argument of its own and a long one. What matters for (P3) and (P4) is that this objection does not defeat them. It declines the question by answering a larger one first. Not one premise needs a decree, and foreknowledge as such takes nothing away, which is precisely what somebody who does not hold the decree reading needs to hear. The larger dispute can stay where it belongs. Some who do hold the decree reading decline the argument from foreknowledge as well. At least one careful Reformed philosopher lists it among the arguments for his own position that he declines to use, on the ground that he does not find it compelling. That is worth knowing, because it shows that saying foreknowledge takes nothing away is not a party line. See also: Philosophy / Free Will: Theological Determinism and the Authorship of Sin Philosophy / Free Will: What the Sovereignty Texts Say, and What They Do Not
+ There is a simpler solution: give up foreknowledge of free choices. God knows everything knowable, future free choices are not knowable, and nothing is lost but a philosopher's puzzle.
1. The motive is the right one. Open theists are not trying to shrink God. They are trying to keep human freedom real, which is exactly what the fatalist's argument threatens, and they have concluded that his argument works. Given that conclusion, dropping foreknowledge is the consistent move to make, and it takes some courage to make it. So the disagreement is narrower than it looks from outside. It is a disagreement about whether the fatalist's argument works, and (P2) through (P4) are the case that it does not. 2. If the argument fails, the trade buys nothing. The only reason anybody pays a price is to get something in return. If foreknowledge never threatened freedom in the first place, then surrendering it purchases exactly nothing, and a doctrine will have been given up for a fallacy. 3. What is given up is not a philosopher's puzzle. Isaiah puts the ability to declare what is coming forward as the credential that separates God from an idol (Isaiah 41:23), and says the announcements were made in advance so that nobody could credit the idols afterwards (Isaiah 48:5). Jesus foretells a betrayal so that the disciples "may believe that I am he" (John 13:19). Peter is told his denials with a count and a deadline (Mark 14:30). Those texts do not merely report that God happened to know a few things. They make advance knowledge a mark of deity and a ground of faith. 4. The open theist has answers to every one of those texts. He is not stuck for a reply, and he has a reading ready for each. - The relenting passages describe a God who genuinely changes course. - The prophecies are announcements of what God will himself do, or forecasts drawn from causes already in place. - Omniscience means knowing every truth there is, and future free acts are not any of them. Each of those deserves an answer at length rather than a paragraph. Whatever their merits, the argument that forced the trade does not work, and so the trade is not forced. See also: Philosophy / Free Will: The Future Is Not Open to God: Answering Open Theism
+ Every proposed solution to this problem has a serious cost. Picking one and calling it the answer is just choosing which bullet to bite.
1. Every answer on offer does cost something. Zagzebski's survey says exactly that, and saying so is a good deal better than pretending the question is closed. Here are the five bills, three of them named for the man whose answer it is. - The Boethian answer, which puts God outside time altogether, needs a theory of timeless knowledge, and that raises difficulties of its own. - The Ockhamist answer, named for the fourteenth-century friar William of Ockham, needs the line between hard and soft facts to be principled rather than merely convenient. - The Molinist answer, named for the sixteenth-century Jesuit Luis de Molina, needs it to be true, right now, what each person would freely do in any situation he might ever be placed in. - The decree reading, on which God knows what you will do because he has settled that you will do it, costs the freedom the whole dispute was about (D6). - Giving up foreknowledge, which is the open theist's route, costs God his knowledge of what free creatures will do. Those who take it say that is no loss to omniscience at all, since on their view there was never anything there to know. Their critics answer that Isaiah puts announcing what is not yet done forward as the credential separating God from an idol (Isaiah 41:23), so what is being surrendered is not a philosopher's convenience. Nobody walks away from this clean. That is the honest state of the field, and it belongs on the table before the bills get compared. 2. A cost is not a refutation. Every position in a hard field carries a price. The question is never whether a view costs something, but what it costs next to what its rivals cost, and that comparison is a judgment anybody can make once the bills are itemized. Of the five, the Ockhamist bill looks like the smallest. Set them side by side: - The Boethian answer asks for divine timelessness and still needs help with a present-tense version of the argument. - The Molinist answer asks that a whole class of if-then statements be true with nothing in the world yet to make them so, which is a real bill even if it can be paid. - The open theist's answer gives up God's knowledge of what you will freely do. - The decree reading gives up the very freedom the whole dispute was about. - The Ockhamist answer asks only that a belief about the future be a fact about the past that depends on the future. It is the one item on the list that gives up nothing whatever about God. 3. Answering one argument does not require tidying the whole field. Nobody has to claim that the relation between divine knowledge and human freedom is now fully understood. What fails is one particular argument, the one running from foreknowledge to fatalism. A negative result of that kind can be perfectly secure while the positive account is still being argued over.
+ Scripture has God regretting that he made man, relenting at Nineveh, and changing his mind after Moses argues with him. That is not a God who knew what was coming.
1. The texts carry real weight. God "regretted that he had made man on the earth" (Genesis 6:6). After the golden calf "the Lord relented from the disaster that he had spoken of bringing on his people" (Exodus 32:14). At Nineveh "God relented of the disaster that he had said he would do to them, and he did not do it" (Jonah 3:10). Hezekiah is told he will die, prays about it, and is then told "I will add fifteen years to your life" (2 Kings 20:6). Anybody who answers all that by saying the texts do not really mean anything has answered nothing at all, and everyone listening will know it. 2. The same Scriptures state the rule these stories are obeying. Jeremiah lays out the standing policy in so many words. When God announces that he will pluck up a nation, "and if that nation, concerning which I have spoken, turns from its evil, I will relent of the disaster that I intended to do to it" (Jeremiah 18:7-8), and the reverse holds for a promise (Jeremiah 18:9-10). That is not an escape hatch invented by commentators with a doctrine to protect. It is the text explaining its own idiom. An announced judgment carries a condition whether or not the condition is spoken aloud, and a condition being met is not news to the one who set it. Hezekiah's fifteen years are the same rule at work: a sentence announced, a prayer, a sentence lifted. 3. The harder texts are the ones about knowing. Jeremiah's policy handles the relenting passages cleanly enough. It does not handle everything, and the texts it leaves standing are the harder ones. At the mountain the angel stops Abraham's hand, "for now I know that you fear God" (Genesis 22:12). Moses tells Israel that God led them forty years in the wilderness "testing you to know what was in your heart" (Deuteronomy 8:2). Those are sentences about knowledge, with no announced threat anywhere in view, and the word "now" in the first of them is doing exactly the work the objection wants done. Anyone who says the objection has no good texts has not read them. Two things can be said in reply, and neither one is a trick. First, a test is how a thing is shown and not only how it is found out. The knife over Isaac made public what had until then been private, and Deuteronomy describes those forty years as forming Israel at least as much as informing anybody about Israel. Second, taking the testing texts as discovery means taking Genesis 18:21 the same way, and then God does not know the present either, which is a good deal further than the objection wants to go. 4. The same books deny the surface reading in the plainest words available. Samuel tells Saul that "the Glory of Israel will not lie or have regret, for he is not a man, that he should have regret" (1 Samuel 15:29). That sentence sits in the same chapter as one of the strongest relenting texts in the Bible. Balaam tells Balak that "God is not man, that he should lie, or a son of man, that he should change his mind" (Numbers 23:19). Any reading of these books has to hold both sets of texts at once. A reading that takes the relenting literally and the denials figuratively has made a choice, and so has a reading that goes the other way. The difference between them is that only one has a stated rule to appeal to. 5. The literalism does not stay where it is put. This is the part that usually decides it. The same method that reads Genesis 6:6 as ignorance of the future has to read God's "I will go down to see whether they have done altogether according to the outcry that has come to me" (Genesis 18:21) as ignorance of the present, and it has to do so in a chapter where God knows what Sarah said to herself under her breath (Genesis 18:12-13). Press the method further and it takes in the eyes, the hands, the nostrils and the outstretched arm, and arrives at last at a God with a body. Almost nobody wants that destination, and the reason nobody wants it is that everybody already reads these texts as speaking of God in human terms. The only question is how far the practice extends, and nobody applies it consistently in the direction the objection needs. See also: Philosophy / Free Will: The Future Is Not Open to God: Answering Open Theism
+ Judas was foretold and Peter was foretold, by name and in detail. Whatever the philosophy says, those two men could not have done otherwise, and one of them was condemned for it.
1. This is the objection with a face. An abstract argument about Jones and his lawn can feel like a shell game. Judas cannot. He is named, the betrayal is announced out loud at the table, and then the woe is pronounced on him for doing the very thing announced: "The Son of Man goes as it is written of him, but woe to that man by whom the Son of Man is betrayed!" (Matthew 26:24). If foreknowledge takes freedom away, this is where it will show. And if it does not, this is where the answer has to hold. 2. The text puts the plan and the guilt in one sentence. Notice what Jesus does not say. He does not say that the betrayal was written and therefore Judas is excused. He does not say that Judas was free and therefore the Scriptures were only guessing. He says both halves in a single breath, and of the two halves the woe is the heavier. Peter preaches the same structure about the crucifixion. Jesus was "delivered up according to the definite plan and foreknowledge of God," and the men listening to Peter had "crucified and killed by the hands of lawless men" (Acts 2:23). The plan does not make the hands any less lawless. 3. (P3) and (P4) explain why both halves stand. Apply them to Judas directly. What was written was written because Judas would do it, so the knowing follows the act rather than shaping it. And had Judas not done it, the announcement would never have been made in those terms in the first place. That is not a dodge. It is the ordinary case showing up in an extreme instance. The prediction is exactly as reliable as it looks and exactly as powerless as any other piece of knowledge. 4. A spoken prophecy is a harder case than a silent belief. There is a sharper version of this objection and it deserves its own hearing. A belief God keeps to himself may well be harmless. A prediction spoken out loud at a supper table is a public commitment, and God's own faithfulness is now bound up with its coming true. Whatever was the case beforehand, the argument runs, once Jesus had said the words, Judas had to go through with it or God's word would fail. Grant the first half of that without hesitation: a spoken prophecy does engage God's faithfulness. Some announcements are declarations of what God will himself do or bring about, and those belong to providence, God's active governing of what happens, rather than to bare foreknowledge. To affirm foreknowledge as such is not to claim that God merely watched. The second half is the part that does not follow. What God commits himself to is what he knows, and what he knows is what Judas would do. Had Judas not done it, the words spoken at the table would never have been those words, any more than a forecast of Thursday's rain would have been that forecast if Thursday were dry. The faithfulness of God is engaged to the truth, and the truth waited on Judas. 5. Providence is a further question. Somebody who thinks God arranged the circumstances, or hardened a heart that was already set, is asking about providence rather than about foreknowledge. Whether God did so is a genuine question with a long history behind it. Suppose he went further still, and resolved to place Judas at that table only if Judas, placed there, would betray. Then there is one sense, and only one, in which no other way lay open to Judas there: God's resolve and Judas's refusal could not both have happened. Even so, nothing would have stopped him from refusing. The betrayal would still be his own, because a resolve of that kind settles where a man stands and not what he does there. Scripture speaks of Pilate, the Roman governor who handed Jesus over to be crucified, in just these two breaths. Jesus tells him that his authority was given from above, and that he still bears a sin: "You would have no authority over me at all unless it had been given you from above. Therefore he who delivered me over to you has the greater sin" (John 19:11). A lesser sin is still a sin of his own. God need not arrange things this way, and in the ordinary run of choices the door stands open. What excuses a man is something other than him settling the deed: somebody or something else making it go the way it went. Nothing in a prophecy, or in God's placing Judas where he stood, did any of Judas's betraying. So foreknowledge settles nothing about Judas. The foretelling as such did not take his act out of his hands, and the Gospel writers plainly did not think that it had. See also: Philosophy / Free Will: Black the Neurosurgeon: Freedom Never Depended on the Open Door Philosophy / Free Will: God Knows What You Would Freely Do, and Plans Around It
+ This has nothing to do with God: if it is true right now that you will do something tomorrow, then you will do it, and that is that. Adding an omniscient knower is decoration.
1. The objection is right that God adds nothing to the reasoning. This is the old Greek fatalism with the theology stripped out of it, and there is no trap here. The argument really does run without God. Necessarily, if it is true that you will do a thing, you will do it; it is true; therefore you will. Add nobody at all to that and fatalism follows just as fast as it did before. Richard Taylor, who defended fatalism, needed no God for it whatever. His version ran on nothing but a statement's being true in advance, and the decade of replies it drew never had to mention God either. The theological version is that same argument with a detour through a believer, and the detour adds nothing at all to the reasoning. 2. The problem was never foreknowledge. Follow the consequence out. If the reasoning works, then the mere truth of a statement about tomorrow makes tomorrow unavoidable, and every true prediction anybody has ever made was a sentence passed on the person predicted about. If the reasoning does not work, then it does not work with God added to it either. Either way, God drops out of the argument and the Christian doctrine of foreknowledge is not the thing on trial. That is a considerable relief. 3. The fatalist's usual repair is harder to make out than it sounds. Pressed on this, the fatalist says that God's holding a belief is a fact about the past in a way that a bare statement's being true is not. He needs a line there, with the belief on the side of the past and the bare truth on the other side. That distinction has to be made out, and attempts to make it out keep running into ordinary usage. - Take a man who believed in 1997 that something would happen. His believing it was a fact about 1997, and the truth or falsity of what he believed was every bit as much a fact about 1997. When the thing failed to happen, it became a fact that what he had believed in 1997 was false. - A statement's changing from true to false looks a good deal like an event, too. "The flight departs in five minutes" is false all morning, true for one moment, and false again afterwards. Sorting all of this into facts and non-facts in a way that saves the argument is work no fatalist has yet done. 4. The right response to both versions is the same. Whatever blocks the inference for the bare statement blocks it for God's belief. Against the simple version, (P2) is the block: from the fact that a thing is truly said in advance it follows that the thing will happen, never that it had to. Against the version that leans on the fixity of the past, (P3) and (P4) are the block, since God's belief waits on the choice just as the statement's truth does.
+ Your answer depends on Molinism, which most Christians do not hold. So it is not available to the ordinary believer who just wants to know whether God's knowing ruins his freedom.
1. The answer is more than two centuries older than Molinism. Molinism, named for the sixteenth-century Jesuit Luis de Molina, holds that God knows what any person would freely do in any circumstance he might be placed in, and that he plans the world around those truths. Philosophers call that middle knowledge, because it sits between God's knowledge of what could happen and his knowledge of what he has decided will happen. Now count the years. The answer (P4) rests on is Ockham's, and Ockham died in 1347. Molina published his account in 1588. Whatever middle knowledge adds to the picture, the reply to the fatalist was in hand well over two centuries before it existed, and it has been held ever since by people who reject middle knowledge outright. 2. Neither premise uses middle knowledge. (P3) uses nothing but the claim that knowledge of what another person does conforms to what he does, and Origen made that claim nearly thirteen centuries before Molina was born. (P4) uses the claim that God's belief about a future act depends on the act, which is the same thought stated about the past. Neither one mentions what a creature would freely do in circumstances he never meets, or which worlds God could have created, or the order in which God knows what he knows. Somebody could accept both premises and hold no view at all about how God knows. 3. Middle knowledge answers a further question. It is fair to say that a full doctrine of providence wants more than the argument against fatalism supplies. How God knows a free act, and how he can govern a world of free creatures in any detail, are questions the fatalism argument simply leaves alone. Middle knowledge is one answer to them. But an answer to a further question is not a premise in this one, and the distinction matters for a practical reason. Somebody who cannot accept middle knowledge is not thereby stuck with fatalism. 4. Nor does the answer lean on divine timelessness. Many Christians decline that doctrine too, and they lose nothing by it. Neither premise requires a verdict on whether God is in time. The Boethian reply is available to those who hold that God is timeless, and no premise assumes it (D2). See also: Philosophy / Free Will: God Knows What You Would Freely Do, and Plans Around It
+ The Smith example proves nothing. Smith merely happens to be right about tomorrow, while God cannot be wrong at all, and an infallible believer's belief is a different kind of fact from a lucky guess.
1. The difference is real. Smith could have been wrong. God could not. That is not a quibble, and an old reply to the fatalist came to grief on exactly this point. It used to be said that human foreknowledge compels nobody, so divine foreknowledge compels nobody either. Jonathan Edwards answered that human foreknowledge is fallible and God's is not, which leaves the two cases unalike precisely where it counts, and on that narrow point he was right. The Smith case was never that argument. 2. The substitution tests the inference, not the knower. Look at what the popular argument actually uses. It uses that God believes, that the belief is true, and that a true belief and the thing believed cannot come apart. It never once uses the claim that the belief could not have been mistaken, because it never needed to. So a believer who merely happens to be right satisfies every premise the argument leans on, and out drops the same conclusion. That tells you the defect lies in the shape of the inference rather than in the credentials of the believer. Think of a kitchen scale reading twelve pounds for a bag of flour. Whether it is a cheap scale that happened to be right or a laboratory instrument that could not have been wrong makes an enormous difference to what you may safely conclude from the reading. It makes no difference whatever to how heavy the flour is. 3. Infallibility by itself does not reach the conclusion. Grant the difference, then, and ask what it buys. It buys this much: nobody can make God's belief a false one. Everybody in the dispute agrees about that, and (P4) concedes it in so many words before anything else is said. To get from there to Jones being unable to refrain, the fatalist still needs that ancient belief to be beyond Jones's reach in every other respect as well, and it is the fixity of the past doing that work, not the infallibility. Which means the objection is not really a defense of the popular argument at all. It is the step from the popular argument to the serious one, and the serious one is where (P4) is aimed. 4. Infallibility is what makes the belief lean forward. The difference cuts the other way as well, and that is the part the objection does not see coming. Smith's belief really is a hard fact about the past. Smith could have been wrong, so his believing on Monday guarantees nothing at all about Saturday, and his belief would have stood exactly as it did whatever Jones goes on to do. God's belief could not have stood whatever Jones does, and the reason is the very one the objection insists on. God cannot be wrong. A belief that carries Saturday around inside it was never a fact locked in the moment it was held in, and that is what (P4) means by calling it soft. So the objection hands over the premise it meant to take away. The fatalist needs both halves at once, a believer who could not have been wrong and a belief locked in a past nobody can reach, and the two halves pull against each other. Take the infallibility away and God's ancient belief is as hard as Smith's, and just as harmless. Leave the infallibility in, and it is the infallibility itself that unlocks the past. 5. An infallible believer tracks the act rather than binding it. Here is what infallibility comes to for Jones, spelled out in full. If Jones mows, God believed he would mow. If Jones refrains, God believed he would refrain. There is no third case, and ruling out that third case is the whole of what infallibility guarantees. What it never supplies is a case in which Jones tries to stay home and finds that he cannot. Until the fatalist produces one, what he has described is a witness who cannot be fooled rather than a guard who will not let Jones past.
+ Grant that knowing causes nothing. It is not up to me what God believed before I was born, or that his beliefs are never false, and what follows from two things not up to me is not up to me either.
1. The move at the end of this objection has a name. This is the sharpest shape the argument from foreknowledge takes, and the step doing the work at the end is called a transfer principle. Start with something you plainly had no say in: whether it rained on the morning you were born. The objection needs "up to you" read in just that way, as having a say. You have no say in a thing when nothing you could do would so much as risk its being otherwise. The transfer principle then says that if you have no say in something, and that something guarantees that a second thing will happen, you have no say in the second thing either. The principle is better known from the Consequence Argument, the case that free will and determinism cannot both be true. Determinism is the view that every choice is settled beforehand by earlier conditions that are enough by themselves to make it go the way it goes. In that argument the untouchable fact is the state of the world before anybody was born, together with the laws that run it. Swap in God's ancient belief and the theological twin appears. There is no fallacy in it, no advance pressure, and no confusion of certainty with necessity, which is exactly why it survives everything said in (P2). 2. The first premise is exactly what is denied. Separate out the two things Jones is said to have no say in, because they are not alike at all. - That God's beliefs are never false is something Jones has no say in, and nobody on either side pretends otherwise. It belongs to what God is. - That God believed eighty years ago that Jones would mow is another matter altogether. The objection says Jones had no say in this either, and (P4) denies that in plain words. Had Jones been going to golf instead, that ancient belief would have been a belief about golf. So the objection is not turned aside on a technicality. It is answered at its first line, and answering it at its first line is what the whole soft-fact case exists to do. 3. The deepest form of the complaint is that a dependence is not an ability. Here the argument gets genuinely hard, and it is worth slowing down for. Suppose every word of (P4) is granted. What has been produced, the objector says, is a relation and nothing more: were Jones to refrain, the belief would have been different. A relation is not a power. Jones was asked for an ability, the ability to stay off the lawn on Saturday afternoon, and what he has been handed instead is an if-then. So ask what that ability was ever supposed to consist in. Jones has to be able to put the mower back in the shed and drive to the golf course. That is an ordinary ability and every man with a lawn has it. The fatalist's entire argument was the claim that in Jones's case it had somehow gone missing, and the claim ran through the fixity of the past. Take the fixity away and the ordinary ability is sitting exactly where it always was. Nothing extra has to be supplied, and no second power over that eighty-year-old belief has to be discovered anywhere. Think of it this way. A footballer can retire at the end of the season, and if he does retire, it will follow that last Saturday's match was his second to last. That consequence is not some additional power over last Saturday which he has to hold on top of the power to hand in his boots. It is a description of what handing in his boots would have meant for last Saturday. What the objector is entitled to is the claim that counterfactual dependence is not itself a power. What he is not entitled to is the conclusion, because the dependence was never offered as a power. It was offered as the reason his fixity premise fails. 4. The transfer principle has to be restricted anyway. Run it on a fact with no God anywhere in it. It was true in 1946 that you would eat whatever you ate for breakfast this morning. That truth looks as far beyond your reach as anything could possibly be, and it guarantees the breakfast. Transfer the lack of say along the guarantee and this morning's breakfast was out of your hands too, which is fatalism arrived at with no believer in the story at all. Refusing a line of the argument is not a modern convenience, either. Laid out one line at a time, nearly every line of the argument has been denied by somebody the church still reads. - Aristotle denied that a statement about a future free act is true in advance at all. - Ockham denied that God's past belief is beyond reach. - Duns Scotus, a Scottish theologian of the generation before Ockham, denied this very transfer. 5. The principle's first published form had a hole in it. The hole was found in a case with no theology anywhere near it. That first version counted a thing as out of your hands whenever nothing you could do would make it otherwise. Here is the case. - A coin sat in your pocket all morning, and you never took it out. It did not come down heads, and nothing you could do would have made it come down heads, since even had you tossed it, it might just as well have come down tails. So by that first count, its not coming down heads was out of your hands. - For the same reason, its not coming down tails was out of your hands too. - Now put those two together, which is exactly what the principle licenses when it passes a lack of say from two things to what follows from them. Out comes the verdict that it was out of your hands that the coin came down neither way. - That is the one thing in the whole story which plainly was in your hands. You could have tossed it, and a tossed coin comes down one way or the other. Peter van Inwagen, who gave the principle its modern form in the Consequence Argument, granted that the version he first stated fails. Its defenders have mended it in more than one way. - One repair counts more strictly: a thing is out of your hands only when nothing you could do would so much as risk its being otherwise. Tossing the coin would have put heads at risk, so on the strict count the coin's not coming down heads was never out of your hands, and the coin case has nothing left to work with. - Another repair, published in 1998, never lets the principle join two facts into a pair at all. The repairs are still argued over, and a step under repair is a thin floor for a proof that nobody is free. 6. The principle stops at facts that lean forward. Everybody has to say where the principle stops, the fatalist included, or it will eat his own deliberations along with everyone else's. The Ockhamist says where it stops and why. It stops at facts that lean forward, because a fact that waits on what you do was never out of your reach to begin with. That holds whichever version of the principle survives, since every version starts from things out of your hands. God's ancient belief waits on Jones's choice, so it is exactly as much in his hands as the choice is, and it can never be the independent thing that takes the choice out of his hands. 7. At most the transfer could take the ordinary mark. Notice how little the principle could take even if it went through. All it does is pass a lack of say along a guarantee, from the thing that guarantees to the thing guaranteed. But a guarantee that brings nothing about settles nothing. The neurosurgeon's idle electrode (C2) guarantees a man's vote and still leaves the vote his, because nothing settled it but him. God's belief does less than the electrode, since it stands ready to do nothing at all. So the most the transfer could reach is the ordinary mark, the ability to have chosen otherwise. The source, the one who settled the choice, would stay exactly where it was (C2). See also: Philosophy / Free Will: If Determinism Is True, Nothing Is Up to Us Philosophy / Free Will: Black the Neurosurgeon: Freedom Never Depended on the Open Door
+ God's knowledge is not downstream of anything. Aquinas taught that it is the cause of things, as a builder's plan is the cause of the house, and that Origen's formula shows only what follows from what, not why God knows.
1. The strongest form of this objection comes from inside the church. This is not a skeptic's objection. It belongs to Thomas Aquinas, the thirteenth-century theologian. His authority for it is Augustine, writing more than eight centuries before him: "Not because they are, does God know all creatures spiritual and temporal, but because He knows them, therefore they are" (Summa Theologiae I.14.8, quoting Augustine, On the Trinity XV). Put plainly, God does not know his creatures because they exist; they exist because he knows them. Aquinas asks whether God's knowledge is the cause of things, and he answers that it is. God's knowledge, he says, is to everything he has made "what the knowledge of the artificer is to things made by his art." An artificer is a craftsman. - Think of a house and the two people who know it best. The buyer walks through it once it is finished and comes to know it. The house is the measure of what he knows: if he thinks it has three bedrooms and it has four, he is the one who is wrong. - The builder knew it before there was any house to walk through, and his plan is the measure of the house. If the plan calls for four bedrooms and the crew builds three, it is the house that is wrong. - Our knowledge is the buyer's kind. God's, Aquinas says, is the builder's. Behind the picture lies a conviction about God himself. Nothing outside him acts on him or adds anything to him, so his knowledge cannot wait on anything he has made, and on Aquinas's view that includes whatever his creatures choose. A God whose knowledge answered to his creatures would be a God they determined. Then Aquinas turns to the very formula (P3) takes from Origen, that a thing does not happen because God knows it, but God knows it because it will happen. Read strictly, he says, it names only what follows from what: "if things are in the future, it follows that God knows them; but not that the futurity of things is the cause why God knows them." If that is right, (P3) has drawn God as the buyer when he is the builder. 2. Aquinas grants that knowing by itself causes nothing. His first words to Origen are a concession. Origen, he says, "spoke in reference to that aspect of knowledge to which the idea of causality does not belong unless the will is joined to it." Knowledge on its own, even God's, brings nothing about. It becomes a cause when the will is joined to it. When God's knowledge does cause in that way, Aquinas notes, it is usually called the knowledge of approbation, that is, of approval: God's knowing, joined to his deciding to make. - Think of an architect's drawing lying in a drawer. It holds everything there is to know about a house, down to the last window latch, and it builds nothing. It becomes the cause of a house on the day somebody decides to build from it, and what made the difference was the deciding. - So on what knowing does, Aquinas and Origen stand together. God's knowing, as knowing, does nothing to what it knows, and that is all (C2) needs to close its first road, on which the knowing would be among the things that bring the act about. Whatever God himself brings about, it is his will that brings it about, never his knowing alone. - Luis de Molina, the sixteenth-century Jesuit, drew the line more sharply still. Willing is not a kind of knowing, he held, so God's knowledge is a cause of things in two limited ways only: as the pattern creatures are made on, and as the map of which worlds he is able to make. The making belongs to the will. 3. A builder's knowledge of a free choice would be knowledge of a decree. Nobody in this dispute doubts that God is the builder of the world. He made it and every creature in it, each with the nature it has, and he chose the world in which each of us stands where we stand. He knows all of that the way its maker does. Augustine's own next sentence says as much: God "was not ignorant of what He was about to create; therefore He created because He knew." The question is about one thing in that world, a creature's free choice, and whether God builds that too. - If God's will, joined to his knowledge, settles the choice itself, then God knows the choice by knowing what he decreed, and the objection has become the decree reading (D6). What settled the choice, and so took it out of the chooser's hands, would then be the decree and not the knowledge. - If God's will does not settle the choice itself, the choice is not a house he builds, and his knowing it is not the builder's kind of knowing. 4. Aquinas answers that God's will does not compete with the chooser's. That is his reply to the first possibility, and his followers, the Thomists, keep it to this day. God, they say, is not one more cause pushing beside the chooser, as though whatever God did had to be taken from what the chooser does. He is the first cause, working on another level altogether. He holds every creature in being and moves each one according to its nature, the stone as a stone and the will as a will. In moving a will, Aquinas says, God does not take away its acting of its own accord; he is the cause of that very thing. Nor, he adds, does being free require being the first cause of one's own act (Summa Theologiae I.83.1). On that view God's will can bring a choice about without taking it out of the chooser's hands, and the builder's knowledge costs the chooser nothing. This is a serious position held by serious people, and it is not fatalism. Those who hold it affirm the conclusion of (C2) by a road of their own. 5. Aquinas's picture of the sun can be read two ways. Aquinas has a picture for how the two causes fit together (Summa Theologiae I.14.13), and everything turns on which way it is read. - Think of the sun over a field of seedlings. Its course is fixed, and yet a seed's coming up can still fail, because whether it comes up is settled nearer the ground, by the seed and its patch of soil. The sun makes the growing possible and leaves to them which seeds come up. - On the first reading, God's causing is of the sun's kind. He holds the chooser and his power to choose in being, and acts together with him in whatever he does, while the chooser settles which way the choice goes. Theologians call God's acting together with a creature concurrence, and on this reading his concurrence leaves the settling to the chooser. - On the second reading, God's will settles which way the choice goes, moving the will to this choice rather than that one. The Dominican school called this physical premotion: a real moving of the will, not a persuading of it, that comes ahead of the choice in the order of causes and settles which choice it will be. "Physical" here means real, not bodily. On the first reading nothing settles the choice but the chooser. God's knowing which way it goes is then no builder's plan, and (P3) is true of it. On the second reading the sun has picked the seed. The question is then no longer how God knows. It is whether a choice his will settles is still settled by nothing but its chooser, and (D6) answers it. A choice is free when nothing settles it but the one who makes it, for reasons of his own. Putting the two causes on different levels explains how God would settle the choice, but it does not make God any less the one settling it. Catholic theologians argued over exactly this around 1600, and in the open. - The Dominicans who followed Domingo Báñez held that God knows free choices in his decree and moves each will to its act, and that the motion takes nothing from freedom, since God moves a will to act freely. - The Jesuits who followed Molina answered that knowledge taken from the decree would have God deciding every choice, and that freedom would not survive it. - A papal commission on grace heard both sides from 1598, and in 1607 Pope Paul V ended the dispute without deciding it: each side might teach its own view, and neither might condemn the other. The answer to the fatalist does not need that dispute settled. On the first reading, and wherever else a choice is settled by nothing but the one who makes it, (P3) and (P4) close both of the fatalist's roads (C2). On the second, what settles the choice is the decree and not the knowledge (D6). The Thomist has his own answer to the fatalist besides, the one Aquinas drew from God's eternity (D2) and from the two necessities (P2). Either way, foreknowledge is not what takes a choice out of anyone's hands. 6. The "because" in (P3) names a measure, not a cause. Aquinas denies more than that God learns from the world. He denies that a thing's being future is what God's knowing rests on at all, since God knows each thing by knowing himself as its cause. Part of that is common ground. God does not gather his knowledge from the world the way the buyer gathers his from the house (D5), and nothing travels backward from Saturday's mowing to a belief God held about it eighty years earlier (P3). (P3) makes a different claim. It says which of the two is the measure of the other: what God's knowledge of a free choice has to match in order to be true. What follows from what cannot decide that on its own. On Aquinas's reading, what follows from what is all that Origen's formula names. And where the knower cannot be wrong, each follows from the other: the knowing from the act, and the act from the knowing. So something more has to give the dependence one direction. For a choice nothing settles but the one who makes it, the choice is the measure, and three facts put it there. - What God knows answers to what will be done: had the act been otherwise, he would have known otherwise (P4). Aquinas grants this in the very sentence the objection quotes, since from a thing's being future "it follows that God knows" it. - His knowing is not what makes the act go the way it goes. Aquinas grants this too, in his concession to Origen. - His will does not make it go that way either, since a choice nothing settles but its chooser is one God's will does not settle. It takes all three facts. Without the third, the first two would leave open which way the dependence runs, because God's will could then stand behind both the choice and his knowing of it, as on the Dominican reading. With all three, nothing on God's side makes the choice go the way it goes, and his knowledge can answer only to the choice. That is more than Aquinas allowed Origen's formula to say. And it is that one-way dependence that makes God's belief a soft fact (P4): the belief waits on the choice, and the choice waits on nothing in the belief. - Think of a sentence written on a Friday: "The bridge will open on Monday." It is true because the bridge will open on Monday. Monday does not reach back and write the sentence, and the sentence does not open the bridge. Nor does that leave God determined by his creatures, as the objection fears. The choice does nothing to God and adds nothing to him, any more than Monday wrote Friday's sentence. What hangs on the choice is which truth there is to be known, and knowing every truth there is belongs to God by nature (D5). So for every choice the fatalist says foreknowledge would take away, a choice nothing settles but the one who makes it, the order in (P3) stands. Whatever God makes, he makes by willing it, and knowing alone makes nothing. See also: Philosophy / Free Will: What Free Will Is, and Why It Matters Philosophy / Free Will: Theological Determinism and the Authorship of Sin Philosophy / Free Will: God Knows What You Would Freely Do, and Plans Around It

Public Theology

Theological Analyses of Societal Issues

Theology of the Family

Where Faith and Family Intersect

Biographies

Notable Works & Great Quotes from Key Figures

Contemporary Voices

1960 AD – present
Playlist
Justified True Belief