Advanced Search

Does an Omniscient God contradict Free Will?

Does an Omniscient God contradict Free Will? Yes, a very age-old question, with many answers. The problem seemed to arise when we thought that if God knows what we will do or "choose" then it's metaphysically necessary for us to choose or do that, because what God knows IS true, thus it's true event A will happen if God knows it will. There's no Free Will because there's no chance that event A can NOT happen, in this view Free Will is just an illusion. But! Some Philosophers have objected by saying that God's knowledge is from or depends on our choice, it's formed by the choices we genuinely (freely?) make for ourselves, because God's omniscience is "logically simultaneous" with our choices. So God's knowledge doesn't write out history, history writes out God's knowledge. (By the way doesn't this make god a contingent being? Thus precluding God from "working" as an answer for the modal ontological and cosmological argument, since God is not a non-contingent being?) But I've never been convinced by...

First, I don't think it matters for the answer to your question whether the omniscient God is or is not the creator of the beings whose conduct He foresees. Thus suppose the rats are created not by God but by some fairy. God observes the rats and, knowing of each whether it is smart or dumb, foresees whether it will end up rewarded or punished. Can this new wrinkle in the story -- that the rats are created by some fairy rather than by God -- possibly make a difference to whether the rats have free will or not? I cannot see a reason for believing this.

(To be sure, the wrinkle makes a difference to God's responsibility: if He creates dumb rats Himself, then he is actively responsible for the suffering they predictably undergo in the punishment zone. If He's merely an onlooker, then He is at most passively responsible insofar as He could but fails to protect the dumb rats from pain.)

Second, I don't understand why there should be felt a tension between foreseeability by others and free will. The only way your conduct can be unforeseeable is for it to be indeterminate: ruled from moment to moment by quantum events, for example. Surely, that's not a good example of free will. On the contrary, I would think that good examples of your free will are quite predictable behaviors. Those who know you know certain things about your future behavior. They know, let's say, that you are deeply committed to stand by your sister. You have carefully thought about this commitment, fully embraced it, adjusted your other values and commitments to it, built your life around it. You and others assume that you could cut your sister loose if you so chose, but you and they (and she) know that you won't. Here the firmness of your commitment seems quite compatible with the freedom of your will (and others' foreknowledge is based on your commitment and not the other way around).

Returning briefly to the first point, suppose now that your firm commitment to your sister was part of a plan your mother hatched before giving birth to you. Knowing of her daughter's frailty, she deliberately had a second child who would stand by her first one. When you were old enough to understand, she explained all this to you and helped you appreciate the wonderful difference you could make to your sister's life. Your mother correctly foresaw that you would be moved by this appreciation and would become committed to standing by your sister. Again, I don't see how adding in this additional information about the history of your commitment to your sister undermines the initial judgment that your foreseeable loyalty to your sister is compatible with your free will.

The difficulty of the free-will problem seems to be not specific to certain scenarios (e.g., free will in a world with an omniscient creator god), but quite general: how to make sense of it at all.

First, I don't think it matters for the answer to your question whether the omniscient God is or is not the creator of the beings whose conduct He foresees. Thus suppose the rats are created not by God but by some fairy. God observes the rats and, knowing of each whether it is smart or dumb, foresees whether it will end up rewarded or punished. Can this new wrinkle in the story -- that the rats are created by some fairy rather than by God -- possibly make a difference to whether the rats have free will or not? I cannot see a reason for believing this. (To be sure, the wrinkle makes a difference to God's responsibility: if He creates dumb rats Himself, then he is actively responsible for the suffering they predictably undergo in the punishment zone. If He's merely an onlooker, then He is at most passively responsible insofar as He could but fails to protect the dumb rats from pain.) Second, I don't understand why there should be felt a tension between foreseeability by others and free will. The...

Can a nation have an official religion and be a democracy?

Can a nation have an official religion and be a democracy?

I would consider Norway and the UK to be examples of this. Here the fundamental equality of citizens is not seriously undermined because the role of the state religion is largely ceremonial. In other countries, of course, citizens who do not share the state religion suffer severe discrimination which can be grave enough to defeat, by itself, the claim that the state in question is democratic.

It makes sense here to think of "being a democracy" as a matter of degree. Most of the states we call democracies fail fully to live up to democratic principles in one way or another. Having a state religion is a shortfall, but can be a relatively minor one if any resulting discrimination is not too severe.

I would consider Norway and the UK to be examples of this. Here the fundamental equality of citizens is not seriously undermined because the role of the state religion is largely ceremonial. In other countries, of course, citizens who do not share the state religion suffer severe discrimination which can be grave enough to defeat, by itself, the claim that the state in question is democratic. It makes sense here to think of "being a democracy" as a matter of degree. Most of the states we call democracies fail fully to live up to democratic principles in one way or another. Having a state religion is a shortfall, but can be a relatively minor one if any resulting discrimination is not too severe.

In his answer to question 2275 (from Sep 7th 2008), Thomas Pogge wrote: “Most

In his answer to question 2275 (from Sep 7th 2008), Thomas Pogge wrote: “Most political leaders do not act well, morally, and in most cases this is because they are not moral persons, not serious about morality. To be serious about morality, one must try to integrate one’s considered moral judgments through more general moral principles into a coherent account of morally acceptable conduct; one must work out what this unified system of beliefs and commitments implies for one’s own life; and one must make a serious effort to honour these implications in one’s own conduct and judgments. Those who are not serious about morality typically do not act well, morally...” I am very interested in the notion of ‘moral seriousness’, and would be interested to know what the other panelists think about the nature of ‘being morally serious’, as opposed to that of merely ‘being moral’ – and whether they agree with Prof Pogge’s account. I would also be grateful if you – Prof Pogge – could elaborate on your previous...

Moral judgments are often distorted by self-interest. A morally serious person must try to combat this danger by thinking beyond the particular case. A very simple way of doing this is to contemplate analogous situations in which roles are reversed (the Golden Rule). By extending one's judgments to a larger set of cases and then aiming for a coherent way of judging these cases, one is beginning to do what I was asking. Philosophers may take this sort of exercise quite far and, as you surmise, I don't think that every morally serious person needs to do this. But a morally serious person will question her or his moral judgments in the ways I sketched, especially when they are "convenient", that is, in accordance with her/his own self-interest.

Moral judgments are often distorted by self-interest. A morally serious person must try to combat this danger by thinking beyond the particular case. A very simple way of doing this is to contemplate analogous situations in which roles are reversed (the Golden Rule). By extending one's judgments to a larger set of cases and then aiming for a coherent way of judging these cases, one is beginning to do what I was asking. Philosophers may take this sort of exercise quite far and, as you surmise, I don't think that every morally serious person needs to do this. But a morally serious person will question her or his moral judgments in the ways I sketched, especially when they are "convenient", that is, in accordance with her/his own self-interest.

My question pertains to two common attributes given to God. Omniscience and

My question pertains to two common attributes given to God. Omniscience and omnipotence. If we use a definition of omniscience that includes knowledge of all future events (as most believers would today due to things like prophecy and revelation) then it follows that God knows all of his future actions with absolute certainty. If this is the case, then God's omniscience is compromised. For example, let's say God knows he is going to create a global flood "x" years in the future. If omniscience is perfect he MUST do that action and is powerless to do otherwise, lest he compromise his knowledge. If he does exercise his power and not flood the earth then he was previously wrong and his omniscience is compromised. Therefor no single entity can be all knowing and all powerful. Is this a good argument? I have never heard it used or refuted in a public debate/piece of literature.

It's an interesting idea, but I don't think it's quite compelling. Your argument assumes that God is in time in much the same way we take ourselves to be in time: experiencing only the present and acting only in the present. But, being omniscient, God would really be experiencing all times at once; and being omnipotent, he would be shaping the entire universe at once, from (temporal) beginning to (temporal) end or throughout an unbounded, infinite duration. An omniscient and omnipotent God engages in only one grand act of creation which is fully transparent to Him. Or so it might be said in response to your argument.

Still, I believe your conclusion can be supported in another way. With regard to omniscience we might ask whether God can really know why He exists. We find this question raised, for example, in a little-known passage in Immanuel Kant's masterwork The Critique of Pure Reason: "We cannot put aside, andyet also cannot endure the thought, that a being, which werepresent to ourselves as supreme amongst all possible beings,should, as it were, say to itself: 'I am from eternity to eternity,and outside me there is nothing save what is through my will,but whence then am I?'"

And with regard to omnipotence we might ask analogously whether God has control over His own existence: does God exist if and only if He so chooses?

Now if we can make sense of an affirmative answer to the latter question, then we may use it to answer the first: Yes, God can know why He exists; God can know that He owes His existence to His own omnipotent will. But I cannot make sense of this idea that God's existence is due to His own decision to exist. So I agree with your conclusion that we have reason to reject the hypothesis of a literally omniscient and omnipotent God.

It's an interesting idea, but I don't think it's quite compelling. Your argument assumes that God is in time in much the same way we take ourselves to be in time: experiencing only the present and acting only in the present. But, being omniscient, God would really be experiencing all times at once; and being omnipotent, he would be shaping the entire universe at once, from (temporal) beginning to (temporal) end or throughout an unbounded, infinite duration. An omniscient and omnipotent God engages in only one grand act of creation which is fully transparent to Him. Or so it might be said in response to your argument. Still, I believe your conclusion can be supported in another way. With regard to omniscience we might ask whether God can really know why He exists. We find this question raised, for example, in a little-known passage in Immanuel Kant's masterwork The Critique of Pure Reason : "We cannot put aside, andyet also cannot endure the thought, that a being, which werepresent to ourselves...

Couldn't we take the "ontological proof" of God's existence to prove that there

Couldn't we take the "ontological proof" of God's existence to prove that there are many God-like creatures? For instance, imagine a creature that has all thinkable perfections except for the fact that it has dirty fingernails. If existence is a perfection, then this creature must have this perfection, since one can both exist and have dirty fingernails. And so, if the ontological proof proves that God exists, then it proves that dirty fingernails-God exists too. Doesn't it?

I'm with Thomas Pogge on what the real issue is here. For what it's worth, I'm also no friend to the Ontological Argument. But let's see if a supporter of the argument might have something to say in response to this challenge...

First of all, what form of the argument are we going to consider? It's been presented in many different ways over the centuries, and some versions have had much more force to them than others. In its simplest form, the argument basically goes as follows. God, by definition, possesses every perfection; existence is a perfection; therefore, God possesses it, i.e. God exists. But this version is notoriously vulnerable to objections like those that, for instance, Kant formulated way back in the eighteenth century.

But there are other versions. One of the best (particularly associated with Leibniz, but formulated by several other people too, both before him and since) basically goes like this. (i) God, by definition, possesses every perfection; (ii) necessary existence is a perfection; (iii) it is possible that a God, thus defined, should exist. Therefore, God does exist. Given a few extremely basic principles of modal logic, it can now be shown that the argument is valid, in the sense that the conclusion really does follow from the premises. How does this work? Well, to say that it is possible that God should exist is equivalent to saying that there is some possible world or other where God does exist. Thus far, this possible world may or may not be the one that we actually inhabit. But we can still consider what will be true at such a possible world, even if it isn't actual. And, given the other two premises, one thing that we can declare to be true at that world is that God necessarily exists. But to say that it is true at a certain possible world that something necessarily exists is equivalent to saying that it is true at every other world that it exists. And our world is certainly going to be among these others. So it is true at our world that God exists. Indeed, from this it follows in turn that God necessarily exists here. So there you have it: the Ontological Argument is logically valid! Sound the trumpets!

But, of course, validity isn't everything. What we really want is 'soundness'. It's not enough for the conclusion to follow logically from the premises: we also hope that the premises themselves might actually be true (and, indeed, that we might have solid grounds to believe them to be true), for only then will we have any solid grounds to accept the conclusion.

So let's now look more narrowly at the 'dirty fingernails' argument that you've raised. The place where a defender of the divine argument is most likely to criticise your reformulation of it will be on the possibility of the existence of the being that you've defined. People like Leibniz went to some lengths (albeit with questionable results) to argue that the perfections they attributed to God were all 'compossible', i.e. that it was possible that a being should have all of them together. And I think this would be how they'd respond to you: they'd deny the compossibility of dirty fingernails with the perfection of necessary existence. No being, they would say, could have both; i.e. they would reject your version of premise (iii). And, since the argument hangs on the interaction between the necessary existence described in premise (ii) and the possible existence described in premise (iii), they would thereby declare that your version of the argument was unsound.

And why might dirty fingernails be regarded as incompossible with necessary existence? Well, plenty of reasons. If we don't reject that compossibility, then we do indeed seem to be facing a very real prospect of being forced to accept the necessary existence of the being you've described. But that would mean that these dirty fingernails themselves necessarily exist, and that just doesn't seem right. Can't we imagine possible worlds where no fingernails exist at all? Moreover, if this alleged being's dirty fingernails are anything like the kinds of fingernails that we are familiar with -- and, if they're not, then why are we calling them 'fingernails' at all? -- then they ought to be every bit as destructible as any other fingernails are. But destructibility and necessary existence definitely don't seem to be compossible. So the being you've described (all-perfections-plus-dirty-fingernails) doesn't seem to be a possible existent; i.e. there is no possible world where such a being exists; and consequently the 'necessary existence at some world' step in the argument, and the move from that to existence in the actual world (and from that back to necessary existence in the actual world), never actually kicks in at all.

I read the question differently from Oliver. The questioner agrees that dirty fingernails are an imperfection, in fact, this is part of the point. We are to imagine a being that is all-perfect except for those dirty fingernails. Now if existence is a perfection, as the ontological argument assumes, then this imagined being has it. So it exists. (And never mind whether it's Divine or Divine-like, that's irrelevant to the point.) And likewise for all the other imaginable beings that are all-perfect except for one imperfection (other than non-existence) -- each of them also exists. And so do all the other imaginable beings that are all-perfect except for two imperfections (other than non-existence). And so on. So I think this is a nice reductio ad absurdum of the ontological argument for God's existence. If the ontological argument proves the existence of God, it also proves the existence of a vast number of other beings whose existence those interested in proving God's existence would have wanted to deny.

Should a political leader let his faith influence his decisions that he makes

Should a political leader let his faith influence his decisions that he makes for the people he leads?

In a society where this faith is shared and influences political decisions in the direction of peace, justice, humanity, and equal citizenship, I see no problem.

Problems arise when either of these two conditions are not satisfied. Faith can lead people to do terrible things, for instance torture and murder those with different religious beliefs. No one should let his or her faith exert such an influence. Rather, when one's religion seems to require conduct that seems wrong, then one ought to re-examine one's religion. This does not mean that one should abandon one's faith, though it may come to that in some cases. Another possible outcome is a reinterpretation of one's religion (for example, many Christian now understand that their faith did not really require the Inquisition). And one may also conclude, on reflection, that one's religion was right, after all, to require the conduct in question (as when Christians in Nazi Germany came to endorse their religious duty to engage in treasonable resistance). These remarks about the second condition apply to political leaders and ordinary people alike.

You are probably more interested in cases where the first condition is not satisfied: where the political leader's faith is not shared and some political decisions this faith favors are controversial. Here a political leader may still allow himself to be influenced by his faith insofar as this faith favors good policies that have a solid secular justification -- as when his faith requires religious toleration, for example, or forbids torture. He should clearly not use state power to obstruct the practice of other religions or to require conformity to his own -- even if his faith suggests that those who practice another religion are doing wrong thereby.

The hard cases are those where the political leader's faith may influence him to endorse, and then to follow, one secular (freestanding) morality in preference to another. For example, quite apart from any religion, it is controversial among moralists whether it is wrong (absent special circumstances such as rape or genetic defect) to have or perform an abortion. A person of Catholic faith will feel drawn to a morality that deems abortion wrong (a crime against an as yet unborn human being) and may then favor some legislation against it as appropriate. Can we fault her for this? I think we cannot, so long as her morality, though supported by her religion, is also independent of it as shown by the fact that many non-Catholics also endorse it.

It is of great importance that political leaders act well, morally, because the decisions they make have such momentous consequences. (By the end of a President's or Prime Minister's term of office, millions have typically died prematurely as a result of decisions he has made.) Most political leaders do not act well, morally, and in most cases this is because they are not moral persons, not serious about morality. To be serious about morality, one must try to integrate one’s considered moral judgments through more general moral principles into a coherent account of morally acceptable conduct; one must work out what this unified system of beliefs and commitments implies for one’s own life; and one must make a serious effort to honour these implications in one’s own conduct and judgments. Those who are not serious about morality typically do not act well, morally, and when they are political leaders, then this can lead to a great deal of harm.

When a political leaders is committed to a religion that has implications for human conduct, and when he is serious about morality in the sense defined, then his understanding of his religion will invariably influence the coherent account of morally acceptable conduct he will develop. And his faith will then, via his morality, influence his decisions. If we voters want to avoid this, we can vote for non-religious politicians. But this may make it even harder to find ones who are morally serious about the momentous decisions they make in our names.

In a society where this faith is shared and influences political decisions in the direction of peace, justice, humanity, and equal citizenship, I see no problem. Problems arise when either of these two conditions are not satisfied. Faith can lead people to do terrible things, for instance torture and murder those with different religious beliefs. No one should let his or her faith exert such an influence. Rather, when one's religion seems to require conduct that seems wrong, then one ought to re-examine one's religion. This does not mean that one should abandon one's faith, though it may come to that in some cases. Another possible outcome is a reinterpretation of one's religion (for example, many Christian now understand that their faith did not really require the Inquisition). And one may also conclude, on reflection, that one's religion was right, after all, to require the conduct in question (as when Christians in Nazi Germany came to endorse their religious duty to engage in treasonable...

From a moral Christian point of view, I cannot understand the idea that we

From a moral Christian point of view, I cannot understand the idea that we should punish anyone. In America, which is a highly Christian-dominated society, there is little resistance to capital punishment from the "right wing." My understanding is that Christians are not supposed to judge. God will judge everyone when their time comes. Isn't Christian morality about tolerance and acceptance, and not revenge? "Turning the other cheek?" "Love thy neighbor/enemy as thyself?" Are Christians simply turning a blind eye to this action?

There is indeed a tension between capital punishment and the teachings of Christ. One can ease this tension somewhat by highlighting the contribution of penal institutions to the protection of innocent people, who are safer when criminals are taken off the street and potential criminals deterred. This does not justify the death penalty, nor our kind of prisons in which inmates are routinely raped and abused, but it does help justify penal institutions of the kind we know from the more civilized states.

I see much greater tensions between Christian teaching and many other policies we pursue, especially internationally. We pressure very poor countries to undertake “structural adjustment programs” -- cutting public funding and raising fees for basic education and health care -- so that they can better service their loans to our banks, which loans are often taken out by brutal dictators who use the money we lend them to buy the arms they needed to stay in power. We allow our banks to help such tyrants and their supporters to transfer embezzled funds into the international banking system. We pressure very poor countries to enforce the intellectual property rights of our pharmaceutical firms by ensuring that their populations have no access to generic versions of medicines on which one of our companies has a patent. We tolerate severe poverty of nearly half the world’s population, which kills about 18 million people annually including over 10 million children, even while such poverty could be eradicated at half the cost of our military budget. We impose severe sanctions on Iraq, and later a botched occupation, each of which have needlessly killed hundreds of thousands of Iraqi civilians. We refuse to intervene in Rwanda, Sudan, and Myanmar, where enormous massacres and suffering could have been prevented at minimal cost. We detain thousands of poorly selected terrorism suspects indefinitely without charge or trial while subjecting them to every kind of recreational and premeditated abuse. Much more than our penal institutions, such policies -- often associated by their perpetrators and supporters with Christian language and values -- are an indelible insult to Christ and His teachings.

We are a Christian-dominated society in that there is a lot of talk about Christ and His teachings. We are not a Christian-dominated society in that many of our policies are plainly incompatible with Christ’s teachings and perpetrated by people who plainly have no expectation that they will ever have to answer to Christ for what they did to further their political ends while invoking His name.

There is indeed a tension between capital punishment and the teachings of Christ. One can ease this tension somewhat by highlighting the contribution of penal institutions to the protection of innocent people, who are safer when criminals are taken off the street and potential criminals deterred. This does not justify the death penalty, nor our kind of prisons in which inmates are routinely raped and abused, but it does help justify penal institutions of the kind we know from the more civilized states. I see much greater tensions between Christian teaching and many other policies we pursue, especially internationally. We pressure very poor countries to undertake “structural adjustment programs” -- cutting public funding and raising fees for basic education and health care -- so that they can better service their loans to our banks, which loans are often taken out by brutal dictators who use the money we lend them to buy the arms they needed to stay in power. We allow our banks to help such tyrants and...

The debate between science and religion has gone on for many years, and many

The debate between science and religion has gone on for many years, and many people think that they must choose one or the other to believe. To me, it's a lot like trying to collide two trains on parallel tracks. If one chooses to believe in God, then that person can still believe in the big bang or evolution while believing that God created the universe, because religion explains what happens on a spiritual level, and science explains what happens on a physical level. The two run parallel. Using this as a way of thinking, can science contradict religion at all, and why has the debate between the two gone on for so long when this explanation reconciles them?

Your idea works fine on a certain modest understanding of religion. If religion were only about the Divine, perhaps with the additional thought that God created the universe, then no explanation given by science of anything in the universe could interfere with religion.

Religions are typically not so modest, however. A typical religion may ascribe certain duties to human beings along with the freedom and responsibility to live up to these duties. And this can raise scientific (and philosophical!) doubts about whether human beings have the requisite freedom.

In response, you might propose dividing human beings over your two levels: into a physical body (brain included) and a spirit or soul. But this proposal raises further puzzles about the relation between these two parts or components of human beings. If religion attributes some of what you do to your soul it may compete with scientific theories that attribute all your conduct and thinking to physical causes. If religion attributes nothing you do to your soul, then it is hard to understand in what sense it is your soul at all or can bear any responsibility for what your body does.

There are philosophical accounts of how the kind of human freedom religions typically assert is compatible with any explanatory account science may end up providing of human thought and conduct. (A famous such account is given by Immanuel Kant.) But such accounts remain contested.

There is another, less obvious way in which science may interfere with religion: Science may provide an explanation of our religious beliefs, and some such explanations may tend to undermine these beliefs. Suppose, for example, that scientists develop a neat evolutionary explanation of why human beings are prone to believe in in gods. Such an explanation would not show that there is no god, but it might well undermine the conviction that there is. An analogy: Your strong belief that you have once spent a summer in Antarctica is likely to be undermined when you are shown a videotape of how you were hypnotized to believe that you once spent a summer in Antarctica and hypnotized to forget the hypnosis. In both cases, to be sure, the scientific explanation of the belief does not refute its content. You might say that God created a world in which religious beliefs emerge through evolution. Still, religious communities typically tell rather different stories about their origins, and these stories may turn out not to fit with the best scientific explanation.

Your idea works fine on a certain modest understanding of religion. If religion were only about the Divine, perhaps with the additional thought that God created the universe, then no explanation given by science of anything in the universe could interfere with religion. Religions are typically not so modest, however. A typical religion may ascribe certain duties to human beings along with the freedom and responsibility to live up to these duties. And this can raise scientific (and philosophical!) doubts about whether human beings have the requisite freedom. In response, you might propose dividing human beings over your two levels: into a physical body (brain included) and a spirit or soul. But this proposal raises further puzzles about the relation between these two parts or components of human beings. If religion attributes some of what you do to your soul it may compete with scientific theories that attribute all your conduct and thinking to physical causes. If religion attributes nothing you...

I have recently become interested in the following philosophical idea, and am

I have recently become interested in the following philosophical idea, and am wondering if it carries much weight. It rests on the idea that there cannot be any such thing as 'religious evidence'. Any religious claim cannot be made without some sort of evidence - this may differ from what a scientist would term 'evidence' as it may involve the mere 'feeling of truth' rather than a demonstratable proof. However, here is the problem that currently interests me. For any religious claim to have some sort of weight, it must rest upon some sort of evidence. The nature of evidence in general is that it is either empirical or theoretical in form - however, the status of the latter is such that it allows for future empirical verification or falsification, and as such does not rule out testing. With evidence, we either demonstrate something to 'be the case' through example, or show how a method carries value. Let me bring in an example of a religious claim: "We look around and see an order and structure to the...

As I understand your argument, much of it depends on understanding the predicates religious and empirical as mutually exclusive. This allows you to infer that, if a claim is empirical, then it cannot be religious -- and that, if evidence is empirical, then it cannot be religious. If I wanted to argue against you, I would dispute that understanding and this inference.

Since you are making an assertion about all religious claims, your opponent is free to present you with any one such claim as a counter-instance. So, let me give you the claim that the prayers of truly pious people are very often answered: What they pray for very often comes true, much more often than what less pious people pray for. I say that this is a religious claim.

Now you ask me for evidence for this claim. To give you evidence, I ask you to join a group of people who together grade a randomly selected population of 2000 self-declared believers in terms of their piety. We do this by interviewing each of these 2000 subjects as well as his/her friends and family and religious guides and so on. After we've assigned a piety score to each of the 2000, we then ask them to tell us, just before any prayer, what they are going to pray for. And then we check each subject's "success rate."

If this "experiment" finds no correlation between assessed piety and success, then my religious claim remains without evidence. But suppose we do find a robust positive correlation of the kind I had claimed. Then (if I understand you correctly) you propose to say: "OK, there's empirical evidence for your claim. Your claim is thus an empirical one. And it is therefore not a religious one."

What if I accept your first two (imagined) sentences and reject the third? In other words, I insist that, empirical evidence notwithstanding, my claim is still a religious one. It is a religious one because it involves a religious concept (piety).

Now you may respond with the assurance that science will provide an explanation of the positive correlation: Some causal influence exerted by pious prayers upon distant events; or some causal influence exerted by distant events upon the prayers of the pious; or some causal influence exerted by some third class of events upon both the prayers of the pious and on the events relating to these prayers' fulfillment. Once such a scientific explanation is on hand, you will continue, my religious claim is transformed into a scientific one (in which the religious term pious will presumably no longer occur).

Let me accept this, for the sake of the argument. There still is the possibility that no scientific explanation is found, that we remain baffled by the fact that the prayers of the pious tend to be fulfilled. If this were to happen, then (I think) we would have a genuinely religious claim for which there is good empirical evidence.

To conclude. I have argued that your a priori argument fails, that there could be genuinely religious claims backed by solid empirical evidence. Nonetheless, I am sympathetic to your conclusion: I do not believe that there are any such genuinely religious claims backed by solid empirical evidence.

As I understand your argument, much of it depends on understanding the predicates religious and empirical as mutually exclusive. This allows you to infer that, if a claim is empirical, then it cannot be religious -- and that, if evidence is empirical, then it cannot be religious. If I wanted to argue against you, I would dispute that understanding and this inference. Since you are making an assertion about all religious claims, your opponent is free to present you with any one such claim as a counter-instance. So, let me give you the claim that the prayers of truly pious people are very often answered: What they pray for very often comes true, much more often than what less pious people pray for. I say that this is a religious claim. Now you ask me for evidence for this claim. To give you evidence, I ask you to join a group of people who together grade a randomly selected population of 2000 self-declared believers in terms of their piety. We do this by interviewing each of...