The Antecedence of Natural Revelations, pt II

I am currently going through finals, so blogging will be limited. Once they are over, I'll write a reply to Aaron Armitage. His latest can be found here and here.

Just a few notes about it: I made a few slips in writing my last piece and Aaron called me on it. Somehow, I had managed to use "claims" instead of "would claim." A minor point, perhaps, but still significant. I thus apologize to Aaron. I will also address his last point now, regarding my belief in the conceptual antecedence of natural revelation.

This view, in my opinion, is profoundly backwards. If we believe we have a revelation from an omniscient Being who never lies (although He may reveal Himself cryptically), and we believe ourselves to be fallen and fallible, then we should go to the revelation first and make sure our philosophical speculations conform to it, not speculate philosophically and then conform revelation to our speculations.

It may be that modern research impacts our understanding of some passages, but if we see a flat contradiction, the thing to do is believe Scripture. For example, it was formerly charged that there were no such people as Hittites and therefore the Biblical histories are unreliable. Who's been vindicated, the scholars, or the "ignorant" believers who kept on thinking Hittites existed just because the Bible said so? Another example: it's been claimed -- and still is in quite a few corners -- that the account of Abraham using camels is anachronistic because camels weren't domesticated yet. Turns out they were.

And those are cases subject to empirical proof. Whether our free will is compatibilist free will or incompatibilist isn't. So Joshua is asking us to impose limits on what we'll accept from the Bible on the basis of limited, fallible human reason where we can't even check our conclusions.

I am not familiar with the specifics of the Hittite case, but it sounds to me like it is an example of bad methodology. Lack of evidence is not positive negative evidence. At the risk of revealing my ignorance of archaeology, I believe that this is the primary reason archaeology is not a science: positive negative evidence is extremely hard to come across. Aaron provides less information about camel domestication, so I will just assume that it is similar.

Independent of the specifics of his examples, I agree with Aaron. However, he has not yet shown that incompatibilism flatly contradicts Scripture. Aaron has set himself up for a tough task. He seems to imply that Joshua 10 and modern astronomy are not in direct contradiction. The definition of contradiction has been strengthened quite a bit by Aaron.

Finally, Aaron accuses human reasoning of being fallible. I fully agree. However, he doesn't say the same about sensation. In fact, Aaron seems to claim that we are more reliable about empirical matters than a priori matters. I disagree. Sensations (seeing, hearing, feeling, etc.) are notoriously misleading. A priori matters can be found through reasoning alone. Empirical matters seem to require both reasoning and sensations. The addition of a less reliable method of gathering information doesn't seem to me to help reliability. Just as (in chemistry) impurities in a sample lower its melting point dramatically, the addition of an even more unreliable method lowers our accuracy.

EDIT: 3:33am 3/5/05 - I added links to Aaron's posts. I am indeed up at 3 in the morning. I'm pretty hungry, but the roommate's asleep, so I'm afraid to get up and get food. And... I don't really have any food left in the room.

Rich Mullins and Freedom

"I say I wanna give You glory, Lord, and I do
But everything that I could ever find to offer comes from You
But if my darkness can praise Your light
Give me breath and I'll give my life to sing Your praise

And I wanna thank You, Lord
More than all of my words can say
And I give my life to sing Your praise
And beyond this I would not beg
For anything except the grace
To give my life to sing Your praise"

-- Rich Mullins, Damascus Road (from Brother's Keeper)

[copyright 1995 edward grant/kid brothers of saint frank publishing]

Rich Mullins is a favorite songwriter of mine. I'm not here to debate his allegiance to compatibilism (I have a feeling he probably was). This song struck me this morning. It explains one aspect of why I believe and hope we have incompatibilist free will. Rich realizes that giving something to God that comes entirely from Him is not much of a gift. Everything in the world comes from God, except for ourselves. This is the only thing we can give to the Lord that has any worth whatsoever. The objective freedom we have is the only thing that is our own. God cannot control it; if He could, what kind of gift would that be? He'd be giving to himself, and it'd be no different than sacrificing manna.

There is no greater gift than laying down our freedom to conform to the will of God.

Prohibitions and Encouragement

This morning in the New York Times, in an article on a new gun law in Florida, some "Chief Timoney said, 'you're encouraging people to possibly use deadly physical force where it shouldn't be used.'" (Here, 8th graf)

The law in question removes prohibitions on shooting an assailant in public. Now, we can argue over whether or not removal of prohibitions causally encourages the action. What I am more interested in is the relation of this argument to sexual laws. Both killing and sex are situations with moral implications. If removing prohibitions on killing encourages that action, then removing prohibitions on sex also encourages that action. Interestingly, usually conservatives and liberals with both oppose this broad stroke. Conservatives will wish to say that removing prohibitions on killing do not encourage it, and liberals will say the same about sexual laws. Perhaps this is a straw man generally, but I have heard both arguments before.

My view is that it is the case that removing the prohibition on an action encourages it. While I definitely prefer to have as few prohibitionary laws as possible, this encouragement factor must be weighed when legislating.

The Problem of Evil and Freedom of the Will

An important reason for believing in incompatibilist free will is that it offers an answer to the problem of evil.

The problem of evil is thus:

1) God exists. God is an omniscient, omnipotent agent.
2) An agent with the power to prevent an evil event has a moral obligation to do so.
3) God has an obligation to prevent all evil events.
4) God does not prevent all evil events.
5) God is not fulfilling all his moral obligations.
6) God is not good.

Any Christian who is serious about the life of the mind must face this problem at some point. Some deny the omniscience and omnipotence of God. Some deny that evil events exist. My view is that God, in granting us incompatibilist free will, voluntarily abrogates his omnipotence.

Calvinism comes under heavy fire from the problem of evil, because the origin and control of one’s character and actions comes from God ultimately. We can imagine a Mr. Rogers world, in which everyone is disposed to do good from birth. Why didn’t God create that world? It seems clearly superior to ours. God’s responsibility in (3) can in part be fulfilled by creating the Mr. Rogers world. It won’t do any good to appeal to the Fall. The Fall is the responsibility of Adam and Eve’s choices. Choices are caused by character and character is caused by God: we are back to square one. We ask God, Why didn’t you create Adam and Eve with different characters? It would have caused us all less pain.

If incompatibilist free will holds, evil initiated by humans is no longer a problem. Because God has restricted his powers by creating us the way we are, the buck stops with humans, and they hold an objective responsibility. The causal chain stops with the person. God cannot intervene in our character formation without violating our autonomy. This does still leave evil events that are causes in the natural world, like tsunamis and earthquakes and poisonous snakes. However, natural evil is clearly very distinct from moral evil, and a different solution must be sought for it.

A Defense of Choosing -- A Second Reply to Aaron Armitage

I would just like to start this post by writing that this unplanned exchange with Aaron has been so far very enjoyable and challenging. His most recent post on the topic can be found here. My goal today is to defend choosing as neither random nor determined. I will do this by appealing to reasons as being enough for an intelligible choice. I have attempted to divide the discussion into a few parts. Each begins with a header and then usually an indented quote from Aaron's post.


Organic vs. Mechanical Metaphors

... the dominant analogy shouldn't be one physical object impacting another and moving it, so it moves another, and so on until we get some final result, like gears moving more gears or a cue ball hitting another billiard ball; instead, it should be a living thing which grows according to its nature.

The dispute over the proper metaphor for determinism is minor. I will continue to nitpick, however. Aaron believes that physical laws are very distinct from biological laws. He wants to use a biological metaphor for determinism as opposed to a physics metaphor. The difference between biology and physics is merely one of size and methods, not of type. Billiard balls follow their nature just as much as do trees. Although Aaron doesn’t come out and say it, he seems to have adopted some flavor of vitalism. Vitalism is, of course, discredited in biology.

The Structure of Choosing

Of course there are such situations, and when it happens a decision still has to get made. How? Or on what basis? Is it something which finally makes one action more appealing? Or is it nothing? If it's something, we're back to determinism. If it's nothing, we're with Bill the robot. Positing "free will" itself won't help because the question is how that free will actually choses in the hard cases. If the free will doesn't choose one over the other because of something, why isn't it random? And if it is random, what does that do to moral responsibility? We can hardly blame someone for something random. This, incidentally, was the crux of the argument I mentioned losing.

If he's restricting his defense of free will to such equally balanced situations, that leaves and awful lot of determinism.

As I understand it, Aaron'’s view of deliberation is this: I sit down and look at my choices. Each choice gets a score on how much it conforms to my character. I then pick the choice with the highest score. Before, I claimed that Aaron’s picture of the soul seemed too simplistic. This is what I mean. The idea of a linear scoring pattern for decisions turns humans into a big computer or automaton. Thus, when Aaron talks about “equally balanced situations,” he is thinking of both choices getting the same score.

My view is admittedly much messier than Aaron’'s. Even if the crude “points” language is removed, his view is very odd. There just seems to be something wrong with this claimed ability to decisively compare choices. I do not want to defend balanced situations; I want to defend incommensurable situations. What exactly are incommensurable choices? Kane defines them this way: they are both desired more than all the other alternatives for different, noncomparable reasons. Although I believe that incommensurable choices face us more than we might think, I am willing to allow a lot of determinism in life. The important point is that in some situations, we can look at our choices, and be free to choose.

This is the intuitive picture of the world that we all have. We believe that we do actually choose. But in Aaron’'s view, we do not choose. For what is choosing if we lack the power to choose the negation?

Choosing for Reasons

...a decision still has to get made. How? Or on what basis? Is it something which finally makes one action more appealing? Or is it nothing? If it's something, we're back to determinism. If it's nothing, we're with Bill the robot. Positing "free will" itself won't help because the question is how that free will actually choses in the hard cases. If the free will doesn't choose one over the other because of something, why isn't it random? And if it is random, what does that do to moral responsibility? We can hardly blame someone for something random.

Aaron continues to offer the dilemma: either we choose based on something, or on nothing. If something, then determinism. If nothing, then we are acting on chance. This is the classic compatibilist argument against libertarian free will. There are interesting aspects lurking below the surface of this story, including the assumption that exercise of free will is an event. However, I will face it head-on and take on one of the horns of the dilemma.

Suppose this general scenario:

1) reason R-1 is sufficient for action A-1 and R-2 is sufficient for A-2
2) R-1 and R-2 cannot be compared in a scoring sense.

When I choose A-1, Aaron claims that I need some reason R-3 that explains why I chose R-1 instead of R-2. If I don’t have R-3, then I am on the same level as Bill the robot. However, this could go on ad infinitum. Why didn’t I choose R-4 as my reason instead of R-3? Is that random, or because of R-5? Determinist or indeterminist, this claim of reasons being necessary for reasons is absurd because of its infinitely regressive nature. The vicious regression poses a problem for both sides. The determinist usually escapes this cycle by passing the buck to outside factors. It is thus disingenuous (to a degree) for Aaron to cite “character” as the reason for action. If his infinite criticism holds, then the ultimate reason for every are not as nice as character or goals, but 1) the initial state of atoms and minds and 2) the causal laws that govern mind and matter.

Take another example of the infinite regression problem: knowledge. Descartes believed that you must always know that you know. In modern epistemology, it is accepted that it is clearly possible to know something without knowing that you know it, or knowing that you know that you know it. Just as it is acceptable to know without knowing you know, it is acceptable to act on a sufficient reason without having a reason for that reason. In many cases, it is better to know that you know, and in the same way, it is better to have reasons for reasons. It just is not required.

My view is this: We do not need reasons for reasons. We only need reasons for choices and actions. The choosing from reasons is properly basic, and need not be explained on a deeper level. (2) entails that I cannot choose the stronger reason, because there is no stronger reason. Aaron believes this to be random. Is it? I believe that as long as an intelligible continuity can be seen along my choices, it is sufficiently non-random. This is ensured by the formation of reasons from within my character. This aspect of creation of the self is important to fend off charges of randomness.

Frankfurt’'s Mistake

Harry Frankfurt wanted to show that we could have moral responsibility even in cases without being able to do otherwise. I have hesitated to post a argument against him, as there are very many, but I am unhappy with various implications of most. I will stake my ground now.

Frankfurt'’s counterexample (and those inspired by him) fail because they still presuppose causal determinism. In the main world, Jones chooses to perform A in accordance with Black’s wishes. In the counterfactual world (which we will deal with), if Jones shows an inkling of choosing not to perform A, then Black “zaps” Jones so that Jones desires to perform A. Black cannot zap Jones after Jones has chosen. For if Black did, it would be a straightforward case of overt coercion, which even compatibilists do not like. The only way for Black to guarantee that his zapping will ensure that A is still chosen is if the zapping is causally determined. If the counterfactual world was not causally determined, then Jones could indeed actually choose A and perform it. Thus, Frankfurt’'s example holds in causally determined worlds only.

This is hard to discuss without access to the original paper, or at least a FSC paper. Fortunately, I found a copy online.

The Role Of Scripture

Being Christians, the problem of scripture must be addressed at some point. I will reiterate what I mentioned in the comments of a previous post. General revelations, both empirical (science, history) and analytical (philosophy), are conceptually antecedent to specific revelation (the Bible). Lyell and Copernicus affected our understandings of Genesis 1 and Joshua 10. To even a greater extent, must philosophy color our understanding of passages in the Bible. Thus, if it can be shown that compatibilism is clearly incoherent, then compatibilism is not an option for interpretation. The debate of free will, then, need not involve scripture at all.

Once again, conceptual antecedence is not chronological antecedence. My view does immobilize believers from reading and learning from the Bible.

In the comments, Robert seems to advocate a Kierkegaardian picture. He sees incompatibilism as rational, and the truth of compatibilism must be revealed through faith. The fideists inspire and challenge me, but I ultimately disagree with them. My view of Christianity is what Victor Reppert terms critically rational. I believe that God is the author of all truth and each truth thus reinforces and supports my faith.

If it is true that salvation is by grace alone (the usual Protestant position), then we cannot and should not be morally responsible. Thus, while I would like the PAP, in some sense, to be satisfied in salvation, it does not severely affect my view if PAP is not fulfilled.

Xon wishes that I discuss the passages. I am hesitant to; lack of knowledge and training may lead me down the wrong paths. With that said, I see no reason why the elect (referred to in Ephesians 1 and Romans 9) cannot refer to the Gentiles as a whole. My reading of the passages then is that God had always planned to save the Gentiles. The snatching out of the hand then refers to those attempting to coerce believers into falling away (through torture or lies). A passage in favor of incompatiblist free will is the Lord's prayer itself. "Thy will be done in earth as it is in heaven." (Matt 6:10) Somehow, God's will is not being done on earth right now. If he is controlling everything, then it must be his will that his will is not done. That is like saying, "Never pay attention to anything I say."

Conclusion

There is an intuitive and yet exciting aspect to this theory. We are in control and responsible for our choices. There is some measure of objective worth that is being met. In addition, when we choose between incommensurable alternatives, it is often at hard times in our lives. In a very real sense, we form our souls at those moments, determining many future actions. We have great power and thus great responsibility.

Compatibilism and Christianity - a response to AJ Armitage

Aaron Armitage, of the excellent blog Calvinist Libertarians, has posted a reply to my initial take on free will. He wants to defend compatibilism from my claims. The thrust of his argument is a strong assertion: compatibilism is required for moral responsibility. He also objects to my second post as attacking a straw man. I am afraid that I have not been as clear as I should in either. This is what comes of a lack of planning.

Aaron starts off by attacking my definition of determinism. I'll try to defend it and clear up any misunderstandings that my readers might have gained.

Aaron's new definition of determinism comes from Dictionary.com: "The philosophical doctrine that every state of affairs, including every human event, act, and decision is the inevitable consequence of antecedent states of affairs."

There are two ways in which determinism can be true. Every event can be deterministically caused, or every event can be known before it happens. My two posts were separated so as to distinguish between the two. Christian and naturalist compatibilists alike can argue that causal determinism is true, but Christian compatibilists can bring to bear another mechanism for determinism: God’s knowledge of the future. As Xon noted in a comment below, it is "accidental" determinism, in some sense. One point on which Aaron is very vague is as to the mechanism for his determinism. He does not believe that either the laws of nature or knowledge of the future is the mechanism.

The state of the world alone is not sufficient to explain determinism. Dictionary.com is silent on this issue probably out of recognition that determinism has room to accommodate many different mechanisms. I should note also that I mean mechanism in a technical philosophical sense, and not as specifically machine-like. The job of a biologist is to explain the mechanism of the fruit from the trees.

Now, onto the arguments. Aaron looks repeatedly to agents post-choice, notes that determinism is required, and then concludes that determinism is required pre-choice. His criticisms miss my argument, but it is most likely my fault for not being clear. So, I will go through his argument and show where I part ways with Aaron. I should note first that I have no objection to the world being deterministic post-choice; it is pre-choice that I believe we need to preserve libertarian free will.

Aaron’s first critique: “...an agent's character, desires, goals, knowledge, and so forth are part of the antecedent state of affairs. How can an agent who doesn't act from his own goals, etc., be said to have free will?”

The mere fact that mental states and dispositions are part of the antecedent state of affairs does not make them deterministic. It seems obvious that after I choose, the link between my choice and my action is causally deterministic. It does not seem obvious that, when faced with multiple sufficient causes or incommensurable belief sets, the choice is determined. There are situations in which two different paths of action can both be consistent with one’s character, some desires, some goals, and so on. Aaron's picture of the human soul seems too simplistic. Why must the outcome of deliberation always be determined before we start?

I noted in an earlier post that an agent could have two sufficient reasons to perform A. Just in the same way, an agent could have a sufficient reason to perform A and a sufficient reason to perform B. Prima facie, this looks like a problem for Aaron.

Aaron’s second critique: “Let's take two hypothetical actors. Bill is a robot. He looks just like a person, but he has no intentions or wishes, and no interior experience of any kind. His programming does take input from sensory data, but this interacts with a complicated algorithm that takes most of its inputs from random quantum events before producing a course of action, which means that Bill behaves randomly. Bob is an ordinary person, but he's a little obsessive. Or rather, a lot obsessive. There's one thing he wants more than anything. Say, a woman. Every day he's eaten up with desire for her. He would, literally, die to get her, and he would consider it the best deal he ever got. She's the only thing he wants or can think of wanting. You get the picture. Now suppose she offered herself to one of them. Bill's "choice" of whether to take her up on it would seem to satisfy the PAP more than Bob's.”

I’ll take these characters one at a time. First, Bill. I do not understand what the problem is with Bill. Sure he satisfies PAP. In fact, Robert Kane might say that, from a purely naturalistic standpoint, Bill is what a free agent would look like. On the natural level, all we can observe in the brain are quantum cascades in neurons. However, PAP is not sufficient for freedom of the will. It is only one of several necessary conditions, another being a will.

Now, Bob. At this point, Bob is clearly not free. If he were to do otherwise, he would be acting out of character. Aaron and I are in agreement so far. Let us rewind the tape of Bob’s life back to when he first gained the desire for this woman. At some point, Bob must choose whether or not to will to pursue this woman. He must go from desire to volition. For, Bob has more than desire now. He has an obsession. Thus, once again, Aaron is looking at agents post-choice and demanding that their actions be determined. I have no objection. But this hardly proves anything pre-choice, which is where I am interesting in having an incompatibilist free will.

Aaron wraps up by noting two reasons why he is a determinist. First, he apparently used to be an indeterminist and lost an argument to an atheist. I’m sorry to hear that I lost an ally. I just want to note that although incompatibilism is currently a minority view in philosophical circles, it is highly respected. The reason for compatibilism being favored is that naturalism is favored in academia, and as I noted, naturalism seems to entail determinism. I’'m not saying that Aaron is a closet naturalist. I am noting that if Aaron is correct, naturalism offers moral responsibility, as well. It seems to me that naturalism cannot offer moral responsibility for incompatibilist reasons.

The second reason Aaron cites is that Christianity requires compatibilism. I generally try to stay out of theological debates, as I have found them to be filled with polemics. I will say a few things here. I am a Protestant like Aaron, but I believe that the Bible requires incompatibilism. I believe this because free will is necessary for moral responsibility.

UPDATE (4/21/05 8:40am): Aaron has posted a reply to this post. You may find it here.

God: Author or Chess Grandmaster?

Lars Walker has a pair of fun, thought-provoking posts over at Brandywine Books about Calvinism and predestination. Predestination is another catchword in the debate over free will. Predestination is foreknowledge with the perspective flipped towards us: God foreknows one’s action A iff one is predestined to perform A. I hope that I have atoned for any sin of omission that I committed in my previous post.

In the first post, Lars is drawn to the analogy of God as an author and humanity as the characters in God’s book. He writes:

It occurred to me that when I build characters I’m extremely ruthless with them. If I create a character to be evil, evil he is. And if I bring him to an evil end, I feel no guilt whatever at punishing him for something I caused. Every evil deed of his sprang originally from my own mind, but I punish him anyway. And it feels perfectly right.

The analogy comes close, but ultimately fails because the hypothetical author has causal powers over the actions of the characters. In the question of CC, causation is unimportant. Note that if Tony Blair knows beforehand that I will eat pancakes, Tony Blair does not in any serious sense “cause” my eating of pancakes. I tried to address the problem of causation in my first post. The only addition to the discussion is a implicit premise that omniscient causation is an appropriate power for God to have. The conflation of foreknowledge and causation tends to reoccur in theological discussions. In addition, the difference between punishment and moral desert is also often conflated, as I think Lars does here.

The proper analogy for the predestination claim is that of a reader (not an author) to a book.

I’d like to comment on Lars'’ second post, but my head hurts too much after reading it. I suggest that you read it and see what you think. My only aside is that if this scenario is true, I think the dragon of theodicy rears its head even more strongly than usual. I do not know that even the wisest of Men can slay it.

Amanda Witt reprints a passage by Phil Yancey about the power and knowledge of God. The critical part is this:

Although I had complete freedom to make any move I wished, I soon reached the conclusion that none of my strategies mattered very much. His superior skill guaranteed that my purposes inevitably ended up serving his own. . . . When a Grand Master plays an amateur, victory is assured, no matter how the board may look at any given time.

Yancey, I believe, describes the middle path between open theism and Calvinism much better than I could. Many of our actions are not determined, but our freedoms cannot significantly alter the way history ends. Here, PAP and the sovereignty of God can both be preserved. We walk through the garden of forking paths and are free to choose our route, but all paths will get to the gazebo. (I hope I'm not carrying the metaphor too far.) In this view, God is still omniscient, as well. While he does not know which path we will take, God still sees all the paths. To take a whimsical view, imagine God writing a book of every action on Earth. It would look a lot like a logic book written entirely in conditionals. The final page would be say: "Jesus returns on the white horse and sets up a second heaven and earth. QED." QED of course means quod erat demonstrandum, or what was to be demonstrated. All of history, despite being written in conditionals, ends up with the same conclusion, logically and historically.