On Being a Catholic Philosopher
Some people have asked me to address the question "Can one be a Catholic and a philosopher?" as a result of various provocations (including a recent blog post in which the question came up, but which I've not had time to read). I paste below a very quick email in response to these requests.
----------------
Haven’t had a chance to look, but it's hard for me to imagine what the point of the question is. The only interesting question I can see in the vicinity is this: do one’s commitments as a Catholic prevent one from having the intellectual freedom to do philosophy as it should be done? The answer to that question is also clearly yes. Every philosopher has certain commitments which are or nearly are “ungivupable” because of what seems most clearly true to them. Sometimes these are called our “intuitions”. Some intuitions are stronger than others and sometimes they shift. For many (I’d guess most) philosophers there are certain intuitions which are pretty strong and stable, for example that there are no true contradictions. I don’t take seriously the idea that there are, but I know there are smart people who think there’s something to paraconsistent logic and I’ve been convinced in the past of the necessity of certain deviations from classical logic, so I’m not 100% certain that there are no true contradictions (though I don’t even at this point know what it would mean to say that there are). An even better example is the Moorean response to skepticism. I’m very convinced that there is an external world. Much more convinced than I am of the disjunction of the conjunction of all the premises of all the skeptical arguments of which I am aware. I don’t take skepticism seriously in that I’m not the *least* bet tempted to believe skepticism. However, I do take it seriously as a *puzzle* I’m convinced that we learn a lot about rationality when we consider the skeptical puzzles. Nevertheless, it’s conceivable to me that things could change. Likewise, I’m quite convinced that Catholicism is true. I’m not very tempted to believe any objection to its truth of which I am aware. However, I do take such objections seriously in that I think there is much of value to learn from pondering them. Still, it’s conceivable that things could change: if I became more convinced of the conjunction of the premises of some argument the conclusion of which entailed the negation of the core claims of the Church, then I’d stop being a Catholic. In short, that Catholicism is true is one of the things about which I count as being rationally convinced. So other, less-certain items have to make room for it (just like they have to—to different degrees—the law of non-contradiction and the existence of the external world). But I’m a good neo-Bayesian, so I don’t have 100% confidence in anything. All propositions are negotiable. A better question is Can a Naturalist be a philosopher? It has been noted that naturalism is not really a philosophical approach to anything, but rather a stance or attitude (see van Fraassen’s essay in Kvanvig’s volume on Plantinga and recently my colleague Ney “Physicalism is an Attitude” but there are others in between such as Perry’s “antecedent physicalism”). It’s essentially a kind of fideism, but fideism has been condemned by the Church so though it is an option for Naturalists and Protestants, it simply isn’t an option for Catholics (modulo what’s said above). The use of the terms "certitude," "certainty," and "doubt" have pretty narrow usages in traditional Catholic epistemology, so don't be mislead by statements before Vatican II on the subject.
----------------
Haven’t had a chance to look, but it's hard for me to imagine what the point of the question is. The only interesting question I can see in the vicinity is this: do one’s commitments as a Catholic prevent one from having the intellectual freedom to do philosophy as it should be done? The answer to that question is also clearly yes. Every philosopher has certain commitments which are or nearly are “ungivupable” because of what seems most clearly true to them. Sometimes these are called our “intuitions”. Some intuitions are stronger than others and sometimes they shift. For many (I’d guess most) philosophers there are certain intuitions which are pretty strong and stable, for example that there are no true contradictions. I don’t take seriously the idea that there are, but I know there are smart people who think there’s something to paraconsistent logic and I’ve been convinced in the past of the necessity of certain deviations from classical logic, so I’m not 100% certain that there are no true contradictions (though I don’t even at this point know what it would mean to say that there are). An even better example is the Moorean response to skepticism. I’m very convinced that there is an external world. Much more convinced than I am of the disjunction of the conjunction of all the premises of all the skeptical arguments of which I am aware. I don’t take skepticism seriously in that I’m not the *least* bet tempted to believe skepticism. However, I do take it seriously as a *puzzle* I’m convinced that we learn a lot about rationality when we consider the skeptical puzzles. Nevertheless, it’s conceivable to me that things could change. Likewise, I’m quite convinced that Catholicism is true. I’m not very tempted to believe any objection to its truth of which I am aware. However, I do take such objections seriously in that I think there is much of value to learn from pondering them. Still, it’s conceivable that things could change: if I became more convinced of the conjunction of the premises of some argument the conclusion of which entailed the negation of the core claims of the Church, then I’d stop being a Catholic. In short, that Catholicism is true is one of the things about which I count as being rationally convinced. So other, less-certain items have to make room for it (just like they have to—to different degrees—the law of non-contradiction and the existence of the external world). But I’m a good neo-Bayesian, so I don’t have 100% confidence in anything. All propositions are negotiable. A better question is Can a Naturalist be a philosopher? It has been noted that naturalism is not really a philosophical approach to anything, but rather a stance or attitude (see van Fraassen’s essay in Kvanvig’s volume on Plantinga and recently my colleague Ney “Physicalism is an Attitude” but there are others in between such as Perry’s “antecedent physicalism”). It’s essentially a kind of fideism, but fideism has been condemned by the Church so though it is an option for Naturalists and Protestants, it simply isn’t an option for Catholics (modulo what’s said above). The use of the terms "certitude," "certainty," and "doubt" have pretty narrow usages in traditional Catholic epistemology, so don't be mislead by statements before Vatican II on the subject.
posted by Trent_Dougherty at Friday, October 19, 2007
45 Comments:
Trent:
Thanks for the thoughtful post!
A quibble:
I'm worried that you may be equivocating on "fideist". The fideism that is condemned by the Catholic Church (e.g., in Vatican I) consists of two doctrines:
1. The existence of God cannot be known by reason, but only by faith.
2. No rational justification can be given for Christian doctrine, and we can only be justified by faith.
Fideism as a doctrine thus contains two parts: that some doctrine cannot be known by reason, and that only a justification by faith is possible.
If this is right, it wouldn't be fideism to accept the basic doctrines of Christianity solely by faith apart from all justification. (Indeed, to be pedantic, it would be a category mistake to say that it's fideism, because fideism is a doctrine, and what we're talking about is a doxastic action--an acceptance.) Nor would it be fideism to claim that faith is sufficient for justifying belief. (In fact, the idea that faith is sufficient for justifying belief is implicit in the teaching of the First Vatican Council.) It would be fideism to add to that the claim that only faith can justify belief.
So the naturalist position is not a fideism unless it comes with a self-conscious epistemology according to which naturalism cannot be justified rationally. Some naturalists take this stance. But others think that they have arguments for why naturalism is rational, in terms of Ockham's razor, or the success of science, or the like. These arguments for naturalism are no good at all. But it won't make Anselm a fideist if the Ontological Argument turns out to be no good at all.
More seriously, I think you're underestimating the traditional views on the certainty of faith. If I am to submit every thought to Christ, I cannot take
"seriously" any objection to, say, the existence of God, if to take an objection seriously means to allow that there is a non-zero epistemic probability that the objection succeeds. Every action of ours ought to be an exemplification (or something like that) of our love for God. But the action of allowing a non-zero probability to the non-existence of God is not an exemplification of our love for God.
We can, of course, take the objection "seriously" in another sense, that of examining it carefully and honestly, maybe even strengthening it, while of course remaining utterly convinced that the objection is unsound.
I do want to say something to the general question. Philosophy is a means to an end, the end being the possession and practice of wisdom. Granted, the means has non-instrumental value, but the primary value is the end. And here I am struck, with St Paul and the Fathers of the Church, at how much wisdom the ordinary, uneducated, serious Christian has. Indeed, the ordinary, devout Christian has a depth of wisdom that the non-Christian philosopher has not heard of. I see the non-Christian philosopher laboring to find wisdom, and this labor is a noble and valiant endeavor, whereas the Christian already has just about all this wisdom. And the having of the wisdom is more important than the search for it, because it is only after one has found wisdom, that life's real task begins--the task of living by this wisdom.
Thanks Alex, I *knew* you were going to pick on my use of "fideism" I was sure enough that I went to the Catholic Encyclopedia and looked it up, and almost quoted some councils and encyclicals, but, alas, I simply didn't have anything like the time. I still think that the fideism I point to is in the *spirit* if the fideism condemned. And one of the encyclicals the CE quoted said something (I'm sure it's an allusion to agustine) that reason preceeds faith and faith is rewarded with reasons. So those who accept "on faith" the existence of God or other claims--and for reasons Saint Thomas points out at the beginning of SCG it may be necessary--I take that to mean on the teaching authority of the Church. But one must have reason so to trust (I don't think it's hard to come by) so that the typical one who accepts dogma "on faith" is in fact not a fideist: they have sufficient reason to trust what the Church says.
Where we have a big difference of opinion is on whether there is anything untoward about allowing for non-zero epistemic probability of a proposition which entails that God does not exist. That Faith entails a certain kind of Certitude is a traditional teaching, but "certitude" is usually defined as the absence of doubt where "doubt" entails the inability to assent. I think this is VERY important, for I can clearly *assent* to propositions even when I recognize non-zero epistemic possibility of propositions that entail their negations. I think this is key for understanding some of the seemingly harsh things that Newman says in the Grammar of Assent. Assent is a propositional attitude more like some uses of "accept" and does not come in degrees. Whether belief comes in degrees, I think rational credence does and it's *that* that I take must be proportioned by the rational agent to the degree of evidential probability.
In my more Bayesian moments I don't even recognize such a thing as "belief". As Glymour wittily said, we don't *give* answers we *rate* them. However, I clearly understand the question (I tend to like Mark Kaplan's explication): Do you believe in God or don't you? My answer to that question is Yes, so I don't have any "real" doubt (and in the traditional Catholic literature this adjective is often present) in the sense that I have no doubts of such a degree as to cause me to fail to assent to the proposition. So, lacking doubt, I have a degree of certitude (the tradition leaves open whether and in what manner certitude comes in degrees, my own view is that we can have different degrees of certitude in that neither of us can have any real doubt--we both assent--yet I'm *closer* to being in a position to fail to assent than you are).
Dear Trent:
Thanks for the thoughtful reply.
It may be that we're talking at cross-purposes. For instance, you might think that we do not assign probabilities to propositions--that credences are not up to us. Hence, a failure to assign probability one to a proposition can't be a failure to love God.
It may also be that maybe I shouldn't even talk of assigning a probability one to the propositions of faith. Maybe instead I should talk of unconditional commitment to the propositions of faith, i.e., a commitment to believe them no matter what should befall (after all, as Vatican I says, "those who have accepted the faith under the guidance of the church can never have any just cause for changing this faith or for calling it into question"). And maybe this unconditional commitment is not the same as assignment of unit probability. It may be that I am conflating epistemic commitment with moral commitment. (I probably am doing that. After all, I don't recognize a distinction between epistemic reasons and moral reasons.)
Still, it does seem to me that to assign a probability less than one to a proposition p typically involves a commitment to rejecting p under some circumstances. And I am not sure how the love of God can give one reason to make such a commitment when p is the proposition that God exists. (It seems clear to me that the love of God, or at least the lovability of God, is supposed to be our reason for everything that we do and think.)
Compare the following. It seems to me that it would be bad if one made a commitment to hate God or to torture the innocent for fun under circumstances C, no matter how unlikely one believed C to be. We should live the love of God wholeheartedly, and I think that this rules out having even hypothetical commitments incompatible with the love of God. Such commitments smack of the prenuptial agreement.
Now maybe you will say that this is all OK if one thinks C is logically impossible or something like that (even if one doesn't think that C has zero epistemic probability). Maybe this way out will work.
Another thought: I think faith gives definitive reasons to believe (cf. the previous quote from Vatican I). So everyone who has faith, has definitive reasons to believe. If fideism* is the idea that by faith we believe some doctrines that we have no epistemic justification for, then fideism* is completely mistaken, since everything that we believe by faith we have epistemic justification for--through faith. Faith is a supernatural virtue, something wrought in us by the Holy Spirit; hence, any false propositions are not actually believed by faith. (One issue here is that a lot of folks in our culture--but I think not you--use "faith" in a different sense, one on which we might say that the Muslim accepts by faith that God is not a Trinity. But that is not faith, since it is not produced by the Holy Spirit.)
My feeling (based on what I've heard about the context) about the "certainty" in the context of Vatican I's teaching that the existence of God can be known with certainty by the light of natural reason is that the Council is talking of the kind of certainty that apodeictic argument starting from self-evident premises induces. (Can one assign a less than unit probability to self-evident premises?) Moreover, the Tradition will, I think, say that the certainty of faith is greater than that.
A final note. I hypothesize, without having any texts at hand to support it, that the Tradition would agree that there is no proposition of reason which is more certain than a proposition of faith. If so, then tautologies are not more certain than propositions of faith. Hence, if tautologies get probability one, propositions of faith a fortiori do so as well.
These are hard questions, and it may be that you're right that I'm mistakenly imposing a modern understanding of the word "certainty".
Trent,
1. "... do one's commitments as a Catholic prevent one from have the intellectual freedom to do philosophy as it should be done? The answer to that question is also clearly yes. Every philosopher has certain commitments ..."
So no philosopher can do philosophy as it should be done? If so, why exactly the commitments prevents this?
2. "I've been convinced in the past of the necessity of certain deviations from classical logic, so I'm not 100% certain that there are no true contradictions (though I don't even at this point know what it would mean to say that there are)."
Very interesting. Would you have some paper, draft, tip, link, etc. about these convincing deviations?
Would not every argument for such a deviation entail that there is no such deviation?
Cf.: "... Quine's own position on truths of logic runs serious risks of incoherence. As Anthony Quinton points out, the revision of a rule of inference such as Modus Tollens amounts to declaring an immense logical amnesty; any propositions discarded because of their failure to fit the web ouf our beliefs will need to be reexamined, since they may have been ruled out on the basis of a (now) obsolete rule of inference. The resulting chaos in the "web of belief" cannot be passed over lightly with the comment that we will usually prefer not change one of the theorems of logic. All of the theorems of ordinary SL are (trivially) interderivable, so in the name of consistency we will have to discard them all. Or will we? Perhaps the reasoning we would use to discard them would rest on Modus Tollens itself? Such a move will not isolate the damage, since any proof by Modus Tollens can be recast as a hypothetical syllogism or even a Modus Ponens. In the resulting confusion, it is not at all clear that the idea of incompatibility which is so vital to the process of criticism - the correction of certain beliefs in the web on the basis of more or less intractale expreriences - can survive. Nothing can count against the principles of logic, since it is only if they are accepted that anything can count against anything else." T. J. McGrew, The Foundations of Knowledge, Lanham 1995, p. 42. (The reference to Quinton: The Nature of Things, New York 1973, pp. 146, 216-7.)
3. "I'm a good neo-Bayesian, so I don't have 100% confidence in anything. All propositions are negotiable."
Cf. the quotation from McGrew. Further, I thought the principles of classical logic are 100% certain background of every Baysianism and every form of probability calculus.
Trent and Alex,
Trent wrote: "The use of the terms "certitude," "certainty," and "doubt" have pretty narrow usages in traditional Catholic epistemology, so don't be mislead by statements before Vatican II on the subject."
4. What EXACTLY, according to you, Trent and Alex, is the usage of the mentioned terms in traditional Catholic epistemology?
5. How should I interpret the Catechism of the Catholic Church, 157?: "Faith is certain. It is more certain than all human knowledge because it is founded on the very word of God who cannot lie. To be sure, revealed truths can seem obscure to human reason and experience, but "the certainty that the divine light gives is greater than that which the light of natural reason gives."
Alex,
6. Do you have 1OO% confidence in (some) principles of classical logic? Do you have 100% confidence in (parts of)the Creed?
Thanks a lot.
In Christ,
V.
V:
I am a bit uncomfortable talking about confidence. Confidence is something psychological, which waxes and wanes. One might wake up at night from a nightmare, in a cold sweat, not feeling confident of anything, not completely sure of who or where one is.
So, I don't know what it means to say that I have 100% confidence in a proposition.
Does it mean that I couldn't cease to believe the proposition no matter what? As a libertarian, I think it is causally possible for me to reject grace, and start living an epistemically and morally reckless life, thinking muddily, to the point of denying the existence of God and the principle of non-contradiction.
Does it mean that I wouldn't cease to believe the propositions no matter what? Since I am not a Molinist, the question doesn't make sense to me, and if I were a Molinist, I wouldn't know the answer to it.
Does it mean that I always act in accordance with the propositions? If so, then my confidence in the propositions of faith is not 100%, because I keep on sinning. But such a measure of confidence neglects akrasia and freedom of the will?
I think the best thing to do is not to talk of something purely psychological, like confidence, but of something normative, like commitment.
I am unconditionally committed to the doctrines of faith. What does that mean? It doesn't mean anything really impressive about me or my confidence. It primarily means that if I should ever disbelieve these doctrines, no matter for what reason, I would be going against my commitment. In this way, the doctrines of faith contrast with the belief that I now have two hands. There, my commitment to hold on to the belief is conditional. If you should show me that I am likely hallucinating, point out drugs that I look like I have taken, etc., and then I were to abandon my belief that I have two hands, my abandonment would not go against my commitment.
Secondarily, that I am 100% committed also means that I am aware of and accept the normative aspects of the commitment. So there is a psychological component in "commitment" as ordinarily understood. But what is of most significance is the normative component.
This is like my marriage vows. I am unconditionally committed to fulfilling them. That does not primarily indicate a psychological state, or mean that I would never break them or anything fancy like that. Barring grace, surely torture and brainwashing could get me to break them, and anyway I am on a daily basis falling far short of the self-sacrificing spousal love that I am committed to. What it primarily means to say that I am unconditionally committed to fulfilling them is that if I should break them, I would be going back on my commitment.
Secondarily, that I am thus committed to my spouse also means that I am aware of and accept the normative aspects of the commitment.
I notice that I didn't answer the question whether I am confident of the laws of classical logic. I don't think I have unconditional commitment there. But at the same time, I am right now completely psychologically confident of them and quite unwilling to revise them. :-) I don't know how that maps on to the probability calculus. I doubt that we have numbers in our heads attached to beliefs.
Thanks, Alex.
I took the term "100% confidence" from Trent. I have meant (epistemic) probability; which is a concept both you and Trent have used here.
So let me reformulate:
6'. Do you rate the (epistemic) probability of (some) principles of classical logic as 100%? Do you rate the (epistemic) probability of (some parts of) the Creed as 100%?
I know you wrote: "I don't know how that maps on to the probability calculus. I doubt that we have numbers in our heads attached to beliefs." However, it seems you also wrote we CAN assign such probabilities: e.g., "it does seem to me that to assign a probability less than one to a proposition p typically involves a commitment to rejecting p under some circumstances." See also my question 7.
7. "the Council is talking of the kind of certainty that apodeictic argument starting from self-evident premises induces. (Can one assign a less than unit probability to self-evident premises?) Moreover, the Tradition will, I think, say that the certainty of faith is greater than that." This seems right, see CCC 157. However, in which sense is "the certainty of faith GREATER" (when no probability can be greater than 1)?
Trent,
Sorry for missing your following claim, relevant to my 4th question: ""certitude" is usually defined as the absence of doubt where "doubt" entails the inability to assent."
This really seems to modern eyes as an idiosyncratic and weak definition of "certitude" (because one can easily dream up a situation when a person assents to a proposition which she rates as not particularly probable, maybe she even could rate is as improbable).
Two, no three things:
First, regarding faith and commitment. I think the sort of extreme expression of commitment which love calls for is indeed something more Pascalian than cognitive. So the things Alex says about that are in the right direction I think.
Second, I honestly don't think the certainty discussed in this context in the discussion is apodictic certainty. In the traditional writings I have read most recently and have the firmest memory of the relevant notion is lack of doubt where doubt is essentially assent-blocking. The Medievals just didn't have a very fine-grained epistemology or philosophy of propositional attitudes: belief is to "think with assent", so the role serious doubt plays is to block belief by blocking assent. So I think it's not a good idea to use a modern fine-grained account of propositional attitudes. Indeed, there is in the tradition an extant debate on whether there can be degrees of certainty or degrees of doubt. The tradition here clearly seems to me to be in the process of development. I do wish to reiterate that Mark Kaplan's assertory Bayesian explication of belief is such that I could have considerable doubt and still believe.
Third, regarding the possibility of apodictic certainty I want to say two things. A. There is in the tradition a distinction between what status it is possible for a *proposition* to have, and there is what status it actually has for any given person. Saint Thomas thinks he's proved the existence of God pretty rigorously, but still admits that it needs to be revealed because it would otherwise take people a lot of time, trial, and error to get it right. So I can say that I believe the proposition *can* be known with certainty but not myself know it. This is a complicated picture because if I've got evidence to believe that, holding the requisite stuff fixed, possibly it can be strictly proven then there are some plausible principles of reflection tells me to have that degree of credence *now*. This seems to me to be a very interesting context for discussions of such principles.
B. Saint Thomas says that Sacred Doctrine is a science because, roughly, it's a deductive system based on first principles. Yet some sciences get their first principles from higher science. The first principles of the science of Sacred Theology are revealed by God (who cannot be mistaken).
The passage is worth quoting.
I answer that, Sacred doctrine is a science. We must bear in mind that there are two kinds of sciences. There are some which proceed from a principle known by the natural light of intelligence, such as arithmetic and geometry and the like. There are some which proceed from principles known by the light of a higher science: thus the science of perspective proceeds from principles established by geometry, and music from principles established by arithmetic. So it is that sacred doctrine is a science because it proceeds from principles established by the light of a higher science, namely, the science of God and the blessed. Hence, just as the musician accepts on authority the principles taught him by the mathematician, so sacred science is established on principles revealed by God.
So this gives us a conditional: If God says it, it can be accepted with certainty. (GS ---> AC). This is some thing of which we can be certain, we can know it as well as we can know anything, so let's use a 'K(...)' operator to signify that: K(GS ---> AC) so, ceteris paribus, justification will transfer across. But of course Saint Thomas himself draws our attention to the distinction between "necessity of consequence" and "necessity of consequent". So if I have rational credence to degree d in GS, I can have no less rational credence in AC. However, the trouble comes in with our credence in GS. It's very hard for me to see how Alex's love argument can apply where one is not even sure of the existence of the other individual. You might be able to show--though I doubt it--that it is unloving of me to have non-zero credence in the possibility of her cheating on me, but if I wasn't even sure I had a wife, it seems a bit of a stretch.
Trent:
On the love argument, try this. Suppose I love Fred but am not completely sure whether Fred exists. This might happen. For instance, I correspond by email and the correspondence comes to me with an @mit.edu address, ending with "Yours, Fred". I come to believe that Fred exists and I come to love him. However, I am also aware that there are lots of clever geeks at MIT, and so I remain somewhat open to the possibility that Fred doesn't exist, and instead I am corresponding with a different computer each time, running the same software, which is a database retrieval system that has been fed an enormous database of email, and that matches my message against a message from the database, and sends the next reply in the thread from the database, but changing names and adding "Yours, Fred" at the bottom. (I carefully tweaked the example so we probably wouldn't say that Fred exists and is an AI system.)
Now, ask this question: Does my love of Fred give me reason to remain open to the possibility of Fred's non-existence, i.e., to set the credence at less than 1? No. For if Fred exists, there is no gain to Fred, or to me, or to our relationship from remaining open to the possibility of Fred's non-existence. And if Fred doesn't exist, then my love of Fred does not give me any reasons at all, since love of non-existent beings is not reason-giving. (Of course such a love gives apparent reasons, but that's beside the point.)
This isn't so controversial. The reason for remaining open to the possibility of Fred's non-existence is not love of Fred, but it's easy to find a different reason, such as love of other people (I need to be open to the possibility of Fred's non-existence, because if Fred turns out not to exist, I am shortchanging the other people by devoting time to correspondence with Fred which I should be devoting to correspondence with them) or maybe love of truth.
By analogy, remaining open to the non-existence of God does not seem to be something for which the love for God gives a reason. But every action of ours should have either God's lovability or love for God as its reason. And God's lovability does not give us a reason to be open to the non-existence of God since God's lovability is not a reason at all if God doesn't exist.
The Catholic Encyclopedia defines "certitude" not just in terms of lack of doubt, but the positive exclusion of doubt. They give the example where you read the morning paper, and you have no doubt that what you read is true. But then you read the evening paper and you withdraw your belief. This means, they say, your belief was not such as to exclude doubt. It seems to me from the example that a case where one is open to revision will not count as a certitude.
The Encyclopedia also says: "The certitude of faith is supernatural, being due to Divine grace, and is superior not merely to moral certitude, but to the certitude of physical science, and to that of the demonstrative sciences."
So the Encyclopedia's reading of the Tradition are different from yours.
On Aquinas, he also says: "Matters of faith are known with greatest certainty in so far as certainty means firmness of adherence. For the believer clings to nothing more firmly than those things which he holds by faith. But they are not known with greatest certainty in so far as certainty implies repose of understanding in the thing known. For the believer's assent to what he believes does not come from the fact that his understanding concludes to the things believed by virtue of any principles, but from the will, which influences the understanding to assent to what is believed. Hence it is that in matters of faith, movements of doubt can arise in one who believes" (De Veritate, 10, ad 6').
This is quite an interesting passage. Aquinas is claiming that the believer accepts doctrines of faith more firmly than any other truths, presumably including self-evident truths such as that 1=1. At the same time, he says that the believer is not immune to movements of doubt, which I take it are distinguished from actual doubt. What is particularly interesting here is that he explicitly brings in the will here. So he does seem to be holding that (by grace--this is clear elsewhere) the will holds on to the doctrines of faith with a submission more firm than that which we give to tautologies. However, it is not like the case of tautologies where we see a proposition and its justification, so that doubt can gain no foothold. We do not see the justification of the doctrine of faith, though we are in fact justified, and so doubt can gain a foothold in the case of faith (though there cannot be actual doubt).
Also, from the commentary on John, Chapter 4:
'Faith should be certain, because one who doubts in the faith is an unbeliever: "Ask with faith, without any doubting" (Jas 1:6). And so their faith was certain; thus they say, and we know. Sometimes, one who believes is said to know (scire), as here, because scientia [science, knowledge in a more perfect state] and faith agree in that both are certain. For just as scientia is certain, so is faith; indeed, the latter is much more so, because the certainty of scientia rests on human reason, which can be deceived, while the certainty of faith rests on divine reason, which cannot be contradicted. However they differ in mode: because faith possesses its certainty due to a divinely infused light, while scientia possesses its certainty due to a natural light. For as the certitude of scientia rests on first principles naturally known, so the principles of faith are known from a light divinely infused: "You are saved by grace, through faith; and this is not due to yourselves, for it is the gift of God" (Eph 2:8).'
Hence, Aquinas really does think that faith is the most certain of things.
Now of course neither Aquinas nor the Encyclopedia are infallible. However, they seem to be a pretty good guide to what the Tradition means by "certainty" or "certitude".
Trent,
Thanks again.
I think there is a conflict between your and my interpretation of the teaching of the Church, e.g. CCC 157 (cited by me above). I think CCC 157 implies that faith is certain about, e.g., the great claims of gospel or about the Creed. I would call it - loosely - the certainty of faith about the 1st order matters. But you say that faith is not certain about such 1st order matters, it's certain on a metalevel - certain about some higher order matters, like about: if God reveals p, then p.
But see CCC 151: "Faith is first of all a personal adherence of man to God. At the same time, and inseparably, it is a free assent to the whole truth that God has revealed. As personal adherence to God and assent to his truth, Christian faith differs from our faith in any human person. It is right and just to entrust oneself wholly to God and to believe absolutely what he says. It would be futile and false to place such faith in a creature." CCC 146: "Abraham thus fulfills the definition of faith in Hebrews 11:1: "Faith is the assurance of things hoped for, the conviction of things not seen"." CCC 177: ""To believe" has ... a twofold reference: to the person, and to the truth: to the truth, by trust in the person who bears witness to it." CCC 182: "We believe all "that which is contained in the word of God, written or handed down, and which the Church proposes for belief as divinely revealed" (Paul VI, CPG # 20)." And, finally, what is the Creed? The content of faith. Thus, when one has a faith, his faith is certain (CCC 157), and it is certain about, e.g., the Creed. Not only about the propositions like: if God revealed "There is one God ...", then there is one God ...
Secondly, even (some) atheists are certain about: if God revealed p, then p; or about: if it is certain that God revealed p, then it is certain that p. Thus, your interpretation of CCC 157 seems to be incorrect. For CCC 157 refers to and describes certainty that is no possessed by atheists.
Alex,
Thanks, both your and Trent's notes are extremely instructive.
Let me raise again my questions (6') and (7) (see above). I think your answers could wipe away a lot of my confusion.
V. read it again. Personal faith also affirms the antecedent, but not with 100% epistemic probability.
Sorry, Trent, you are right. Still, the objections by Alex remain.
Alex,
I incline to agree with you that "a case where one is open to revision will not count as a certitude" in the sense of traditional Catholic epistemology.
However, what would be a rejoinder to the following three apparent objections to faith which not open to revision?
1. If the divinely infused light of faith is construed in such a way that it always swamps the opposing evidence, then this evidence seems irrelevant.
2. Assigning an indefinitely large evidential value to the testimony of the Holy Spirit (or to the divinely infused light of faith, as Aquinas says) would make Christianity indefeasible—a position non-believers find frustrating (at best) or intellectually dishonest (at worst).
3. Finally, this evidence may seem unfair since it is “private.” Unbelievers will hardly be impressed by alleged evidence to which they have no access.
(Adapted from http://www.johndepoe.com/
Evidentialism_RE_Holy_Spirit.pdf )
Trent,
Do you think these objections to faith which is not open to revision are serious?
Well, given the testimony of faith, there is a sense in which the arguments are evidentially irrelevant. But they are still useful. I prove a mathematical theorem. And then I get empirical confirmation (maybe some physical system that obeys the axioms is seen as obeying the statement of the theorem). That's nice. It at least adds to my psychological confidence. And it may help convince someone who can't follow the proof of the theorem.
Likewise, the arguments are useful even given the light of the Holy Spirit. They are useful for apologetics. And they are useful for adding to our psychological confidence.
Moreover, and this is significant, the arguments are useful for a deeper understanding of the theology
Alex, I think the Aquinas passage is key and I'd love to try to parse "movements". Also, as you point out, there is the typical Thomistic subtlety here (often flagged by the phrase "There are two ways in which a thing may be said to be ____..."). In once sense articles of faith are most certain, but in another, and he is equally clear on this, they are not. Clearly, philosophers--especially epistemologists--especially especially Bayesian epistemologists--are going to focus on the latter sense. I think the statements I've been making are true statements about that sense.
How would I explicate the latter sense? Well, once tried to do this in a paper for Swinburne. The idea is somewhat Pascalian, yet different from standard Pascalian wagering. So I have intellectual doubt about the existence of God, more about the identity of Jesus (though they are very close), etc. By I *believe* them. I intellectually assent to them. I don't think assent comes in degrees (I Kaplan-believe them, and Kaplan-belief does not come in degrees, that's the point of it, to be an explication of belief simpliciter which a Bayesian can live with.). Now, believing it, I'm going to cling to it more than anything, there is nothing I'm going to try harder to live by, etc. However, I do realize that, theoretically, the intellectual part could change. Yet that doesn't cause me to hedge my bets any more than I am going to do so with my wife just because I intellectually realize the non-zero possibility of her cheating on me. This would open up the love analogy you used earlier.)
The no-hedge view of volitional certitude floats freely of the assent stuff. I have previously posted on this conception of faith at The Counsel of Trent under "Religious Posts".
Finally, the CE does note debate on degrees of certitude. I suspect a development of doctrine here. You may recall our discussion at Princeton on development and dissent (I want to write a book called _The Diffident Dissident_. Development differs from dissent, but each require saying where the tradition did not have it quite right, each owe an explanation of how they ended up with the alleged shortcoming. I think it is very relevant that the Medievals lacked a fine-grained account of propositional attitudes (at least one with any systematicity and at times it seems completely lacking).
My views might turn out to be somewhat untraditional. I'd prefer that they not--I'm very much more concerned with what Saint Thomas teaches than the illustrations and examples of the CE, though I do take those seriously as well. Or it might be that they merely appear untraditional when they are really super-traditional in that they represent part of a development of the doctrine on certainty. My own view is that they are an admissible precisification of one vague disambiguation of a term used analogically. :-) (seriously though, that's what I think.)
I want to write about this more than anything else in the world, so, deo volente, you can count on a book on it from me, but, alas, for now I've got to go back to grading mid-terms. Thank you both for the very VERY stimulating and probing questions!!
Trent:
Thanks for your thoughtful answer. The quote from Aquinas came from a big search of the NLX Aquinas database. The impression I got from quickly scanning a bunch of passages about faith and certainty is that he sees the relations between faith and scientific knowledge (in his sense of "scientific", which is deductive) as follows:
1. scientific knowledge yields more understanding of what is going on (we not only see that p is true, but we see why p is true); and
2. faith is more certain (there is more "firmness of adherence") than scientific knowledge.
I think (2) is meant in pretty much the modern sense of "certain". From (1) he gets the "movements of doubt", but given that he gets them from (1), I don't think it follows that these imply any lesser certitude.
And here, I think, Aquinas is on to an interesting bit of psychology. When we think we thoroughly understand why p holds, doubts about p are less likely to arise than if we know p by some other means, even if the other means is epistemically superior. I take apart a mechanical contrivance, and now I understand why it is that lifting lever A causes the wooden gopher to jump out of hole B. This makes me much more confident in thinking that by lifting A, I will make the gopher jump out of hole B. Indeed, if the contrivance is see-through, it might be impossible for movements of doubt to arise. However, suppose that I know the contrivance is a black box, but I know on the testimony of ten mechanical engineers of absolute probity that they have examined the device, found it easy to understand within their professional competence, and are absolutely certain that if I lift A, the gopher will jump out of B. It may well be that I assign a higher credence here. For instance, I may well realize that my own understanding of mechanical devices is limited, and that things might interact in ways that I do not fully understand. My commitment to the claim based on my own observation may be significantly less.
Nonetheless, even if I assign a higher credence to the inference from the evidence of the engineers, movements of doubt may arise, particularly if the result in question is a surprising one.
It is rather interesting that movements of doubt do not match up with degrees of commitment. But it is true. Suppose that Fred knows he is half-drunk and he seems to see something surprising. Then, after he sobers up, he vividly remembers that surprising thing. He may find it difficult to make movements of doubt arise, even though his actual level of commitment is significantly less than one. On the other hand, suppose that Fred learns of the surprising thing from a very, very reliable friend. His actual level of commitment may well be significantly higher, but his susceptibility to movements of doubt may also be higher.
At least, this seems to me to be the phenomenon Aquinas is describing.
All that said, I think the CE quote is relevant, since it reflects the Church's understanding of the term "certainty", a term that the Church used in an important and relevant context in the First Vatican Council. It seems very likely that the Fathers of the Council meant more by "certainty" than just absence of actual doubt.
Alex,
A minor question: don't you SEE THAT and WHY p (i.e., if I lift A, the gopher will jump out of B) IS TRUE even in the case of the testimony of the engineers? You see it through your seeing that ten engineers of absolute probity testify that they have examined the device, found it easy to understand within their professional competence, and are absolutely certain that p.
If so, why such a cognition should not be called scientific knowledge?
Secondly, though I'm afraid of being tedious, let me reiterate my questions (6') and (7).
6'. Do you rate the (epistemic) probability of (some) principles of classical logic as 100%? Do you rate the (epistemic) probability of (some parts of) the Creed as 100%?
I know you wrote: "I don't know how that maps on to the probability calculus. I doubt that we have numbers in our heads attached to beliefs." However, it seems you also wrote we CAN assign such probabilities: e.g., "it does seem to me that to assign a probability less than one to a proposition p typically involves a commitment to rejecting p under some circumstances."
7. You wrote: "the Council is talking of the kind of certainty that apodeictic argument starting from self-evident premises induces. (Can one assign a less than unit probability to self-evident premises?) Moreover, the Tradition will, I think, say that the certainty of faith is greater than that." However, in which sense is "the certainty of faith GREATER" (when no probability can be greater than 1)?
Trent,
What is the definition of a Kaplan-belief (which does not come in degrees)? (I wasn't able to find it on the net.) Thanks!
Vladimir:
I have a commitment to not giving up any de fide proposition, and so I do assign a credence of 100%. I have a lesser commitment to classical logic, though for at least some of the rules of it my credence is 100%. (Propositions with the same numerical credence can still differ in commitment. If a dart with an infinitely thin tip will be thrown at a board, the probability that it will land somewhere is 1. The probability that it will land somewhere other than some specified point x. But the former I am more confident of than the former.)
I just picked up a batch of mid-terms, so all I can do for now is tell you that a Kaplan belief is essentially willingness to assert in the context of inquiry, formally: if S's only goal was to assert the truth and her options were to assert that p, to assert that not-p, or not to assert, she would assert that p.
That's from memory, so apologies if it's not exactly right. I think it does not commit Shope's conditional fallacy which most subjunctive definitions do.
A reply to Alex will have to wait until Tuesday.
Thanks, Trent.
Now, I think, my last question: how to define "assert" in that definition of Kaplan belief?
The definition is supposed to be a definition of a "Kaplan-belief", which you also called "ASSENT"; you also said that it differs from "rational credence" which, according to you, comes in degrees, whereas "Kaplan-beliefs" ("assents") do not.
I just mean, it is somewhat uninformative for me if I am given a definition of "assent" in terms of the verb "to assert".
Thanks for your patience.
Alex,
If you assign de fide propositions a creedence of 100% (assuming one can assign creedences that way), are you being irrational? How can one (did you) rationally "move" from assigning de fide propositions a credence of less than 100% to assigning them 100%? And if you have creedence 100% it follows, doesn't it, that you could never rationally lower your credence in them? This seems odd. Are you saying that if you die and go to some sort of pagan afterlife your belief in de fide propositions won't be shaken a bit? This can't be what the Church requires, right? Maybe what I want here is more of an account of what it is (according to you) to assign a creedence of 100% to a proposition. I apologize if my questions could only arise out of ignorance.
John
John:
You can move to credence of 100% for some proposition p by coming up with a sound argument for p from premises one assigns a credence of 100% to, or by making a leap of faith under the influence of grace, or some combination of these (e.g., accepting some premises by faith and others by reason, or having overdetermination: both faith and reasoning yielding the same credence, etc.)
"Are you saying that if you die and go to some sort of pagan afterlife your belief in de fide propositions won't be shaken a bit?"
That's like asking: "Are you saying that if you die and then by a direct vision of truth see that p and not-p, your belief in the law of excluded middle won't be shaken a bit?"
When I am absolutely committed to denying some proposition, I need have no commitments about how I would doxastically act were I to come to see that proposition as true.
In any case, I do not understand credence in terms of counterfactuals. There are imaginable circumstances where I would deny the faith. E.g., I can imagine being brainwashed, and God withholding grace, and my losing faith then. I can even imagine not being brainwashed and my denying the faith simply because I choose to be morally and epistemically vicious (I am a libertarian).
I am committed to believing the de fide doctrines no matter what, but that does not mean that I would believe them no matter what. For me to claim the latter would be a form of pride. (I don't even think there is necessarily a fact of the matter about the counterfactual, since I am not a Molinist.)
Alex
Thanks Alex, that helps a lot. One quick follow up: I'm not interested in just any counterfactuals, I'm interested in what you are rationally allowed to do once you assign those props 100% credence. No brainwashing or epistemic viciousness (or any that entails irrationality). Assuming you don't assign 100% credence to anything now that entails that de fide props are false, then it seems like your belief in them is rationally unshakable: i.e., if you go to a pagan afterlife you should believe you're being tested by God or the devil or hallucinating, no matter how long it seems like you're there.
I only use this fanciful example because somewhere above others were being accepted as relevant. I think one could construct one not so fanciful: the Roman Church could be destroyed (every priest was killed, all the writings burned, etc.). I don't know if it is IMPOSSIBLE for that to happen (given that Catholicism is true), but since I think God wouldn't allow that to happen, if it did happen I think I would have to lower my credence in de fide propositions. Less fanciful still: certain protestants start "performing" miracles.
Anyway, I'm not trying to argue against your commitment--it's a commitment I share or at least try/hope to. But I'm struggling with cashing out that commitment (at least in part) in terms of assigning 100% credence.
J
I think Catholic faith should be rationally unshakeable barring sin or brainwashing.
John,
You wrote to Alex: "Assuming you don't assign 100% credence to anything now that entails that de fide props are false, then it seems like your belief in them is rationally unshakable: i.e., if you go to a pagan afterlife you should believe you're being tested by God or the devil or hallucinating, no matter how long it seems like you're there."
I think Alex DOES believe that Pr(De fide props are true/Pagan afterlife) is VERY LOW. Or do you interpret him otherwise, John?
(Of course, it seems Alex also believes that Pr(Pagan after life) is 0. His own example is similar. According to Alex, Pr(LEM/I really see p and not-p) is very low or, more accurataly, 0. However, according to Alex, Pr(LEM) is 1, and Pr(I really see p and not-p) is 0.)
... Maybe Alex would say that still better would be to claim that Pr(LEM/I really see p and not-p) is (not 0, but) undefined.
... So, I think, Alex would say:
Pr(De fide props/Pagan afterlife and K) is undefined, because Pr(Pagan afterlife and K) is 0. Still, it is true that: if (Pagan afterlife and K), then (not-De fide props). "K" is Alex's background knowledge.
Pr(LEM/I really see p and not-p, and K) is undefined, because Pr(I really see p and not-p, and K) is 0. Still, if (I really see p and not-p, and K), then (not-LEM).
But maybe I misinterpret Alex.
I thought Alex was claiming that the faithful must/should assign credence 1 to de fide propositions and credence 0 to any proposition or possibility that is incompatible with a de fide proposition. I can't see how one could do this an be rational (in large part because it seems like it commits one to thinking one is hallucinating if one ever sees something incompatible with a de fide prop), which I take to be an objection to Alex's account of the requirements of faith. Or else I'm not sure what Alex means by 'credence'. But these are subtle matters and I should think them through more carefully. Sorry about the delayed reply...perhaps more later.
John
John:
Why is it irrational? It is not irrational to think one is hallucinating if one sees something that contradicts, say, the claim that one is human (e.g., if one looks at oneself and sees a reptilian body), or the claim that 1=1, or some other like claim.
It would be hard to say why it is irrational (if it is). The human example is a helpful one. If I "woke up" and had the following experience I think it might be irrational for me to continue to have credence 1 that I am human. I wake up and have an alien body and aliens tell me how I was "planted" on Earth for some reason, and we go off into space and back home to planet X and years pass. Of course in some sense I want to say I assign 0 credence to that happening. But really it is just a negligible credence: if I did have that experience I would not be certain that I was in the middle of an extremely long lasting hallucination.
In a nutshell that seems to me to be the problem with trying to cash out the certainty of one's faith in terms of credence. I'm as certain as I can be that I am human but I don't think that I can have credence 1 in it. It is a negligible difference in most contexts, however. But not negligible in a philosophical discussion about my credences.
I'm heading out to Milwaukee tomorrow where I'll be meeting up with Alex (who will be presenting his mentor and one of my heroes Nick Rescher the Aquinas medal at the meeting of the American Catholic Philosophical Association). I'm sure we'll talk about this there, but I wanted to add one thing before that that I've been waiting to say.
First, I think Alex is right on track in his discussion of how most "doubts" aren't really "intellectual" and stem mostly from the imagination (my terms).
However, I think it is a big mistake to think there is one probability function at work in the comparisons made above. Interpretations of probability functions is a pretty core research area for me so I'll be working on a specific proposal, but in the mean time I want to call into question the usefulness of treating Pr(...) as univocal when applied as above. In other words, I think some of the arguments on both sides commit a "four-term" fallacy.
One last thing that was not stressed as it should have been. The traditional teaching which the Catholic encyc discusses refers to the *possibility* of certainty, not it's actual possession by each believer. Aristotle and Aquinas both speak of things "more evident in themselves" vs. "more evident to us".
More later.
Trent:
Where do you see the encyclopedia as talking of just the possibility of certainty, at least in the case of faith? Here's one quote: "The term natural certitude is sometimes used in another sense, in contradistinction from the certitude of Divine faith, which is supernatural certitude, and which, according to theologians generally, is greater than any degree of certitude to be had in science, because it rests not upon human reason, which is liable to be mistaken, but upon the authority of God, who cannot err."
Note, though, the "according to theologians generally" which may well signal that this has lesser authority than something de fide.
I should add that I don't particularly care about probability assignments (I am rather dubious that there is any fact of the matter as to what number in [0,1], or range of numbers, is to be attached to each belief). What I really care about is unconditional commitment: the commitment to believe the doctrines of faith no matter what happens. (Oh, and I don't really want to distinguish epistemic commitment from moral commitment, just as I don't want to distinguish epistemic reasons from moral reasons.) So it might be that there is a way of neutralizing our disagreement...
Trent,
"Interpretations of probability functions is a pretty core research area for me so I'll be working on a specific proposal, but in the mean time I want to call into question the usefulness of treating Pr(...) as univocal when applied as above. In other words, I think some of the arguments on both sides commit a "four-term" fallacy."
I look forward your specific proposal. Let me (us) know about it. Thanks!
OK I'm in the airport now waiting to fly to Milwaukee so I've only got a bit of time.
Alex, I'm not concerned with numbers but with meaning. I'm saying there's a huge disconnect between the things said above in quasi-formal language about the difference in attitudes between logical principles and de fide propositions.
Two things: 1. I distinguish between degrees of belief, degrees of certainty, and degrees of confidence. I think these distinctions could be applied helpfully here. Degrees of confidence I take to be something like Savage's personal probabilities which describe betting behavior.
I favor a comparative conception of epistemic probability, but for present purposes I take degree of certainty to be something like implicit estimate of degree of "intellectual surprise" at falsehood. You'll just have to give me that cryptic remark for now. So these two things are consistent: I would be more "intellectually surprised" if there were true contradictions than that there was no God. However, I'd never bet on the absence of true contradictions what I'm betting on God. I'm giving up far far more (though far far less than I should) for my commitment to God (and belief I should give up absolutely everything without exception (that doesn't involve paradox)) than I would ever give up for belief in some mundane proposition about logic (even if it were that than which I could not be more surprised at its falsehood). This *might* put us closer together.
As for the possibility quotes, I'm pretty sure I know where they are, they'll be forthcoming. And I definitely don't think the tradition is done developing on this: how could it be, probability is so nascent. Much more on that ASAP, hopefully at O'Hare!
I like that: the betting behavior is closer to the commitment stuff I care about.
Alex, I'm in room 1221.
I'll leave my cell number at the front desk (if they'll let me).
I wanted to publicly apologize for questioning your orthodoxy and to thank you for our conversations.
I wrote a paper after our lunchtime conversation, summarizing one of my arguments (and adding a variant). It is here. I think I'll send it to Faith and Philosophy. I think you should write a response to it and send it to them, too.
Here's a different idea. To figure out what the Church means by certainty, one could look at theologians' reactions to Descartes' use of the term.
Thanks for the paper, Alex. I like it.
According to you, natural morality in conjunction with the Christian tradition requires that Pr(Christianity)=1. Of course, many will apply Modus Tolles and deny the Christian tradition. And of course, Trent's strategy will be different: he will deny the conditional link.
I am now thinking that there is something subtly wrong with my argument. My argument depends on some principle like that one should never commit oneself to do something doing which would be wrong. But this principle seems to have counterexamples. I am thinking through these counterexamples...
Post a Comment
<< Home