Defensor Fidei

Secularists often criticise the notion of “faith,” which they take to mean belief without evidence, or at least belief without sufficient evidence. I think this is a mistake.

I do not mean to deny that many doctrines put forward as articles of faith are in fact propounded without sufficient evidence. All that I deny is that their being so is part of the meaning of the word “faith,” either in ordinary language or in theology. Note that I am not saying that being based on evidence is part of the meaning of “faith,” but only that not being based on evidence is not part of the meaning of “faith.” Faith can either be well-grounded in evidence, or not. It is not the purpose of this post to affirm (or for that matter to deny) that any particular article of faith is true, or reasonable, or justified by evidence. I’m in formal rather than material mode here.

Consider some of the following secularist characterisations of faith:

1. Leonard Peikoff defines faith as “blind acceptance of a certain ideational content, acceptance induced by feeling in the absence of evidence or proof.” (Ominous Parallels, ch. 3.)

2. Barbara Smoker of the Council of Secular Humanism defines faith as “firm belief in the absence of evidence,” and adds: “If there were objective evidence for its doctrines, it would no longer be faith; it would be knowledge.”

3. And Sam Harris opines: “It is only when the evidence for a religious doctrine is thin or nonexistent, or there is compelling evidence against it, that its adherents invoke ‘faith.’ … Faith is nothing more than the license religious people give themselves to keep believing when reasons fail.”

Do these represent an accurate description of how the word “faith” is actually used by religious believers – or indeed by people generally (secularists included, when they’re not waving their secularist banners)? I don’t think so.

Fides It seems to me that what the word means in ordinary language is not belief that goes beyond the evidence, but rather belief that goes beyond one’s personal experience. To someone of skeptical tendencies these might of course come to the same thing, but for most of us they do not. My belief that Stonehenge exists is not based on personal experience (nor on demonstrative deduction therefrom), but surely I have, by all but the sternest skeptical standards, sufficient evidence for it.

Suppose I say that I have faith – or, as we might equally say, trust (in fact the Greek word pistis translated as “faith” might more helpfully be translated as “trust,” or “confidence,” which better captures the implication that one is prepared to rely on the object of one’s faith) – in someone. Or, more specifically, suppose I say that I have faith/trust that Bonzo has turned off the stove. I wouldn’t speak of faith if I had actually seen Bonzo turn off the stove. But that doesn’t mean that my reliance on Bonzo’s having done so is somehow blind or irrational. If Bonzo tells me that he did turn off the stove, and if in my experience he has always shown himself to be reliable and truthful, then it seems I have sufficient evidence – based on experience – that he has turned off the stove, even if I didn’t actually experience his turning off the stove. (On the other hand, Tibor Machan writes: “One has faith in someone one no longer can trust – as a wife may have faith in a repeatedly philandering husband, despite all the evidence. It takes faith to believe that this man will never repeat his betrayals.” Well, yes, that’s one way the term is used in ordinary language; but it’s surely not the most common way.)

In this sense, then – and contrary to what is often asserted – faith plays a central role in the empirical sciences. I am not talking about “faith in the senses” or “faith in reason” or any such rot; I am talking about the widespread practice of relying on the results of other scientists without testing them oneself. After all, scientific inquiry is a cooperative enterprise; a scientist cannot personally test for herself all the theories and principles on which one relies. (Otherwise she wouldn’t even be able to use a thermometer!) Hence reliance on testimony is a pervasive feature of the scientific enterprise.

“Ah, no,” you say, “scientists don’t accept other scientists’ findings until they’ve been tested and replicated!” Yes, quite true, but they don’t necessarily test and replicate the results themselves; instead they rely on the testimony of group B that it has experimentally confirmed the testimony of group A. There’s nothing blind about this sort of faith; but faith it remains nonetheless.

This is also, I claim, what theologians generally mean by “faith” – not belief on the basis of insufficient evidence, but rather belief that depends on trusting someone’s word or testimony for some claim that we have not ourselves experienced or demonstratively proven. This is entirely consistent with thinking that such testimony counts as good evidence – or that we have good evidence for regarding the source of the testimony as reliable. St. Paul (or whoever wrote the Epistle to the Hebrews), for example, defines faith (at Hebrews 11:1) as the evidence of things unseen (though admittedly there’s room for debate as to precisely how to translate elegkhos in this context).

But aren’t scientific conclusions supposed to be held tentatively, in contrast to the firmness with which articles of religious faith are to be held – even in the face of countervailing evidence if need be? Yes, there’s some genuine difference there (though the tentativeness with which scientists hold their conclusions is somewhat exaggerated by people – scientists include – who talk about science). Think, by analogy, of your faith in a friend who’s been accused of murder. You’d be a funny sort of friend if you took too “scientific” an attitude toward your friend’s guilt – if, for example, you said “I regard the probability of your innocence as 87%,” or “The conjecture of your innocence has not yet been falsified.” Perhaps religious faith is more like loyalty to a friend than it is like a scientific conclusion.

But this doesn’t mean that your faith in your friend is unconnected to all evidence. To be sure, if you are a loyal friend you will continue to believe in your friend’s innocence even in the face of a fair bit of evidence to the contrary; but that may be because you regard your trust as well-grounded in your previous experience – and you may well be quite justified in doing so. Your loyalty is thus based on evidence, and could in principle be overturned by enough contrary evidence – so despite the emotional attitude with which you hold it, it’s not really blind faith. Of course loyalty to a friend can be blind or irrational, it can persist in the face of sufficient evidence to the contrary; but it’s not part of the concept that it be so. Ditto for religious faith.

It is true, of course, that some theologians, called fideists, do hold that we should believe regardless of evidence. But this has always been a minority view within Christianity (e.g., Tertullian’s famous doctrine Credo quia absurdum, “I believe because it is absurd,” is traditionally regarded as heretical), and as far as I can tell within most other major religious traditions as well. (For example, in Hindu epistemology the category of Śabda-pramāna, or valid-cognition-on-the-basis-of-testimony, includes both scriptural revelation and the ordinary testimony of reliable witnesses.)

Doubting ThomasThe story of Doubting Thomas (John 20:24-30) is sometimes taken as evidence of the fideistic nature of Christianity. When told by his fellow disciples that Jesus has returned from the grave, Thomas replies: “Except I shall see in his hands the print of the nails, and put my finger into the print of the nails, and thrust my hand into his side, I will not believe.” When Jesus finally shows up and Thomas is convinced, Jesus rebukes him, saying “Because thou hast seen me, thou hast believed: blessed are they that have not seen, and yet have believed.” On a popular interpretation, Thomas is someone who refuses to believe without evidence, and the meaning of Jesus’s rebuke is that Thomas ought simply to have believed, evidence or no.

But is this a plausible interpretation of the passage? Remember, the incident presupposes that Thomas has been hanging around Jesus for several years; he’s seen him heal the sick, multiply the loaves and fishes, walk on water, command the tempest, and even raise the dead. (You may think both the resurrection and the earlier miracles are historical events, or you may think both the resurrection and the earlier miracles are fictional; but it would be odd to claim that the resurrection is a historical event but the earlier miracles are fictional. Hence the context of Jesus’s post-resurrection conversation with Thomas presupposes the earlier miracles as background.) So when the other disciples tell Thomas that they’ve seen Jesus resurrected, he’s receiving this testimony in a context that makes it reasonable to believe it. Jesus’s rebuke seems perfectly justified: Thomas has been withholding assent in defiance of the evidence.

An additional reason for rejecting the fideistic interpretation of the Doubting Thomas story is that Jesus has already said somewhat earlier in the same gospel: “The works that I do in my Father’s name, they bear witness of me. … If I do not the works of my Father, believe me not. But if I do, though ye believe not me, believe the works.” (John 10:25-38) This seems like a clear rejection of fideism; Jesus is telling his followers not to have faith in him unless he provides a certain kind of evidence, and that it is appropriate for their faith to rest upon this evidence. This is reminiscent of the Buddha’s decidedly non-fideistic advice to his own disciples: “As the wise test gold by burning, cutting and rubbing it, so should you accept my words after testing them, and not merely out of respect for me.”

The Wikipedia page on fideism is, I suspect, somewhat confused about what does and does not count as fideism. It notes, correctly, that the Catholic Church officially rejects fideism. But it illustrates this rejection by citing the Catholic insistence that God’s existence can be rationally demonstrated. This invites the reader to suppose that if the Catholic Church had instead held that God’s existence could not be rationally demonstrated then they would have a fideist position on the matter; but that would be a mistake. There are indeed some Catholic articles of faith which are held not to be susceptible of rational demonstration (the Trinity, for example), but since the Church does not decline to offer reasons for accepting Catholic teaching as a reliable source of testimony as to the Trinity, this is not a fideist position. (“S says p, and here are some strong reasons for regarding S as a reliable source of information as to whether p” is a perfectly appropriate, albeit non-demonstrative, form of argument for p.) The distinction between demonstrable and indemonstrable theological doctrines is not a distinction between doctrines based on evidence and doctrines to be accepted regardless of evidence.

John Locke in the Essay clearly distinguishes between belief without evidence and religious faith. He defines faith as “assent to any proposition, not thus made out by the deductions of reason, but upon the credit of the proposer,” and in the present case “as coming from God.” But so far from suggesting that this assent should disregard the evidence, Locke insists that “faith is nothing but a firm assent of the mind; which if it be regulated, as is our duty, cannot be afforded to any thing but upon good reason,” since anyone who “believes, without having any reason for believing, may be in love with his own fancies; but neither seeks truth, as he ought, nor pays the obedience due to his Maker, who would have him use those discerning faculties he has given him.” Hence “whatever God hath revealed, is certainly true” and “the proper object of faith,” but “whether it be a divine revelation or not, reason must judge.”

Lest it be thought that Locke is idiosyncratically modern and rationalist here, consider the paradigmatically orthodox Thomas Aquinas, who writes in defense of faith:

[I]f one were willing to believe only those things which one knows with certitude, one could not live in this world. How could one live unless one believed others? How could one know that this man is one’s own father? Therefore, it is necessary that one believe others in matters which one cannot know perfectly for oneself.

When Aquinas describes knowing who one’s father is as a matter of faith, he is of course not denying that one can have good reasons to think that such-and-such a man is one’s father; all he is saying is that (in the days prior to DNA testing) such reasons involve crucial reliance on testimony. (The argument is incidentally borrowed from Augustine.)

Thomas Aquinas Part of the confusion may stem from the fact that Aquinas here denies that what is held on faith is known – whereas we might want to say that we know perfectly well (albeit via testimony) who our father is. But what Aquinas and most other mediævals meant by “knowledge” (or by the word scientia, which we translate as “knowledge”) was something like deductive demonstration from self-evident premises. Mere justified true belief (plus whatever is needed to meet the Gettier problem) will count as knowledge by our lights, but not as scientia by Aquinas’s lights. (Indeed knowledge is not even a species of belief on Aquinas’s view.) So when Aquinas denies faith the status of scientia, he is not denying it the status of what we would call knowledge. The notion that faith and knowledge are mutually exclusive applies to the mediævals but not to us.

Here, for example, is Aquinas’s argument for its being reasonable to accept scriptural testimony on faith:

[I]f a king sends letters signed with his seal, no one would dare to say that those letters did not represent the will of the king. In like manner, everything that the Saints believed and handed down to us concerning the faith of Christ is signed with the seal of God. This seal consists of those works which no mere creature could accomplish; they are the miracles by which Christ confirmed the sayings of the apostles and of the Saints.

If, however, you would say that no one has witnessed these miracles, I would reply in this manner. It is a fact that the entire world worshipped idols and that the faith of Christ was persecuted, as the histories of the pagans also testify. But now all are turned to Christ – wise men and noble and rich – converted by the words of the poor and simple preachers of Christ. Now, this fact was either miracle or it was not. If it is miraculous, you have what you asked for, a visible fact; if it is not, then there could not be a greater miracle than that the whole world should have been converted without miracles.

In short, Aquinas is offering Jesus’s miracles as evidence of the reliability of his testimony. And to those who doubt whether the miracles occurred, Aquinas is offering the success of Christianity as evidence for the reported miracles – or more precisely for the disjunctive claim that either the reported miracles occurred or else a different miracle, likewise supportive of Christianity, occurred, namely Christianity’s success despite the absence of the reported miracles!

Now I’m not claiming that this is a good argument. (In particular, the appeal to the impressive spread of Christianity would seem to apply just as well to other impressively-spreading religions such as Buddhism and Islam; and Aquinas certainly knew about the latter.) But it is an argument, and so shows that for Aquinas an appeal to faith does not preclude an appeal to evidence of the trustworthiness of the source in which one is being asked to have faith.

John Locke Or consider Locke’s argument in The Reasonableness of Christianity to the effect that Jesus’ moral teachings, being “such, as though reason of itself had not clearly made out, yet it could not but assent to, when thus discovered,” constitute evidence of the trustworthiness of their author, so that Jesus’s other pronouncements are likewise to be believed. Now this is not such a great argument either. (For one thing, it assumes that Jesus’s moral teachings and his theological teachings indeed originated from the same source, rather than, say, being combined from different sources by later copyists; I don’t say that’s necessarily an indefensible assumption, but Locke doesn’t defend it.) But my present concern is not with the merits of the evidence offered on behalf of faith but simply with the fact that evidence is offered.

It may be objected that even if I am technically correct in my claim that the concept of faith does not preclude evidence, the fact that the claims put forward as articles of faith so often do rest on insufficient evidence, combined with the fact that in practice those whose faith falters are so often met with censure rather than argument, implies that there’s some nasty entanglement between the concept of faith and the fideistic attitude after all.

But is that a fair conclusion? After all, the same point could be made about the concept of reason. Many of the claims put forward as deliverances of reason have likewise turned out to be inadequately grounded, and those who depart from them have likewise often been met with censure rather than argument. It is true enough in practice that, in Sam Harris’s words, faith often gets invoked when reasons fail; but reason often gets invoked when reasons fail too. In short, it is perfectly possible to treat reason in a fideistic manner. (Just look at the history of the Randian movement.) But that doesn’t mean that all appeals to reason are thereby fideistic. Ditto for faith.

, ,

15 Responses to Defensor Fidei

  1. William H. Stoddard January 15, 2008 at 2:33 am #

    I don’t feel able to address the substance of this post, at least not on a first reading; but I’m moved to make a small copy editing point:

    You refer to a Greek word that you transliterate as elegkhos. It’s true that that word has a gamma preceding the chi. But it was a convention of ancient Greek spelling that the velar nasal (the sound at the end of “king”), for which Greek did not have a distinct letter, was represented by a gamma preceding another gamma, a kappa, a xi, or a chi. Because English represents the velar nasal by an n preceding the equivalent letters (as in “king,” “think,” and “sphinx”), it’s customary to transliterate gamma in that specific context as n, rather than g. For example, the Greek word spelled letter by letter as aggelos, meaning “messenger,” is customarily transliterated as angelos.

    I recommend using this transliteration, not simply out of traditionalism, but because in fact it better represents the sound to an English speaker, and leads to a roughly correct pronunciation of the Greek word, whereas the use of “g” leads to a hopelessly incorrect pronunciation, and also obscures the etymological connection with English words derived from Greek (such as “angel,” derived from angelos).

  2. JOR January 15, 2008 at 8:30 am #

    This is great stuff. I’m waiting for some lunk to chime in and exlaim, “…but that makes the word ‘faith’ meaningless!!!!1”

  3. Administrator January 15, 2008 at 5:00 pm #

    Re elegkhos: I know it’s traditional to render it elenkhos or elenchos (or indeed just to Latinise it as elenchus), but it seems to me that ease of correspondence to the original is a desideratum of any system of transliteration. Ideally when one sees a Greek word it shouldn’t be needlessly ambiguous how to render it back into the Greek alphabet; now if someone sees elenkhos they won’t be sure whether it corresponds to ??????? or to ??????? whereas if they see elegkhos it can only be ???????.

    Now with some languages one can’t really achieve this kind of correspondence without producing some highly weird-looking results. (I’m looking at you, Russian!) But one can do it OK with Greek.

    Of course aiding pronunciation is also a desideratum of transliteration, but I think it’s one that should come in at the level of deciding the overall system of transliteration rather than being a reason for deviating from the overall system for particular cases.

  4. Geoffrey Allan Plauche January 15, 2008 at 6:06 pm #

    I’m curious how you think negative coherentism solves the Gettier problem. A quick skim didn’t reveal a section on it in the Wikipedia article.

  5. DELMAR ENGLAND January 15, 2008 at 6:40 pm #

    “Secularists often criticise the notion of “faith,” which they take to mean belief without evidence, or at least belief without sufficient evidence. I think this is a mistake.”

    Of course it is. Unless one is omniscient, one is obliged to admit there are a lot of facts out there the evidence of which is presently unknown. This makes the real issue believing contrary to the evidence known and demonstrable, yet denied. For instance,
    literally, all evidence shows that value is subjectively created, that is, subjectively attributed. Yet, Mr. Piekoff apparently believes that value is objective and discovered in spite of universal evidence to the contrary.

    What is the philosophical, epistemological and psychological difference between “God’s will” and “life proper to man?” Ans. None. No matter what undefined and self-deluding labels are attached, every philosophy with the identical underlying elements named above are fundamentally the same. Neither lip service to a worthy admonition, nor individual spin on a common fallacy creates truth.

    Word games of non definition is what sustains beliefs in all concepts of the illusion, objective value. (How do you think Rand got “ought from is?”) Since the concept, objective value, is the most popular and protean fallacy that has ever come down the pike, to claim that “Objectivism” and the idea of objective value is new is the epitome of self-delusion and absurdity.

    As for “faith vs reason” or “reconciliation” of the two, argument is pointless until and unless an objective differentiation is made to separate the implied two modes of thought. After that, what remains is to identify the mode by which all known truth is acquired. The other is consistently false. It won’t make you omniscient, but it sure will cut down on the thinking errors.

    Delmar England

  6. William H. Stoddard January 15, 2008 at 10:23 pm #

    Re elegkhos: I know it’s traditional to render it elenkhos or elenchos (or indeed just to Latinise it as elenchus), but it seems to me that ease of correspondence to the original is a desideratum of any system of transliteration.

    I agree that it’s a desideratum. But I believe there is never any doubt about how to reverse transliterate Greek words back into the Greek alphabet if you use gamma => n in the context we’re discussing. This is because there are no Greek words in which the letter nu precedes the letter gamma, kappa, xi, or chi. The nasal sound before a velar stop or fricative is always assimilated to it in Greek, becoming a velar nasal rather than a dental nasal; and the velar nasal in Greek is always written as gamma. So if you see an n before a g, k, x, or kh, there is only one Greek letter that it could represent.

  7. Administrator January 15, 2008 at 10:48 pm #

    Of course it is. Unless one is omniscient, one is obliged to admit there are a lot of facts out there the evidence of which is presently unknown. This makes the real issue believing contrary to the evidence known and demonstrable, yet denied.

    I don’t understand this inference. From the fact that there are plenty of facts for which evidence is unknown, it doesn’t follow that there isn’t a difference between believing something on sufficient evidence and believing something on insufficient evidence.

    For instance, literally, all evidence shows that value is subjectively created, that is, subjectively attributed. Yet, Mr. Piekoff apparently believes that value is objective and discovered in spite of universal evidence to the contrary.

    Well, remember what Peikoff and other Randians mean by “objective.” “Objective” value in the Randian sense is value that consists in a relationship between consciousness and external reality; it contrasts on the one hand with “subjective” value (which exists in consciousness apart from external reality) and on the other hand with “intrinsic” value (which exists in external reality apart from consciousness). Given that that’s what Peikoff means by objective value, what is this “universal evidence” to the contrary? (For that matter — hey, I’ll bite — what do you take to be the “universal evidence” against even intrinsic value, again in Peikoff’s sense?)

  8. Administrator January 16, 2008 at 12:49 pm #

    But I believe there is never any doubt about how to reverse transliterate Greek words back into the Greek alphabet if you use gamma => n in the context we’re discussing

    Agreed — never any doubt to someone well-versed in Greek. But someone well-versed in Greek also won’t be confused about how to pronounce elegkhos, so spelled. It’s the beginner in Greek who may be misled pronunciation-wise by elegkhos (your worry) and misled reverse-transliteration-wise by elenkhos (my worry).

    But if one must choose between these two possible misleadings, I think it’s more important for the beginner in Greek to be able to find the word when they look it up in a Greek dictionary (a desideratum facilitated by elegkhos and frustrated by elenkhos) than to pronounce it correctly.

    Plus I confess I just find it more elegant for the transliteration to mirror, where feasible, the structure of the Greek.

  9. William H. Stoddard January 17, 2008 at 12:40 am #

    How much of a beginner in Greek are we talking about? If you mean someone who has actually studied a little bit of Greek, enough to know the alphabet and use a lexicon (I have the shorter Liddell & Scott on my shelves), well, my copy of Chase and Phillips explains the pronunciation of gamma as ng in chapter 1, note 4. That was enough for me to figure out the Greek spelling of words such as “angel.” If they don’t even have that much knowledge, I doubt they’ll be looking anything up in a Greek dictionary in the first place.

    Still, this is partly a matter of conflicting senses of elegance—because what really bothers me about elegkhos is that it seems inelegant.

  10. Administrator January 17, 2008 at 12:08 pm #

    Well, sure, elegkhos is inelegant, but only because ελεγχος itself is inelegant. Why should the transliteration paper over the oddness of the Greek?

  11. Administrator January 17, 2008 at 12:10 pm #

    I’m curious how you think negative coherentism solves the Gettier problem

    Negative coherentism (as I see it) is a theory of justification, not a theory of knowledge; more precisely, it’s a theory of the justification-component of knowledge, not the Gettier-component of knowledge. So I don’t think it needs to address the Gettier problem.

  12. Gene Callahan January 18, 2008 at 4:02 am #

    Have you read Collingwood’s An Essay on Metaphysics? It has quite a bit to say on “faith.”

  13. Administrator January 18, 2008 at 1:00 pm #

    I’ve read it, though at the moment I can’t remember what he says about faith.

  14. Gene Callahan January 18, 2008 at 7:07 pm #

    The basic idea is that every metaphysical system pre-supposes certain postulates, and that those are its articles of “faith” — “scientific” and “religious” worldviews alike.

Trackbacks/Pingbacks

  1. All the Court Also Laughed Loudly Thereat | Austro-Athenian Empire - July 17, 2019

    […] the mediævals generally meant not “blind” faith, i.e., faith without evidence, but rather belief on the basis of testimonial evidence rather than either perceptual evidence or rational demon…), it is still nobler to seek for a rational proof and understanding of the doctrines of faith, […]

Leave a Reply to Administrator Click here to cancel reply.

Powered by WordPress. Designed by WooThemes