Faith and theory

by PhilGoetz5 min read26th Mar 201140 comments


Personal Blog


Faith is often described as belief without evidence.  The famous definition in Hebrews, in its best-known form, is close to that:

"Now faith is the substance of things hoped for, the evidence of things not seen." - Hebrews 11:1 (King James Version), often attributed to St. Paul

This is the way the term "faith" is used by religious people when they argue against the primacy of reason, as demonstrated in these quotes, which is the context I am concerned with.  (It's also the meaning used by atheists arguing against religious faith, eg. Sam Harris in The End of Faith.)  But the New International Version, which is less pretty, but translated more carefully by better scholars from more and older texts, says:

"Now faith is confidence in what we hope for and assurance about what we do not see."

The Wikipedia and Plato entries on faith give information on the use of words translated as "faith" in English in different religions and philosophies.  Wikipedia cites the New American standard exhaustive concordance of the Bible as saying,

In English translations of the New Testament, the word faith generally corresponds to the Greek noun πίστις (pistis) or the Greek verb πιστεύω (pisteuo), meaning "to trust, to have confidence, faithfulness, to be reliable, to assure".[22]


Scientific theory is also assurance about things unseen.  (If you can observe something directly, you don't need science.)  Science builds an abstract mental structure that interprets data and makes predictions from it.  It is also an epistemology for belief in things we can't see, like atoms, oxygen, radio waves, vast distances, or circulation of the blood.

The history of faith and theory

The most popular belief appears to be that faith is ancient, and scientific theory came along later to supersede it.  But I'm not aware of evidence for this.

I've read many accounts by Christian missionaries of how they persuaded people living in tribal cultures to convert to Christianity.  Faith is not a factor in these debates.  Tribal cultures don't need to protect their religions with rationalizations.  In these narratives, the tribe members often seem unaware that not believing their religion is a physical possibility; they are sometimes less surprised that the white-skinned stranger descended from the skies, than that he pretends to be unfamiliar with their gods.  Tribal people are more reasonable about the matter than people from advanced civilizations; they never defend their religion by making arguments against reason itself.  They are pragmatic and opportunistic:  Can the missionary cure the sick better than the tribal shaman?  Yes?  Then his god is stronger!

There are numerous verses in the old Testament about faith, and many more passages describing the same sentiment.  But "faith" in these passages seems synonymous with "obedience", or possibly "trust", when it isn't something else entirely that was lost in translation or history (e.g., "And there shall be faith in thy times: riches of salvation, wisdom and knowledge" ... what?)

Nothing described in the Wikipedia or Plato pages on faith definitely antedates Buddhism1.  The concepts of faith, and of scientific theory, both may have developed primarily between the 6th century BC, when mathematics, philosophy, Buddhism, large parts of Hindu philosophy, and Phariseeic (intellectual) Judaism were being invented; and the time when Jesus spoke and Paul wrote about faith as a kind of power and a means of justification (a technical Christian term).

By the time Jesus spoke of faith, it had one main feature distinguishing it from theory:  The truth of the belief was proportional to the strength of the belief.  Luke 17:6 (NIV): "If you have faith as small as a mustard seed, you can say to this mulberry tree, 'Be uprooted and planted in the sea,' and it will obey you."2  Paul's chapter on faith (Hebrews 11) describes faith as having causal power.

How many people living during that time could have distinguished between faith and theory?

Faith as reaction to theory

One theory of mine is that faith developed as a reaction to theory.  How to keep a religion alive, when these pesky philosophers are corrupting the youth with reason?  Declare that reason is not the final arbiter!  Faith is thus a sort of early post-modernism.

You could alternatively see it as the post-philosophical epistemology for people who value winning over being right; it lets them retain adaptive beliefs that their society had evolved, even after philosophers found disproofs of those beliefs.

Faith as popularized theory

"Christianity is Platonism for 'the people'".  - Nietzsche, preface to Beyond Good and Evil

Another theory is that faith is a faulty theory about theory.  Ancient people could have observed proto-scientists, and their ability to make seemingly-impossible predictions - such as the time of eclipses.  The first prediction of a solar eclipse was perhaps in 585 BC.  What did non-astronomers believe the astronomers were doing?  Did they think that the power of the astronomer's belief caused the eclipse?  How many Pythagoreans understood why the sum of the squares of the sides of a right triangle equalled the square of its hypotenuse, or why the square root of 2 could not be rational?  Did they take these claims on faith, like most high-school geometry students do today?  If you imagine that understanding has magical power, it would be easy to read what Paul wrote about faith, and substitute the word "understanding".  Is that what some Gnostic Christians were doing?

Faith as proto-theory

Another theory is that theory is faith++.  The mental commitment of faith is very similar to the mental commitment of theory, if not the same.  People practiced in having faith in things unseen, would be more likely to be able to have theories about things unseen.  They would just have better heuristics for what to have faith in.  This suggests that faith is a necessary historical precondition for theory.

Faith ≈ theory

Perhaps my favorite theory - and these theories are not all mutually exclusive - is that many pre-modern thinkers did not distinguish between faith and theory.  It's not clear that Gnostic Christians, Hermetic philosophers, or alchemists were able to make that distinction.  Many Gnostic Christians may have thought they were practicing a kind of natural philosophy.  Many Renaissance alchemists may have regarded their studies as a spiritual practice.  Therefore, in modern debates between faith and science, when people cite pre-modern texts and try to claim their authors for their side based on their usage of terms like "faith" or "reason", it may be those authors weren't even aware there were different sides to be on.


1. The Hindu trans-rational epistemology referred to by Wikipedia may be an older variant of faith.  Its age, and its relation to what I'm calling faith, are unclear to me.

2. Faith performs work, and is therefore a type of energy.  A mustard seed weighs about .002g, which is the mass equivalent of the energy of 43 tonnes of TNT.  So Jesus was being conservative.


40 comments, sorted by Highlighting new comments since Today at 12:41 AM
New Comment

I think you lost me when you assumed "faith" ought to be a meaningful word with a coherent definition.

I think the best definition to give for faith is a practical one: faith is the word people use as a combination semantic stop-sign and applause light when asked why they believe in religion. If someone then goes all philosophical on them and asks them what exactly they mean, they then use whatever plausible explanation seems appropriate.

I've heard faith described as:

  • exactly the same as inductive reasoning; thus, you have "faith" that the sun will rise tomorrow.
  • similar to the concept of credibility; the Bible's always been right before, so I imagine it will be right on this one issue here, even though I'm not exactly sure how.
  • similar to the concept of trust: God is a good being, I'm sure He knows what He's doing.
  • a direct meddling of the Holy Spirit in the internal workings of your brain, so that you are able to mystically come to the right answer about whether or not to believe in religion despite insufficient evidence.
  • a decision to follow whatever idea seems intuitively true or most pleasant, and to give evidence less weight than these intuitions
  • a decision to believe something even though you have no reason for doing so; also the thing those people who ask the obvious followup question "but why did you choose to believe that for no reason, as opposed to something else for no reason" should feel bad for not having.
  • Not comprehensible to the limited human mind so don't talk about it

I don't think any of these definitions are "the correct definition"; I just think they're different ways that people in different situations and with different degrees of philosophicalness cash out the idea of "I believe in religion and you can't tell me not to and I feel pretty good about it"

As such, I don't believe there's a concept called "faith" which it is necessary to distinguish from theory in the first place.

I think the best definition to give for faith is a practical one: faith is the word people use as a combination semantic stop-sign and applause light when asked why they believe in religion. If someone then goes all philosophical on them and asks them what exactly they mean, they then use whatever plausible explanation seems appropriate.

Calling something a "semantic stop-sign and applause light" can be an important first step towards a definition. But it's not a definition itself, not even a merely "practical one". A practical definition needs to explain how the concept of faith works as a semantic stop-sign. Just which peculiarities of human cognition does faith exploit? And how, in practice, can we keep it from exploiting them?

Why a definition of a word must include explanation of biases which correlate with its use? If I haven't heard the word "faith" before, after reading Yvain's definition I would be able to use it and understand what other people mean by it, which is exactly what I expect to learn from definitions of words.

Yvain's "definition" is as follows: It starts by saying that "faith" is a "semantic stop-sign and applause light". That alone would be inadequate, of course, because it doesn't distinguished faith from other semantic stop-signs and applause lights, of which there are many. So Yvain goes on to distinguish faith from the rest as follows: Faith is that which, when people who claim it are asked to explain what they mean, they proceed to give one of the items on Yvain's list of different ways that different people

cash out the idea of "I believe in religion and you can't tell me not to and I feel pretty good about it".

Well, I suppose that that's a definition in some sense. But, as Yvain would agree, it's not a definition in the sense sought in the OP. The OP seeks a description of the contents of the concept that sits in the minds of faith-holders. Yvain, as I read him, is denying the existence of any such concept. For Yvain, "faith" is just a sort of verbal defensive behavior. In fact, if you knew only Yvain's definition, you would think that the word "faith" was only used to defend one's beliefs from attack by others.

This appears to me to be an inaccurate picture of what is going on in the minds of faith-holders. So, insofar as Yvain's definition encourages this picture, the definition is inadequate.

In fact, if you knew only Yvain's definition, you would think that the word "faith" was only used to defend one's beliefs from attack by others.

It seemed quite accurate to me, but that can be because I am not much familiar with religious thinking. Can you provide an example of relatively recent use of "faith" outside an apologetic argument?

Here's a page from a book that I got to by following a citation to the Wikipedia article on Faith. I'm looking at the first complete paragraph.

The text is written by a Christian for Christian readers. As near as I can make out, the page is not arguing that the reader is required to have faith, or that the writer's faith can survive all criticism. So, I wouldn't call it apologetics. The tone is more like, "So, we all have faith here. That's not at issue. But, just what kind of animal is this faith thing that we have? Where does it come from? What role does it play in the fate of our souls?"

You are right. They don't use the word to defend their beliefs, even if I can't figure out for sure what "faith" is supposed to mean there. It seems still to play the role of a sort of stop-sign, with approximate meaning of "acceptance of Catholic dogma".

Still, I am not sure whether there is a concept behind "faith" distinct from "belief immune from scrutiny". I have found a Christian definition where they basically say that faith is a belief which is

  • not based on factual evidence, but rather hearsay
  • absolutely certain
  • motivated by God's personal qualities
  • a supernatural act

That seems to vindicate the naïve atheist view of faith as a belief firmly held in spite of evidence.

...why did it not occur to me that the word "faith" might be a semantic stopsign?


I think you lost me when you assumed "faith" ought to be a meaningful word with a coherent definition.

I didn't read the article as doing that, nor any of the four hypotheses at the end. But I do read Phil's response to you as doing so. Huh.

I think the best definition to give for faith is a practical one: faith is the word people use as a combination semantic stop-sign and applause light when asked why they believe in religion. If someone then goes all philosophical on them and asks them what exactly they mean, they then use whatever plausible explanation seems appropriate.

That is pretty close to be Phil's first hypothesis "Faith as reaction to theory"

But Phil is saying that this is not an individual reaction, but a historical reaction. Protestants go on about faith all the time on their own, without any philosophers present.

If you don't distinguish having faith from having a theory, how do you talk to religious people?

I understand the point you're making, though I think you're going too far when you say there isn't a concept called "faith". I should have explained that I'm responding to the use of "faith" in only 2 contexts:

  • faith as a technical Christian term: What did Jesus and Paul mean by faith? There is a large body of literature on this, and each denomination of Christianity has a pretty good idea what they mean by it.

  • faith as it is used as an argument against rationality.

Only the fourth and sixth definitions you listed above occur in those contexts.

If you don't distinguish having faith from having a theory

That doesn't follow from what Yvain said at all. There being no singular thing designated with the words "having faith" doesn't mean those words always designate the same thing as the words "having a theory". In fact, if they did there would be a singular thing they designate, which would directly contradict Yvain's point!

I've been toying with a way of comparing theory and faith, but I'm not yet sure how well it works. Summary: Theory is to faith as our concept of physical necessitation is to that of social obligation.

We think of theory and faith as both imparting an obligation to believe. But the sense of "obligation" is different in each case. Theory borrows its sense of "obligation" from how we think about the natural world. Faith borrows its sense of "obligation" from how we think about the social world.

The sense of "obligation" in theory is that of natural necessity. We have an intuition that, in the natural world, forces act on things to necessitate changes in their states. (I'm talking about a primitive folk notion of "necessitate", here, not the determinism of Newtonian mechanics.) Analogously, we are obliged to believe a theory because the force of reason, like a natural force, pushes us to do so.

The sense of "obligation" in faith is that of duty, trust, and deference to those who deserve it. If someone deserves our trust, then it feels wrong, or insolent, or at least rude, to demand independent evidence for their claims. Certain kinds of social relations impose a duty on one party to act as though the word of the other party were beyond doubt. Faith is the duty that we owe to God to trust him, which duty we have in virtue of the fact that we are indebted to him for our existence. And this faith is rewarded, just as one would expect to be rewarded for showing due deference to one's social superiors.

This is nice. I wonder, though, why people wouldn't have then used some existing, familiar term referring to feudal obligation, instead of "faith". People were very conscious of the parallels between feudal relations, and relations between man and God. Yet I haven't heard faith described in those terms.

In support of this idea, remember that, while in the 1st thru 4th centuries AD, as well as today, there were/are a lot of questions about who wrote various books and whether they were inspired by God, these questions weren't often asked publicly in the middle ages. (There was inquiry into this during the Protestant Reformation.) If someone said in 1500 that you must have more faith in the Bible than in your observations, they might not have meant that there was this thing "faith" justifying belief in the Bible. Of course the Bible was inspired by God. Doubting the words of the Bible was perhaps not interpreted as doubting that the Bible was inspired by God, but as doubting that God was telling the truth. So faith wouldn't have needed to have anything to do with epistemology.

This is nice. I wonder, though, why people wouldn't have then used some existing, familiar term referring to feudal obligation, instead of "faith".

Because "faith" was the existing familiar term for those relations. The feudal concept of fealty comes from the same Latin root, fidelis, that's translated into English as "faith". The word "faith" is still occasionally used in its original sense, e.g., "faithful servant" or "I have faith in you".

This is a much more useful way to think about faith than just thinking of it as somehow the opposite of reason.

[-][anonymous]10y 0

Faith is belief that unconditionally prevents understanding views inconsistent with it. A simple, illuminating definition, rendering faith a defined variety of viral meme. For the mechanism, based on psychologist Daniel T. Gilbert's "Spinozan" theory of comprehension and belief, see "Unraveling the mystery of morality: The unity of comprehension and belief explains moralism and faith"

Why is this important?

You're at a party, and for some reason are arguing about the nature of religion.

After a few minutes of stereotypical debate, the religionist claims that

You just have to take it on faith.

You respond with

Well, faith is a fairly modern invention. Originally, people thought that their beliefs actually had to correspond with things that they observed in reality. People would convert because their God would lose to someone else's God in some observable field, or because their rituals were less effective. The way you're using 'faith' is mostly based on some mistranslated Greek words, and only seems to have appeared in response to religion being shown so wrong across the board that the only possible defense is to deny thinking itself.

Now, the religionist is probably going to be pissed off at you, and isn't going to change their beliefs. But every agnostic around will entirely lose respect for the "Faith" argument, and maybe turn outright atheist.

It also would defuse statements like "Humans are wired to have faith."

Nick Szabo has some good essays about interpreting tradition and why it's useful:

Hermeneutics: An Introduction to the Interpretation of Tradition

Objective Versus Intersubjective Truth

This is somewhat tangential to the post, for which I apologize, but it's in response to a number of different comments, so I figured I'd just respond to the post itself.

For my own part, I see no reason to treat the theological use of "faith" as something with sharp linguistic borders that separate it from other uses of the same word.

In the privacy of my own mind, I use "faith" to denote an emotional sense of confidence in a proposition, connotatively one that overcomes countervailing emotions.

For example, when I was recovering from my stroke, I experienced a lot of pain, anxiety, anger, fear, and depression. All of those things made it very difficult to actually engage in my therapy, take my medication, and do the other things that I needed to do in order to complete my recovery. And that motivational framework reinforced itself by adding weight to the prediction that I wasn't going to recover my full capabilities anyway. (After all, only ~10% of major stroke patients do, according to some sources.)

Something that helped was to explicitly invoke an emotional state that presumed I was going to recover, which made doing work towards my recovery feel much more like a worthwhile thing to do.

I have since then talked to a number of theists who went through similar experiences, and it seems pretty clear to me that the emotional state that I invoked is pretty much the same one they refer to as "faith."

In their case, it was explicitly "faith in God," though for many of them "God" didn't seem to mean much of anything in particular... it was more a parent node for everything they valued or endorsed. In my case, it was faith in my eventual recovery, without reference to God or any similar umbrella node.

But the common thread is an emotional state that exists with respect to a concept X, which increases the weight of X and, consequently, of behavioral impulses consistent with an expectation of X.

So I mostly consider epistemological discussions of faith -- both in the secular and the theological sense -- to be misguided. Faith isn't a belief, it isn't anything like a belief. It's a function that alters the motivational structure around my existing beliefs, just like fear and hope and love and hate and anger and other emotions are.

Of course, we have this cognitive habit of reifying the objects of our emotional functions. If I'm scared of X, there's a strong tendency to think and behave and talk as though X really exists, even in the absence of evidence, or in the face of counterevidence. Ditto if I love X, or hate X, or am angry at X, or have faith in X, or whatever.

And I accept that the epistemological model of faith in particular (though this is also true to a lesser extent for other emotions) has become pervasive at least in modern English usage, to the point where I pretty much cannot use the word without invoking an epistemological frame. That is, if I use the word "faith" in most communities I'm inviting a conversation about evidence and certainty, rather than a conversation about motivation and commitment, and I know that and act accordingly.

But I nevertheless consider it misguided.

Would you say, then, that "faith" as you've defined it (related to motivation and commitment) would also be related to the feeling of certainty ?

Related, sure... though, again, I would say that people frequently mistakenly interpret "that feeling of certainty" as a fact about the world, rather than about themselves.

I use "faith" to denote an emotional sense of confidence in a proposition, connotatively one that overcomes countervailing emotions.

This is essentially correct according to LDS theology.

I'm surprised to find people questioning the importance of the relation between faith and reason. It's the central issue in the debate between religion and rationality. Google faith vs. reason and start reading if you're unaware of this tradition.

Maybe people mapping this post to some sort of religious apologia?

[-][anonymous]11y 0

Who cares? Reason won. Next!

It's the central issue in the debate between religion and rationality.

There is no debate between religion and rationality.

Don't understand the downvotes. This is correct. There is no debate between gambling advice and probability theory, either.

Don't understand the downvotes.

Vladimir_Nesov appears to be trying to sound deep by asserting that a piece of conventional wisdom is false without providing any kind of explanation.

This is correct. There is no debate between gambling advice and probability theory, either.

I don't understand this analogy.

Gambling advice is based on probability theory, so the analogy seems to be saying that religion is based on rationality.

Not the gambling advice that has been offered to me.

So there's apparently no chance of having a knowably correct answer to what people 2000 years ago thought about faith and science.

And anyway, what would it change in our modern understanding if we knew what said dead guys thought?

Why does this post have only 5 up votes? This is an honest question. If you didn't up vote it, why not?

If you didn't up vote it, why not?

Didn't read it, didn't care. I reasonably good conception of the concepts and issues surrounding faith and belief and expected the post to include annoying semantics. That is, annoying to me albeit potentially interesting to others.

I did read Yvain's reply, which gives a good summary. I also just saw the second footnote from the post which is an amusing anecdote. My Christian friends should appreciate that one!

Faith: having an unprovable axiom in your belief system that simply states that everything will work out in the end.

What I like about this definition is that it separates all religion and almost all spirituality from the discussion of faith, leaving a small pearl that you can hold onto, knowing full well it's unprovable and not subject to reason.

A couple nitpicks: The author of the book of Hebrews is not known, and this book is not normally attributed to Paul even in Christian tradition.

Also, your second quote appears to be from the New Revised Standard Version, not the New International Version that you cite.

Hebrews is "sort of" attributed to Paul in Christian tradition (it doesn't say it's written by Paul, but it's very Paul-like); but this authorship is controversial. I wasn't aware of how controversial it was. I'll revise my reference.

New Revised Standard Version: "Now faith is the assurance of things hoped for, the conviction of things not seen."

Interesting post. Not sure you said anything new, but this area has been worked over so thoroughly that it would be hard to find a truly novel interpretation.

Why is this so controversial?

[-][anonymous]7y -3

It's pretty plain that when people say they believe in something they don't neccersarily see it before them or in the past - nobody bothers to say I believe in dogs, for example, but are super keen to say they believe in gods - which they probably haven't seen.

Wikipedia parses the meaning of belief into 2 categories. Not that categorisation can neccersarily be defended in any meaingful way until it because ''incorporated by obliteration '' into scientific literature.


Commendatory - an expression of confidence in a person or entity, as in, “I believe in his ability to do the job.”
Existential claim - to claim belief in the existence of an entity or phenomenon with the implied need to justify its claim to existence. It is often used when the entity is not real, or its existence is in doubt. “He believes in witches and ghosts” or “many children believe in Santa Claus” or “I believe in a deity” are typical examples.[13]