Q: Do you need faith to believe in science?

Mathematician: This question could mean a few different things, depending on what is meant by the word “faith.” Let’s start with the dictionary definitions.

1. Strong belief in God or in the doctrines of a religion, based on spiritual apprehension rather than proof

This has nothing to do with science.

2. Complete trust or confidence in someone or something.

You certainly don’t need complete trust or confidence to believe in the results of science. In fact, this is one of the great things about science: since results can be independently confirmed, you don’t have to trust any individual scientist much at all. In good science, researchers will check each other’s work to see if it holds up, attempting to refute what was claimed by conducting further experiments. And it would be silly to have complete trust in what science has figured out so far. When done right, the process works well at sifting out true ideas from false ones, but even then mistakes will occur which may not be caught for some time. However, when multiple experiments by competent, independent teams confirm a result, we can say that it is very likely to be true (or at least, represent an accurate model of reality).

Sometimes, when people say “science requires faith”, what they are trying to get at is the idea that scientists have to rely on assumptions that they can’t prove. For instance, scientists have to assume that induction works (e.g. that you can generalize about the future laws of the universe by looking at the past laws). If tomorrow the laws of physics were suddenly different than they ever were before, science would be in pretty deep water. The thing is though that all methods for drawing conclusions about the world rely on some hidden assumptions, so saying this is true for science isn’t saying much. In fact, the deep rooted assumptions that science relies on are pretty modest.

When people are able to build working satellites, lasers, bridges and computers using certain methods for acquiring and applying knowledge,  it’s strong evidence that the assumptions made by the methods can’t be that unsound.

Physicist: No.  Exactly no.

In fact having an unshakable belief, even in a particular scientific idea, is detrimental (scientifically speaking).

At the risk of making science sound like the sport of jerks; a good scientist is someone who trusts nothing and no one and is willing to drop their deepest held beliefs as though they were a bucket full of red-hot cobras.  “Science” is nothing more than looking carefully at the world, while trying not to delude or trick ourselves too much, and seeing what’s what.  As a result you find that, unlike conclusions based on faith (which I’m not knocking, good on you if you have them) conclusions based on careful consideration of the world tend to show up independently and frequently.

So while the Conquistadors and Aztecs may have had some subtle differences of opinion about feathered gods and whatnot (who can remember), they already agreed on a lot of stuff involving seasons, water pumps, and astronomy, among other things (though not astrology, oddly enough).

But it’s easy to lose track of all that.  When Sagan says “We are made of star stuff“, as both inspiring and accurate as that is, it still sounds a bit like some kind of ancient creation myth (it also doesn’t help that Sagan was rocking the ’70s vibe).  When you get right down to it, the universe is really weird.

So the claims/findings made by science seem about as crazy as the claims made by everybody else‘s religions (not yours or mine of course): time slows down when you move fast, all matter and energy is made of waves, your mind controls reality*, there are an infinite number of parallel universes**, every living thing is descended from goo or something, the universe is billions of years old, our bodies are made of trillions of semi-independent living things, most of the stuff in the universe is invisible and ghostly, the Earth was once ruled by gigantic monsters that were destroyed by a rock from the sky, the world is really a sphere that’s whipping through an infinite void at hundreds of miles per second while in the company of other spheres some of which are so much larger that Bambi v. Godzilla seems kinda fair, and on and on.

(*Not even remotely true, but you still hear it attributed to “science”.  **This is so misquoted and misunderstood that it’s safer to say that it’s false, but it is something science people say.)

It’s easy to see why science seems like just some wacky new belief system that you have to have faith in to believe.  The difference is, if you don’t believe it (and you’re properly motivated), then you can go out and test it. To be fair, most people do take science on faith.  It’s much harder to test things yourself, than it is to trust that the kind of people who can figure out how to fly around in space and build fancy computers have things pretty well sorted out.  But keep in mind; the option is there.

By the way, here are some fairly interesting, slightly dangerous, things you can test yourself.

Scientists, being a lot like people, have a hard time believing the same weird stuff that bothers everyone else.  I mean, seriously, gigantic monsters?  Something becomes “scientific knowledge” after many people have tried their damnedest to prove it wrong and ended up verifying it instead.

For example, nobody really believed that time slowed down for things that move fast, so (and this was just one of many tests) a couple of dudes put some ridiculously accurate clocks on some airplanes and flew them around to check.  And why would anybody think that we’re made of lots of tiny living things?  If you don’t believe it (and why would you?), get a microscope and a little skin or blood and take a look.  It’s hella gross.

Biology: that’s inside you right now.

Even Schrödinger (of equation, cat, and trance-techno fame) didn’t believe several of the clearly impossible implications of his own equation (like quantum tunneling) until they were experimentally verified.

Some of the more obscure stuff takes a bit more work (money), but at the end of the day when something is “scientific fact” it’s been verified many, many, many times, until the strange and inescapable facts about the universe are forcibly inflicted upon the pitiable scientists who study it.  Science isn’t about making bizarre pronouncements and then having everybody nod sagely and agree on faith.  It’s about making bizarre pronouncements and then throwing it to your colleagues to fall upon and mercilessly tear apart, like ivory tower hyenas.  However, physical reality always has the last word.  If an idea doesn’t stand up to observation and experiment, it’s gone.

Faith is about knowing and certainty, while science is about learning and doubt.  You don’t need either for the other.

This entry was posted in -- By the Mathematician, -- By the Physicist, Philosophical, Skepticism. Bookmark the permalink.

30 Responses to Q: Do you need faith to believe in science?

  1. Jim W says:

    This is in response to “does science require faith”. I would appear that anyone who spouts “there is a concensus” and then rejects skeptism is rallying for faith in one theory, whilst bullying others for thinking.

  2. Jordy B says:

    Great article 😀 I just stumbled upon your website and I find your articles very amusing and educative. I am now doing a thesis on relativity and the CERN Neutrino experiment for my last year of VWO* (I must admit I chose it, because it is a very recent discovery and it got in the news exactly at the moment I had to choose a topic, which I otherwise would not have found). I find your articles very helpful, so props for that!
    * http://en.wikipedia.org/wiki/Vwo

    Just one more thing: your description of science, sadly, is not true in every field of science…
    Especially the last part you wrote, which should a ‘scientific rule’, is not always lived by: “then throwing it to your colleagues to fall upon and mercilessly tear apart, like ivory tower hyenas. However, physical reality always has the last word. If an idea doesn’t stand up to observation and experiment, it’s gone.”

  3. The Physicist The Physicist says:

    @Jim W

  4. The Physicist The Physicist says:

    @Jordy B
    If you’re talking about the softer sciences (lookin’ at you psychology), then fair enough. Seems to me a science that doesn’t stand up to careful and rigorous testing, can’t rightly be called science at all.

  5. minator says:

    This is a great article 🙂 Just wanted to take the time to thank you guys for all these awesome posts. Let them keep coming 🙂
    Greetings from Germany!

  6. Will says:

    I think he’s probably more talking about how scientists are only human and sometimes they try and skip or rig the whole ‘peer review’ part.

  7. I just have one question. Regarding parallel universes and evolution. Because it would literally take FOREVER to molecules to organize themselves into the first ever replicating and able-to-evolve cell, would it be okay to say that because there are an infinite number of universes, with every possible combination of everything in each one, that the reason we exist is because we JUST HAPPENED to be in one of them where this kind of evolution just happened to happen? (as an alternative to, say, intelligent design)

  8. Will says:

    “Because it would literally take FOREVER to molecules to organize themselves into the first ever replicating and able-to-evolve cell”

    This is quite incorrect, we have one data point of life appearing from non-life that took approximately one billion years. It’s not a very good data point and it’s a bit fuzzy and you really need more than one data point to form any kind of conclusion, but the one conclusion we can form is that it does not take literally forever to happen.

  9. When one talks about “science” requiring faith (or, rather, belief — technically faith and belief are different concepts, but the distinction isn’t important here), it’s usually in regards to theories and hypotheses, not about the experiments and experimental facts. In Philosophy of Science that takes the form of a distinction between two basic approaches one can take towards the former: realism and instrumentalism.

    Scientific realism, roughly speaking, is to consider and understand the conceptual entities that are part of a given theory, such as, e.g., atoms, particles, space-time, DNA etc., to exist as such, hence the theory to be a literal description of an aspect of reality.

    Scientific instrumentalism, on the the other hand, is to consider and understand those same conceptual entities as artifacts of the theory, useful shortcuts when talking about what would otherwise be purely mathematical constructs, the whole of the theory being no more, and no less, than a very precise way to correlate macroscopic phenomena.

    I’m an instrumentalist, so IMHO, those who adopt a scientific realist approach towards any theory, no matter how well established, is applying faith, yes. Not that I think that’s particularly bad, after all, a realist scientist is no less a scientist just because he believes in these abstractions as if they were real. But I think it’s better for a scientist to be an instrumentalist, as it greatly helps him avoid becoming attached to something that, although useful, is at best secondary.

  10. wrf3 says:

    Can you list the “the deep rooted assumptions that science relies on”?

  11. @wrf3: You asked the blog authors, but let me (also) provide a list of such assumptions.

    First, it’s important to distinguish between science as a practice from scientists, as persons with their own opinions, philosophical or otherwise. A relativist such as Feyerabend would probably say the former is part of the later and it’s all basically opinions, which is a complex discussion onto itself, but I’ll avoid that to keep things simple.

    So, science as a practice has really very few assumptions. Two, to be more precise: a) repeatable events follow patterns; b) these patterns can be described. And that’s it.

    On top of this scientists add lots of additional assumptions. A few examples: that pattern descriptions must hold true for new similar events (inductive reasoning); that math is a valid tool for building such descriptions (this used to be called “scientific Platonism” in the 17th century, mostly by those who saw it as a kind of magical thinking, but nowadays it’s so much part of the common sense that no one questions it anymore); that of the four kinds of causality in classical philosophy, namely material, formal, efficient and final, only the efficient one, i.e., time-bound succession of logically-connected events can and should be used in pattern descriptions (mechanicism); that only math-based pattern descriptions hold value, and hence that non-math-based and math-refusing descriptions either aren’t really science, or at least are “softer” kinds of science; that the conceptual components used to construct pattern descriptions exist as such in reality (realism, although there are a lot of agnostics on this: the instrumentalists); as a possible consequence of the previous two points, that only that which is measurable, i.e., numerable, is actually real (physicalism); that complex patterns can and must be re-described as correlations of simpler patterns; that the hierarchy of causalities starts from simpler patterns, which combine themselves into complex patterns, never the other way around (reductivism); that such combination is also time-bound (evolutism); that one-of-a-kind, non-repeatable events don’t exist, or at least aren’t to be taken into account in the construction of pattern descriptions; among others. But all of these, while integral to the way most scientists out there think, and in fact tacitally taught in scientific classes, are actually philosophical in nature, hence separate and independent from science as a practice, and not actually needed for one to “do” science.

    The above isn’t comprehensive by any means, but it can give you a notion of how the discussion in Philosophy of Science goes. And, evidently, some of my own bias shows.

  12. Will says:

    Technically you only need to make one assumption to use the Scientific Method to find something out; that our senses are accurate.

    So long as you start at the assumption that our senses are accurate you can extrapolate everything else from there. Of course what we’ve found out since is that our senses aren’t always accurate, but that ends up going in a bit of a loop.

  13. vasiln says:

    Scientists study science, not philosophy; it’s not surprising that they’re frequently unaware of their axioms.

    The simplest axioms of science are the axioms of logic: for instance, the principle of noncontradiction. That is, if A, then not not-A. This isn’t something vulnerable to science! You just have to assume it. (Some people claim not to assume it. That seems crazy to me, but whatever.)

    It can certainly be said that to practice or understand science, you need faith in the principle of noncontradiction.

  14. Darkstar says:


    Actually, we should not assume that our senses are accurate (nor our instruments) – in fact, we know that our senses are easily fooled in many ways. That’s why we often construct machines that perform the measurements for us and then numerically analyze them. That way, we minimize the error influence from our senses.

    In addition to our raw senses, we humans suffer from numerous cognitive biases, http://en.wikipedia.org/wiki/List_of_cognitive_biases

    These must also be identified and accounted for and the scientific methodology is usually constructed around trying to avoid falling pray to these biases.

    And finally, we use logic, reasoning, and mathematics and these must be constantly cross checked for consistency and applicability to observations of Nature.

    When I describe science I say it is the amalgamation of the best methodologies we know that help us minimize or eliminate sources of error, bias, and illogic from our conclusions.

    And, as we found out with the recent neutrino measurements, extreme care must be taken at extreme levels of details because, despite the very best efforts of a very large number of highly skilled people, a “simple” mechanical source of error slipped through and was very difficult to detect.

    The skepticism in the results was warranted and eventually the source of the error was found. Cross checking results is one way to make sure that we’re actually measuring what we think we’re measuring (highly accurate measurements are never as simple as they might seem from the outside). It gets to the point where the vibrations caused by a car driving by a mile away can interfere with your measurements if you don’t take precautions.

    For example, we used to think the rotation of the Earth was a pretty good way to measure time – but as our measurements got more accurate we found out that it wobbles too much. Now we mostly use the vibrations of Cesium atoms for accurate timekeeping, and even that isn’t accurate enough for us and new, more accurate methods are being researched and trialed.

  15. Alex says:

    @ vasiln

    I would recommend reading the history of science. Philosophy, religion and science have all been growing together for a long time. However, you are saying science and philosophy and two different areas, and I would ask to provide a definition of the two. Since philosophy created science, and for you to even start science, you need to start non-empirical (i.e., philosophical) assumptions to even get started.

  16. Anshul says:

    If science does not need faith to be believed in, how do you explain axioms? There’s no “proof” that proves mathematical axioms, you either believe in them or you don’t. Indeed, you may be able to “prove” the validity of these axioms by extrapolating them and doing experiments that convince us that they are “true.” However, that’s no different than me saying that a coin will always land on heads, and if this happens, let’s say 50 times in a row (unlikely, but possible), I could call it an axiom and start believing it until the coin someday lands on a tail (which, in the case of our axioms, we don’t know if that “law-breaking event” has happened yet). We just don’t know if or when the axiom will start to break, but the initial belief in the axiom seems purely like an act of faith.

    It’s not like these axioms have been extracted from nature, so they’re purely man-made phenomena. I can see how a law extracted from nature can be self-evident, but calling a man-made phenomena self-evident is akin to faith, wouldn’t you agree?

    On Wikipedia, it says that an axiom is “a premise or starting point of reasoning. As classically conceived, an axiom is a premise so evident as to be accepted as true without controversy.” How do you explain the fact that a supposedly faithless idea such as a scientific axiom needs to be accepted as true without controversy?

  17. Mike Max says:

    I would agree that Science requires Skepticism, but submit that Skepticism leads to Faith.

    This does not necessarily mean religious faith. Science is the employment of the scientific method, i.e. a form of probabilistic induction. This requires acceptance of induction. Accepting induction requires faith in the senses to provide a useful account of reality.

    You don’t need faith to believe in the results of science, but you need faith to believe in the method of science.

    There is no rock solid foundation on which to build knowledge. You have to start somewhere, and this act of starting somewhere is always an act of faith. Everything requires some degree of faith. Why is this important? Because denying the basis for one’s own system shows a lack of understanding.

    Science is great precisely because it places itself below truth. And faith positions itself below science, as its support. The Mathematician had it close.

  18. Robert Mastragostino says:

    I think the discussions here are getting a bit pedantic, and that the Physicist was trying to answer the question in the context in which it’s normally asked. I.e, “science is just another religion”. While these assumptions are a form of faith, they’re minimal. The assumptions of science are essentially the minimal criteria for an ability to deduce highly likely results that can be reliably shown to others. In other words, all you’re assuming is the minimum requirement necessary for gathering knowledge to be a meaningful exercise.

    Any mathematical system starts from axioms. Does that mean I “have faith” in the number system? Maybe to a point, but mostly no. It’s the one I *choose* to study. Similarly, any physical model of the universe is almost necessarily incomplete, and is something we *choose* to study. It would be ‘having faith’ (in a meaningful sense) to assume that we had found the whole truth at any point. That is, we develop a system (classical mechanics) that describes what we want it to (classical motion), and use it to solve problems. The system is a chosen formalism, and working with it does not require faith. Does taking said system and applying it to the universe require faith? Maybe, but not really: after all, it has to be experimentally tested.

    I get what people are saying when they say “science requires faith”, but if you define faith that way then faith is essentially meaningless. I have faith that I’m typing in a coherent language. I have faith that most things I know today will be true tomorrow. This level of ‘faith’ is essentially a bare minimum for *any* idea you could possibly come up with, so saying science ‘has it’ as some kind of meaningful property is essentially useless, since I would posit that there’s nothing to contrast it to. If you say science needs faith, then there is no idea (no, not even the famous ‘cogito ergo sum’, see Kierkegaard) that does not, rendering this reduced concept of faith essentially meaningless.

  19. cordeg says:

    Alas, both Mathematician and Physicist overplayed their hands as a result of misstating what “faith” means.

    Quite simply, take dictionary definition #1 and replace God with Whatever; i.e., “Strong belief in Whatever, based on something other than proof.”

    Clearly, then, belief in science in any complete sense requires faith. And not merely as someone suggested merely a trivial kind of “faith” that underlies one’s ability to say “I have faith that I’m alive and typing on a computer”, but real honest to goodness faith in first principles not backed by proof, but which form the basis of a whole complex web of understanding built upon them.

    I believe in the descent of man, although there is no proof. There is lots of evidence. All makes sense in one frame of reference. But not proof. I believe in the “big bang”, although there is no proof. Lots of evidence. Makes a certain sense. No proof. Also a certain amount of non-sense (requires the laws of physics to be quite different at a convenient time to allow it). [Coincidentally, Father Georges Edouard Lemaitre, a Jesuit priest and man of that other kind of faith, first theorized this notion in 1927.] Dark Energy? Dark Matter? Without scientific “faith”? Really? Best you reconsider your notion of “proof” and “faith”. [And if you think Mathematics is a “science”, you need to re-assess even more — though as Feynman put it in his borderline Seinfeld way, Not that there’s anything wrong with Mathematics not being a “science”; it’s still Math, but Mathematics “proof” is not scientific “proof”.]

    Plenty of serious and significant scientists have admitted the same: Feynman went so far as to say that most people fundamentally misunderstand science in relation to the meaning of “proof.” Neil DeGrasse Tyson famously said that the only way he could believe in all sorts of tenets of modern science was “faith”. One could go on all day.

    Certainly, faith in science does not require faith in God, though it also does not preclude it. When I go to a football game, my understanding of what is happening and what is likely to happen is based on my understanding of the rules of football and not in any way contradicted by my understanding of the rules of baseball (and vice-versa). The two things are not at odds. They are different reference frames in which different rules apply and different notions of “proof”. Scientific proof explains nothing about religious matters. Religious proof explains nothing about scientific matters. Neither scientists nor clergy should be concerned nor defensive about this. It would be as silly as sending a baseball umpire to run a football game — or a line referee off to run a baseball game.

  20. RM says:

    @cordeg, that seems to be talking about specific results. Yes, it takes faith to believe in any specific scientific result as being some kind of revealed truth. But you can’t talk about science as a complete endeavour in that same way. To believe that science has revealed some specific truth will take a certain amount of faith, but to say that believing in science *as a process* takes faith makes less sense. It only takes the faith that we are in a coherent world, have means to observe it accurately, and can communicate these findings. To even have a conversation on the subject at all implicitly accepts these ideas (albeit only in a certain field of relevance), so to talk about rejecting them seems oxymoronic.

  21. Bret Lenehan says:

    Science obviously requires faith, the faith that our human brain, our senses show us reality as it actually is. Is there really an external world at all. Science cannot prove there is. Hawking uses the term model dependent reality he cannot just say reality. However unlikely we may think it ,we could be brains in a vat, a computer simulation from a higher civillization ect. Science likes to think it has all the answers but philosophy and even quantum physics show us science has its limitations. I had not even thought about the assumption of the laws of physics being the same in all places and times, quite an assumption. I won’t even get into miracles or the paranormal but most scientifically inclined dismiss them out of hand.

  22. CBG says:

    Beyond the “faith” that we exist, that we can observe, that we can reason snd draw conclusions, no “faith” is needed. Quite the opposite. Skepticism is the requirement. Science is built on subjecting explanations to critical inquiry. Weak ideas get weeded out and better ideas eventually replace them. Do that long and often enough and one builds a better view of whatever is being studied.

  23. Tony says:

    Science is the study of the natural, it has nothing to do with anything other than the natural. Absolutely nothing. It cannot prove the existence nor the nonexistence of what we call the supernatural, meaning the existence of God. It’s power lies in the ordinary, nothing else.

  24. Steve says:

    Piltdown Man, after 40 years of acknowledgement from the science community as a link to early evolved man, was shown to be a fraud. Absolutes do not require Faith, absolutes are infinitely true and self evident. Science says “we’re learning new things all the time”. Therefor science cannot be absolute because it can and does change. Piltdown Man had the unfortunate affect of creating 40 years worth of false research, false hypothesis, and in the case of Sir Arthur Smith Woodward a limp and meaningless legacy. I challenge any scientist here to argue faith is not needed when “evidence” and “proof” are not absolutes BY YOUR OWN ADMISSION.

  25. Jim Walton says:

    I have “faith” in the scientific method, not religious faith, but faith that truth will rule the day, eventually. The Piltdown Man escapade is just one of many examples of how ‘evidence’ goes wonky, but the method (experiment/verification) works.

  26. Alanna Tu says:

    Good article, and a great read.

    — Alanna Tu


  27. April Xin says:

    i enjoyed reading this! Thanks. It was a very thought provoking article!

    —April Xin


    (Xinyi Xin)

  28. Angel says:

    The commenters on this thread must have a bad understanding of the definitions of “belief”, “faith” and “knowledge”. You are arguing that everything requires faith. Your arguments do one of two things: to render science as utterly useless, or to render the concept of faith (actually, belief, really) as totally irrelevant. And if that, then I don’t know why do you even argue. But by saying that all forms of knowledge require belief (which is literally all you have actually said, just in many different ways) and that nothing can be known, you just defeated your own arguments, as they depend on reasoning itself, which cannot be fully trusted according to your own conclusions! Logic seems useless in that context.

  29. Joe Dyson says:

    Do you need faith to believe in Science ?

    Yes, of course you do – the scientific method for example relies on our natural laws always behaving in the same way. Until we can ‘prove’ that this is impossible then science will always need faith.

    Most high profile scientists have actually accepted this now.

  30. Darin says:

    You have to have faith in everything/anything…to whoever is reading this right now, how do you know you are really who you think you are? How do you know you are not some little girl from some prehistoric world in a space ship with a bunch of aliens that have wires hooked up to you making you live in a ‘matrix’ type world, where you think there is an Earth and galaxy and the laws of physics, etc. ?? You don’t that what you think is reality is reality. It takes faith to believe in ANY reality that you think is real. For some really smart scientist that is reading this right now, how do you know you really ARE a really smart scientist and you are not some poor guy in an asylum right now believing in some fantasy world that is all in your head? Everything requires faith. I have in God…some have faith in science….you can choose what to have faith in, but EVERYTHING requires faith.

Comments are closed.