On Science as a Form of Religion
Jason Kuznicki on Nov 28th 2007
The thesis of this Paul Davies piece seems to be that science is ultimately another form of religion. Superficially, the claim appears profound, and I’ve often heard people – superficial ones – drop this idea into conversation hoping to stop in its tracks any talk about science versus religion.
But with all pithy philosophical pronouncements, as the words become fewer, their definitions become more important. Much depends on what we mean by “religion,” and on what we understand science to be. A misdefinition can be fatal, and similarities do not imply identity. This last I think is the real problem with Paul Davies’ argument.
“Science is just another form of religion” actually has considerable truth to it: Sociologically, science and religion are quite similar. There are codes of behavior, rituals, professional organizations, degrees of rank and honor, and even (on a very materialist level) special places, forms of dress, and texts. Each of these things acts to bind a group together (the root of the word “religion” means “binding”), and to mark the in-group as distinct from outsiders.
But Davies’ piece is not about sociology. Neither are most claims about science as a form of religion. Davies is writing about metaphysics, and he’s not doing it particularly well.
Davies posits that descriptions of observed or expected patterns in physical phenomena necessitate belief in unseen things. I think he is simply wrong on this point; an observation of a pattern may be explained in any number of different ways, testable or not. Some of the best scientists (Newton for example) just observed patterns without offering any explanation at all for why they were there. Newton freely admitted he didn’t have a mechanism in mind for gravity–all he had were equations describing what it did. Those equations worked astonishingly well, and thus he was a remarkably good scientist.
Yet the act of observing for example that the laws of gravitation are invariant does not commit one to any metaphysical explanation for why they should be that way. Discovering that phenomena happen in regular patterns says nothing in particular about metaphysics, or about why patterns exist, or about the idea of “pattern” itself.
Descriptions of observed regularities, and descriptions of regularities that we think we might be able to observe if we looked for them, are the work of science. Beyond that is metaphysics, and, because metaphysical claims can’t be observed or tested, science has nothing to say about them. They are the province of religion and philosophy alone.
But it’s nonetheless appealing to the religious to be told that science, too, is a kind of faith. That way they can dismiss all that silly talk about people coming from monkeys: Buddhists believe strange things, too, after all, and just as we need not believe in Buddhism, we also need not believe in science. It’s just another religion, and it’s not my religion.
A related fallacy from the article may help to illustrate the point: “If the laws of physics were just any old ragbag of rules, life would almost certainly not exist,” Davies writes. True, but then, if a frog had wings, he wouldn’t bump his tail when he hopped. Speculation based on alternate laws of physics that are not the laws of this universe is most emphatically not the province of science. It’s metaphysics, and science can’t contribute anything useful to it.
Consider similar speculations in another context and the problem becomes clearer. As a historian, I would be ridiculed if I offered a counterfactual of a like nature. For example:
“If George Washington had died of cholera as a child, we would still be a British colony. God is protecting the United States of America.”
Now both of these sentences may actually be true for all I know (I happen to be an atheist, but my belief on this point may after all be mistaken). Yet these propositions are not the data that historians have to work with, and the relationship between them must remain totally outside the realm of serious history. Insofar as history is a science, which is admittedly to a limited extent, we historians confine ourselves to what actually happened. True history frowns on making positive claims from counterfactuals. So do all true sciences.
Given that scientists have only ever observed one universe, and only a very small part of it, it’s entirely ridiculous for scientists to speculate about other universes, or about the process of how other universes were built, or about the type of Entity that would create a universe like our own. (Side note: If God had wanted a universe tailor-made to humanity, why does it appear that over 99.9999999999% of it is perfectly uninhabitable? Those who take an anthropocentric view of this universe and its life-sustaining properties should think twice. After all, God might have ordered things so that the whole visible universe was just endless Tahiti, which would be a whole lot more pleasant than an endless frozen vacuum. Our universe is certainly NOT “just right for life,” whatever may be contained in Davies’ book. It’s a vast desolate wasteland, with an infinitesimal island of relatively habitable space that, we now believe, will one day disappear. It’s not very hospitable at all. Unless you’re a hydrogen atom; these, apparently, God loves most abundantly.)
After this digression, statements like the following from Davies should immediately seem suspect:
Clearly, then, both religion and science are founded on faith–namely, on belief in the existence of something outside the universe, like an unexplained God or an unexplained set of physical laws, maybe even a huge ensemble of unseen universes, too. For that reason, both monotheistic religion and orthodox science fail to provide a complete account of physical existence.
The problem here is that science need not be based on any of these things. Science isn’t compelled to make any sort of predictions on any things that aren’t observed. On the contrary, these speculations aren’t part of proper science at all, since they are not subject to observations given any known or even potential technology or experimental apparatus. Science simply reports on observed phenomena and tries to identify patterns within them. Occasionally it proposes phenomena that might be observable, and then it tries to do so. To the extent that a scientist talks about something he cannot possibly observe, and that no potential experiment or evidence could ever illuminate, he has stopped following the scientific method.
Maybe our universe really is the only one out there. Maybe it’s part of a multiverse. Maybe God made the whole thing, and maybe he didn’t. Since these things cannot be observed, they are not a part of science, and the author is wrong to say that science entails belief or unbelief in any of it.
Filed in The Belfry, The Biosphere
Jason,
The first part of this statement (up to the last “or about the type”),
Given that scientists have only ever observed one universe, and only a very small part of it, it’s entirely ridiculous for scientists to speculate about other universes, or about the process of how other universes were built, or about the type of Entity that would create a universe like our own.
is not strictly true. Cosmological speculation which may involve other universes or universe creation may have implications regarding the creation of our universe which may be measured. For example, George Gamow noted that the big bang theory had an implication, e..g, the microwave background radiation which was an observable consequence and is observed. Likewise speculation into the theoretical requirements of universe creation may have topological or other features which may have side effect or observable consequences in our universe.
There is another inherent difference between science and religion that Davies is missing - the nature of that which is “unexplained.” Whereas religion consigns most or all “unexplained” things to the realm of the unknowable and potentially the supernatural, science takes that which is unxeplained as a transitory state. By generating theories about the natural universe and testing them through observation and experiment, the unexplained becomes knowable, or at least predictable. In fact even the potential of the multiverse comes not from a metaphysical platform, but from observed phenomena that could be explained by multiple universes interacting as part of a larger structure. Some string theorists even posit observable phenomena that could be used to demonstrate the existence of these other universes through their effect on ours.
Excellent post, Jason!
This is reminiscent of the Anthony Flew flap a couple of years ago. Apparently entering senility and not well-versed in evolutionary biology, he bought into (possibly temporarily) the argument from design. Davies seems to flirt briefly with a version of the anthropic argument, but doesn’t consummate the relationship. Of course, the parallel is at most partial since Davies is well-versed in physics; but perhaps the first possible point of similarity maintains.
The op-ed piece is a textbook example of what’s often wrong with this sort of argument. As Mr. Kuznicki observes, definitions are critically important; so is logical consistency. On both counts, the piece fails.
Dr. Davies opens with the (implicit) definition of “faith” as “belief without evidence”. Now I find it hard to believe that he really accepts this primitive definition. (FWIW, I consider faith to be a special case of belief if the latter is properly defined.) But let’s assume he does.
He then states, “science has its own faith-based belief system”. By his own definition, this expands out to “belief system based on belief without evidence”, which seems incoherent. But again being generous, let’s assume it means something even if we aren’t sure what.
The rest of the paragraph can be abbreviated to “[S]cience … assum[es] that nature is ordered in a rational and intelligible way … [a]nd so far this faith has been justified.” But notice the verbal alchemy: an assumption verified by centuries of evidence has been magically transformed into an article of faith. Surely a prominent physicist understands the difference between postulates and the logical conclusions drawn from them. Or the difference between “reality” and the mathematical models used to explain and predict aspects of that “reality”.
Or maybe not. Later on we get “the very notion of physical law is [] theological” and “physicists think of their laws as inhabiting an abstract transcendent realm of perfect mathematical relationships”. This smacks of whether mathematics is discovered or created. Again, it would help to have a definition of “physical law”. I am neither physicist, scientist, nor philosopher (lowly engineer) and may be entirely wrong, but I tend to think of a “physical law” as being behavior (eg, bodies attract) and the formula describing that behavior as only an approximate mathematical model. Now if Davies means the former, then I can see some parallel between physical laws and Platonic forms, which for all I know are viewed by the cognoscenti as “theological”. But if he means the latter, then my (somewhat informed) bias toward the “created” position leads me to conclude that the “laws” are temporal. Unfortunately, given his cryptic language (”abstract transcendent realm”, etc”), it isn’t clear (to me) which he means.
Next, Davies creates what is IMO a strawman scientist. Asked “why the laws of physics are what they are”, colleagues stumble in answering; possibly because seeing the question as perhaps a trick and not wanting to be embarrassed, they eschew the obvious answer: “who knows”. From this stumbling, Davies infers abandonment of reason in the clutch and concludes that if in seeking the “bedrock of reality - the laws of physics” one finds “that reason [] deserts us, it makes a mockery of science”. The “if … then” logical statement may be arguably true (though grammatically awkward), but the proffered evidence in no way supports the hypothesis.
All of which leads to some (IMO) marginally coherent explanation of why this has emerged as a critical issue. It’s here that the anthropic argument appears, followed by some material about multiverses that is way over my head, and presumably over the heads of even the typical elite reader of the NYT. And finally, the bottom line: “both monotheistic religion and orthodox science fail to provide a complete account of physical existence”; therefore, both are based on faith.
This is perilously close to the “god of the gaps”, although to his credit, Davies doesn’t actually go there. But it is also perilously close to logical nonsense, how close depending on the definition of “religion”. A deist’s faith in God may well be in some sense analogous to a scientist’s faith in the “bedrock of reality” (whatever that is). To which my response would be: “And so what? Even if true, why is that insight helpful to anyone?”
But in the case of a fundamentalist Christian, not only is there no analogy, the conclusion is clearly wrong. A scientist necessarily assumes the current scientific account of reality is incomplete; a fundamentalist Christian presumably assumes the Biblical account to be complete. And the two postures affect us in dramatically different ways; IMO, positively and negatively respectively.
All of which leads to the question: what was gained by arguing a close parallel between science and religion? It seems clearly wrong in the generality presented. Careless references to God by prominent scientists invite claims by religious apologists: “see, s/he’s one of us”. And the claim (true or not) that there is an element of “faith” in science can be argued without reference to religion (just as the post argues against the claim with minimal references to religion). All in all, a very strange piece.
- Charles
Thanks for responding to Davies’ piece. Your essay is better than most which are on the Edge website, and certainly deserves to be reproduced there.
>
Not a valid argument since we neither know the conditions necessary to evolve Man neither the other purposes God may have in creating the universe.
Also, a vast Universe and small Earth gives to Man a notion of relation between himself and God (this is from CS Lewis)
>Not a valid argument since we neither know the conditions necessary to >evolve Man neither the other purposes God may have in creating the >universe.
If we don’t even know the conditions for producing Man, then we certainly can’t conclude that this universe offers the best possible way to get those conditions.
On a related point I have never understood the first part of the fine tuning argument, that it is unlikely for the universe to have the constants it does. Can anyone explain to me how you can talk about the likelihood or probability of something when you have a sample of one?
“how you can talk about the likelihood or probability of something when you have a sample of one?”
If you know a little probability theory, you obviously can’t. But most people don’t and then it’s easy. (:>)
I’m not very clear on the anthropic (fine-tuning) argument either, but my impression is that it isn’t about probability in a quantitative sense, only qualitative. It seems to be (logically) a variant on the argument by design. Just like some things are (supposedly) so complex that they are “unlikely” to have evolved without help, our existence is so sensitive to the precise values of various physical constants that those values (supposedly) are “unlikely” to have occurred without help.
But in any event, why the god whose existence is “proved” by such an argument is of other than purely academic interest? Even if I were convinced by the argument, I don’t see what I would do differently. To motivate a change in behavior, it would be necessary to connect that god to some doctrine. Which one and how?
- Charles
ctw says:
But in any event, why the god whose existence is “proved” by such an argument is of other than purely academic interest? Even if I were convinced by the argument, I don’t see what I would do differently. To motivate a change in behavior, it would be necessary to connect that god to some doctrine. Which one and how?
The Doctrine of Fear.
Fear of God.
Fear of death.
Fear of the unknown.
Fear of solitude.
Fear of responsibility.
Actually, I believe it goes deeper than that. When physicists are looking at specific constants and physical constraints of the universe, it turns out that very small differences, and we’re talking minute here, could result in a radically different universe. For instance, if some very minute changes in the early universe had occurred, it might have been impossible for any matter to have formed - the universe would have been a great energy field only (IIRC). Thus, man, plants, stars, etc., could only exist with the given constants of the universe.
Is this evidence of intelligent design? Well, on the other hand if the constants and constraints had not been exactly as they are, we would not be here to discuss it - in other words, they were as likely as this universe, so perhaps we just got lucky. Or, perhaps there are multiple universes where all the potential constraints and constants have/will occur, and we can only live in this one universe.
CPT_Doom:
The behavior you describe is what I meant by the phrase “so sensitive to the precise values”.
My reference to the argument by (sic - should be “from”, I think) design may have mislead you. I didn’t mean to imply that the fine-tuning is a case of ID, just that I think the argument in both cases is of the form “this phenomenon is so unlikely that it must had a helping hand”. My inept way of making the distinction was adding the parenthetical “logically”.
But having said that, the only reason I separate the anthropic and design arguments is that they seem to be treated as separate by people who actually know what they’re talking about. From my unsophisticated viewpoint, they look pretty much the same.
BTW, the position you express in comment 2 seems so obvious that I’m always amazed that anyone thoughtful either disputes or overlooks it. I know it’s a bit subtle for the clueless, but Dr. Davies clearly isn’t. It’s a mystery.
- Charles
I concur with Mark Olsen’s comment. Multi-verse hypotheses can be, and probably are, both speculative and scientifically derived in the sense that it may be possible that future measurement based evidence will be found (probably by expensive ground or space based instruments) that supports predictions made by one or more hypothesis that incorporates a multi-verse scenario. This is a more mathematical, theoretical type of science than the science of Newton, but it would still be a conclusion derived from the application of similarly strict logical and observational constraints and thus still qualifies as science even if the conclusion remains beyond more direct observation and thus remains more speculative.
Regarding fine tuning, since we don’t know why the physical constants are as they the fine tuning argument for god sounds to me like another god of the gaps type of argument. God did it just doesn’t work as an explanation, its synonymous with saying we don’t know. We can only explain the otherwise unexplained in terms of what we do know, and we don’t know “god”, so god is not a basis for any explanation. And that is the big problem with Davies argument, he assumes a-priori that god is a valid explanation. Once you accept that false premise, then of course “god”, however one defines that, is one “unexplained explanation” that is at least as good as any other possible “unexplained explanation”. But, again, that is a mistaken premise. The physical constants exist and anything that exists is a proper target for seeking an explanation for that existence, but ill-defined catch-all declarations rooted in the unknown are not explanations, so the proper conclusion is that we don’t have an explanation. Davies wants to use our not having an explanation as a justification for giving any ill-defined catch-all declaration rooted in the unknown the status of a valid explanation. That is a circular, closed, fatally flawed argument.