More "Self-evident" Premises of Science

Nathan Richardson

In a previous post, “‘Self-evident’ Premises of Science,” I cited six premises listed in the opening essay of the BYU physical science textbook. The authors explain, “All reasoning must rest upon assumptions, and the scientific method … assumes basic philosophical ideals as a foundation. … There are some assumptions that are so logical and basic that we present them here as six “self-evident truths.”1 I questioned each assumption as defined in the text and gave my reasons for doing so with three of them: non-contradiction, existence, and causality. I will now address the other three:

Position symmetry. The laws of nature are the same everywhere in the universe.
Time symmetry. The laws of nature have remained the same through time. They are the same now as they were in the distant past, and they will be the same in the future.
Simplicity (Occam’s Razor). If alternative explanations of any phenomenon are available, where each are logical and explain the phenomenon equally well, then the simplest explanation shall be chosen.1


Simplicity. Also called the principle of Parsimony, Occam’s razor is stated in various ways, sometimes saying that, all else being equal, “Simpler explanations are more likely to be true than complex ones.”1 Simply stated, there is no way of knowing that is true. When exploring unknown regions of the physical universe, a person does not know what law governs the phenomena he is studying. So how can she assume anything about the final explanation, when she hasn’t reached it yet? Simplicity seems to me more like a practical premise that is often true, than a self-evident premise that is always true. It may be very useful for expediting scientific inquiry, but I would never let it prevent me from considering possibilities just because a simpler explanation that was available also explained all the data. As Jeff explained in “Odd Realities and Moral Imperatives,” reality is often something we could never have guessed.

Position symmetry. With the exception of divine revelation, the only way I can think of that a person could claim to know whether the laws of nature are not different at different locations would be to experience those laws functioning at every point in the universe. In other words, God himself would have to tell you this, or you’d have to have some god-like capabilities to test it. Position symmetry seems a likely possibility to me, but I have no way of showing that it’s true or not. Maybe the laws of nature are different at different locations. Maybe certain parts of the universe “govern all those which belong to the same order” by different natural laws (Abr. 3:3, 9). I hesitate to cut myself off from the full range of possibilities, especially when we have studied such a small fraction of the universe.

Time symmetry. It seems to me that the scriptures contradict the way people usually interpret time symmetry. Heavenly Father has said he is consistent in his attributes, such as wisdom, power, justice, and mercy. But how can we say, “The laws of the universe do not change with time,”1 given what we know about the fall of Adam? Excepting the fall, “all things which were created must have remained in the same state in which they were after they were created,” which sounds to me like many fundamental changes occurred (2 Ne. 2:22). We have no idea what all was entailed in the fall. Many prophets have made statements that the fall affected not only Adam and Eve, but the entire planet. Death entered the world, and none of us knows what changes may have happened at a biological, chemical, or physical scale. So how can I be sure that carbon-14 has always had a regular decay rate, or a steadily-changing decay rate? There is no way I can ever demonstrate that. I can see why geologists assume it; it gives regularity and organization to their observations. But I would once again call it a pragmatic premise, not a self-evident one.


As I said before, some of these premises seem true in many ways, but I hesitate to call them all “self-evident” truths. And because of that, I can’t help but mentally qualify every conclusion based on them. I can see someone replying, “Well yes, of course we are aware of exceptions that the restored gospel brings up, but these statements are mostly true, in most situations.” To me, it seems short-sighted to use something as a premise when we acknowledge exceptions. When we know of exceptions, it is not a strong premise, and treating it as such is an unwarranted self-limitation. At best, we cut ourselves off from many possibilities when understanding the world around us. At worst, we abandon revealed truths in favor of speculation, merely because the speculations have been repeated for so long by so many.


1. Physical Science Foundations, 2nd ed., BYU College of Physical & Mathematical Sciences.


  1. I like the simplicity assumption. And just because “reality is often something we could never have guessed” does not mean that this “something” is not always the most simple explanation. Scientifically speaking, this assumption is delicious to me. I see your point about these assumptions being more pragmatic than self-evident.

  2. Right, I totally agree when you call it delicious. 🙂 It’s incredibly useful in sorting through explanations. I think it may even likely be true. But the point is, there is no way to ever demonstrate it scientifically. Its scope is way too broad. That said, I’m still in favor of using it as a procedural principle. I’ll just call it pragmatism, not logic or sensory empiricism.

  3. I worry that the first two “assumptions” are not really assumptions at all, but hypotheses which have been subjected to, and have withstood, testing.

    I also think that the principle of parsimony could use a better defense than you have given it. Let’s think of it this way: For any phenomenon, we can come up with literally an infinite number of explanations for it simply by multiplying entities. Now out of all these explanations, does it not stand to reason that those with the lowest number of theoretical entities more likely to be true? And this is in addition to the pragmatic value of dealing with fewer theoretic entities and variables.

  4. Jeff, I believe space and time symmetry have withstood a lot of testing—in the very limited scope that we’ve been able to test them. That scope is pretty much within our solar system, and within the past few centuries or millennia. Judging the size and age that most people attribute to the universe, how can we reasonably say that we’ve thoroughly tested even a fraction of the time and space that we assume to be symmetrical? I agree with you that they’re probably better labeled hypotheses; to use them as foundational premises, as was done in the textbook which I quoted, seems premature to me.

    This is in addition to the pragmatic value of dealing with fewer theoretic entities and variables.

    I agree with the pragmatic value of Parsimony. That’s why I said, “Simplicity seems to me more like a practical premise that is often true. … It may be very useful for expediting scientific inquiry.”

    Does it not stand to reason that those with the lowest number of theoretical entities more likely to be true?

    I see what you’re saying, and it seems to hold in most cases. I just balk at calling it a self-evident premise, because it’s easy for me to imagine a situation in which the third-simplest theory turned out to be the true one, but it was rejected because it had a few more entities than were deemed necessary.

    But when it comes down to it, it also seems strange to use odds when talking about explanatory theories. For example, of ten mutually exclusive theories of motion, only one can be true. It’s not the most likely to be true; it is true or it isn’t. The other nine theories aren’t less likely; they’re wrong. OJ Simpson isn’t 87% likely to have killed his wife; he either did or he didn’t.

    Maybe odds are a useful decision-making tool, but as I think about it, to talk about abstract principles as likely or unlikely seems, well, odd.

  5. Nathan, I don’t think you can say the testing is just within our solar system. Most of the ‘tests’ come from cosmological studies involving observations of the entire universe.

    Of course we can always say there is more to learn. Just look at new phenomena like dark matter which only a few decades ago wasn’t even a gleem in an astrophysicist’s eye. I think we need be careful though to assume the fact we don’t have all knowledge entails an ‘anything goes’ hermeneutic. (I’m not saying you are doing this—just that it is very common)

  6. Yeah, I agree about the “anything goes” hermeneutic; I’m hoping to encourage the “revelation goes” hermeneutic. My main reason for examining these premises is my concern when some saints so casually dismiss prophetic statements in favor of “undeniable” science. I hope this discussion helps them see that if a prophet makes a statement that doesn’t match current theories, there’s no need for a crisis of faith. But in the absence of revelation on any given matter, I think these premises can be very useful.

  7. I think the problem is that when people pit revelation against science what they are typically doing is pitting a particular reading of revelation against science. And typically the hermeneutic used in interpreting the revelation outright ignores science and what we know of the world.

    The obvious worse case example of this is new earth creationism or anti-evolution readings of Genesis. But there are lots of other examples.

    The question of interpretation simply can’t be neglected. And, historically, readings that contradicted science are almost always eventually taken as wrong. (I’m not aware of a contrary example)

  8. To add, I’m not going to say a revelation can’t contradict current scientific thinking. Obviously it could. I am saying that I’m just not aware of a revelation that ought be read that way.

  9. The main reason I like this post is because (although I don’t presently disagree with the present usefulness of the assumptions in this post) I think it’s important to recognize that science is based upon assumptions… not empirically verified facts alone, but empirical facts interpreted through the lens of particular naturalistic assumptions that are far from indubitable (even though they may be true).

    Another reason I feel this is important is because there are other assumptions of science (not listed in this post) that I do disagree with. People treat them as proven fact, when they are simply assumptions.

  10. One thing to remember though is that science is a social practice as much as anything. Thus within the community of scientists not everyone shares the same assumptions. Indeed that is crucial for the development of science.

  11. Clark,

    I agree with you that science is a social practice as much as anything else. I also agree that within the community of scientists not everyone shares the same assumptions. However, I’ve seen that line of argument used way too often as a dodge to evade having to admit that there are certain basic assumptions (whether explicit or implicit) that are shared by many or most scientists. I’ve been in many debates where in the face of sound criticism and because of mounting unease with the clear implications of endorsing commonly accepted assumptions, my opponent responds with something more or less like this:

    “Well, that doesn’t apply to me. I don’t assume that. So, what you claim about what scientific psychologists claim cannot be true. After all, I’m a scientific psychologist and I don’t believe that (at least, I don’t anymore now that you pointed it out to me and have made me uncomfortable). Thus, what you are arguing is not relevant to me or my work, so just leave me alone and let me get back to business as usual.”

    (Note: The following observation hinges on the questionable assumption that contemporary experimental psychologists are in fact scientists at all.)

    I think there is excellent evidence that the majority of mainstream experimental psychologists assume that the universe of human behavior and experience is governed by certain natural laws of behavior best characterized in terms of efficient causality and solely discoverable via carefully controlled experimental means. All you have to do is read the most widely used textbooks, the most cited journal articles, or listen to a few high profile conference presentations sometime to see that I’m not exaggerating. Now, I’m sure you can find some experimental psychologists out there who would either disagree with this or would have a more finely nuanced position on the nature of causality and natural law (indeed, I work with some of them here at BYU). Nonetheless, I would argue that one would be on very solidly defensible ground to claim that a basic assumption of the mainstream of contemporary scientific experimental psychology is efficient causalism.

    Having said that, let me again say that I agree with your two first observations. I also agree with your concluding observation (i.e., it is crucial for the development of science that not everyone share the same assumptions). I believe that science (or any intellectual endeavor, for that matter) can only make real progress when there is a tolerance for alternative viewpoints and competing assumptions sufficient to allow some genuine (even heated) dialogue and self-reflection to take place. Unfortunately, for the most part, my experience—both in psychology and in my interactions with scientists in other disciplines—there has been little real tolerance for alternative perspectives that question root assumptions or entrenched practices. This is nowhere more true than in the realm of academic publishing.

    Of course, one would expect that sort of thing if science were less a detached exercise in pure rationality and more—as you say—a social practice.

  12. Ed, I don’t want to say there are no generally shared assumptions. Efficient causality obviously being one example, although even there I think one needs be careful. Simply due to the often problematic nature of causality. Consider the Dirac equation and it’s evolution in simple QM. How do we conceive of causality?

  13. BTW—getting back to the original assumptions listed. Symmetries aren’t really assumed, despite what the textbook said. There is a very interesting mathematical theorem called Noether’s theorem that relates symmetries and conservation law. While due to some complex things in GR you can’t simply apply this to cosmology you do find scientists investigating these issues. So for example in cosmology it’s generally acknowledged that there isn’t an universal time symmetry. Likewise the ‘lumpiness’ issue has led some to question spatial symmetries of various sorts.

  14. Clark: It’s generally acknowledged that there isn’t an universal time symmetry.

    My understanding is that this allowance usually only refers to the very first moments after the Big Bang, but that after that, time symmetry is assumed. Are there theories that assume time asymmetries beyond the initial stages of the Big Bang?

    The main example I know of that employs time symmetry is the uniformitarianism assumed in geology. “The present is the key to the past,” and, “Geological processes are occurring at the same rate today as they were in the distant past” are two ways this assumption is usually articulated. I don’t see how that can ever be experimentally demonstrated.

    Likewise the ‘lumpiness’ issue has led some to question spatial symmetries of various sorts.

    That’s really interesting. What more do you know about that? Are they saying that there might be regions of space where, for example, the gravitational constant is lower or higher?

  15. The obvious time asymmetry is the arrow of time.

    But the early moments of symmetry breaking in the universe are others. Certainly most of time in the universe was reasonably stable. Whether it is stable enough is another question.

    The lumpiness issue isn’t related to constants (which appear pretty constant) but just is a lack of symmetry.

Leave a Reply

Your email address will not be published. Required fields are marked *