Science and truth

In Discussion on March 11, 2010 by Pete Abram

The question of how well science is at finding truth is one we’ve discussed a few times during the break in class, so I figured I’d start up a thread for discussion about it. Todd and I firmly believe that although scientific theories claim to aim at truth, the matter of the fact is that they just plain can’t. I don’t want to put words in his mouth, so anything from here on out is all me. I don’t think that science can ever get to the fundamental thing-in-itself because of the limitations of our methods. Every method we have of testing/observing/deconstructing an object in science will depend on certain aspects of the object e.g. how it reflects light or how it reacts when bombarded with electrons. There isn’t a method of study in science that does not depend entirely on the interactive properties of an object, so we never know what the object actually is, we merely know how it acts. Now I’m sure the objection could be leveed that how an object acts is the same thing as what that object is, and someone can bring that up if they’d like.

I also don’t think we have particularly good reason to think that any scientific theory is correct for the following reason: every single scientific theory that’s existed before our current ones has been false (geocentrism, Newtonian mechanics, plum pudding atom model, &c.). Basic inductive reasoning tells us that there will be some new theory that better describes how everything interacts and proves the previous theories to be laughably absurd. And any response that says “we don’t have reason to think that it’s false yet” will just be an argument from ignorance. Think of the scientific mode of finding facts like a person. Everything that this person has told you so far has seemed to be correct, but has actually been wrong. Why should we think that what he says now is true, even if all of the available evidence confirms it? I guess a decent way of posing this second concern is as follows: Should we trust reputations/induction or current evidence when past evidence has proven to be misleading?



8 Responses to “Science and truth”

  1. If we adopt a pragmatic orientation then truth is not important. Instead, we become concerned about whether a theory is instrumental (or useful) in solving human problems.

    Newtonian mechanics help us solve a huge class of problems albeit with some anomalies. Creating a theory that solves these anomalies is great and often allows a new set of problems to be solved. Thus, science is said to progress.

    Note that theories may not always progress – solving one set of anomalies may “unsolve” previous problems (cf. Laudan).

    There is also no guarantee that new elegant theories can solve ANY practical problems, thus, raising the question of the “end of science” (see Horgan, “The End of Science: Facing the Limits of Knowledge in the Twilight of the Scientific Age”).

    The tricky part is that we never know when some seemingly abstract theoretical contribution may have a profound practical application (E=mc2 anyone?)

    However, my gut tells me that we are over-investing in fundamental research right now. For instance, most research papers generated by lower-tier universities will never be cited by others in the field. Science has been spectacularly successful in solving important human problems. However, it doesn’t mean it should get a free pass to unlimited investment.

  2. There is a fairly large literature on the realism vs. anti-realism debate in science. And, good news, there are good textbooks that are accessible and current. I particularly like James Ladyman’s textbook and his treatment of realism:

    This is good too: It is by Boyd, a respected realist.

    One of the most influential “pragmatist” is Bas vanFrassen. His work is worth seeing. He’s an “empiricist”.

    Larry Laudan has a nice book that I’ve used in general Phil Sci courses (maybe this is the one to which Steve refers?):

    The only thing of substance I will say here (I think the questions are too big to cover in one comment post) is that to be a realist you don’t have to believe that any particular theory reliably tracks the way the world is. Rather, you can believe that while we may never achieve that ideal, we’re approaching it.

  3. The following is from Yasha and Collin:

    Pete, you make some bold and interesting claims in this post. First, you say, “I don’t think science can ever get to the fundamental thing-in-itself because of the limitations of our methods.” It is unclear to us what the “fundamental-thing-in-itself” is supposed to mean here. Are we looking for essences? If so, hasn’t science discovered the essence of numerous things; e.g. water is H20 or that gold just is having atomic number 79. What more could we request of science? Anything more seems to set an unreasonably high standard for scientific truth.

    You go on to say, “There isn’t a method of study in science that does not depend entirely on the interactive properties of an object, so we never know what the object actually is, we merely know how it acts.” Although we interact with objects through their properties, it is unclear why we should think that this means we can’t learn about the objects themselves (see Johnathan Schaffer’s paper, “Quiddistic Knowledge”). Indeed, we don’t know what it even means to say that we might learn about objects (essences?) directly. That is, how else could we come to know things about objects if not through interaction with their properties? Is there some kind of spooky metaphysical core that objects are underneath their properties; i.e. there is more to being water than being H20? And even if there are core essences, what is the argument for thinking that we can’t learn about them through their properties?

    You also say, “I also don’t think we have particularly good reason to think that any scientific theory is correct for the following reason: every single scientific theory that’s existed before our current ones has been false.” First, it is unclear what exactly a scientific theory is, but perhaps it is simply a set of assumptions. If so, does a theory fail to say anything true just in case at least one of its assumptions is false? Just because a theory has some false assumptions does not entail that the theory fails to capture anything true about the world. If theories fail to capture anything true about the world, then how do we explain their ability to make successful predictions most of the time? Is it just luck?

    • Yasha & Collin,

      We actually don’t ‘know’ that water is composed of two hydrogen atoms and one oxygen atom. That is a conclusion based upon the current scientific paradigm. The evidence that led to modern atomic theory is all circumstantial and had to do with the appearance of the ratios of masses in the process of creating chemical compounds. When one looks at the early atomic theory literature, it is a little disturbing how circumstantial the case it. Similarly, science does not tell us something demonstrably true by defining gold as something with a particular atomic weight; it is simply a definition that is part of a larger paradigm that appears sufficiently consistent with known phenomena.

      The point is that there is every chance that water is not composed of any real things that currently have any remote relation to our notion of hydrogen and oxygen atoms. All alleged knowledge of these atoms is indirect and circumstantial. This is evidence, but not proof. It is a possibility, not a truth.

      This theory is not merely coincidence, but has been crafted as our best (current) explanation for our knowledge and understanding of the diverse and complex phenomena experienced in life. We have created conventions for choosing simple arguments and suggesting they are the best, and there is much utility in the theories created by science in that they have some degree of predictive accuracy. The difficulty is that, if we are true to the skepticism of science, we must admit of the possibility, if not the probability, that atomic theory will be eventually replaced a theory that better explains our deepening and unfolding understanding of phenomena.

      The difficulty I have with much of the conversations going on in the philosophy of science literature is the discussion of ‘truth values’. Science is more of a process that seeks to account for and explain, and makes no claims to ultimate truth. Things begin to feel like ‘truths’ when the theories have been around for quite a while, like atomic theory. But to be Cartesian we must remain ever open to the possibility that the current dominant theory is wildly wrong. In the newer sciences that are still in their infancy, like evolutionary theory, ecology, and psychiatry, we should expect the potential for radical paradigm shifts as the massively complex data is analyzed. In the social ‘sciences’, we should hope for radical realignments as those practicing them begin to realize they are doing nothing that even remotely resembles science (but to be fair to them- the information being dealt with does not easily lend itself to scientific analysis, but they need to recognize this).

      A hypothesis is a testable conjecture that either seeks to explain or codify the potential regularities of the observable phenomena. A theory is merely a hypothesis that has stood the test of time. Though theories do not necessarily explain the fullness of our understanding of the phenomena, they are generally kept until we have another set of hypotheses that address our growing difficulties with the previous dominant theory. There is no way to know which hypothesis in its imperfect explanation of phenomena is actually ‘closer’ to reality because reality is unknowable. Thus there is no ‘truth’ value in the dominant scientific theory.

      The difficulty in your argument is that you start with the assumption that the dominant scientific atomic theory is ‘true’ and thus you can be told that water is made up of two hydrogen and one oxygen atoms and the atomic weight of gold is whatever it is. There is only circumstantial evidence to support this theory (but- it should be noted- direct evidence is also merely evidence and not proof of anything). Science tells us about atoms, but it does not tell us if it is true that matter is made up of atoms. Atomic theory provides a reasonable explanation for the diversity of phenomena that is experienced, but there is no way to be certain of its truth. Good science requires we be willing to willing to discard atomic theory if we can produce an explanation that better accounts for phenomena.

  4. Todd: here again you would benefit from reading more into the basic philosophy of biology literature. You worry that there are no direct observations of water’s atomic compositions, only circumstantial evidence. But, not all circumstantial evidence is poor. Consider the role of inference to the best explanation in scientific confirmation. inference to the best explanation allows us to postulate unobservables. The relevant atomic hypothesis could be something like this: if atomic theory of water W is true then it makes the experimental observations probable.

    You seem to know this at some level: look at your very last sentence.

    Now, of course, there is a literature that denies that the approach I’m offering provides us any evidence of a “truth”. I suggest you read vanFrassen for that.

    Here’s my message to you: In your posts you assert propositions as true without recognizing that there are oodles of debates surrounding those propositions. I like your style–boldly assert. But, the next step towards philosophical “enlightenment” is to seek out the literature that brings your issues to the fore.

  5. Andre,

    I do not think what I am writing is merely assertion. Cartesian skepticism is such that doubt should extend to all. I am not suggesting that there is not good reason to act as if atomic theory is correct, but it is merely the best current explanation for phenomena, and it is impossible to confirm its relation to alleged reality.

    It strikes me as what you, Collin, Yasha, and others are doing is assigning some truth value to what conforms to your rules for ‘best explanation’. Semantically, that implies the existence of at least two other explanations. A change in the understanding of the details of phenomena will potentially change which you all would select as best. Do you now have a new ‘truth’?

    Personally, I find enlightenment when those who understand the arguments can demonstrate why I am wrong. The process of discussion is the search for enlightenment. But I will look up the Van F dude.

    I think you are mistaking my refusal to call atomic theory ‘truth’ for any lack of belief that it is the current best explanation. I just retain a healthy skepticism and acknowledge that there is a significant possibility that it has no bearing on reality.

  6. Todd: I urge you to do some background reading. For instance, philosophers would recognize that invoking Cartesian skepticism to make your point won’t help you at all. First, Descartes wasn’t a skeptic. Second, the skeptic to which Descartes answered believed that you know nothing at all, not a posteriori, not a priori beliefs. To that Descartes responded that the only thing we can know (certainly and indubitably) are beliefs about ourselves as thinking things, and our appearances. That’s not a good foundation to build up any scientific claims at all!

    I would be happy to discuss whether you are wrong or not but the problem is, I think you are not clearly articulating the proposition at issue. I can’t see how Cartesian skepticism helps question inference to the best explanation. Just doesn’t make much sense.

  7. Andre,

    I have now looked over some on realism and Van Fraasen. I feel I have invoked Descartes properly (potential misuse of the word skeptic aside) in the way you describe. I also see why my position is difficult to get at as it mostly accepts the realist position (with a significant rejection) and goes to Cartesian lengths in doubting evidence that goes much beyond Van Fraasen’s view that there is a difference in circumstantial evidence. I wrote about 1600 words clarifying my position, now posted on my blog at:

    While I have not reviewed even a significant portion of what is out there, my position seems sound and practical to me, and the above link tries to address the issues in reference to the realist position. The post might really annoy epistemologists, as it explicitly relies on no ideas less than two hundred years old.

    The short version and main objection is this- scientific realism strikes me as being a bit arrogant to think that the current scientific moment is significantly less likely to be overturned by something along the lines of a Copernican revolution. I practically believe and act as if atoms are true, but I have no faith atoms will be a scientifically viable theory in the future. Thus I refuse to speak of the truth of atoms, though I accept them as a scientific fact or a practical fact for now.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: