Note: I wrote this for a class, so it deals with a specific text in the introduction, but the ideas should be applicable even if you haven't read Feyerabend. Also, note that when I say science, I mean what we all know of and generally think of as Science (with a capital S). Not the general method of assessing evidence to reach conclusions, but Science as is done by Western physicists, chemists, etc.
At the start of his essay, How to Defend Society Against Science, Feyerabend introduces a general worry about science’s privilege as a method for reaching truth and forming correct beliefs about the world. The worry seems two have two parts: first, that science has become an ideology whose standards of truth rely more on a dogmatic acceptance of the scientific method’s accuracy than any real notion of truth; and second, that even if science has found the real truth, we do not have to accept the truth, but rather have a choice between pursuing truth versus other important values (like freedom). Initially, this view seems absurd — of course science gets us the truth, that’s why we have airplanes, computers, and a cure for polio. Why shouldn’t we privilege the real truth that science discovers over truths with less logical and empirical support? Ultimately, I feel that Feyerabend does not flesh out his general worry enough to combat these objections, and so I aim to carry on in his spirit and deny any inherent privilege science should have in forming truthful beliefs. Specifically, the view I wish to argue against can be roughly stated as follows: Science gives us the most accurate conception of the world, and so if we want to form the correct beliefs about the world, we ought to privilege science’s method over other systems. I will argue, first, that science’s empirical method does not necessarily give us the most accurate understanding of the world. Second, I will maintain that even when science’s method does seem to give us the most accurate understanding of the world based on its evidential support, it may be desirable to weigh different sorts of evidence in our analysis and conclusions. Importantly, my overarching claim is that science deserves no inherent privilege in matters of reaching truth, and should only be privileged when doing so is useful. I am not arguing for the complete rejection of science, nor will I be able to give a complete catalogue of when it is useful to privilege science in assessing truth and when it is not.
The first of my arguments, against science’s ability to reach an accurate understanding of the world, is roughly borrowed from Schopenhauer. The idea is that science necessarily presupposes certain facts about the universe, without being able to explain why those assumptions are necessarily true. For example, imagine asking a scientist why there are four fundamental forces. They might respond by explaining that the nature of the elementary particles in the universe necessitates the four forces that we have discovered. But, we can continue by asking why the elementary particles have the nature that they do. We may be met by further explanations, but if we continue to question the scientist in this way, they will eventually respond, “that’s just how it works.” As we keep asking “why,” the scientific explanation must either eventually end by referencing a fundamental assumption for which no further “why” can be asked, or must infinitely regress such that we can keep asking “why” forever. If the explanation can go on infinitely, then it seems that science hasn’t explained anything; we never reach a satisfying answer to our question of “why.” If, on the other hand, the explanation stops at some fundamental assumption whose validity is assumed rather than explained, then it seems that science may not present a completely accurate view of reality. Since science cannot explain why the universe is fundamentally governed by this set of forces, or why these forces necessarily operate in the way they do, we cannot be sure whether an assumption of these forces truly produces an accurate understanding of the world, or whether they are simply the best estimates that we can come up with using current methods. Maybe there are certain forces that science does not yet know how to detect; maybe these fundamental forces merely appear to be separate based on our collected data, but are actually one unified force; until science provides a sufficient explanation for the necessity of the operation of the fundamental forces, we cannot be certain that they accurately depict the nature of the world. Even more, this will be true no matter how much science’s assumptions are refined to account for newly observed phenomena. All of science’s assumptions are true only to the extent that they are able to explain observed data without making false predictions, and so we will never know whether those assumptions are actually correct, or whether there will be new data and predictions that call for a refinement of the current assumptions.
It remains possible, however, that even though we cannot be certain of the accuracy of science’s assumptions without sufficient explanation, those assumptions may still be, in fact, accurate. Perhaps, some day in the future, the assumptions made by science will be so complete that they are able to account for all experimentally reproducible phenomena in the universe, with no false predictions; even though we may not be able to explain why those assumptions are necessary, can we not be certain that those assumptions are correct? I now aim to argue against this objection, and claim that science only allows certain types of evidence in its analysis, and that it may be desirable to allow different evidential criteria in our collection and analysis of data. Vine Deloria, for example, offers tribal systems of thought an illustration of how new knowledge can be grasped through different methods of interpreting data. As Deloria sees things, tribal systems of thought are as “systemic and philosophical” as science; they simply allow different kinds of evidence and do not share in science’s goal of “determining the mechanical functioning of things.” The Sioux system, for example, denies science’s view of knowledge as absolute, separate from humans, and waiting to be discovered. Where science rejects any evidence that is not experimentally replicable to an objective observer, the Sioux do not disregard any experience, and derive their conclusions from “individual and communal experiences in daily life”, “emotional experiences”, “keen observation of the environment”, and “interpretive messages…received from spirits in ceremonies, visions, and dreams.” The system fundamentally asserts that all experience is valid, and its willingness to accept and interpret a wider range of experiences does not make it less systematic or true than science; the Sioux simply wish to interpret and account for a greater variety of experience than is reproducible in experimental settings. Instead of searching for abstract principles to understand and explain the structure of the world, Sioux knowledge is directed at discovering the best way for people to lead an ethical life. All of our experiences have content and validity to the extent that they contribute to the moral framework of the universe, and so we cannot disregard anything in our search for moral understanding. As a result of this motivation and epistemology, all events and beings are believed to be related in the moral community. Everything has a responsibility to fulfil itself and participate in the creation of moral and experiential content — “nothing has incidental meaning and there are no coincidences”. Entities are, in fact, viewed as communities themselves; everything has a personality that can be used to guide one’s moral understanding of the universe. Ultimately, the evidence allowed by science may be more useful for an understanding of the mechanical function of the universe, while the evidence allowed by the Sioux may be more useful in understanding its moral content, and it may be circumstantially desirable to accept the latter evidential criteria over the former. Of course, Deloria provides only one example of alternative evidential criteria, but I believe that the example illustrates that there may be circumstances where it is desirable to reject the evidential criteria accepted by science.
One final worry is that, upon accepting new standards of evidence, we are not changing our standards of truth but rather disregarding truth in place of something more useful (like a better moral understanding). Science tells us what’s really happening, but we would rather believe something false in order to further our other goals. I feel that this objection misses my (and Deloria’s) point: why prefer science’s conclusions as what’s really happening? Why take science’s evidential criteria as the most sure way toward truth? Even if science may be able to most comprehensively explain objectively replicable phenomena and the mechanical structure of the universe, aren’t there other phenomena and structures that it hasn’t considered? Does it give a satisfying account of the most important phenomena? My point, much like Feyerabend’s, is that we have a choice: science may lead us to one truth, and another system may confirm another truth, and we can decide which truth seems more circumstantially appropriate. Either way, we accept the conclusion of one system as true, and reject the other system as lacking in evidence; we change our evidential criteria for assessing a belief to be true, in such a way that science’s evidence may no longer be convincing. Ultimately, I grant that I have not provided complete specificity as to when we ought to privilege other systems over science, which systems we ought to privilege, and why. I believe that I have shown, however, that we do have a choice in how to assess truth and reach and understanding about the world. While science’s methods have achieved the accuracy to allow for airplanes and computers, there are circumstances in which we may question the relevance of science’s evidential criteria, and prefer the method of another system; when collecting data and assessing truth, science deserves only circumstantial — rather than inherent — privilege over other systems.