(Not that this is news to anyone, but Holiday + Sickness + Laziness = Procrastination on Intellectual Projects. But I have overcome that equation, and can finally present my first summary. It should be noted that I have revised my project somewhat and expanded my page limit to 2 pages. While I could probably trim these precis down to one, I think I'd have to sacrifice too much nuance to do so. Besides, that would be more like an abstract than a precis anyway. :-p)*****
Scientific Explanation(original length: 10 pages)
Science is a fundamentally human enterprise that has much in common with other academic disciplines. As an explanatory mechanism, science aims to account for phenomena in terms of causal structures. Science differs from other kinds of explanation in virtue of its subjects of inquiry and its methodology: It focuses on the natural world and emphasizes repeatable testing and empirical data. Science is not our only source of understanding, but rather one strand of the greater web of human knowledge. Because science is only one tool of the knowledge enterprise, it can both inform and be informed by other intellectual disciplines. It is in this role, as an important but not domineering contributor to the sum of human knowledge, that science finds its proper place.
The primary virtue of causal models of explanation is that they can account for explanatory asymmetry. Explanations have directions, and explanatory direction mirrors causal direction. For example, the height of a flagpole causes (and explains) the length of its shadow, but the length of the shadow does not cause (or explain) the height of the pole. This asymmetry has proved incredibly difficult to account for with other (non-causal) models of explanation.
One of the most apparent problems with causal models is the question of relevance—that is, of knowing which causal influences are relevant to the explanation at hand. (Are the gravitational forces exerted on me by distant stars relevant to how hungry I am?) On a “difference-making” model, potential
explanantia are relevant if and only if they made a difference to whether or not the
explanandum occurred. (If the gravitational forces of distant stars had not acted on me, I would still have been hungry, so those forces did not make a difference; therefore they were not causally relevant.) This conception has the benefit of matching our intuitive understanding of what causation means.
A more challenging problem for causal conceptions is the difficulty of accounting for the ground of causal laws, which provide the structures that connect
explanantia to their
explananda. Do such laws describe the nature of reality, or do they merely report observed regularities and repetitions? Do all regularities result from laws? If not, how do we distinguish laws from mere regularities?
Three elements can help distinguish laws from regularities. First, laws are universal in scope; they are not about particular circumstances or entities, but rather about every instance of some kind of circumstance or entity. Second, laws express necessary relations. Finally, laws support counterfactual truths by distinguishing between contingent and necessary states of affairs. For example, it is true that there are no solid spheres of pure iron weighing more than 100,000 kilograms; but there
could be such spheres, so this is a contingent truth. It is also true that there are no solid spheres of pure plutonium weighing more than 100,000 kilograms, but this is necessarily true; the nature of plutonium makes it physically impossible that there should be such spheres, so the lack of such spheres is a result of a natural law.
The nature of causal laws remains a problem for causal models of explanation (and for the philosophy of causation as a whole), but it is also a problem for most other models. Because the causal model can account for explanatory asymmetry, it remains the best model of how explanation works.
As mentioned above, science deals with natural, observable entities, seeking to provide an increased understanding of the natural world. In actual practice, the “Scientific Method”—with its idealized step-by-step process—is neither exclusive to science nor rigorously employed by scientists. The self-policing system of repeated, peer-reviewed experimentation has analogues in most other academic disciplines. And while the general elements of the Method—observation, hypothesis, experimental testing, evaluation, and revision—are indeed typical in scientific practice, the order in which they are applied varies from case to case.
Though science relies on observations, we cannot make observations independently of ideological presuppositions. Our current theories and beliefs inform the way we perceive new things. This is particularly noticeable when it comes to observational instruments, for even a tool as simple as a magnifying glass is built and used with certain theories about the world in mind.
Presuppositions play an even greater role in interpretation than in observation. Consider two scientists who hold competing hypotheses (X and Y) to explain some phenomenon. Given a set of data that seems to confirm X at the expense of the Y, the scientist who supports Y might well interpret the data as anomalous or inconclusive. Indeed, his presuppositions may not have led him to make the observations in question in the first place. In contrast, the scientist who supports X will accept the data as they are and claim increased support for X.
This is the problem of underdetermination—multiple hypotheses can account for any given set of experimental data. In such cases, science alone (governed as it is by observation and testing) cannot make a value judgment between competing proposals. We must look outside of science—using extra-scientific criteria, such as simplicity or predictive power—to fully understand such phenomena, and different presuppositions lead to different conclusions from the same data. But such presuppositions are themselves nonscientific. Indeed, the problem arises even before science begins, for we have to decide what to observe, and such a decision simply cannot be made independently of background beliefs. Science is therefore ineluctably linked to the broader realm of human knowledge (notably philosophy).
This connection has spurred controversy over the proper practice of science, particularly as it relates to theology. Numerous scientists, judges, and philosophers insist that God cannot be included as an element of scientific explanation, but many of the same people would use science to prove that God does not exist. We cannot have it both ways: If God cannot be a factor in scientific explanations, then neither can science disprove the existence of God.
Because science is an inescapably
human endeavor, it is no more infallible than any other academic discipline. Indeed, the history of science shows that most ideas that were once held to be true were eventually shown to be false, or at least incomplete (e.g. phlogiston, ether, and the “plum-pudding” model of the atom). Science is messy, emotional, and susceptible to political and ideological dogmatism (as are all human endeavors).
Furthermore, because science deals with natural subjects, there are subjects it cannot deal with directly (including God, art, and history, among others). But God and other agents can influence the natural world, thereby influencing the outcomes of scientific inquiries. Science cannot operate in an intellectual vacuum (independent of philosophical foundations); it overlaps with other academic fields to varying degrees. When different fields seem to contradict each other, we must evaluate the results of each field on its own terms. We don’t use scientific testing to evaluate the validity of historical conclusions any more than we would use historical methods to determine whether scientific results were valid. When conflicting claims still seem equally viable, we can use inference to the best explanation (abduction) to help us decide which claim is most likely to be true. However, since such quests for knowledge are human endeavors, it is always possible that we simply do not have enough information (or intellectual capability) to understand certain topics.
*****
Discuss.
Labels: Philosophy/Thinking, School
