And Why Does It Matter to Scientists? We’ve asked a specialist at UNIGE (first part here)
Professor Dr. Marcel Weber
Head of Biological Interest Group
Department of Philosophy
University of Geneva
Scientists, it seems, have little reason to care about philosophy. Isn’t this an outmoded way of trying to know things that only scientific research can reveal? Or even worse, isn’t it a form of pretentious literature written in an incomprehensible and pompous language that only masochists or experts in French or German literature would want to read?
While it must be admitted that this kind of philosophy is widespread, there are also other kinds that are perhaps less well-known but that deserve to be taken seriously even or especially by scientists. The philosophy of science is one such discipline. Its goal is to better understand scientific reasoning, perhaps even to improve it, and to provide explicit justification for it. The importance of this quest can hardly be exaggerated at a time when scientific rationality is under attack from both extremes of the political spectrum (e.g., anti-vaxxers from left and right, global warming skeptics from the right).
However, is it really philosophers who are in the best position, or in any position, to defend science as the most reliable way of knowing about the world? I would like to show here that they can at least make an important contribution.
Philosophers have studied reasoning in general and scientific reasoning in particular for a long time. They have developed or adopted for their purposes a variety of tools that are useful for this task, tools that include deductive logic, Bayesian confirmation theory, or the conceptual analysis of such notions as evidence, truth, explanation, reduction, model, cause and effect, probability or biological function, to name just a few. Some of these notions can mean different things in different contexts, therefore it is important to keep them apart in order to avoid confusions.
For example, “these results provide evidence for the hypothesis” could mean that the results support a hypothesis and are accurate or true, thus providing good reasons for accepting the hypothesis. However, it could also mean that the results support the hypothesis but they are not accurate or true, for example, because they are the result of an experimental artefact or a biased data-collecting procedure. Obviously, in the latter case the results provide no good reason for accepting a hypothesis, but they are nonetheless called “evidence” if they support the hypothesis. The reason for this is that sometimes scientists may not know if the results are true or accurate, but the fact that they support a hypothesis makes them relevant and they don’t want to throw them out as evidence (yet).
In any case, it is important to be clear if we mean “evidence” in a sense that implies truth or accuracy or not in a given situation. This is just to give an example that should be close to scientists’ practical concerns.
Danger of False Hypothesis
Another issue where philosophers can help to clarify things has to do with facts and values. Usually, it is assumed that science ought to conform to the ideal of value-freedom and be concerned with facts only. However, this issue turns out to be more complicated. One reason is that scientists must sometimes or even often make risky inferences. For example, they must extrapolate from a sample or from an experimental situation (e.g., a randomized controlled trial) to real-world systems or populations. This is risky, that is, there are several ways in which the inference, even if it uses the best statistical methods or whatever methods are the best in the given discipline, can go wrong.
In other words, even if scientists apply the best reasoning that there is and make no mistakes, they may end up accepting a false hypothesis, or rejecting a true one. This is inductive risk. Unlike in mathematics, this is always possible in empirical science. Reasons for such errors may include unrepresentative samples, unknown confounding factors, or sometimes just plain bad luck.
Scientists Shouldn’t Be Alone
How much inductive risk is acceptable for a given scientific investigation? And who gets to decide what is an acceptable level of inductive risk? How should we weigh such risks against other risks and benefits? Should we try to minimize false positive or false negative results (it turns out that they cannot be minimized both)? Such questions cannot be answered without an appeal to values, and they should not be left to scientists alone. In this sense, science cannot and should not be value-free. But it is important to know exactly which questions require an appeal to social values and which ones don’t. This is just one kind of problem where the philosophy of science can help both science and society at large.
If you got interested in philosophy of science and would like to see what is going on about it at UNIGE, do not hesistate to contact Marcel or visit one of their IgBIG group meetings:
This September UNIGE holds a major European Philosophy of Science Symposium!