account the possibility of different interests. That is, to assess a scientific claim, the needs and desires of the investigator are relevant. When asked if a forest fire should be allowed to burn, an ecologist might answer that the dominant trees in the ecosystem of concern are pyrophytic, requiring fire to reproduce, so “let it burn.” But another ecologist might answer that the last surviving population of the endangered yellow-spotted salamander lives in the ponds within the burning forest and the fire should be extinguished to conserve this species. And an ecologist working for a forest products company might insist that jobs would be lost if the fire is not controlled.
You can see that perspectivism admits that scientists may have differing values and, in fact, there is no value-free science (at a minimum, scientists value knowledge). If the forest fire is allowed to burn it is true that the plants will recover (a long-term perspective valuing natural processes), that the amphibian will go extinct (a perspective valuing the existence of species), and that people will be harmed (a short-term perspective valuing economic well-being). So, while logically incommensurable claims are not countenanced from a given perspective (e.g., it cannot be the case that the salamander is both endangered and flourishing at this time), incommensurable conclusions can be drawn from scientific data when perspectives diverge (e.g., forest health might be harmed in the short-term and benefited in the long-term by the same event).
1.5 Conclusion
This completes our brief survey of some central concepts in epistemology, such as the concepts of epistemic reasons, knowledge, and truth. As you can imagine, there is much more to be said on all of the topics we touched on. The suggested readings provide further and deeper treatments of many of those topics, if you are interested. Having shown that scientific knowledge most often involves some form of inductive reasoning from evidence, we next go one step deeper and turn to a fundamental question in the philosophy of science: What is the nature of evidence, and how can we access evidence?
Notes
1 1 The debate between a dynamic theory of heat and the defenders of the caloric account was an important stage in the development of thermodynamics. Early defenders of the dynamic theory included Galilei Galileo, Isaac Newton, and Robert Boyle, who thought that heat is due to the internal motion of particles. This however was difficult to reconcile with the discovery of latent heat (i.e., the heat absorbed by a substance that undergoes phase change, e.g., melting, but doesn’t rise in temperature) in 1757 by Joseph Black. For details, see Robert D. Purrington, Physics in the Nineteenth Century, Rutgers University Press, 1997, pp. 75–101.
2 2 Elliott Sober, Philosophy of Biology. Westview Press, 1993, p. 32.
3 3 A form of inference related to inductive inferences is reasoning by analogy. We’ll say more about this in Chapter 9.
4 4 A very simple Gettier case involves a clock whose hand is stuck at 5. If you don’t know the clock is broken, you are justified in believing the time it indicates (it has worked reliably so far). Moreover, if you just happen to look at that clock when in fact it is 5 o’clock, you have a true belief that is also justified. But few would say that this justified true belief counts as knowledge of the time, because your belief is only accidentally true.
Annotated Bibliography
Ronald Giere, 2006, Scientific Perspectivism. Chicago: University of Chicago Press.
Giere argues that scientific claims are conditioned on highly confirmed theory or reliable observations such that if the assumptions of the underlying theories or the biophysical properties of the instrument/observer changes, then the result can be a different, even incommensurable, accounts of the natural world. There is no complete, objective, or singularly correct account of the world through correspondence how the world really is.
Alvin I. Goldman and Matthew McGrath, 2015, Epistemology. A Contemporary Introduction. New York and Oxford: Oxford University Press. This book contains advanced discussions of the structure and nature of justification, the Gettier problem, skepticism, and contextualism. It also covers methodological questions concerning the use of intuitions as evidence as well as important new developments in social epistemology which are important for understanding how knowledge emerges from research teams, as opposed to individual scientists.
David Manley, 2019, Reason Better: An Interdisciplinary Guide to Critical Thinking (Tophat Monocle) https://app.tophat.com/e/455176/assigned This recent and very accessible book on critical thinking covers many of the concepts we encountered in this chapter, such as reasoning and inference. In addition, it contains illuminating discussions of various biases, inferential fallacies, and cognitive illusions that can interfere with successful reasoning.
2 EVIDENCE, OBSERVATION, AND MEASUREMENT
2.1 The Promises of Evidence
What’s so great about evidence? Science is often regarded as the paradigmatically rational way to gain knowledge about the natural world. In particular, it has been argued that in contrast to ideology, religion, and mysticism, science is both appropriately responsive to how the world actually is and allows for a rational resolution of disputes between different theories about the world. Both of these features of science are intimately tied to its use of evidence. Evidence, at least ideally, is publicly accessible, sharable, and can serve as an arbiter in the competition between different theories. In other words, in contrast to disputes between different religious beliefs, political worldviews or economic ideologies, which typically cannot be resolved by appealing to a neutral standard, science solves such disputes in a rational manner, because it has access to a neutral standard: evidence.
Let’s look at this in more detail. Generally speaking, evidence typically indicates (although we’ll see important exceptions) particular facts about the world. Tracks in the snow, for example, may indicate the prior presence of a mule deer, and those tracks are thereby evidence for the mule deer’s earlier presence in the area. How do those tracks provide evidence for that fact? The answer is straightforward: The deer caused the tracks that you are seeing in the snow. This means that at least this one form of evidence is grounded in causal relations between different features of the world. In our case, the causal relation was one between the deer’s hoofs and the tracks. And there are numerous similar cases: The victim’s blood on John’s shirt is evidence for the hypothesis that John was at the murder scene, because the injuries of the victim caused her blood to end up on John’s shirt. The so-called Koplik spots on a person’s gums are evidence for the hypothesis (often called a diagnosis in this context) that the person has the measles, because having the measles causes those spots to appear. In fact, there are so many cases in which we have observable indicators (i.e., evidence) for features of the world we might not directly observe (deer, measles, etc.) that one might be tempted to simply define evidence in terms of a causal indicator relation, maybe along the following lines:
An observable feature O of the world is evidence for the existence of an unobserved or unobservable feature U of the world if and only if U caused O, and O thereby indicates the existence of U.
A brief remark on the distinction between unobserved and unobservable features, invoked above, is in order. In our first example, the deer was simply unobserved, but is of course observable. But there also seem to be clear cases where we can’t observe that for which we might have observable evidence. Newtonian gravitation is a good example. We can observe its effects when we see a football drop into the outstretched arms of the receiver. However, we can’t observe the gravitational force directly. In fact, it seems we can’t observe any of the fundamental forces of nature directly. For this reason, many have argued that forces are prime examples of entities that are in principle unobservable, and that the only reason we have for believing that they exist at all is the evidence manifest in their actions. We can see neither the strong nor the weak nuclear force, but we