The Scientific Attitude

by Lee McIntyre

  Conclusion

  In previous accounts, some philosophers of science have felt that the best way to defend science is to come up with a logical justification of its method, rather than look at how science is actually done. When this resulted in a criterion of demarcation that purported to sort all and only science to one side of the ledger—and all and only that which was not science to the other—trouble ensued. For this reason, it seems a virtue of the scientific attitude that it is flexible enough to capture why scientific explanation is distinctive, yet robust enough to ensure that even without a decision procedure, we can still tell whether inquiry is scientific. Having a scientific “attitude” toward evidence may seem soft, yet it captures the essence of what it means to be scientific. Whether this is an observation about the context of discovery (how science actually works) or the context of justification (a rational reconstruction after the fact) matters little. For I think that the greatest threat to science’s credibility comes not from some philosophical distinction between the way that scientists do their work and the method we use to justify it, but instead from the improper introduction of ideological commitments into the scientific process. And the scientific attitude is a bulwark against precisely this sort of ideological infection.39

  The idea behind the scientific attitude is simple to formulate but difficult to measure. It nonetheless plays a crucial role both in explaining how science operates and in justifying the uniqueness of science as a way of knowing. Science is successful precisely because it embraces an honest and critical attitude toward evidence (and has created a set of practices like peer review, publication, and reproducibility to institutionalize this attitude).40 Of course, science is not always successful. One can have the scientific attitude and still offer a flawed theory. But the power of caring about empirical evidence is that we (and others) may critique our theory and offer a better one. When we are trying to learn about the empirical world, evidence must overrule other considerations.41 The evidence may not always be definitive, but it cannot be ignored, for the check that it gives us against reality is the best means of discovering (or at least working toward) the truth about the world.

  This highlights again the tentative nature of any scientific theory. The scientific attitude is fully consonant with the idea that we can never be sure that we have the truth: all theories are provisional. But this is as it should be, for what is distinctive about science isn’t the truth of Newtonian or Einsteinian theory, but the process by which those theories came to be warranted. Scientific theories are believable not merely because they fit with the data of our experience, but because they are built through a process that respects sensory evidence and the idea that our untutored hypotheses can be improved by a clash with the searching criticism of other scientists who have seen the same data.

  We are reminded again of the fragility of science, for it depends on the willingness of its practitioners to embrace the scientific attitude. No matter how reliable our method, science could not work without the forthright and cooperative spirit of scientists. If we want to be scientists, our commitment cannot be to any given theory (even our own) or any ideology, but must be to the scientific attitude itself. Other factors may count, but in the end they are and always should be trumped by evidence. While this may sound meager, it is the heart of what is distinctive about science.

  Notes

  1. For the classic argument, see Norwood R. Hanson, Patterns of Discovery: An Inquiry into the Conceptual Foundations of Science (Cambridge: Cambridge University Press, 1958). For an excellent overview, see Thomas Nickles, “Introductory Essay: Scientific Discovery and the Future of Philosophy of Science,” in Scientific Discovery, Logic, and Rationality, ed. T. Nickles (Dordrecht: Reidel, 1980), 1–59.

  2. The quotation above is widely attributed to Feynman, but it is only a piece of what he had to say on the subject. For more see his delightful essay “What Is Science?” Physics Teacher 7, no. 6 (1968): 313–320.

  3. Some would maintain that the problem here is much more severe—that the evidence is always ambiguous because there are infinitely many possible theories that could in principle fit the same data. See Helen Longino, “Underdetermination: A Dirty Little Secret?” STS Occasional Papers 4 (London: Department of Science and Technology Studies, University College London, 2016). For more background, see Paul Horwich, “How to Choose between Empirically Indistinguishable Theories,” Journal of Philosophy 79, no. 2, (1982): 61–77, and Larry Laudan and Jarrett Leplin, “Empirical Equivalence and Underdetermination,” Journal of Philosophy 88, no. 9 (1991): 449–472. For a general discussion of this problem, see Lee McIntyre, “Taking Underdetermination Seriously,” SATS: Nordic Journal of Philosophy 4, no. 1 (2003): 59–72.

  4. I will have more to say about the practices by which scientists do this in chapter 5.

  5. Though we can certainly look for some telltale markers of poor reasoning. Denialists and pseudoscientists may say that they care about evidence, but it matters a great deal how one demonstrates this. Wishful thinkers may hope that there is evidence to back up their theory, but this is not sufficient. Those who engage in hasty generalizations are likewise disrespecting what it means to care about evidence. To give lip service to empirical support—for instance by cherry picking only the data that fit one’s theory—is not to truly care about evidence. One demonstrates care for scientific evidence by working hard to see not just whether there is evidence that fits a theory, but also whether there is anything that is capable of refuting it. For more on this issue, see the discussion of denialism and pseudoscience in chapter 8.

  6. And occasionally a whole scientific field can get off track. What does one do when the individual is right and the scientific community is wrong? This issue will be discussed in chapter 8 with the example of Harlan Bretz.

  7. Peter Achinstein, ed., Scientific Evidence: Philosophical Theories and Applications (Baltimore: Johns Hopkins University Press, 2005), 1. Some of the different concepts of evidence that Achinstein discusses include (1) falsificationist, (2) inductivist, (3) explanationist, (4) Bayseian, and (5) “anarchist” accounts.

  8. See in particular Deborah Mayo, Error and the Growth of Experimental Knowledge (Chicago: University of Chicago Press, 1996). Mayo’s “error-statistical” model offers a stirring challenge to the dominant Bayesian approach. For other perspectives, see Peter Achinstein, The Book of Evidence (Oxford: Oxford University Press, 2003), and Clark Glymour, Theory and Evidence (Princeton: Princeton University Press, 1980).

  9. Roughly speaking, adherents of subjectivist approaches feel that it is reasonable to rely on our background knowledge of prior probabilities as a starting point for assessing the likelihood of a hypothesis, with modification over time based on experience. Frequentists dispute this and argue that it is folly to attach a probability estimate to a hypothesis before it has been tested against experience. Mayo argues that scientific knowledge arises only from “severe tests” of a theory.

  10. I will have much more to say about this issue in chapter 4.

  11. One such example occurred recently when astronomers argued that the BICEP2 telescope at the South Pole had discovered “direct evidence that in the first fraction of a second after the big bang, the universe underwent a bizarre exponential growth spurt.” It was later shown that the finding was an artifact of microwave radiation from background dust. At first, the researchers were reluctant to abandon their conclusion, but later came around when the evidence became overwhelming. Adrian Cho, “Curtain Falls on Controversial Big Bang Result,” Science, Jan. 30, 2015, http://www.sciencemag.org/news/2015/01/curtain-falls-controversial-big-bang-result.

  12. I will have much more to say about the critical role of community judgment in the evaluation of scientific theories in chapter 5.

  13. Note that these brief examples are here included to illustrate what I mean by the scientific attitude. Much more detail will be provided in chapter 6 (where I will discuss how the scientific attitude transformed modern medicine) a
nd in chapter 5 (where I will discuss how cold fusion undermined the scientific attitude by refusing to facilitate group scrutiny of individual work).

  14. Carl Hempel, Philosophy of Natural Science (New York: Prentice Hall, 1966), 3–8.

  15. See Noretta Koertge’s fascinating essay, “Belief Buddies versus Critical Communities,” in Philosophy of Pseudoscience, ed. M. Pigliucci and M. Boudry (Chicago: University of Chicago Press, 2013), 165–180, where she argues that the critical appraisal of one’s ideas by one’s scientific peers is valuable in improving them and bringing them to the attention of the wider community.

  16. Roy Porter, The Greatest Benefit to Mankind: A Medical History of Humanity (New York: Norton, 1999), 369.

  17. Porter, Greatest Benefit, 369.

  18. Porter, Greatest Benefit, 369–370; W. F. Bynum et al., The Western Medical Tradition 1800–2000 (Cambridge: Cambridge University Press, 2006), 156; Hempel, Philosophy of Natural Science, 3–8.

  19. Another example of a lone scientist who was vindicated after years of resistance is Galileo. For a more modern example—which bears on the question of what happens to the scientific attitude when the individual is right but the group is wrong—see my discussion in chapter 8 of Harlen Bretz and his theory of megafloods in the Eastern Washington scablands.

  20. Koertge, “Belief Buddies,” raises the intriguing point, though, that even the lone genius’s ideas could be improved by membership in a critical community.

  21. For further discussion of cold fusion within the context of peer review, see chapter 5. See also Lee McIntyre, Dark Ages: The Case for a Science of Human Behavior (Cambridge, MA: MIT Press, 2006), 19–20.

  22. See also Robert Merton’s important essay “The Normative Structure of Science” (1942) (reprinted as chapter 13 in The Sociology of Science, ed. Robert Merton [Chicago: University of Chicago Press, 1973]), which has some enormously insightful things to say about the role and importance of values in scientific inquiry.

  23. This idea may be more important than it sounds. Is it possible that Popper missed the point of his own criterion of demarcation? Perhaps what matters most is not whether a theory is falsifiable, but whether the scientists who put it forward seek to falsify it. For an intriguing take on this idea, see Janet Stemwedel’s essay “Drawing the Line between Science and Pseudo-Science,” where she writes, “The big difference Popper identifies between science and pseudo-science is a difference in attitude. While a pseudo-science is set up to look for evidence that supports its claims … a science is set up to challenge its claims and look for evidence that might prove it false. In other words, pseudo-science seeks confirmations and science seeks falsifications.” Scientific American, Oct. 4. 2011, https://blogs.scientificamerican.com/doing-good-science/drawing-the-line-between-science-and-pseudo-science/.

  24. Karl Popper, “Remarks on the Problems of Demarcation and of Rationality,” in Problems in the Philosophy of Science, ed. Imre Lakatos and Alan Musgrave (Amsterdam: North-Holland, 1968), 94.

  25. The Philosophy of Karl Popper, ed. P. A. Schilpp (LaSalle: Open Court, 1974), 29.

  26. K. Brad Wray, “Kuhn’s Social Epistemology and the Sociology of Science,” in Kuhn’s Structure of Scientific Revolutions—50 Years On, ed. W. Devlin and A. Bokulich (Dordrecht: Springer, 2015), 175–176.

  27. Thomas Kuhn, The Road since Structure: Philosophical Essays, 1970–1993, with an Autobiographical Interview, ed. J. Conant and J. Haugeland (Chicago: University of Chicago Press, 2002), 101.

  28. Though some arguably did. Note again Merton’s discussion of the values of science. He lists four of them (Communalism, Universalism, Disinterestedness, and Organized Skepticism), which Sven Hansson—in his essay “Science and Pseudo Science” in the Stanford Encyclopedia of Philosophy—says have been underappreciated for their role in the demarcation debate. It is intriguing to note that Kuhn says some very laudatory things about Merton in his Structure of Scientific Revolutions.

  29. Note here Kuhn’s statement (see page 211, note 20) where he says that we must not try too hard to look for a criterion of demarcation.

  30. Popper, “Science: Conjectures and Refutations,” in Conjectures and Refutations (New York: Harper Torchbooks, 1965), 52.

  31. See http://www.earlymoderntexts.com/assets/pdfs/bacon1620.pdf.

  32. Rose-Mary Sargent, “Virtues and the Scientific Revolution,” in Scientific Values and Civic Virtues, ed. Noretta Koertge (Oxford: Oxford University Press, 2005), 78.

  33. Noretta Koertge, ed., Scientific Values and Civic Virtues (Oxford: Oxford University Press, 2005), 10.

  34. Alasdair MacIntyre, After Virtue: A Study in Moral Theory (South Bend: University of Notre Dame Press, 1981), 1.

  35. In their important paper “The Virtues of Scientific Practice: MacIntyre, Virtue Ethics, and the Historiography of Science,” Daniel Hicks and Thomas Stapleford recognize that “although history and philosophy of science is not the focus of MacIntyre’s virtue ethics, science serves as a prominent exemplar in his writing.” Isis 107, no. 3 (Sept. 2016): 4.

  36. As we will see in chapter 8, it is also possible for the scientific community to make a mistake.

  37. At the end of their paper, Hicks and Stapleford argue for precisely this sort of vision, where the virtues of science are recognized for their role in community practice, by analogy with MacIntyre’s vision for virtue ethics.

  38. Abrol Fairweather, ed., Virtue Epistemology Naturalized: Bridges between Virtue Epistemology and Philosophy of Science (Dordrecht: Springer, 2014).

  39. Is there a relationship between the content of a theory and the behavior of the people who advance it? Traditionally the problem of demarcation has focused on the former, yet perhaps there is some role for the latter. See here Martin Curd’s review of Pigliucci and Boudry’s Philosophy of Pseudoscience in Notre Dame Philosophical Reviews (July 22, 2014), where, citing Boudry’s article, Curd raises the question of whether the behavior of pseudo-scientists is relevant to the question of demarcation: http://ndpr.nd.edu/news/philosophy-of-pseudoscience-reconsidering-the-demarcation-problem/. Action, after all, is motivated by values. How one approaches evidence matters. One’s intentions matter. This seems consonant with the scientific attitude.

  40. We will explore these practices of science in detail in chapter 5.

  41. Clearly, we must also care about the logical coherence and consistency of our theory. If a theory is self-contradictory, no evidence can save it.

  4    The Scientific Attitude Need Not Solve the Problem of Demarcation

  If the scientific attitude is our best means of articulating what is special about science, the question inevitably arises: could it be the long-awaited solution to the problem of demarcation? If so, would it need to provide a set of necessary and sufficient conditions for telling the difference between science and nonscience? It is my contention that the philosophy of science need not get sidetracked into this issue, when all we really need to do is identify one simple necessary condition for good science: caring about evidence and being willing to use that evidence to revise one’s theories. If some potential area of inquiry fails to do this, then it is not science. Thus, in keeping with the theme of this book, one sees that we can learn quite a bit about science by looking at what it is not. One need not prove that anything with the scientific attitude is science; one need only show that anything without the scientific attitude is not. Whether this constitutes a full-blooded criterion of demarcation in the most robust sense of the word seems less important than that it gives us what we need to understand what is most distinctive about science.

  A Short Detour through the Vexed Question of Necessary and Sufficient Conditions

  Recall here Laudan’s admonition: without specifying a set of necessary and sufficient conditions for science, one cannot hope to solve the problem of demarcation. Yet given the repeated failure of philosophers of science to do this, he argues, the demarcation problem is dead. Of course, even if Laudan is right, this does not mean that on
e could have no fruitful discussion over whether there is something special about science. But it is also possible that Laudan is incorrect in saying that providing a set of necessary and sufficient conditions is itself necessary (or sufficient) for solving the problem of demarcation.1

  The key insight here is to realize just how high one is putting the bar to say that we have to provide a set of necessary and sufficient conditions. In saying that A is necessary for B we are saying “if B, then A.” Likewise, in saying that A is sufficient for B we are saying “if A, then B.” Any student of logic will recognize the equivalences at work here and understand that saying “A is sufficient for B” is equivalent to saying “B is necessary for A,” and likewise that “B is sufficient for A” is equivalent to saying “A is necessary for B.” But this is where the magic happens, for when you combine these two statements into one that says “A is both necessary and sufficient for B”—which of course means the same thing as “B is both necessary and sufficient for A”—you get the logical equivalence of A and B. And this is the strongest relationship one can get in logic.2

  This sets a formidable task for demarcation and the implications are staggering, for it means that in searching for the necessary and sufficient conditions for science we are searching for some criterion that is logically equivalent to science. The best illustration of the cost of this comes from Popper. Recall that Popper equivocates over whether his notion of falsifiability—as a solution to the problem of demarcation—is intended to provide a necessary condition for science, a sufficient one, or one that is both necessary and sufficient.3 Laudan takes Popper to task for his sufficiency condition, arguing that it does not do enough to maintain the high standards of science: What should we do about fields like astrology that make demonstrably false claims (which are surely therefore also falsifiable)? Must we admit them as science? Likewise, others have questioned the necessity interpretation of Popper’s criterion, arguing that it is too strict: What should we do about scientific fields that seem not to make any falsifiable claims?4 Are they not sciences? But I would now like to point out that even if Popper did offer falsification as both a necessary and sufficient criterion, he is still in hot water. For later in life, when Popper wrote “a sentence (or theory) is empirical-scientific if and only if it is falsifiable,”5 he is not solving the problems above, but merely combining them. By committing himself to the logical equivalence of science and falsifiability, he suffers criticism both from those who think his account is too strict and from those who think it is too liberal.6