Why Trust Science?

by Naomi Oreskes

  14. July 9, 1954, Ernest Pollard to Thomas E. Murray, Commissioner to the Atomic Energy Commission, copy to Smyth, in Henry DeWolf Smyth Papers, American Philosophical Society, Philadelphia.

  15. Galison, “Removing Knowledge.”

  16. Wang, American Science in an Age of Anxiety; Wang, “Physics, Emotion, and the Scientific Self.”

  17. See Steinberg below.

  18. Wang, American Science in an Age of Anxiety; Moore, Disrupting Science; Bridger, Scientists at War.

  19. Freire, “Science and Exile.”

  20. Smyth kept a file of public responses to his work and ideas. In Papers of Henry deWolf Smyth, American Philosophical Society, Philadelphia.

  21. See particularly Steinberg’s letter summarizing the accusations and rumors, December 11, 1953, Steinberg to Dr. Sydney Farber at the Children’s Cancer Research Foundation, where he had been considered for a position, in Papers of Arthur Steinberg, American Philosophical Society. Steinberg was not offered the position because the dean had heard rumors of his Communist sympathies, though none of these rumors, he said, were correct.

  22. See documents in Papers of Arthur Steinberg, American Philosophical Society Archives, Philadelphia. Scientists had in the course of the twentieth century even been imprisoned, like the geneticist Richard Goldschmidt in World War I, suspected of German sympathies, and like the cytogeneticist Masuo Kodani who did rubber research at the Japanese Internment Camp at Manzanar in World War II (Richmond, “A Scientist during Wartime”; Smocovitis, “Genetics behind Barbed Wire”).

  23. For a discussion of Luria’s perspectives on the tensions of Cold War science, see Selya, Salvador Luria’s Unfinished Experiment.

  24. Probstein, “Reconversion and Non-Military Research Opportunities,” 52.

  25. Gusterson, Testing Times; Aaserud, “Sputnik and the ‘Princeton Three’ ”; Cloud, “Imaging the World in a Barrel.”

  26. See forthcoming research by my PhD student Kathryn Dorsch.

  27. Engber, “The Grandfather of Alt-Science.”

  28. Haraway, “Situated Knowledges,” 598.

  29. Haraway, “Situated Knowledges,” 579.

  30. Shapin, “What Else Is New?”.

  31. Edgerton, The Shock of the Old Technology and Global History since 1900.

  32. For a compelling study of freezing things in science, one that has nothing to do with frozen peas, see Radin, Life on Ice.

  Chapter 4. What Would Reasons for Trusting Science Be?

  1. Hume, A Treatise of Human Nature, bk. 1, pt. 3, sec. 6; Hume, An Enquiry Concerning Human Understanding, secs. 4–5.

  2. Descartes, Meditations on First Philosophy.

  3. Sextus Empiricus, Against the Logicians, p. 179.

  4. Sellars, “Empiricism and the Philosophy of Mind,” sec. 38.

  5. Lange, “Hume and the Problem of Induction.”

  6. Sellars, “Some Reflections on Language Games,” p. 355; cf. Lange, “Would Direct Realism Resolve the Classical Problem of Induction?”

  7. My thanks to an anonymous referee for kindly encouraging me to make this point more explicitly.

  8. Kuhn, The Structure of Scientific Revolutions.

  9. Galilei, Two New Sciences, p. 167.

  10. Meli, “The Axiomatic Tradition in Seventeenth-Century Mechanics.” In October 1643, Marin Mersenne mentioned these rival proposals in a letter to Theodore Deschamps. In his reply, Deschamps gave the same argument to show that neither Fabri’s nor Le Cazre’s proposal was correct. See Palmerino, “Infinite Degrees of Speed: Marin Mersenne and the Debate over Galileo’s Law of Free Fall,” pp. 295–96. I have followed Palmerino’s elegant method of displaying the argument.

  11. For more details on dimensional homogeneity, see Lange, Because without Cause, ch. 6 and the references given there.

  12. Galileo’s odd-number rule is not the only proposal regarding falling bodies that achieves dimensional homogeneity. Whereas the odd-number rule says that in the nth interval, the body covers 2n-1 times the distance covered in the first interval, consider the rule that in the nth interval, the body covers 3n2–3n + 1 times the distance covered in the first interval. On this proposal, the distances traversed in successive intervals are 1s, 7s, 19s, 37s, 61s, 91s,.… Like Galileo’s rule, this proposal is dimensionally homogeneous. For example, in time intervals that are twice as long, the distances covered on this proposal are 8s (= 1s + 7s), 56s (= 19s + 37s), 152s (= 61s + 91s) …—and the ratio of 8 to 56 to 152 is the ratio of 1 to 7 to 19. So although Galileo’s dimensional argument rules out some rivals to the odd-number rule, it does not suffice to rule out every rival.

  13. Newport, “In U.S., 46% Hold Creationist View of Human Origins.”

  14. Horowitz, “Paul Broun: Evolution, Big Bang ‘Lies Straight From the Pit of Hell.’ ”

  15. Hoffman, “Climate Science as Culture War.”

  16. Goodman, Fact, Fiction, and Forecast, pp. 59–83.

  17. Laudan, “The Demise of the Demarcation Problem.”

  18. Laudan, “A Confutation of Convergent Realism.”

  Chapter 5. Pascal’s Wager Reframed: Toward Trustworthy Climate Policy Assessments for Risk Societies

  1. See https://www.epa.gov/sites/production/files/2017–10/documents/ria_proposed-cpp-repeal_2017–10.pdf, accessed November 30, 2017.

  2. Beck, Risk Society, 21.

  3. Kowarsch et al., “A Road Map for Global Environmental Assessments,” showed that many decision-makers and stakeholders—having set ambitious mitigation goals in the Paris Agreement in 2015—want future assessments by the Intergovernmental Panel on Climate Change (IPCC) to focus more explicitly on the assessment of solution options, in particular policies, rather than problem analysis.

  4. Sarewitz, “How Science Makes Environmental Controversies Worse.”

  5. Some mistakenly believe that the social sciences and humanities cannot, or should not (for moral reasons), provide objective assessments of policy pathways and their practical implications. This is argued for different reasons, including: (1) the theoretical assumption that it is impossible to deliberately steer social processes resulting in some form of general policy skepticism, (2) ethical relativism and radical constructivism, implying that any policy assessment is based on highly questionable, purely “subjective“ value judgments, (3) the exclusive focus on politics and power structures in much of the science and technology studies in recent years, etc. We do not regard these as compelling reasons against engaging in scientific policy assessment to facilitate policy learning processes.

  6. This wager was conceptualized by the French philosopher, mathematician, and physicist Blaise Pascal (1623–62) in a theological context.

  7. IPCC, Climate Change 2014.

  8. E.g., Koch et al., “Politics Matters.”

  9. Kowarsch et al., “A Road Map for Global Environmental Assessments.”

  10. Edenhofer and Kowarsch, “Cartography of Pathways.”

  11. Kowarsch, A Pragmatist Orientation for the Social Sciences in Climate Policy, ch. 5.

  12. Kowarsch, A Pragmatist Orientation for the Social Sciences in Climate Policy, ch. 6.

  13. Oreskes mentions the vivid example of astronomy promoted by the Catholic Church to calculate the exact Easter date with greater precision.

  14. For more details see Putnam, The Collapse of the Fact/Value Dichotomy and Other Essays and Kowarsch 2016, sec. 6.2.3.

  15. Edenhofer and Kowarsch “Cartography of Pathways”; Kowarsch, A Pragmatist Orientation for the Social Sciences in Climate Policy.

  Chapter 6. Comments on the Present and Future of Science, Inspired by Naomi Oreskes

  1. Bhattacharjee, “The Mind of a Con Man.”

  2. Bem, “Feeling the Future.”

  3. Yong, “A Failed Replication.”

  4. Lehrer, “The Truth Wears Off.”

  5. Vul et al., “Puzzlingly High Correlations.”

  6. Lehrer and Vul, “Voodoo Correlations.”

  7. Zimbardo, “The Stanfo
rd Prison Experiment.”

  8. Reicher and Haslam, “Rethinking the Psychology of Tyranny.”

  9. Festinger, A Theory of Cognitive Dissonance.

  10. Lord, Ross, and Lepper, “Biased Assimilation and Attitude Polarization.”

  11. Miller et al., “The Attitude Polarization Phenomenon.”

  12. Carey, “Many Psychology Findings Not as Strong as Claimed.”

  13. LaCour and Green, “When Contact Changes Minds.”

  14. Carey and Belluck, “Doubts about Study of Gay Canvassers.”

  15. Mathews, “Papers in Economics ‘Not Reproducible.’ ”

  16. See, e.g., Barone, “Why Political Polls Are So Often Wrong.”

  17. Baker, “Biotech Giant Publishes Failures to Confirm High-Profile Science.”

  18. Open Science Collaboration, “Estimating the Reproducibility of Psychological Science.”

  19. Prinz, Schlange, and Asadullah, “Believe It or Not.”

  20. Walter, “Call to Arms on Data Integrity.”

  21. Koricheva and Gurevitch, “Uses and Misuses of Meta-analysis in Plant Ecology”; Jennions and Møller, “Relationships Fade with Time.”

  22. Freedman, Cockburn, and Simcoe, “The Economics of Reproducibility in Preclinical Research”; Freedman, “Lies, Damned Lies, and Medical Science.”

  23. Ioannidis, “Why Most Published Research Findings Are False.”

  24. John, Loewenstein, and Prelec, “Measuring the Prevalence of Questionable Research Practice.”

  Chapter 7. Reply

  1. Oreskes, Science on a Mission: American Oceanography from the Cold War to Climate Change.

  2. On the issue of retaining control of the intellectual agenda, see Forman, “Behind Quantum Electronics.”

  3. Bloor, The Enigma of the Aerofoil, gives several examples.

  4. “Perceptions of Science in America.”

  5. Wazeck, Einstein’s Opponents.

  6. Oppenheimer et al., Discerning Experts; and Wolfe and Sharp, “Anti-Vaccinationists Past and Present.”

  7. Cook, Ellerton, and Kinkead, “Deconstructing Climate Misinformation to Identify Reasoning Errors”; Cook, Lewandowsky, and Ecker, “Neutralizing Misinformation through Inoculation”; Linden et al., “Inoculating against Misinformation”; Linden et al., “Inoculating the Public against Misinformation about Climate Change.”

  8. Layton, “Mirror-Image Twins.”

  9. Oreskes and Conway, Merchants of Doubt.

  10. Oppenheimer et al., Discerning Experts.

  11. Proctor, Value-Free Science?

  12. Pope Francis, Encyclical on Climate Change and Inequality.

  13. On the replication crisis in psychology see: Yong, “Psychology’s Replication Crisis Is Running out of Excuses”; Bishop, “What Is the Reproducibility Crisis in Science and What Can We Do about It?”; “Oxford Reproducibility Lectures.”

  14. On the retraction crisis see Brainard and You, “What a Massive Database of Retracted Papers Reveals about Science Publishing’s ‘Death Penalty.’ ”

  15. Gonzales and Cunningham, “The Promise of Pre-Registration in Psychological Research”; Nosek and Lindsay, “Preregistration Becoming the Norm in Psychological Science.”

  16. https://www.nature.com/articles/d41586-019-00857-9.

  17. One particularly misguided response, in my view, was the 2018 proposal by a group of psychologists and other scientists to solve the replication problem by raising the threshold for statistical significance from 0.05 to 0.005: https://psyarxiv.com/mky9j/?_ga=2.29887741.370827084.1500902659-399963933.1500902659. One can only hope that these scientists read the 2019 Nature article.

  18. Lewandowsky et al., “Seepage and Influence: An Evidence-Resistant Minority Can Affect Scientific Belief Formation and Public Opinion”; Lewandowsky et al., “The ‘Pause’ in Global Warming in Historical Context”; Lewandowsky et al., “Seepage”; Lewandowsky, Risbey, and Oreskes, “The ‘Pause’ in Global Warming”; Lewandowsky, Risbey, and Oreskes, “On the Definition and Identifiability of the Alleged ‘Hiatus’ in Global Warming”; Risbey et al., “A Fluctuation in Surface Temperature in Historical Context”; Risbey et al., “Well-Estimated Global Surface Warming in Climate Projections Selected for ENSO Phase.”

  19. Kennedy, “Why Did Earth’s Surface Temperature Stop Rising in the Past Decade?”

  20. Risbey et al., “A Fluctuation in Surface Temperature in Historical Context”; Mooney, “Ted Cruz Keeps Saying That Satellites Don’t Show Global Warming. Here’s the Problem”; Richardson, “Climate Change Whistleblower Alleges NOAA Manipulated Data to Hide Global Warming ‘Pause’ ”; Taylor, “Global Warming Pause Extends Underwhelming Warming.”

  21. One study by Daniele Fanelli, “How Many Scientists Fabricate and Falsify Research?” found that out of more than eleven thousand scientists 2% (1.97, to be exact) had committed research fraud at least once in their career. I do not know how this number compares to doctors, lawyers, accountants, or investment advisors, but if it is correct, then 98% of scientists are thoroughly honest, which strikes me as a rather good figure. On the other hand, the same study reported that, “up to 33.7% admitted other questionable research practices.” Clearly this invites further scrutiny of what constitutes questionable practices.

  22. http://www.bbcprisonstudy.org/. It would be interesting to investigate the background to the BBC’s sponsorship of this research.

  23. One attempted replication concluded that an important factor involved self-selection by participants, and that small changes in the wording of the advertisement could affect outcomes. https://journals.sagepub.com/doi/abs/10.1177/0146167206292689?casa_token=6YVE-o6G9BsAAAAA%3AwT8rDXdHa6jJp7vrqXo2bnPFOiCM5w7FFgrF26XsBlrJ7uJicqAlf3w3d3SLLxPWaeuyn-QMViuC. This also reminds us that psychology may be more vulnerable to the problem of non-replication of results, than, say physics, chemistry, or geology, because of how variable and suggestive human beings are. It would not be unreasonable to conclude, for example, that the Stanford experiment told us something interesting about how some people responded at a particular time to a particular set of circumstances, while acknowledging that small changes in those circumstances might yield different results.

  24. Phillips, “The Female Mathematician Who Changed the Course of Physics—but Couldn’t Get a Job.”

  25. Alberts et al., “Self-Correction in Science at Work.”

  26. This is an important reason why the tobacco industry long insisted that the science was not settled: if that had been true, then it might have been reasonable for the government to hold back on regulating tobacco. See Brandt, “Inventing Conflicts of Interest.” On the other hand, if there was already significant evidence that tobacco was likely harmful, then it might have well made sense for the government to begin to act to protect public health even in the absence of complete scientific agreement.

  27. Hill, “The Environment and Disease: Association or Causation?”

  28. For a more detailed exposition of this argument, drawing on practices in the oil and gas industry, see Oreskes, “Reconciling Representation with Reality.”

  29. Frederickson and Losada, “ ‘Positive Affect and the Complex Dynamics of Human Flourishing’: Correction to Fredrickson and Losada (2005).”

  30. Brown, Sokal, and Friedman, “The Complex Dynamics of Wishful Thinking”; Brown, Sokal, and Friedman, “The Persistence of Wishful Thinking.”

  31. For a full discussion, see Friedman and Brown, “Implications of Debunking the ‘Critical Positivity Ratio’ for Humanistic Psychology.”

  32. Steen, Casadevall, and Fang, “Why Has the Number of Scientific Retractions Increased?”

  33. Fang, Steen, and Casadevall, “Misconduct Accounts for the Majority of Retracted Scientific Publications.”

  34. I am grateful to my colleague Alex Csiszar, who has studied the history of scientific publication, for his perspectives on this.

  35. Steen, Casadevall, and Fang, “Why Has the Number of Scientific Retractions Increased?”

  36. “Ret
raction Watch.”

  37. Retractions of journalistic announcements of findings in paleontology are fairly common, but they are generally cases of outright fraud (fake fossils) that are revealed by scientists after non-science media have reported unpublished reports. E.g., Pickrell, “How Fake Fossils Pervert Paleontology.” One of the most famous hoaxes in the history of science occurred in paleontology—the Piltdown Man—reinforcing the idea that fraud and hoaxes may be more common in arenas that attract extensive public interest.

  38. Siegel et al., “Methane Concentrations in Water Wells Unrelated to Proximity to Existing Oil and Gas Wells in Northeastern Pennsylvania.”

  39. Oreskes et al., “Viewpoint”; Tollefson, “Earth Science Wrestles with Conflict-of-Interest Policies.”

  40. Darrah et al., “The Evolution of Devonian Hydrocarbon Gases in Shallow Aquifers of the Northern Appalachian Basin.”

  41. I might be too generous here. There is lasting damage if industry-sponsored work keeps open a debate that would otherwise have appropriately closed. Of course, this is extremely difficult to judge, since there can be no “controlled experiment” in which the same data were collected and vetted absent industry involvement.

  42. Myers et al., “Why Public Health Agencies Cannot Depend on Good Laboratory Practices as a Criterion for Selecting Data”; Saal et al., “Flawed Experimental Design Reveals the Need for Guidelines Requiring Appropriate Positive Controls in Endocrine Disruption Research.”

  43. A controversial case is the paper by Gilles Seralini and colleagues of the effects of genetically modified maize and the Roundup herbicide on rats. The paper was retracted on the grounds that the findings were “inconclusive.” Many scientists objected that this was inappropriate grounds for retraction; the paper was republished in a different journal. See Oransky, “Retracted Seralini GMO-Rat Study Republished.” It was later revealed that the retraction had been orchestrated, or at least heavily influenced, by Monsanto, the manufacturer of Roundup. See McHenry, “The Monsanto Papers.”