The Scientist Makes Value Judgements
An Essay on the Necessity of the Value-Laden Nature of Science
Introduction
This paper shall begin by assuming that scientists do not make value judgements; it shall assume that scientists make decisions based only on epistemic values, and nothing more. Then, it will be shown how this conclusion must be absurd by exploring the dichotomy between epistemic and non-epistemic values, emphasizing that the former relies on a theory of truth that the scientist must assume to be true to engage in scientific inquiry. Then it shall be argued this presupposition of a theory of truth must be a non-epistemic value judgement, for otherwise such an accepted definition of ‘epistemic value’ becomes circular. Finally, this paper shall anticipate the argument that such a value judgement is not done qua scientist, but instead qua philosopher. A rebuttal to this argument shall be given by citing a historical scenario where scientists did actively engage in what a theory of truth should look like, and by showing that even if not actively discussed, the scientist still passively makes a value decision by holding certain epistemic values.
The Dichotomy between Epistemic and Non-Epistemic Values
Before exploring the relationship between epistemic values and theories of truth, one must first understand what an epistemic value is, and how it is distinguished from a non-epistemic value. The paper shall then examine an argument against a separation between epistemic and non-epistemic values given by Kuhn and Douglas.
In “Epistemic Values and the Argument from Inductive Risk,” Steel distinguishes between epistemic and non-epistemic values by arguing that the distinction is not one of “fact and value” as is commonly believed.[1] He states epistemic values “promote the attainment of truth.”[2] Thus, Ayer’s criterion of verifiability,[3] Popper’s criterion of falsification,[4] and Kuhn’s five criteria[5] are considered epistemic values, as they promote the acquisition of truth. This definition of epistemic value has been supported by major philosophers of science, such as Popper,[6] Goldman,[7] and Rooney.[8]
It is worth noting Kuhn’s recognition of how epistemic values seem to be mediated in some way by non-epistemic values during scientific practice. Listing common epistemic values he takes all scientists to hold (and have always held),[9] he states that even if all scientists acknowledge these epistemic values, “the criteria are imprecise,” as well as that “when deployed together, they repeatedly prove to conflict with one another…”[10] Kuhn then concludes that scientists do make value decisions,[11] as it is impossible to separate the social conditions of a scientist from the epistemic values they choose to hold.[12] This is similar to the position held by Heather Douglas, who states that “[…] one cannot usefully or coherently distinguish the cognitive [epistemic] from the social. There is no principled way to draw such a distinction.”[13]
An example of such social factors would be normative ethical theories that an individual chooses to follow or has been raised to follow. Suppose one had two theories identical in content except that one is more accurate, and the other has a broader scope.[14] Depending on one’s ethical theory, a scientist may choose one or the other. For example, suppose one followed a weak rule utilitarian moral framework, whereby the choice is between two theories based on which one would, in the average situation, produce the greatest amount of utility (happiness or pleasure).[15] In this scenario, it would appear, in most cases,[16] that the moral option would be to choose the theory that has a broader scope, as this would seem to imply that it could be applied to the most “observations, laws, or subtheories,” thus resulting in a greater utility.[17]
The Relationship Between Epistemic Values and Theories of Truth
If we assume scientists only make epistemic value judgements, then it would follow that all value judgements are made to maximize truth. However, there appears to be vicious circularity in this argument—for what is truth? If, as this paper shall aim to show, an answer to this question utilizes or depends on a non-epistemic value, then the thesis outlined in the introduction must be false.
Steel acknowledges the question in his paper, stating that the epistemic values that he has outlined (including Popper’s, Kuhn’s, etc.) all have relied on the correspondence theory of truth.[18] He elaborates that non-Realist[19] theories of truth are not compatible with many of the epistemic values delineated, at least insofar as they refer to the same thing. For example, accuracy will change meaning as one shifts from a correspondence theory of truth (that is Realist in nature), to a Peircean theory of truth, in which pragmatism dominates. In the former, accuracy refers to correspondence to a mind-independent world, while the latter typically is concordance with a community’s investigational results.[20] Under the former theory, Ptolemaic astronomy was never accurate; under the latter theory, it was accurate during the period scientists investigated it and came to a universal conclusion. Herein we see an example of what Kuhn and Douglas are referring to when discussing the ways in which a social (or subjective) factor can influence what an epistemic value means—the meaning of an epistemic value changes depending on what theory of truth is taken to be true.
As an example, consider Rudner’s argument: a scientist must make value judgements when deciding the error-rate of a scientific experiment.[21] Levi’s explanation of Rudner’s argument reveals the following premise: “[…] No amount of evidence ever completely confirms or disconfirms any (empirical) hypothesis but only renders it more or less probable.”[22] This is an epistemic value known as falsification (or testability to Steel),[23] harkened by Popper, that theories are never confirmed but merely corroborated.[24] Levi explains that he shall not attempt to refute this premise, as it “is a cardinal tenet of an empiricist philosophy of science and will not be questioned in this paper…”[25] Regardless of whether an individual accepts Levi’s refutation of Rudner, one should now see why Levi’s argument disregards the justification of a premise that cannot be ignored; in order to justify falsification, one must appeal to a theory of truth. This tenet is a logical theory of confirmation, and it is posited to maximize truth—obviously, then, the scientist, if they assume such a tenet, is deciding to subscribe to a theory of truth of some form. Indeed, now we see many philosophers of science abandoning the logical theory of confirmation in favor of subjective interpretations, justified popularly by appeals non-epistemic value,[26] in lieu of Goodman’s new problem of induction.[27]
The Relationship between Theories of Truth and Non-Epistemic Values
One can now analyze the question: what is truth? Specifically, how can a theory of truth be true? With epistemic values, one can judge if something is true based on whether it allows one to maximize whatever truth may be; however, this answer is not available to one inquiring about the status of a theory of truth. Therefore, this seemingly indicates that our thesis that scientists do not make non-epistemic value judgements is false, for the judgement to choose one theory of truth over another cannot be answered with an appeal to truth because this is circular. Thus, the scientist does make value judgements: they judge which theory of truth should be chosen (on what basis is outside the scope of this paper.)
Scientist qua Scientist?
One can make an objection to this argument: does the scientist qua scientist make the judgement when choosing a theory of truth, or is it the scientist qua philosopher? Not wishing to devolve into a semantical discussion, this paper shall instead attempt to show that scientists have actively engaged in such meta-epistemological discussions, and that, even if they did not, they would still be making value judgements passively.
A specific example of an instance where scientists disagree over a theory of truth arose in the controversy surrounding diagnosis criteria in the then new DSM-III.[28] A conflict arose between traditional psychoanalysts who sought to “understand and treat the unconscious source of the problem,” versus later psychiatrists who advocated for “assess[ing] symptoms to arrive at a correct medical diagnosis.”[29] The latter evidently were attempting to utilize a correspondence theory of truth in the sciences, wherein the goal of a psychologist is to analyze the symptoms in order to reach the actual condition the patient is suffering from in the real world. The former, however, appeared to advocate for a different theory of truth: perhaps a non-Realist theory like Peirce’s—the cognitive end of the psychologist is to attain utility by allowing the patient to ameliorate themselves. The ‘true condition’ of the patient is determined based on what will allow for the best success in treating them. Indeed, in McWilliams’ ground-breaking contemporary psychoanalytic text Psychoanalytic Diagnosis, she states that a diagnosis must meet the following criterion to be correct: “Does the careful, nonabusive application of psychodiagnostic concepts increase a client’s chances of being helped?”[30]
Further, even if not actively arguing for or against a theory of truth in some form, a scientist always passively makes a value judgement when operating within the scientific field; otherwise, nothing would be accomplished. For example, by making hypotheses under the umbrella of clinical psychology, a scientist is making a value judgement, even if not explicitly saying so, for they are utilizing a correspondence (or semantic) theory of truth and not a non-Realist theory.
Yes, it is possible to restrict the duties of a scientist to everything underneath the axioms of truth; however, such historical instances seem to indicate that this is not what scientists do. They discuss the issues of what should be “true” in scientific journals and amongst each other in scientific circles. By this same reasoning, one could say that politicians do not make value judgements (qua politician) if one designates acts qua politician as those that only further a politician’s party’s stance—the evaluation of such a stance could be designated qua philosopher or qua political theorist. Based on this, I suggest we incorporate the act of truth evaluation into the acts of what we consider to be qua scientist.
Bibliography
Ayer, A. J. Language, Truth, and Logic. New York City: Oxford University Press, 1936.
Douglas, Heather, and Peter Machamer. “Cognitive and Social Values.” Science & Education 8 (1999): 45-54.
Goodman, Alvin. Knowledge in a Social World. Oxford University Press, 1999.
Goodman, Nelson. Fact, Fiction, and Forecast. Cambridge: Harvard University Press, 2011.
Kuhn, Thomas. “Objectivity, Value, and Theory Choice.” In The Essential Tension. Chicago: Chicago University Press, 1977.
Levi, Isaac. “Must the Scientist Make Value Judgements?” The Journal of Philosophy 57, no. 11 (1960): 345-357.
McNally, Richard J. What is Mental Illness. Cambridge: Harvard University Press, 2011.
McWilliams, Nancy. Psychoanalytic Diagnosis, Second Edition: Understanding Personality Structure in the Clinical Process. Guilford Publications, 2014.
Mill, John S. Utilitarianism. Hackett Publishing Company, 1979.
Peirce, Charles S. Collected Papers of Charles Sanders Peirce, Vols. 1-8. Cambridge: Harvard University Press.
Popper, Karl. Conjectures and Refutations: The Growth of Scientific Knowledge. Routledge, 2002.
Rooney, Phyllis. “On Values in Science: Is the Epistemic/Non-Epistemic Distinction Useful?” Proceedings of the Biennial Meeting of the Philosophy of Science Association 1 (1992): 13-22.
Rudner, Richard. “The Scientist Qua Scientist Makes Value Judgements.” Philosophy of Science 20, no. 1 (1953): 1-6.
Steel, Daniel. “Epistemic Values and the Argument from Inductive Risk.” Philosophy of Science 77, no. 1 (2010): 14-34.
[1] Daniel Steel, “Epistemic Values and the Argument from Inductive Risk,” Philosophy of Science 77, no. 1 (2010): 21.
[2] Ibid., 15.
[3] A. J. Ayer, Language, Truth, and Logic (New York City: Oxford University Press, 1936), 6.
[4] Karl Popper, Conjectures and Refutations: The Growth of Scientific Knowledge (Routledge, 2002), 33-36.
[5] Thomas Kuhn, “Objectivity, Value, and Theory Choice,” in The Essential Tension (Chicago: Chicago University Press, 1977), 357.
[6] Karl Popper, Conjectures and Refutations: The Growth of Scientific Knowledge (Routledge, 2014), 296-297.
[7] Alvin Goldman, Knowledge in a Social World (Oxford University Press, 1999), 221.
[8] Phyllis Rooney, “On Values in Science: Is the Epistemic/Non-Epistemic Distinction Useful?” Proceedings of the Biennial Meeting of the Philosophy of Science Association 1 (1992): 14.
[9] Accuracy, consistency, scope, simplicity, and fruitfulness.
[10] Thomas Kuhn, “Objectivity, Value, and Theory Choice,” in The Essential Tension (Chicago: Chicago University Press, 1977), 357.
[11] Insofar as a scientist makes a value decision via. their interpretation of a criterion, or via. the accentuation of one over another.
[12] Ibid, 360-361.
[13] Heather Douglas, and Peter Machamer, “Cognitive and Social Values,” Science & Education 8 (1999): 45.
[14] I assume that accuracy does not necessarily entail greater usefulness or happiness due to question regarding scientific realism that are outside the scope of this paper.
[15] John S. Mill, Utilitarianism (Hackett Publishing Company, 1979), 19.
[16] Disregarding specific circumstances detailed in the 14th footnote.
[17] Thomas Kuhn, “Objectivity, Value, and Theory Choice,” in The Essential Tension (Chicago: Chicago University Press, 1977), 357.
[18] Daniel Steel, “Epistemic Values and the Argument from Inductive Risk,” Philosophy of Science 77, no. 1 (2010): 21.
[19] The ‘R’ is capitalized in accordance with Kirkham’s Theories of Truth.
[20] Charles S. Peirce, Collected Papers of Charles Sanders Peirce, Vols. 1-8 (Cambridge: Harvard University Press), 5.407.
[21] Richard Rudner, “The Scientist Qua Scientist Makes Value Judgements,” Philosophy of Science 20, no. 1 (1953): 3.
[22] Isaac Levi, “Must the Scientist Make Value Judgements?” The Journal of Philosophy 57, no. 11 (1960): 347.
[23] Daniel Steel, “Epistemic Values and the Argument from Inductive Risk,” Philosophy of Science 77, no. 1 (2010): 15.
[24] Karl Popper, Conjectures and Refutations: The Growth of Scientific Knowledge (Routledge, 2002), 47-48.
[25] Isaac Levi, “Must the Scientist Make Value Judgements?” The Journal of Philosophy 57, no. 11 (1960): 347.
[26] See for example Reichenbach’s vindication of induction.
[27] Nelson Goodman, Fact, Fiction, and Forecast (Harvard University Press, 1983), 74.
[28] This paper assumes that psychology and psychiatry is a science.
[29] Richard J. McNally, What Is Mental Illness (Cambridge: Harvard University Press, 2011), 18.
[30] Nancy McWilliams, Psychoanalytic Diagnosis, Second Edition: Understanding Personality Structure in the Clinical Process (Guilford Publications, 2014), 8.