Skip to main content

The Prejudice of Freedom: an Application of Kripke’s Notion of a Prejudice to our Understanding of Free Will


This essay reframes salient issues in discussions of free will using conceptual apparatus developed in the works of Saul Kripke, with particular attention paid to his little-discussed technical notion of a prejudice. I begin by focusing on how various forms of modality (metaphysical, epistemic, and conceptual) underlie alternate forms of compatibilism and discuss why it is important to avoid conflating these forms of compatibilism. The concept of a prejudice is then introduced. We consider the semantic role of prejudices, in particular conditions in which prejudices turn out to express metaphysically necessary truths. With that as background, I discuss a set of prejudices involving the notion of choice. We consider the role these prejudices might play, should they turn out to be true, in determining the answer to various compatibility questions concerning the nature of moral responsibility and choice.

This is a preview of subscription content, access via your institution.


  1. 1.

    Kripke (1980: 39) stipulates that “an analytic statement is, in some sense, true by virtue of its meaning and true in all possible worlds by virtue of its meaning.”

  2. 2.

    If X or Y contains free variables the analysis of their conceptual compatibility will have to be rephrased in terms of satisfaction: For any assignment to its variables, “⋄C (X & Y)” is satisfied iff it is not the case that “~(X & Y)” is satisfied in all possible worlds in virtue of its meaning.

  3. 3.

    Usually when I speak of epistemic possibility, I will leave the values for “P” and “t” unstated. Sometimes I may index epistemic possibility to a group rather than an individual (as in “for all scientists know it may be the case that life first evolved on Mars”).

  4. 4.

    In saying this I am not assuming that conceptual incompatibility by itself entails epistemic incompatibility. Rather I am saying if X and Y are conceptually incompatible and have been shown to be conceptually incompatible, then (for those to whom this has been shown) X and Y are epistemically incompatible.

  5. 5.

    Free will defenses vary and it goes beyond the scope of this paper to discuss the details of free will defenses. My purpose here is merely to highlight the importance of being sensitive to the modality of compatibility claims in discussions of the problem of evil. For a more detailed treatment of the free will defense as it bears on these issues see (Cain 2004).

  6. 6.

    In fairness to Lewis, I should note that his paper also presents arguments that attempt to raise difficulties for the free will defense even if we assume that its libertarian views of free will are correct.

  7. 7.

    In his essay, Gómez-Torrente summarizes and discusses Kripke’s unpublished lectures on color. Though Kripke makes extensive use of the notion of a prejudice in these lectures, Gómez-Torrente’s discussion is the only published account of this topic of which I am aware. A video of Kripke's lecture “No Fool's Red? Some Considerations on the Primary/Secondary Quality Distinction,” in which he develops his notion of a prejudice (and at which Gómez-Torrente is the commenter), is available at the Saul Kripke Center website: In the video Kripke covers much of the material on prejudices that Gómez-Torrente summarizes in his essay.

  8. 8.

    Eddington (1933: 1-5) was particularly impressed by the separation of the subatomic components of atoms. According to Eddington the space occupied by those components is miniscule compared to the size of the whole atom. “The atom is as porous as the solar system. If we eliminated all the unfilled space in a man’s body and collected his protons and electrons into one mass, the man would be reduced to a speck just visible with a magnifying glass.” (Eddington 1933: 1-2)

  9. 9.

    The fact that Kripke is willing to treat the claim that solids are non-gappy as a prejudice that carries some initial weight (even if we do not ultimately accept it) shows that a prejudice is not to be understood as a stereotype (in Putnam's (1970) sense), for the linguistic competence of a speaker with respect to the term “solid” does not require that the speaker have a stereotype that has implications with respect to the micro-structure of solids (e.g., that there are not gaps between subatomic particles in the way Eddington thought). Nor should prejudices be understood as simply conveying features by which we pick out paradigm cases of the application of an expression. While it might contribute to our picking out solids that they are not observably gappy, we don't pick them by their being fully non-gappy—after all, the paradigms of solidity we pick out are all gappy.

  10. 10.

    Cf. the previous quote from Gómez-Torrente on the semantic role of prejudices

  11. 11.

    The qualification “we ultimately appeal to” has been added to exclude from PF(ii) consideration of choices for which we hold the agent merely derivatively responsible.

  12. 12.

    I am skeptical about the truth of SPF. In addition to the above worry about torture, examples given by Eleonore Stump (1999: 323) cast serious doubt on SPF.

  13. 13.

    I do not want to give the impression that our prejudices together with the discovered empirical facts need always fully determine an answer to this question. See the quotation from Gomez-Torrente with which I introduced the notion of prejudice. It speaks of prejudices as guiding our semantic decisions.

  14. 14.

    For examples of approaches to discussions of free will and free action that give a central place to an appeal to paradigms, see Flew’s (1955)

    classic statement, and Heller's (1996). Heller (1996: 336) holds that if there is a single kind of which the paradigms of free actions are all instances, then if determinism holds compatibilism is true; if there is “no single kind of which the paradigms are all instances”, then it is not actually the case that free acts exist; and what counts as essential to an action's being free is determined by the kind (if it exists) of which the paradigms are instances (and thus it could turn out that “what is in fact essential to an action’s being free is that there be some undetermined event in that action’s causal history”). Heller correctly warns us not to confuse epistemic possibility and metaphysical possibility: “Assuming that we do not know determinism to be false and even assuming that we do know that ‘free act’ refers, what follows is that there is an epistemic possibility of a deterministic world in which our actions are free. It does not follow that there really is any such possible world.”

  15. 15.

    Perhaps an analogy may help here. Suppose that “bacteria” was initially introduced as a natural kind term intended to pick out a kind of organism that had been viewed under microscopes. At that time, it may have been an open question whether there exist bacteria with a much different superficial appearance (say, that of being doughnut shaped) than the appearance of the paradigms used to fix the reference. Being doughnut shaped would have been epistemically compatible with being a bacterium, and—given that bacteria form a natural kind—it may also have been metaphysically compatible. But had it turned out that what were initially taken to be paradigms of bacteria were not microscopic organisms, but were instead scratches on microscope slides, then the term “bacteria” would have been withdrawn and not recognized as a successful natural kind term. It would lack reference in every possible world. Even if doughnut-shaped microorganisms were discovered later, they would not count as falling under the extension of the term “bacteria” as it was originally introduced.

  16. 16.

    An anonymous reader has suggested that there might be some “compatibilistic” prejudices (e.g., uncaused = random = unfree) pertaining to the determination of actions or events that would clash with libertarian accounts of freedom. If so, then if it were discovered that what are taken to be paradigm cases of freely choosing are not caused by anything prior, there would be a clash between the compatibilistic prejudices and the prejudice that what are taken to be paradigms of free choice really are cases of free choice.

  17. 17.

    I am unconcerned with whether the claims I consider regarding determinism, choice and responsibility are a priori. But it will help to illustrate the problem I have in mind to see how a similar problem arises with Kripke’s treatment of a prioricity.

  18. 18.

    Aristotle (1984: 922; History of Animals, Bk.VIII, 588b20) took them to be intermediate between plants and animals.

  19. 19.

    To be precise I would have to spell out, as part of the antecedent, the details of the evolutionary theory that I have in mind, otherwise I could not know a priori that the conditional was true—my knowledge of the conditional would depend upon a posteriori knowledge of what the current evolutionary accounts are. If we let E abbreviate a statement of current evolutionary theory, then I would have a priori knowledge that if cats evolved as suggested by E forming reproductively viable populations of animals, then it is metaphysically necessary that to be a cat a thing must be an animal. Perhaps further qualifications would be needed, but, for present purposes, my approximate statement should be adequate.

  20. 20.

    If it were an analytic truth that choice is undetermined, then (2) would be analytic. I am assuming that neither SPF nor the claim that choices are undetermined is analytic.

  21. 21.

    As I will use the term, to say that an exercise of strong regulative control is an instance of agent causation requires that in that exercise the agent, as a substance, acts as a cause, where the agent’s role as a cause is not ontologically reducible to causation by some event, property, or state involving the agent. Furthermore, in agent causation the agent initiates a new causal chain that is not simply the working out of past causal influences. See Clarke (2019) for a discussion of the notion of irreducible substance causation, as well as a review of several versions of agent causation.

  22. 22.

    I would like to thank Rebecca Bensen Cain and various anonymous reader for helpful comments on earlier versions of this paper.


  1. Aristotle. (1984). History of animals. In J. Barnes (Ed.), The Complete Works of Aristotle: The Revised Oxford Translation. Princeton: Princeton University Press.

    Google Scholar 

  2. Cain, J. (1991) Are analytic statements necessarily a prioriAustralasian Journal of Philosophy, 69 (3), 334–337.

  3. Cain, J. (2004) Free will and the problem of evil. Religious Studies, 40 (4), 437–456.

  4. Clarke, R. (2019). Free will, agent causation, and “disappearing agents”. Noûs, 53(1), 76–96.

    Article  Google Scholar 

  5. Eddington, A. (1933). The Nature of the Physical World. New York: The MacMillan Company.

    Google Scholar 

  6. Fischer, J. M. (1999). Recent work on moral responsibility. Ethics, 110(1), 93–139.

    Article  Google Scholar 

  7. Fischer, J. M. (2006). My way: essays on moral responsibility. New York: Oxford University Press.

    Google Scholar 

  8. Fischer, J. M. (2012). Semicompatibilism and its Rivals. The Journal of Ethics, 16(2), 117–143.

    Article  Google Scholar 

  9. Fischer, J. M. 2016. Libertarianism and the problem of flip-flopping. in free will and theism: connections, contingencies, and concerns, ed. K. Timpe and D. Speak, 48-61. Oxford: Oxford University Press.

  10. Fischer, J. M., & Ravizza, M. (1998). Responsibility and control: a theory of moral responsibility. Cambridge: Cambridge University Press.

    Book  Google Scholar 

  11. Flew, A. (1955). Divine omnipotence and human freedom. In A. Flew & A. McIntyre (Eds.), New Essays in Philosophical Theology (pp. 144–169). New York: Macmillan.

    Google Scholar 

  12. Gómez-Torrente, M. (2011). Kripke on color words and the primary/secondary quality distinction. In Saul Kripke ed. A. Berger (pp. 290–323). Cambridge: Cambridge University Press.

    Chapter  Google Scholar 

  13. Heller, M. (1996). The mad scientist meets the robot cats: compatibilism, kinds, and counterexamples. Philosophy and Phenomenological Research, 56(2), 333–337.

    Article  Google Scholar 

  14. Kane, R. (2007). Response to Fischer, Pereboom, and Vargas. In J. M. Fischer, R. Kane, D. Pereboom, & M. Vargas (Eds.), Four views on free will (pp. 166–183). Oxford: Blackwell Publishing.

    Google Scholar 

  15. Kripke, S. (1980). Naming and necessity. Cambridge: Harvard University Press.

    Google Scholar 

  16. Lewis, D. (1993). Evil for freedom’s sake? Philosophical Papers, 22(3), 149–172.

    Article  Google Scholar 

  17. O’Connor, T. (2000). Persons and causes. New York: Oxford University Press.

    Google Scholar 

  18. Putnam, H. (1970). Is semantics possible? Metaphilosophy, 1(3), 187–201.

    Article  Google Scholar 

  19. Stebbing, L. S. (1937). Philosophy and the physicists. London: Methuen & Co. Ltd..

    Google Scholar 

  20. Steward, H. (2012). A metaphysics for freedom. Oxford: Oxford University Press.

    Book  Google Scholar 

  21. Stump, E. (1999). Alternative possibilities and moral responsibility: the flicker of freedom. The Journal of Ethics, 3(4), 299–324.

    Article  Google Scholar 

  22. van Inwagen, P. (1983). An essay on free will. Oxford: Oxford University Press.

    Google Scholar 

Download references

Author information



Corresponding author

Correspondence to James Cain.

Ethics declarations

Conflict of interest

On behalf of all authors, the corresponding author states that there is no conflict of interest.

Additional information

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Rights and permissions

Reprints and Permissions

About this article

Verify currency and authenticity via CrossMark

Cite this article

Cain, J. The Prejudice of Freedom: an Application of Kripke’s Notion of a Prejudice to our Understanding of Free Will. Acta Anal 36, 323–339 (2021).

Download citation