Skip to main content
Log in

Should have known

  • Published:
Synthese Aims and scope Submit manuscript


In this paper I will be arguing that there are cases in which a subject, S, should have known that p, even though, given her state of evidence at the time, she was in no position to know it. My argument for this result will involve making two claims. The uncontroversial claim is this: S should have known that p when (one) another person has, or would have, legitimate expectations regarding S’s epistemic condition, (two) the satisfaction of these expectations would require that S knows that p, and (three) S fails to know that p. The controversial claim is that these three conditions are sometimes jointly satisfied. I will spend the majority of my time defending the controversial claim. I will argue that there are (at least) two main sources of legitimate expectations regarding another’s epistemic condition: participation in a legitimate social practice (where one’s role entitles others to expect things of one); and moral and epistemic expectations more generally (the institutions of morality and epistemic assessment being such as to entitle us to expect various things of one another). In developing my position on this score, I will have an opportunity (i) to defend the doctrine that there are “practice-generated entitlements” to expect certain things, where it can happen that the satisfaction of these expectations requires another’s having certain pieces of knowledge; (ii) to contrast practice-generated entitlements to expect with epistemic reasons to believe; (iii) to defend the idea that moral and epistemic standards themselves can be taken to reflect legitimate expectations we have of each other; (iv) to compare the “should have known” phenomenon with a widely-discussed phenomenon in the ethics literature—that of culpable ignorance; and finally (v) to suggest the bearing of the “should have known” phenomenon to epistemology itself (in particular, the theory of epistemic justification).

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Similar content being viewed by others


  1. The more general phenomenon is actually the should have believed phenomenon—that phenomenon whereby one ought to have had certain evidence, and ought to have believed on that basis what the evidence supported. The epistemically interesting should have known is a special case of the should have believed. Still, I will continue to talk of the should have known. (With thanks to Jennifer Lackey and to Daniel Silvermint, who, independently, urged this point on me.)

  2. As we will see, Gideon Rosen (2002, 2004) defends a view that might be taken to be in the vicinity of the claim that they are impossible. I will address Rosen’s position below, in Sect. 3.

  3. Various people have stressed this point to me in conversation, including Matthew McGrath and Juan Comesana. (I should say that their view appears to be, not that there are no such cases, only that the cases have no interest for epistemology proper.)

  4. Variants of this case are discussed in Smith (1983, pp. 544–545), Rosen (2004, p. 303), FitzPatrick (2008, p. 597), and Miller and Record (2013, p. 122).

  5. Below, I will discuss the relationship between the should have known phenomenon and the phenomenon of culpable ignorance. Here I merely assume—what I take to be plausible—that if S’s ignorance is culpable, S should have known.

  6. There are other examples in the literature on culpable ignorance in which the culpability of the ignorance reflects the fact that the subject did not have relevant alternatives in mind. See, for example, Fischer and Tognazzini (2009, p. 546).

  7. Compare Fitzpatrick (2008, p. 597), who writes that the doctor’s ignorance will be culpable whenever the doctor acquires a false belief about a patient, where this acquisition is the result of “a culpable failure to discharge procedural epistemic obligations” in connection with her role as doctor. Compare Smith (1983, pp. 544–545) as well.

  8. I will return to the relation between blame and “should have known” allegations below in Sect. 3, below.

  9. Many authors speak of the legitimacy of our expectations of others, and of the relationship between these expectations and allegations of “culpable ignorance.” This is a theme in the ethics literature on culpable ignorance, for which see e.g. Montmarquet (1992, pp. 336–337) and (1999, p. 845); Moody-Adams (1994, p. 291) (although she does not speak of entitlements); Rosen (2002, p. 79); and FitzPatrick (2008, pp. 603, 612). It is also a theme in the epistemology literature on normative defeaters to knowledge and justification. See e.g. Pollock (1986, p. 192); Meeker (2004, pp. 162–163); Senor (2007, p. 207) (although Senor does not speak of entitlements); Record (2013, pp. 3, 8); and Miller and Record (2013, pp. 122, 124). In short, there would appear to be some agreement among ethicists and epistemologists who think about related matters, to the effect that people are entitled to certain epistemic expectations regarding others. Below I will try to give an account of the basis of this entitlement.

  10. For discussions about the nature of epistemic entitlements, see Burge (1993, 1996, 1997, 2003), Dretske (2000); Peacocke (2004); Wright (2004), and Goldberg (forthcoming).

  11. The second condition is meant to cover cases in which what is expected of S is that he has been epistemically responsible in such-and-such a way, where, were S to have been epistemically responsible in the relevant way, he would have known that p. In such cases, S* herself may not know the contents of the knowledge S ought to have; what S* expects, rather, is that S was responsible in such-and-such a way.

  12. We might say that the first two conditions identify the conditions on its being the case that S should know that p, and the third condition is what (together with the first two) transforms a should know into a should have known.

  13. Although Ross speaks of assumptions where I speak of expectations, I think this difference is negligible.

  14. Alternatively: were S to form her belief through the process which makes her belief reasonable.

  15. ‘Practice-generated entitlement’ is my term, not Ross’s, but it is clear that this is what he has in mind.

  16. Think of things that are generally true of the practice without being “internal to” it, in the sense that they are not mandated by the standards of the practice. For example, most soccer players run fast; but it is no part of the rules of soccer that players have to be fast, and so those rules do not entitle you to expect that (most) soccer players run fast. So, too, for other examples from the world of sports: lefties don’t play third base in baseball; gymnasts are flexible; basketball players are tall; figure skaters have excellent balance; and so forth.

  17. There are very interesting questions to be asked about cases in which no sanitation worker saw the bins. It is tempting in such cases to accuse the Sanitation Department itself as being such that it should have known that the bins needed emptying. This raises all sorts of interesting issues regarding group knowledge and group ignorance that I can’t address here. (With thanks to Jennifer Lackey for indicating the relevance of group ignorance to the should have known phenomenon, in conversation.)

  18. This point is emphasized by Smith (1983, p. 544), Rosen (2004, pp. 301–303), FitzPatrick (2008, p. 597), and Miller and Record (2013, p. 122), among many others.

  19. My account of the “should have known” in connection with professional and institutional responsibilities has some affinities with the notion of a “role ought” found in Feldman (2004). As I will develop my account, role “oughts” are a special case of a more general phenomenon, which is to be understood in terms of practice-generated entitlements.

  20. See Meeker (2004, p. 162) and Record (2013, p. 3), among many others.

  21. See Smith (1983, p. 551).

  22. This point is emphasized in Moody-Smith (1994, p. 291).

  23. Indeed, our expectations go beyond merely what we can expect of others insofar as they are playing certain professional or institutional roles; we also are entitled to certain epistemic expectations of our friends and lovers [for which see Smith (2005, pp. 236, 244–245) and Ebels-Duggan (2008)], as well as people more generally, as that they recognize other people’s feelings (Adams 1985, p. 18). I will address these matters below.

  24. Most of the cases to be found are found in the ethics literature.

  25. ‘Arguably’ cannot be handled by \(\hbox {SHK}_{\mathrm{PGE}}\): if our moral and epistemic expectations of one another can be plausibly rendered as reflecting legitimate moral and epistemic practices, then \(\hbox {SHK}_{\mathrm{PGE}}\) would be able to cover the range of cases. I myself have toyed with the idea that our moral and epistemic expectations can be so rendered; but colleagues who have thought much longer and harder than I have about the moral case (Kyla Ebels-Duggan, Ellie Mason, and others) are dubious, and their worries have given me pause.

  26. Hacking’s interest in the case was what it tells us about epistemic modals.

  27. I thank Ellie Mason for raising this point, and for urging that I consider whether such cases can always be handled by appeal to \(\hbox {SHK}_{\mathrm{PGE}}\).

  28. See also Goldberg (2011), where I tried to identify what we might reasonably expect of one another regarding how quickly information from other sources reach us, and on what topics this is likely.

  29. “Observing” is meant to cover apprehension through any of the sensory modalities (not merely vision).

  30. “All” is not equal when basic reading competence alone does not suffice to ensure comprehension of the text in question—say, because the text is hard to read, or requires special expertise.

  31. See (Goldberg (2015), Chaps. 2 and 3) for a full discussion.

  32. Here I bracket the difficult question whether moral standards can ever pertain to oneself and oneself alone—whether it is possible to have moral expectations of oneself that have nothing to do with how one treats others. This is a difficult question which, happily, need not be settled for my purposes.

  33. This case may not count as an epistemically interesting case: after all, the subject did have the evidence in question. But the next case I describe is epistemically interesting.

  34. Objection: this assumes, without argument, that any ignorance of a moral injunctions is culpable. Reply: no it doesn’t. I haven’t said a word about culpability or blameworthiness. More on this in Sect. 3.

  35. Or at least that if this person doesn’t recognize the principle, her ignorance will at that point be culpable.

  36. I thank both Ellie Mason and Kyla Ebels-Duggan for raising this worry, in slightly different ways.

  37. More accurately: if you are entitled to expect something whose satisfaction requires that they know.

  38. Of course, in that case we will want to know what the point of “should have known” allegations is—a matter I will discuss in Sect. 3.

  39. This is an issue that is relevant to Moody-Adams (1994) discussion of the nineteenth century American slaveowner. Moody-Adams’ view is that the slaveowners in question were not in good faith when defending the practices of slavery. On the contrary, she regards them as motivated by “affected ignorance” (1994, p. 296). In an unpublished manuscript, Ellie Mason discusses the case of mid-twentieth-century sexists. It is worth noting, though, that both of these authors are discussing the issue of culpability, whereas I want to distinguish the issue of culpability from the issue of the should have known.

  40. In this connection it is worth citing the distinction between moral reproach and moral blameworthiness, as found in Calhoun (1989). This distinction highlights something very close to a (moralized) version of the difference between should have known and culpably ignorant.

  41. The point that there can be such cases appears to have been anticipated in FitzPatrick (2008, p. 602). He brings out the point in passing, in connection with his criticism of Rosen’s (blamelessness constraint) view of culpable ignorance (cited above). But—apparently in contrast to FitzPatrick—I think that the present point, to the effect that “S should have known that p” does not entail “S’s ignorance regarding p is culpable,” need not antagonize Rosen’s account of culpable ignorance. This is for the reason I noted in Sect. 2, above: Rosen can allow that the allegation that S should have known does not always cast blame on S herself. Then again, if there are epistemically interesting cases of the “should have known” phenomenon that involve epistemic blame of the subject in question, these would antagonize Rosen’s view. (I haven’t argued in this paper that there are such cases, however.)

  42. For dissenting views, see Moody-Adams (1994) and FitzPatrick (2008, pp. 606–608). For a hybrid view, according to which we ought to distinguish moral reproach from moral blameworthiness, see Calhoun (1989). And for a hybrid view in epistemology, according to which there can be gaps between epistemic blameworthiness and epistemic praiseworthiness, see Weatherson (2008).

  43. These cases raise very interesting questions about collective epistemic subjects. I hope to be able to return to this in future work. (I thank Jennifer Lackey for bring to my attention, in conversation, how culpable ignorance and the should have known phenomenon are connected to issues in collective epistemology.)

  44. Objection: if my analysis yields the result that S should have known, despite the fact that there was no reasonable way S could have known, then my analysis violates the “ought implies can” principle. Reply: I am suggesting that the “should” in “should have known” often tracks what is rightly expected of one, and it is not clear why we would want to impose the “ought implies can” constraint on these sets of expectations. If we did impose this constraint on these expectations, this would undermine the possibility of epistemically interesting cases of the “should have known” phenomenon. Since there are clear examples of such cases—the ones with which I started in Sect. 1 of this paper, for example—we should reject the move to insist that this principle applies to these cases. (In this respect my reply is a bit like the reply Feldman (2004) made in response to a similar objection, in which he appealed to “role oughts”; only my reply is based on an account of the relevant “ought” of which his “role ought” is a special case.)

  45. See Sect. 2 for details.

  46. Named after Clifford, famed for his debate on the “ethics of belief” with William James (1896/1979). Clifford’s (1879/1999) view, roughly, was that one’s only epistemic duty was to believe in accord with the evidence one has.

  47. This point can be rejected if one accepts that only knowledge warrants assertion (in which case only those beliefs that are the expressions of knowledge are worthy of belief by others). However, there would appear to be some tension between Extreme Cliffordianism, which tells one to tailor one’s beliefs to one’s evidence, and the knowledge norm of assertion, which tells one to assert only what one knows. This tension is seen in cases in which one’s belief conforms to one’s evidence yet one fails to know. In such cases, the combination of Extreme Cliffordianism and the knowledge norm of assertion entails that one who expresses one’s belief in an assertion does something which would appear to be both permissible (in line with one’s evidence) and wrong (in violation of the norm of assertion). Perhaps this position can be made out (e.g. by distinguishing the norm of belief from that of assertion); but in any case it faces serious challenges of motivation.

  48. See Harman (1973), Pollock (1986, p. 192), Meeker (2004, pp. 162–663), Record (2013, p. 3), and Miller and Record (2013, pp. 122, 124).


  • Adams, R. (1985). Involuntary sins. Philosophical Review, 94(1), 3–31.

    Article  Google Scholar 

  • Alston, W. (1989). Epistemic justification. Ithaca: Cornell University Press.

    Google Scholar 

  • Brown, J., & Cappellen, H. (2011). Assertion: New philosophical essays. Oxford: Oxford University Press.

    Book  Google Scholar 

  • Burge, T. (1993). Content preservation. Philosophical Review, 102(4), 457–488.

    Article  Google Scholar 

  • Burge, T. (1996). Our entitlement to self-knowledge. Proceedings of the Aristotelian Society, 96, 91–116.

    Article  Google Scholar 

  • Burge, T. (1997). Interlocution, perception, and memory. Philosophical Studies, 86(1), 21–47.

    Article  Google Scholar 

  • Burge, T. (2003). Perceptual entitlement. Philosophy and Phenomenological Research, 67(3), 503–548.

    Article  Google Scholar 

  • Buss, S. (1997). Justified wrongdoing. Nous, 31(3), 337–369.

    Article  Google Scholar 

  • Calhoun, C. (1989). Responsibility and reproach. Ethics, 99, 386–406.

    Article  Google Scholar 

  • Clifford, W. (1879/1999). The ethics of belief. In The ethics of belief and other essays. New York: Prometheus Books.

  • Dougherty, T. (2010). Reducing responsibility: An evidentialist account of epistemic blame. European Journal of Philosophy, 20(4), 534–547.

    Article  Google Scholar 

  • Dougherty, T. (forthcoming). The ‘Ethics of Belief’ is ethics (period): Reassigning responsibilism. In J. Matheson, R. Vitz (Eds.), The ethics of belief: New essays. Oxford: Oxford University Press.

  • Dretske, F. (2000). Entitlement: Epistemic rights without epistemic duties? Philosophy and Phenomenological Research, 60(3), 591–606.

    Article  Google Scholar 

  • Ebels-Duggan, K. (2008). Against beneficence: A normative account of love. Ethics, 119(1), 142–170.

    Article  Google Scholar 

  • Feldman, R. (2004). The ethics of belief. In E. Conee & R. Feldman (Eds.), Evidentialism. Oxford: Oxford University Press.

    Chapter  Google Scholar 

  • Fischer, J., & Tognazzini, J. (2009). The truth about tracing. Nous, 43(3), 531–556.

    Article  Google Scholar 

  • FitzPatrick, W. (2008). Moral responsibility and normative ignorance: Answering a new skeptical challenge. Ethics, 118, 589–613.

    Article  Google Scholar 

  • Gibbons, J. (2006). Access externalism. Mind, 115(457), 19–39.

    Article  Google Scholar 

  • Goldberg, S. (2011). If that were true I would have heard about it by now. In Goldman & Whitcomb (Eds.), Social epistemology: Essential readings (pp. 92–108). Oxford: Oxford University Press.

    Google Scholar 

  • Goldberg, S. (2015). Assertion: On the philosophical significance of assertoric speech. Oxford: Oxford University Press.

    Book  Google Scholar 

  • Goldberg, S. (forthcoming). Extended epistemic entitlements. Philosophical Issues.

  • Goldman, A. (2001). Internalism exposed. In H. Kornblith (Ed.), Epistemology: Internalism and externalism. Cambridge: MIT Press.

    Google Scholar 

  • Graham, P. (2006). Liberal fundamentalism and its rivals. In E. Sosa & J. Lackey (Eds.), The epistemology of testimony (pp. 93–115). Oxford: Oxford University Press.

    Chapter  Google Scholar 

  • Hacking, I. (1967). Possibility. The Philosophical Review, 76(2), 143–168.

    Article  Google Scholar 

  • Harman, E. (2011). Does moral ignorance exculpate? Ratio, 24(4), 443–468.

    Article  Google Scholar 

  • Harman, G. (1973). Thought. Princeton: Princeton University Press.

    Google Scholar 

  • James, W. (1896/1979). The will to believe. In W. James (Ed.), The will to believe and other essays in popular philosophy. Cambridge: Harvard University Press.

  • Kornblith, H. (1983). Justified belief and epistemically responsible action. Philosophical Review, 92(1), 33–48.

    Article  Google Scholar 

  • Levy, N. (2007). Doxastic responsibility. Synthese, 155, 127–155.

    Article  Google Scholar 

  • Le Morvan, P. (2012). On ignorance: A vindication of the standard view. Philosophia, 40, 379–393.

    Article  Google Scholar 

  • Littlejohn, C. (forthcoming). The unity of reason. In D. Starr (Ed.), Epistemic norms. Oxford: Oxford University Press.

  • McKenna, M. (2008). Putting the lie on the control condition for moral responsibility. Philosophical Studies, 139(1), 29–37.

    Article  Google Scholar 

  • Meeker, K. (2004). Justification and the social nature of knowledge. Philosophy and Phenomenological Research, 69(1), 156–172.

    Article  Google Scholar 

  • Miller, B., & Record, I. (2013). Justified belief in a digital age: On the epistemic implications of secret internet technologies. Episteme, 10(2), 117–134.

    Article  Google Scholar 

  • Montmarquet, J. (1992). Epistemic virtue and doxastic responsibility. American Philosophical Quarterly, 29(4), 331–341.

    Google Scholar 

  • Montmarquet, J. (1995). Culpable ignorance and excuses. Philosophical Studies, 80, 41–49.

    Article  Google Scholar 

  • Montmarquet, J. (1999). Zimmerman on culpable ignorance. Ethics, 109, 842–845.

    Article  Google Scholar 

  • Moody-Adams, M. (1994). Culture, responsibility, and affected ignorance. Ethics, 104, 291–309.

    Article  Google Scholar 

  • Peacocke, C. (2004). Explaining perceptual entitlement. In R. Schantz (Ed.), The externalist challenge (pp. 441–480). Berlin: De Gruyter.

    Google Scholar 

  • Peels, R. (2011). Tracing culpable ignorance. Logos and Episteme, 2(4), 575–582.

    Article  Google Scholar 

  • Pollock, J. (1986). Contemporary theories of knowledge. Maryland: Rowman and Littlefield.

    Google Scholar 

  • Record, I. (2013). Technology and epistemic possibility. Journal for General Philosophy of Science, 28(1), 1–18.

    Google Scholar 

  • Rosen, G. (2002). Culpability and ignorance. Proceedings of the Aristotelian Society, 103(1), 61–84.

    Article  Google Scholar 

  • Rosen, G. (2004). Skepticism about moral responsibility. Philosophical Perspectives, 18, 295–313.

    Article  Google Scholar 

  • Ross, A. (1986). Why do we believe what we are told? Ratio, 28(1), 69–88.

    Google Scholar 

  • Senor, T. (2007). Preserving preservationism: A reply to Lackey. Philosophy and Phenomenological Research, 74(1), 199–208.

    Article  Google Scholar 

  • Smith, A. (2005). Responsibility for attitudes: Activity and passivity in mental life. Ethics, 115, 236–271.

    Article  Google Scholar 

  • Smith, H. (1983). Culpable ignorance. Philosophical Review, 92(4), 543–571.

    Article  Google Scholar 

  • Smith, H. (2011). Non-tracing cases of culpable ignorance. Criminal Law and Philosophy, 5, 115–146.

    Article  Google Scholar 

  • Teller, P. (1972). Epistemic possibility. Philosophia, 2(4), 303–320.

    Article  Google Scholar 

  • Weatherson, B. (2008). Deontology and Descartes’s demon. Journal of Philosophy, 105(9), 540–569.

    Article  Google Scholar 

  • Vargas, M. (2005). The trouble with tracing. Midwest Studies in Philosophy, 29, 269–291.

    Article  Google Scholar 

  • Wright, C. (2004). Warrant for nothing (and foundations for free)? Aristotelian Society Supplementary, 78(1), 167–212.

    Article  Google Scholar 

  • Zimmerman, M. (1997). Moral responsibility and ignorance. Ethics, 107, 410–426.

    Article  Google Scholar 

Download references


Special thanks to colleagues Kyla Ebels-Duggan, Daniel Immerman, Cristina Lafont, and Ellie Mason, for giving me extensive written and/or spoken feedback on an earlier version of this paper. (I can’t pretend that I have met all of their concerns, though I have tried!) Thanks as well to the very many people with whom I have discussed these matters, and/or who gave me a good deal of verbal feedback on earlier versions of this paper. This includes two anonymous referees for this journal, as well as Paul Bloomfield, Jessica Brown, Adam Carter, Matthew Christman, Juan Comesaña, Amy Flowerree, Mitch Green, Steve Hales, Allan Hazlett, Ted Hinchman, Jenn Jhun, Casey Johnson, Andrea Kruse, Jennifer Lackey, Michael Lynch, Matt McGrath, Rik Peels, Bryan Pickel, Duncan Pritchard, Baron Reed, David Ripley, Blake Roeber, Susannah Siegel, Daniel Silvermint, Martin Smith, Juli Thorson, and Sam Wheeler. Thanks also to the audiences at Bochum University, the University of Connecticut, Notre Dame, and the University of Edinburgh, where I have given this paper as a talk.

Author information

Authors and Affiliations


Corresponding author

Correspondence to Sanford C. Goldberg.

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Goldberg, S.C. Should have known. Synthese 194, 2863–2894 (2017).

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: