1. 21 U.S.C. §841(a)(1).

2. Under the Model Penal Code (“MPC”), “[a] person acts knowingly with respect to a material element of an offense” when he is “aware” or “practically certain” that the element obtains (depending on what sort of element it is). MPC §2.02(2)(b). Thus, in the criminal law, “knowledge requires both belief, or subjective certainty, and the actual truth or existence of the thing known.” Charlow, Robin, Wilful Ignorance and Criminal Culpability , 70 Tex. L. Rev. 1351, 1374–1375 (1992)Google Scholar. The criminal law notion of knowledge thus is weaker than the philosophical notion of true justified belief plus restrictions to avoid Gettier cases.

3. “A person acts recklessly with respect to a material element of an offense when he consciously disregards a substantial and unjustifiable risk that the material element exists or will result from his conduct.” MPC §2.02(2)(c).

4. All the federal courts of appeals have endorsed some version of the willful ignorance doctrine. See Global-Tech Appliances, Inc. v. SEB S.A., 131 S. Ct. 2060, 2070 & n.9 (2011) (collecting cases).

5. United States v. Heredia, 483 F.3d 913, 917 (9th Cir. 2007) (en banc) (discussing such jury instructions).

6. The doctrine I am concerned with thus should not be confused with the distinct evidentiary rule recognizing that evidence tending to show willful ignorance can also constitute evidence from which a jury might infer actual knowledge. See, e.g., Global-Tech, 131 S. Ct. at 2073 (Kennedy, J., dissenting) (observing that “[f]acts that support willful blindness are often probative of actual knowledge”).

7. Id. at 2069.

8. Husak, Douglas N. & Callender, Craig A., Wilful Ignorance, Knowledge, and the “Equal Culpability” Thesis: A Study of the Deeper Significance of the Principle of Legality , 1994 Wis. L. Rev. 29, 53 (1994)Google Scholar.

9. See id. at 58.

10. For example, negligence can substitute for recklessness “if the actor, due to self-induced intoxication, is unaware of a risk of which he would have been aware had he been sober.” MPC § 2.08(2). Similarly, while murder typically requires purpose or knowledge with respect to the victim's death, recklessness can suffice if the circumstances manifest “extreme indifference to the value of human life.” MPC § 210.2(1)(b). See generally Robinson, Paul H., Imputed Criminal Liability , 93 Yale L.J. 609, 616–620 (1984)CrossRefGoogle Scholar (discussing several mens rea substitution principles).

11. Luban, David, Contrived Ignorance , 87 Geo. L.J. 957 (1999)Google Scholar.

12. Michaels, Alan, Acceptance: The Missing Mental State , 71 S. Cal. L. Rev. 953 (1998)Google Scholar.

13. Hellman, Deborah, Willfully Blind for Good Reason , 3 Crim. L. & Phil. 301 (2009)Google Scholar.

14. Sarch, Alex, Willful Ignorance, Culpability and the Criminal Law , 88 St. John's L. Rev. 1023 (2014)Google Scholar; Sarch, Alex, Beyond Willful Ignorance , 88 U. Colo. L. Rev. 97 (2016)Google Scholar.

15. See Heredia, 483 F.3d at 920 (construing the willful ignorance doctrine to require only willful ignorance in what I call the “basic” sense); United States v. Ferrarini, 219 F.3d 145, 154 (2d Cir. 2000) (same); United States v. Flores, 454 F.3d 149, 155 (3d Cir. 2006) (same); United States v. Geisen, 612 F.3d 471, 485–486 (6th Cir. 2010) (same); United States v. Tanner, 628 F.3d 890, 904 (7th Cir. 2010) (same).

16. See, e.g., Larry Alexander, Culpability, in The Oxford Handbook of Philosophy of Criminal Law (Deigh & Dolinko eds., 2011), at 219 (“acts are culpable when they manifest insufficient concern for the interests of others”); Larry Alexander & Kimberly Ferzan, Crime and Culpability (2009), at 67–68 (arguing that “insufficient concern [is] the essence of culpability”); Nomy Arpaly & Tim Schroeder, In Praise of Desire (2014), at 170 (“a person is blameworthy for a wrong action A to the extent that A manifests ill will (or moral indifference) through being rationalized by it”); Victor Tadros, Criminal Responsibility (2005), at 250 (“if [a defendant] is convicted of a serious offence, the state communicates. . .that [his] behaviour manifested an inappropriate regard for other citizens and their interests”); see generally id. 71–99; Peter Westen, An Attitudinal Theory of Excuse, 25 Law & Phil. 289, 373–374 (“a person is normatively blameworthy for engaging in conduct that a statute prohibits if he was motivated by an attitude of disrespect for the interests that the statute seeks to protect”).

17. The insufficient regard theory is similar in spirit to the theory that an action is culpable to the degree that “it is a product of a faulty mode of recognition or response to reasons for action.” Gideon Yaffe, Attempts (2011), at 38. See also Markowitz, Julia, Acting for the Right Reasons , 119 Phil. Review 201 (2010)Google Scholar (discussing the theory that “my action is morally worthy if and only if my motivating reasons for acting coincide with the reasons morally justifying the action”); Graham, Peter, A Sketch of a Theory of Blameworthiness , 88 Phil. & Phenomenological Research 388, 407 (2014)CrossRefGoogle Scholar (arguing that one is blameworthy iff the reactive emotions are appropriate, and this, in turn, is true iff “in φ-ing, X has violated a moral requirement of respect”).

18. I leave it open whether criminal culpability is same as moral blameworthiness. On my view, defended elsewhere, the legal notion of criminal culpability is a simplified, idealized version of the notion of moral blameworthiness. See Alex Sarch, Who Cares What You Think? Criminal Culpability and the Irrelevance of Unmanifested Mental States (forthcoming in Law & Phil). While the moral notion is understood in terms of insufficient regard for morally relevant considerations, the legal notion can be understood in terms of insufficient regard for considerations that the law either does recognize (if we're interested in explaining posited law) or should recognize (if we want to critique the law).

19. See supra note 2.

20. United States v. Svoboda, 347 F.3d 471, 477–478 (2d Cir. 2003) (willful ignorance instruction “permits a finding of knowledge even where there is no evidence that the defendant possessed actual knowledge”) (internal quotation marks omitted); United States v. Freeman, 434 F.3d 369, 378 (5th Cir. 2005) (“deliberate indifference charge permits ‘the jury to convict without finding that the defendant was aware of the existence of illegal conduct’”); Global-Tech, 131 S. Ct. at 2072 (Kennedy, J., dissenting) (“[w]illful blindness is not knowledge”).

21. Charlow, supra note 2, at 1390 (“most definitions of wilful ignorance delineate a mens rea that is the equivalent neither of knowledge nor recklessness”); Robbins, Ira P., The Ostrich Instruction: Deliberate Ignorance as a Criminal Mens Rea , 81 J. Crim. L. & Criminology 191, 226 (1990)CrossRefGoogle Scholar (suggesting “that deliberate ignorance is not knowledge”); Hellman, supra note 13, at 302 (2009) (“contrived ignorance itself is not a form of knowledge”); Husak & Callender, supra note 8, at 51 (arguing that “wilfully ignorant defendants do not possess knowledge”).

22. This is arguably the strategy employed by the MPC. As the MPC commentary explains, its definition of knowledge in “[p]aragraph [2.02](7) deals with the situation British commentators have denominated ‘wilful blindness’ or ‘connivance.’” MPC 129–130 (Tent. Draft No. 4, 1955).

23. Husak & Callender, supra note 8, at 37–38.

24. Glanville Williams, Criminal Law: The General Part (2d ed. 1961), at 157.

25. Global-Tech, 131 S. Ct. at 2070.

26. It is not clear if a person can be willfully ignorant of a proposition that is false. However, this need not concern us, since (barring abuses) one is not likely to be charged with a knowledge crime unless the inculpatory proposition is plausibly true. (Granted one might still be charged with the attempt to do the knowledge crime. But since the mens rea for attempt is intent, not knowledge, willful ignorance seems unlikely to figure into many attempt prosecutions.)

27. I leave it open that what counts as a “sufficiently high probability” here might vary depending on features of the context, e.g., the magnitude of harm that is at stake.

28. Note also that Larry Alexander and Kimberly Ferzan have argued that willful ignorance may be regarded as a type of recklessness: “The prototypical willfully ignorant actor is, of course, reckless. The risk he is taking—of, say, smuggling drugs—is an unjustifiable one.” Alexander & Ferzan, supra note 16, at 34. (Hellman criticizes this view, but acknowledges that Alexander and Ferzan can respond. See Hellman, supra note 13, at 311.) Still, the question I address in this paper arises for Alexander and Ferzan's recklessness account, too. They are correct that the willfully ignorant actor generally is at least reckless. But some cases of willful ignorance seem especially culpable—indeed, as culpable as the analogous conduct done with knowledge (practical certainty) of the inculpatory proposition. And one might want to know what it is, precisely, about opting to remain in ignorance about the risks of one's conduct that could increase the actor's culpability from the level of a reckless wrongdoer up to the level of a knowing wrongdoer. It is precisely this question that my account seeks to answer. Thus, my account can be seen as one way of fleshing out Alexander and Ferzan's view that willful ignorance is a form of recklessness. (For related reasons, my account also differs from Tadros's account of willful blindness. See infra note 104.)

29. 131 S. Ct. at 2070.

30. See, e.g., United States v. Jewell, 532 F.2d 697, 700 (9th Cir. 1976) (en banc) (requiring “conscious purpose to disregard the nature of that which was in the vehicle, with a conscious purpose to avoid learning the truth”) (emphasis added); Heredia, 483 F.3d at 918 (observing that “when Congress made it a crime to ‘knowingly . . . possess with intent to manufacture, distribute, or dispense, a controlled substance,’ . . . it meant to punish not only those who know they possess a controlled substance, but also those who don't know because they don't want to know”) (emphasis added). See also Williams, supra note 24, at 157 (noting that a party is willfully ignorant if he “has his suspicions aroused but then deliberately omits to make further enquiries because he wishes to remain in ignorance”) (emphasis added).

31. United States v. Ferrarini, 219 F.3d 145, 155 (2d Cir. 2000) (concluding “that the jury was properly instructed that conscious avoidance could. . .be used to infer knowledge of the conspiracy's unlawful objectives”).

32. The Supreme Court's statement of the “traditional rationale” employs this broad version of the thesis. Global-Tech, 131 S. Ct. at 2069 (noting that the willfully ignorant “are just as culpable as those who have actual knowledge”). See also Jewell, 532 F.2d at 700 (“deliberate ignorance and positive knowledge are equally culpable”); Heredia, 483 F.3d at 926 (Kleinfeld, J., concurring) (“wilful blindness is ‘equally culpable’ to. . .positive knowledge”); United States v. Stadtmauer, 620 F.3d 238, 255 (3d Cir. 2010); United States v. One 1973 Rolls Royce, V.I.N. SRH-16266 By & Through Goodman, 43 F.3d 794, 808 (3d Cir. 1994) (discussing “the mainstream conception of willful blindness as a state of mind of much greater culpability than simple negligence or recklessness, and more akin to knowledge”); United States v. Rivera, 944 F.2d 1563, 1570 (11th Cir. 1991) (acts of “deliberate ignorance and acts. . .with positive knowledge are equally culpable”).

33. See Hellman, supra note 13, at 305–312; infra notes 39–45 and accompanying text; Michaels, supra note 12, at 986.

34. Larry Aleander and Kimberly Ferzan adopt what essentially amounts to this principle. They “are inclined to say that culpability does vary with the actor's estimate of the probability” of the relevant risks. Alexander & Ferzan, supra note 16, at 38. Charlow also suggests that the criminal law endorses such a principle: “the more certain [the defendant] is that some significant fact exists that will make his conduct criminal, . . . the more blameworthy he is if he goes ahead and acts despite his awareness of that fact.” Charlow, supra note 2, at 1394–1395.

35. To be clear, this principle employs one's subjective confidence in the relevant proposition.

36. Cf. N.Y. Penal Law §150.15.

37. United States v. Verners, 53 F.3d 291, 295 (10th Cir. 1995) (affirming conviction for aiding and abetting where defendant was aware of the drug operation out of her home).

38. The defendant in Heredia argued that she should not be convicted on willful ignorance grounds because she declined to investigate whether her car contained drugs only for safety reasons, given that she was driving on the highway. 483 F.3d at 920. The Ninth Circuit rejected this argument, but noted that “[a] decision influenced by coercion, exigent circumstances or lack of meaningful choice is, perforce, not deliberate. A defendant who fails to investigate for these reasons has not deliberately chosen to avoid learning the truth.” Id. However, this view can only be taken so far. Clearly, it can't be the case that whenever investigating entails risks or costs, then the decision not to investigate is non-willful. After all, almost any decision not to investigate will be the result of perceived pressures against investigating. But this cannot mean that hardly anyone is ever willfully ignorant. Instead, only truly exigent circumstances or coercion (etc.) could plausibly make deciding not to investigate count as non-willful.

39. See Hellman, supra note 13, at 305–312.

40. Id.

41. Id. at 305–306.

42. Id. at 308–309.

43. Hellman thinks these justifications are supposed to be “objective” in the sense that they apply regardless of whether they actually motivated the actor to remain in ignorance. See id. at 307. I discuss this feature of Hellman's view below in Section IV. I want to remain neutral on this aspect of her view for purposes of this article.

44. See Sarch, Alex, Willful Ignorance, Culpability and the Criminal Law , 88 St. John's L. Rev. 1023, 1066–1069 (2014)Google Scholar ; see also Hellman, supra note 13, at 316.

45. To further bolster our intuition about this case, we might also suppose that Patty would not have helped Charles were she certain it would involve transporting illegal items. Thus, she does not satisfy the counterfactual test discussed below in Section III. However, for the reasons explained there, we must be careful about using intuitions about counterfactual conduct when assessing one's actual culpability.

46. To this, one might object that Patty perhaps does not purposefully remain in ignorance, but remains in ignorance only knowingly. As a result, she is perhaps not truly willfully ignorant. In that case, this would not be a counterexample to UECT. As noted above, however, some courts do allow knowingly preserving one's ignorance to count as willful ignorance—especially courts that allow “conscious avoidance” to satisfy the knowledge element of a crime. See supra note 31 and accompanying text. Moreover, even assuming arguendo that willful ignorance requires the purpose to remain in ignorance, Patty still qualifies as willfully ignorant. As stated, Patty deliberately chooses not to investigate whether her son's crate contained illicit items. This is one of her conscious objectives. She aimed at ignorance as the means to achieving her other ends. (In general, to count as bringing about an end E purposefully, one need not aim at E as an end in itself; rather, it suffices for one to aim at E as a means to other ends one has. For example, if one kills one's uncle merely as the means to getting one's inheritance, the killing still would count as purposeful. See also Warren Quinn, Actions, Intentions and Consequences, 18 Phil. & Pub. Aff. 334, 336 (suggesting that bombing civilians as the means to ending a war is an intentional action that is harder to justify than bombing a munitions factory for the same reason while merely knowing that it will cause civilian casualties).) Since Patty aims at ignorance as the means to preserving her relationship with her son, she remains in ignorance purposefully. So she counts as willfully ignorant.

47. I mean for this phrase—“have good reasons that help justify the decision to remain in ignorance”—to be neutral between Hellman's objectivist view and the subjectivist view that one's actual motives matter to the legal notion of criminal culpability. Perhaps having such a reason requires only being in circumstances that help justify remaining in ignorance, or perhaps it requires that the justifying considerations actually were what motivated one to decide not to investigate. Either view is covered by this phrase. See Section IV for more discussion.

48. Husak & Callender, supra note 8, at 40 (emphasis added).

49. Id.

50. United States v. Willis, 277 F.3d 1026, 1032 (8th Cir. 2002) (requiring that the defendant, to be willfully blind, had to have “purposely contrived to avoid learning all of the facts in order to have a defense against subsequent prosecution”); United States v. Delreal-Ordones, 213 F.3d 1263, 1268 (10th Cir. 2000) (same); United States v. Puche, 350 F.3d 1137, 1149 (11th Cir. 2003) (same).

51. Hellman, supra note 13, at 310–311.

52. Id.

53. I defend the motive-insensitivity of the legal notion of culpability in other work. See Sarch, supra note 18, at n.89.

54. If Hellman's objectivist view is adopted, we can put the point by saying that some actors who preserve their ignorance because they desire to preserve a defense might nonetheless be in justifying circumstances that support their ignorance (regardless of whether these circumstances actually are what motivated them to remain ignorant).

55. Sarch, supra note 44, at 1073–1074.

56. Id.

57. See, e.g., Jandro v. Ohio Edison Co., 167 F.3d 309, 316 (6th Cir. 1999) (“employer [wa]s virtually certain that harm [wa]s about to occur but cho[se] to ‘look the other way’ in the interest of continuing the job”); Fiore v. Comm'r, 105 T.C.M. (CCH) 1141, at *30 (Tax Court 2013) (holding that the defendant committed tax fraud because he was willfully blind “for the purpose of getting and keeping clients”).

58. One might try to save RECT1 from this problem by recasting it just as sufficient condition for equal culpability. But that would be unsatisfying because then it could no longer plausibly be a comprehensive account of the conditions in which willfully ignorant conduct is as culpable as knowing misconduct. Moreover, it would still suffer from my first objection.

59. Michaels, supra note 12.

60. Id. at 957.

61. Id. at 995.

62. Luban, supra note 11, at 968–969.

63. Id. at 969.

64. Id.

65. Id.

66. Id.

67. Id. at 969 (The fox, “who aims to do wrong and structures his own ignorance merely to prepare a defense, has the same level of culpability as any other willful wrongdoer—the highest level, in the [MPC] schema.”).

68. Id.

69. Id.

70. Simons, Kenneth W., Does Punishment for “Culpable Indifference” Simply Punish for “Bad Character”? Examining the Requisite Connection Between Mens Rea and Actus Reus , 6 Buff. Crim. L. Rev. 219, 267–275 (2002)Google Scholar.

71. Simons notes that appealing to “counterfactual culpability is highly problematic.” Id. at 269. To illustrate, he uses a version of Alexander's case of Deborah who engages in Sunday driving within the speed limit. Id. at 269–270. Suppose she is so callous that “even if she were to come to believe that the risks imposed by her Sunday driving were much higher than she actually believes them to be, she would continue to drive notwithstanding that belief. . . . Deborah is not acting recklessly at this moment and is therefore not culpable. She has an undesirable character trait, one that might dispose her to act recklessly if the circumstances were right, but her counterfactual culpability does not translate into present culpability.” Id.

72. Simons, supra note 70, at 233–234.

73. Id. at 232.

74. In Luban's terms, Jeff is akin to a fox who simply would not be able to do the actus reus with full knowledge.

75. See Hellman, supra note 13, at 312.

76. Id. at 313.

77. Id. (emphasis added).

78. To be fair, Hellman might be read as aiming only to lay out the conditions under which willful ignorance is culpable at all—not to specify when willful ignorance followed by a subsequent reckless actus reus is as culpable as the analogous knowing misconduct, which is my focus here. Thus, it seems Hellman plausibly has identified a necessary condition for the equal culpability thesis to be true—namely, that one's decision to remain in ignorance is not justified—but it is not a sufficient condition.

79. Hellman, supra note 13, at 310–312.

80. Alexander and Ferzan endorse this view as well. Alexander & Ferzan, supra note 16, at 60–61. I also defend such a view. See Sarch, supra note 18.

81. Hellman, supra note 13 at 315 (emphasis added). See also id. at 313 (“we ought. . .to ask whether there are good reasons for the actor to choose blindness rather than focus on what reasons actually motivated the actor in choosing blindness”).

82. See Sarch, supra note 14.

83. Note one limitation of my account, which has been implicit in the discussion thus far. I only aim to explain what makes defendants with a sub-knowledge level of confidence in the inculpatory proposition, p, as culpable as a knowing wrongdoer. Some have suggested that one can be willfully ignorant despite knowing p. See Husak & Callender, supra note 8, at 48. I am skeptical of this view, however, because if one has enough certainty in the inculpatory proposition to count as knowing it, then it seems implausible to describe one as ignorant of that proposition. Moreover, someone who actually knows p can be straightforwardly convicted of the knowledge crime in question (provided the evidence suffices). Indeed, it is only when one has a sub-knowledge level of certainty in p that any puzzle arises about how one can be as culpable as a knowing wrongdoer. This is the puzzle I aim to answer.

84. Arpaly & Schroeder, supra note 16, at 237.

85. Id.

86. Id.

87. Others have discussed similar duties to investigate, but none have systematically developed the precise duty I defend here. Holly Smith mentions the case of a driver who “should have checked his mirror earlier, but given that he did not, he should check it now rather than back down the driveway.” Smith, Holly, Culpable Ignorance , 92 Phil. Rev. 543, 546 (1983)CrossRefGoogle Scholar. See also George Sher, Who Knew? Responsibility Without Awareness (2009), at 111–112 (discussing the idea that one's moral obligations can give rise to secondary epistemic duties to be or become aware of morally relevant features of one's situation); William K. Clifford, The Ethics of Belief, in The Ethics of Belief and Other Essays (T. Madigan ed., 1999), at 70–96 (famously defending—in 1877—a broad and weighty moral “duty of inquiry”).

Tadros also posits a duty to investigate, and like me, he thinks the duty can be reduced in stringency where investigating is dangerous or difficult. Tadros, supra note 16, at 246–247. However, Tadros's duty to investigate is much broader than mine, covering not only actors who actually have suspicions of the risk to be investigated but also actors who do not. Thus, his duty to investigate also covers some instances of negligent unawareness of the risk (or what U.K. courts call objective recklessness). By contrast, my duty to investigate applies only where one already has some suspicions or awareness of the risk (i.e., is subjectively reckless). Thus, my account focuses on the failures of inquiry that plausibly can provide a basis for treating one as a knowing wrongdoer. (See also infra note 104.)

88. See, e.g., Keating, Gregory, The Priority of Respect over Repair , 18 Legal Theory 293, 309 (2012)CrossRefGoogle Scholar (observing that “[r]emedial responsibilities arise out of the breach of antecedent primary duties” and criticizing the corrective justice theory of tort for not capturing the priority of primary duties over the secondary duties they generate); Gardner, John, What Is Tort Law For? Part 1: The Place of Corrective Justice , 30 Law & Phil. 1, 33–35 (2011)CrossRefGoogle Scholar.

89. One might think that the duty to investigate arises not just when one intends or plans to perform the underlying risky action, but even when one merely seriously considers whether to perform it. I leave that question open here.

90. More precisely, there is such a duty here only assuming the law is just. The duty to reasonably investigate, to be normatively defensible, ideally should not be triggered if the criminal statute in question is substantively unjust. Of course, the law often is unjust. Thus, the law might impose what amounts to a duty of investigation in conditions when, normatively speaking, it should not. (Perhaps this is the case with certain harsh drug laws in the United States, for example.) Does this mean courts must evaluate whether the law is just before applying the willful ignorance doctrine? (Thanks to an anonymous reviewer for this worry.) The answer is no. Courts themselves should not have to make such a determination. Given their role, their main task is to apply the law as given. However, my question as a theorist is different from the court's. Since I am concerned with normative culpability (i.e., the amount of culpability that the law should attribute), all I am claiming is that there should (from the theorist's point of view) be a legal duty to investigate only where the law actually is just (again, as evaluated from the theorist's point of view).

91. It is conceivable that the investigations one has a duty to perform are not just external investigations aimed at collecting more information. I want to leave open the possibility that the required investigations merely involve further processing or reflection on information one already has. For example, one might see several red flags, but then consciously decide not to think about the matter any further to prevent oneself from putting the pieces together and becoming certain that one's planned conduct would cause harm. Perhaps this could count as willful ignorance.

92. The term “planning” is included here because it may well be possible that some lesser degree of commitment, short of an irrevocable intention to do the underlying act, also triggers the duty to investigate.

93. One might worry that this gives Brad an easy way to sidestep his duty to investigate: perhaps he could simply withdraw his intention to burn down the building until the very last moment, and thereby avoid incurring any secondary duty to investigate before starting the fire. However, on closer inspection, such a strategy would never succeed. Were Brad to adopt such a scheme, he would not genuinely be abandoning his plan or intention to burn down the building. Instead, he would merely be executing an elaborate plan to burn it down that involves a generous helping of self-deception. Accordingly, the intention that triggers the duty to investigate would remain in place.

94. Because the duty of investigation is triggered only when the defendant believes the act he is planning would pose a substantial and unjustified risk of harm or illegality (or an unjustified risk that the inculpatory proposition is true), DRI can accommodate cases where exigent circumstances prevent the need to investigate from arising at all. Suppose Laura knows the building must be burned down to halt a fire that is rapidly spreading toward town, but she realizes that there is little time to check if someone is inside. It is not impossible to do so, but it would leave less time for other important precautions (e.g., warning the town authorities). One might think the best way to describe this situation is not to say that Laura would be justified in breaching the duty of investigation, but rather that such a duty does not arise at all. It is to leave room for this possibility that DRI says that the duty to investigate arises only if the substantial risk one is aware of is unjustified.

95. Thanks to Kimberly Ferzan for pressing me on this issue.

96. What about the case where one actually does not know one can investigate, but even if one did, one wouldn't do it? DRI entails that if one reasonably should be aware of available investigative steps, one has a pro tanto obligation to use them. But I take it that the counterfactual truth that one would not investigate even if one knew one could doesn't directly bear on one's culpability—for the same reasons I rejected the counterfactual test in RECT2.

97. One might doubt that this is true in the event that one only unreasonably believes one can investigate. This is a difficult issue that I do not have room to fully settle here. I have formulated DRI so that one's pro tanto obligation to investigate is triggered even by the unreasonable belief that there are feasible ways to do so, but I acknowledge that this might be controversial. My main argument in any case does not depend on it.

98. Arpaly & Schroeder, supra note 84, at 238.

99. A related worry is discussed below. See infra notes 105–108 and accompanying text.

100. It is a simplifying assumption of my model that when the inculpatory proposition p is true, investigating would lead one to acquire greater certainty that p, while when p is false, investigating would lower one's certainty that p. However, this will not always be accurate. It is possible that p is false, but investigating under the circumstances would yield evidence that raises one's credence in p. However, I set aside this complication here to keep the model simple. It can be expanded to capture this additional wrinkle if needed.

101. Cf. Alexander & Ferzan, supra note 16, at 171 (noting that under current law, “resulting harm makes an actor” subject to greater punishment, but arguing that this is problematic).

102. This might help explain why some courts require that the defendant take affirmative steps to avoid knowledge in order for a willful ignorance jury instruction to be appropriate. See, e.g., United States v. Giovannetti, 919 F.2d 1223, 1228 (7th Cir. 1990). It would be another way to home in on especially egregious cases of willful ignorance.

103. See Sarch, Alex, Beyond Willful Ignorance , 88 U. Colo. L. Rev. 97, 121–122 & n.79 (2016)Google Scholar (discussing different ways the equal culpability thesis could be understood).

104. This highlights how my account of willful ignorance is importantly different from Tadros's, although it also relies on a duty to investigate. Cf. Tadros, supra note 16, at 246–247, 258–261. My account, premised on the narrow investigative duty DRI, is designed to answer a specific puzzle about how it can be that some reckless actors (who are aware of a risk that the inculpatory proposition is true) can nonetheless be treated as if they possessed knowledge. Thus, my account singles out a particularly egregious kind of (subjective) recklessness that can substitute for knowledge—as is needed to vindicate the willful ignorance doctrine. By contrast, Tadros is interested in whether willful blindness ever “manifests the degree and kind of vice for which the attribution of criminal responsibility is appropriate,” and he argues that it can. Id. at 258. I do not take issue with this claim. But Tadros is not concerned, as I am, to home in on the worst cases of willful ignorance, which surpass mere recklessness and are as culpable as the analogous knowing misconduct. His account seeks to distinguish cases that are not appropriate for punishment at all from those that are. My account, on the other hand, focuses on cases that are already within the proper scope of criminal liability and defends a claim about precisely how culpable they are.

105. Thanks to Jake Ross for this objection.

106. See, e.g., Jackson, Frank, On the Semantics and Logic of Obligation , 94 Mind 177 (1985)CrossRefGoogle Scholar (discussing the “standard approach” to the semantics of ought claims, and conditional ought claims in particular).

107. Others have raised related worries for the standard view of conditional ought claims. See id. at 179–180.

108. Thanks to Gideon Yaffe for this response.

109. Thanks to Jake Ross, Steven Schaus, and Gideon Yaffe for pressing me on this point.

110. See supra notes 70–72 and accompanying text.

111. Alex Sarch, The Moral and Legal Contours of Culpable Ignorance (manuscript on file with author), at ch. 6 (defending a general theory of mens rea substitution that builds in a relatedness constraint of this sort).

112. Holly Smith has also raised an interesting objection to duties to inform oneself, at least insofar as they are “subjective duties”—i.e., ones that arise in virtue of what one believes. See Smith, Holly, The Subjective Moral Duty to Inform Oneself Before Acting , 125 Ethics 11, 23–31 (2014)CrossRefGoogle Scholar. She argues that it can't be the case that the sole reason we subjectively ought to investigate is that it would enable us to subsequently do the action that has the most “deontic value.” After all, even though keeping a weightier promise has more deontic value than keeping a less weighty promise, it does not follow that one has an obligation to go around making more weighty promises. Likewise, one does not have a duty to go around investigating things just so that one will be able to do actions with more deontic value. I think Smith is right that there has to be more to it than that.

However, my view avoids Smith's objection. The trouble with the investigative duty she considers is that it takes the duty to investigate to be purely derivative of the deontic values of the subsequent acts that investigating, or not, would enable one to perform. However, as we saw, my view makes the duty to investigate non-derivative in Smith's sense. That is, I argued that there is an independently existing reason to inform oneself: namely, that not doing so can manifest insufficient regard for the protected interests or rights of others. Thus, on my view, there is something in itself good (or at least culpability-avoiding) about informing oneself in reasonable ways. It shows a kind of seriousness, conscientiousness, and respect toward others.

113. Such jury instructions generally may be given only if it can reasonably be inferred from the trial evidence that the defendant was willfully ignorant. See, e.g., United States v. Anthony, 545 F.3d 60, 64 (1st Cir. 2008); United States v. Reyes, 302 F.3d 48, 55 (2d Cir. 2002); United States v. Abbas, 74 F.3d 506, 513 (4th Cir. 1996).

114. I discuss some such simplifications elsewhere. See Sarch, supra note 44, at 1094–1101.