Where attainable knowledge could have changed the issue, ignorance has the guilt of vice.
—Alfred North Whitehead
We make moral judgments about a wide range of things—motives, acts, practices, emotions, relationships, character traits, and so on—and among them are moral judgments about beliefs, knowledge, and ignorance. “She should not have known about the merger.” “He should have known where his child was.” These judgments go beyond prudence to imply an epistemic ethic in which we may be morally responsible for what we do and do not know. Indeed, in the examples given, one may be legally responsible as well: insider trading and parental negligence are crimes.
Four factors affect our moral assessment of knowledge and ignorance: process, content, purpose, and context.1 Each can be morally scrutinized; together they color our overall assessment. Process refers to the means by which one pursues and acquires knowledge, or to the causes of ignorance. The way in which one comes to know may be reprehensible: ethical violations may occur in research procedures, in violations of confidentiality or privacy, in the theft of proprietary information, and in other aspects. The negative judgment about one’s ignorance will be more severe if the ignorance was calculated or willful.
Cognitive content may be ethically significant independent of the methods used to acquire it. The knowledge may be forbidden or restricted or hazardous in itself; it may be worthless, trivial, or disgusting; or it may be knowledge that, however acquired, is embarrassing because it violates someone’s privacy. Ignorance may be ethically significant when one should have known specific information that is salient to a decision or action.
One’s purpose in seeking the knowledge is also significant. One’s ultimate purpose for knowing various ways to make a bomb or how to clone a human being may taint the moral worth of knowing; prurient interest taints clinical knowledge. Similarly, since nescience is purposeful, one’s purpose—self-protection, fear, privilege, manipulation of others, and so on—is ethically relevant. The strategic ignorance of a leader who seeks deniability while directing subordinates to act may be ethically questionable.
Context includes many subtleties, including, among other factors, our relationships and roles; the rights of privacy and confidentiality of others; the nature of the immediate social context; and the standards and expectations of the relevant epistemic community. A professional is responsible for knowing specialized information that the public is not; a father should know things about his children that his neighbor need not know, perhaps should not know. Our moral judgment of their ignorance will likely vary depending on their roles and relationships and other such contextual aspects. Context also affects the appropriateness of delivering information or expressing one’s knowledge. “Telling tales out of school,” divulging classified information, and similar acts of knowledge dissemination are judged inappropriate or unethical based on context—however truthful they may be.
Keeping these four factors in mind, let us turn to a hierarchy of claims one can make regarding knowledge or ignorance, and to the ethical considerations that arise. Our concern is not with the epistemology of morality, but with moral epistemology; and that concern begins with belief.
Beliefs are factive; they aspire to truth. It would be absurd, as the British philosopher G. E. Moore observed, to say “It is raining, but I don’t believe it is raining.”2 To believe is to take to be true. Beliefs may be false, however, and they may be false without being morally wrong. Yet there are beliefs we judge to be morally wrong. Among likely candidates: beliefs that are sexist, racist, or homophobic; the belief that proper upbringing of a child requires “breaking the will” and severe corporal punishment; the belief that the elderly should routinely be euthanized; the belief that “ethnic cleansing” is a political solution, and so on. Note that we condemn not only the potential acts that spring from such beliefs, but the content of the belief itself, the act of believing it, and thus the believer.
Making these moral assessments and holding individuals responsible for their beliefs imply that believing is a voluntary act. Or so it seems. It requires special circumstances to consider someone responsible for an act that was not voluntary—negligence, for example, or complicity through the foreknowledge of an act, or action from ignorance that is itself reprehensible. But some beliefs, like some personal values, seem not to have been chosen; they are “inherited” from parents and “caught” from peers, acquired inadvertently, inculcated by institutions and authorities, or assumed from hearsay. For this reason, I think, it is not always the coming-to-hold-this-belief that is the problem; it is the reflective maintaining of such beliefs along with the refusal to disbelieve or discard them that may be voluntary and ethically wrong.
If the content of a belief is judged morally wrong, it is also thought to be false. The belief that one race is clearly inferior or not fully human is not only a morally repugnant, racist tenet; it is also thought to be a false claim—though not by the believer. The falsity of a belief is a necessary but not sufficient condition for a belief to be morally wrong. Neither is the ugliness of the content sufficient for a belief to be morally wrong. There are morally repugnant truths, but believing them does not make them so. Their moral ugliness is embedded in the world, not in one’s belief about the world.
There are irresponsible beliefs; more precisely, there are beliefs that are acquired and retained in an epistemically irresponsible way. One may disregard evidence; accept gossip, rumor, or testimony from dubious sources; ignore incoherence with more embedded beliefs; or display cognitive bias—these are examples of what might be called doxastic dereliction.3 I do not mean to revert to the evidentialism of the mathematical philosopher William K. Clifford, who claimed, “It is wrong, always, everywhere, and for anyone, to believe anything upon insufficient evidence.”4 Clifford was trying to prevent irresponsible “overbelief,” in which faith, wishful thinking, or sentiment (rather than evidence) stimulate or justify belief. But as William James showed, some of our most important beliefs about our world and the human prospect must be formed without the possibility of sufficient evidence. In such circumstances—which are sometimes defined narrowly, sometimes more broadly in James’s writings—one’s “will to believe” entitles us to choose to believe the alternative that projects a better life.5
Unfortunately, in today’s culture of ignorance, many people seem to have taken great license with the right to believe, flouting epistemic responsibility. The willful ignorance and false knowledge that are defended by the assertion, “I have a right to my belief,” do not meet James’s requirements. Rather, the right to believe is proclaimed as a negative right; that is, its intent is to fend off epistemic challenges, to enjoin others from interfering with the formation and holding of the believer’s beliefs. But, as Clifford also remarked, “No one man’s belief is in any case a private matter which concerns him alone.”6 Beliefs guide motives, choices, and actions. Usually, the public-sphere claimant is arrogating more than belief-rights, and also wants the freedom to act on those beliefs. The “right to believe” is used as a shield against the normal interactions of an epistemic community; it is meant to obstruct or shut down discourse.
The ethical issues that arise for ignorance concern vincible ignorance, the sort that prefigures knowledge that might be gained. The moral issues of not-knowing are correlated with, but not exhausted by, the implications of knowing. For vincible ignorance, the possibility of knowing permits an array of claims and judgments that form a hierarchy or progression.7 Let’s look at this progression in simple schematic form, then quickly move to examples. (You may ignore the parenthetical designations in this discussion without loss of meaning, if you choose.)
For any individual or group (S) and any knowable thing (X), one may claim that:
One may deny any of these claims, of course, claiming that “S does not have (the option, desire, need, or right) to know X.” (Label these ~A to ~E, respectively.) Logically different and more interesting for our purposes, however, are these affirmations of ignorance:
These affirmations of ignorance serve to protect the border of our knowledge, even to defend the barricades, thereby constructing or maintaining ignorance.8 It will require patience to parse these differences, but we need a simple scenario, a nonschematic description that illustrates the progression clearly. So let us imagine Tom, who is exploring an unfamiliar cuisine and has selected a soup with an exotic name, which he is now sipping. At least in principle, Tom has the option of knowing the ingredients of the soup. This implies, of course, that Tom also has the option of not knowing. (Statements A and An are practical equivalents; that is the nature of options.) In practice, Tom may also be denied the option of learning the ingredients (~A): if, for example, the recipe is kept secret or if the soup was the culinary improvisation of a forgetful chef.
If Tom has an interest in cooking, he may desire to know the ingredients (B). Of course, if he is preoccupied or indifferent to his food, he may have no such desire. But not desiring to know (~B) is different from desiring not to know (Bn). Though both seem directed toward ignorance, there is no desire present in the former; there is a specific desire in the latter. Tom may be enjoying the taste of his soup, but—being squeamish in general and suspicious about what he is now ingesting—Tom desires not to know what is in the soup (Bn). The presence of desire is motivational: in B, it is directed toward learning; in Bn, it is directed toward deliberate ignorance. (Statement ~B, not desiring to know, suggests a neutral affect and carries no epistemic motivation.)
Suppose now that Tom needs to know the soup’s ingredients (C). We commonly distinguish between a desire or want and a need—a distinction familiar to any parent who has taken a small child shopping—though we may conflate them when our desire is intense. (“I really need those wild shoes!”) A genuine need, however, is tied to one’s interests. One’s needs and desires can be quite divergent. Tom may need to know the ingredients of the soup if has a severe food allergy—whether he particularly wants to know or not. He may also need to know them if his job is to review the restaurant, or if he is tracking the use of salmonella-infected produce, or if he is preparing a pot of the soup as substitute chef. These latter examples show that not only one’s interests but also one’s purposes may give rise to needs, especially those purposes that are related to roles and obligations and secure the means to a sanctioned end. Our needs also vary in their scope (how much it would take to satisfy the need) and in the value of their fulfillment (how important it is to meet that need). We could imagine an intense need devolving from a project that is not really important. Needs form hierarchies that range from basic survival needs to more elaborated and contextual needs; they are arrayed in hierarchies of value and sequence.
The health authorities responsible for emergency medicine and public health need to know what was in Tom’s soup if everyone who ate it became ill. It is essential information to enable them to function competently and fulfill the purpose of their roles. Note that asserting a need always carries an implicit conditional: one needs to know what was in the soup if one is to determine the cause and treatment of the resulting illness; or, at the most fundamental level, one needs oxygen if one is to survive. Thus, although the assertion of a need seems to be a descriptive report, it is value laden: the need is asserted and evaluated in reference to the importance of the conditional. The need for oxygen carries great weight only because we value life so highly, and life is basic to the fulfillment of other needs. If Tom claims that he needs to know the ingredients in the soup in order to replicate it at home, his need carries little moral weight: his intended project is not vital; his cardinal interests are not at stake.
The concept of need serves both as a solicitation and a restraint. While one “in need” may seek its satisfaction or even have thereby a moral claim on resources, one’s need may require certification. “Demonstrated need” is often a requirement for various forms of assistance, and the burden is on the needy. Corporate, military, and intelligence organizations may share information only on a “need to know” basis, and justification of such a need will be required of anyone seeking to know. A desire may require explanation, but when one elevates it to a need, justification may become necessary.
Can we imagine that Tom needs not to know what went into the soup (Cn)? Yes: for example, he may be attempting a blind taste test to determine whether a more expensive ingredient makes a difference in taste. Research in medicine, psychology, and the social sciences often relies on double-blind experiments and trials in which there is methodological need for subjects not to know key elements of the experiments—whether they fall in the placebo group or not, for example; or if the experiment actually tests something different from what they are told, because the research design requires their ignorance as a precondition for valid data.
Claims about having options, desires, or needs to know or not to know are descriptive assertions—though “need” may, as I say, gain normative force from its implied conditional. But standing alone, they do not imply that it is good or bad for one to learn X. A burglar may truly need to know the combination to the safe if he is to steal the jewels.
When we move to claim a right, however, the assertion has more normative force—and how much force depends on the nature of the right. Different sorts of rights are established in different contexts: contractual rights, parliamentary rights, legal rights, parental rights, moral rights, and human rights differ in their grounding and in the sort of normative weight they carry. In any event, having a need to know does not lay a claim on others in the way that having a right to know does (D). Asserting a right declares an entitlement and enjoins others to divulge the relevant information or to refrain from interfering in one’s inquiry. And having a right not to know (Dn) serves to enjoin others not to reveal the information, at least not to the rights-holder.
Let’s get back to Tom. Tom may indeed have the right to know what is in the soup (D), but we would need to imagine a context that creates that entitlement. Perhaps the soup made Tom ill, and he suspects that, despite denials, the chef added monosodium glutamate, a flavor enhancer to which he allergic. So far, we have only Tom’s need to know. But if Tom gains a court order requiring the chef to specify all the ingredients, he would have the (legal) right to know. Farfetched as this example may be, there are many cases in which courts determine whether an individual or group has a right to know something. Does an adult who was adopted as an infant have the right to know the identity of her birth parents? Does a divorcée have the right to know the full value of her ex-husband’s assets? Do police who stop a speeding car have a right to know what is stored in the trunk? Courts create and enforce legal rights; moral rights are “recognized,” either for a specific context or as a concomitant of human life.
Sometimes we speak broadly of a universal human right to know. It is essentially an affirmation of freedom of inquiry, our freedom to learn and to choose what we will seek to know. This, in turn, obligates others not to prevent the pursuit or acquisition of knowledge, not to restrict education and inquiry. Within the catalog of human rights, the right to know seems to be fundamental. But its scope is surely not universal: everyone does not have the right to know everything. In the first place, no individual can learn “everything.” We cannot be entitled to something that is impossible. That is why the general right to know devolves to a right to free inquiry. More importantly, an individual’s seeking to know may be trumped by certain rights or obligations of others, such as the right to privacy and the obligation to confidentiality. In some cases, I may have a prima facie right to know that is overridden by other considerations: for example, although Sheila has a prima facie right to know the identity of her biological father, a priest may refuse to divulge the information, under the obligation of confidentiality. But in many cases, it seems more appropriate to say that the right to know did not exist, even prima facie. Do I really have a right to know whether my neighbor plans to install a swimming pool or what my friend paid for her new car? Surely not. In these sorts of cases—matters that are relatively inconsequential for me—the urge to know is hardly a right, not likely a need, and more like a nosy desire.
What are the factors that create one’s right to know a particular X? The relevance of the knowledge to one’s self-interest or well-being seems an obvious condition, though it is less obvious how one is always to determine the relevance of information before actually obtaining it. It is difficult to predict what known unknowns, let alone unknown unknowns, will be salient to my self-interest; however, information in certain spheres of personal importance, such as health, personal finance, family relationships, and conditions of employment, is clearly pertinent. Relevance to personal well-being is the rationale for the “Right to Know” movement that was inspired by Rachel Carson’s environmental classic, Silent Spring.9 Indeed, the legal principle that an individual has the right to know the chemicals to which she may be exposed in her job or in daily living because some are dangerous to well-being has been used to frame federal workplace law and community environmental law.10
Rights are also created by speech acts and their written elaborations: promises, agreements, contracts, treaties, by-laws, constitutions, and so on. In addition, as intimated earlier, they may be attached to social or official roles: parents, doctors, teachers, counselors, judges, jurors, buyers, government officials, club officers—all these have special rights, including epistemic rights that devolve from their respective roles.
Do we also have a right not to know? It may seem that this is implicit in the right to know: after all, just because one has the right to know X doesn’t mean one has the obligation to know X. But, to be logically precise, that inference conflates not having the right to know X (~D) with having the right not to know X (Dn). Notice that the former (~D) denies a right, while the latter (Dn) affirms a right. The right not to know may require special affirmation, since it would enjoin others not to inform or to divulge specific information. In such cases, informing me would violate my right.
The right not to know (Dn) is articulated and defended today largely in bioethical contexts. A patient may assert her right not to know whether she has a degenerative genetic condition such as Huntington’s chorea, arguing that, if she were to learn of a positive diagnosis, the knowledge would create extreme stress and hamper her remaining life. The rationale for a right to ignorance is thus tied to protecting one’s autonomy, not just in the choice of not-knowing, but also in regard to the self-determination that would be altered afterward by the knowing. Knowledge affects us, and some knowledge about ourselves can be debilitating; it can sap our confidence and hope, diminish our motivation, and drain the joy and fullness of our experience. Once known, it cannot be unlearned. Especially if there is nothing we can do about it anyway, we might well assert our right not to know the truth, our right to ignorance. Like other rights claims, the entitlement not to know is subject to the rights and obligations of others—a circumstance that has made it a contested claim in bioethics. Some genetic information is relevant not only to the patient, but also to the patient’s children, spouse, or other family members. Their right to know, which does not depend on their explicitly claiming it, may trump the patient’s right not to know.
Consider this case:
Barbara, a 35 year old woman and mother of two children, has a family history of breast cancer. Urged by her relatives, she decided to undergo the BCRA1/2 testing. If Barbara has the mutation, she has 80% risk of developing breast cancer. Three days later, depressed by the difficult decisions she would have to make in case the mutation was found, she asked the doctor not to inform her about the test results.11
Several epistemic claims are implicit in this case. Barbara asserts her right not to know; moreover, she assumes that right overrides her doctor’s duty to inform. The relatives who urged Barbara’s screening have a stake in the results, so a case might be made for their right to know the results. They would be affected by a cancer diagnosis and its prognosis, especially so her spouse and children. If Barbara does have cancer, her nescience will likely have dire consequences—which might have been prevented or ameliorated. Her test results may bear on the medical prospects of her female relatives. If so, Barbara may have the obligation to inform them, which entails her obligation to know. Whatever the results, they will otherwise remain in anxious ignorance. From this perspective, Barbara is “in denial,” avoiding the truth out of fear, and violating her epistemic responsibilities.12 In most medical situations (but not all), the decision to inform others is left to the patient; the doctor’s obligation to inform is directed only to the patient (though the doctor may urge a patient to tell affected others). Clearly, such situations, which are quite common, are ethically complex.
In recent years, there has been formal, even legal, recognition of the right not to know. In 1995, the World Medical Association, in its amended “Declaration on the Rights of the Patient” declared that “the patient has the right not to be informed on his/her explicit request, unless required for the protection of another person’s life.” Two years later, the European Convention on Human Rights and Biomedicine stated: “Everyone is entitled to know any information collected about his or her health. However, the wishes of individuals not to be so informed shall be observed.” And the UNESCO Universal Declaration on the Human Genome and Human Rights proclaimed: “The right of every individual to decide whether or not to be informed of the results of genetic examination and the resulting consequences should be respected.”13
These declarations are similar, though not congruent. The shared assertions are: (1) that there is a right not to know; and (2) that it applies to patients in regard to their own medical information (though the UNESCO assertion is restricted to genetic information). The rationale, more or less explicit, is that the patient’s right to self-determination underlies these assertions. It also seems that, while the right to know does not require an explicit claim, the right not to know does: the patient must explicitly request the doctor to withhold the information.
In some situations, however, the patient cannot make the request in advance. Suppose a child needs a kidney replacement and family members are tested as possible donors; the tests disclose incidentally that the father is not the child’s biological parent. If the father anticipated this possibility beforehand, he might choose not to know; if he did not imagine this result, he has no opportunity to claim his right. There is no way now for the doctor to place the choice before the patient without implying that the information is significant; and the doctor may wonder whether the father or the mother has the greater stake in the information.
While I accept the recognition of a right not to know in specific circumstances, I think one’s claim to such a right is deflated when one decides to undergo medical testing. So, it is not an inalienable right; it can be undone by one’s own actions. Moreover, I am wary of the extension of the right not to know into other contexts. It easily becomes a justification for epistemic irresponsibility, for willful ignorance and epistemic injustice.
Are there spheres other than the medical in which the right not to know might apply? Might a student claim a right not to know his grades during a course on the grounds that knowing them would affect his performance? Unlikely, perhaps; but it’s possible. Might a bride claim a right not to know the appraised value of her rings, not wanting the knowledge of their monetary value to affect her sentiments? In adopting a child, might prospective parents claim a right not to be told anything about the child’s history or biological parents—whether the information offered is genetic, medical, or socioeconomic—on the grounds that it would bias their parenting? What is apparent in such cases is how easily a desire not to know something may be elevated to a need and asserted as a right that obligates others.
My wariness about the right to ignorance turns our attention to the epistemic claims of the greatest weight: the obligations to know and not to know (E and En). Asserting the obligation to know (E) denies the right not to know X (Dn). Having the obligation to know X means that one is responsible for knowing X and would be in violation or at fault for not possessing that knowledge. The parent who doesn’t check the label of a cleaning product and therefore takes no precautions to prevent a toddler’s poisoning is negligent. Whitehead’s principle applies: “Where attainable knowledge could have changed the issue, ignorance has the guilt of vice.”14 Imagine the worst scenario in the case of Barbara’s medical tests: the suppressed information is that she indeed has breast cancer, and her ignoring it results in her death. If receiving the results would have led to treatment that “changed the issue” and saved her life, her ignorance is reprehensible.
Obligations are created within the same contexts as rights; they may be attendant to agreements and promises (implicit or explicit, formal or informal), endemic to various roles and positions, established in law, or incurred as a moral agent. As we have seen, normal interaction among adults in an epistemic community presupposes that members know certain things. Ignorance of this common knowledge may be merely embarrassing to oneself and surprising to others, but—especially if the ignorance has the potential to harm oneself or others—it becomes a more serious lapse, a failure to know what one should have learned. We may, in short, have both an obligation to know X and the responsibility that comes with knowing it. One might argue that Barbara has an obligation to know the test results and to inform others. And remember our Tom? One might argue that Tom is obligated to know the ingredients of the soup if he is serving it to a group known to include ethical vegetarians and individuals with food allergies.
There are times when we hold persons morally and/or legally responsible for their ignorance. The “ignorance of the law is no excuse” principle applies broadly across many jurisdictions. The same applies morally: obliviousness to the harmful impact of one’s actions on others does not deflect blame—it may intensify it. In many cases, ignorance is reprehensible because one could reasonably be expected to have known: “Any normal person would have known” that remark was insulting, that vehicle was not safe, or that tactic would produce civilian casualties. In other situations, however, the ignorance is a violation of a clear obligation to know: the physician should have known the drugs were incompatible; the teacher should have known the child showed signs of abuse. There is also a large domain in which we are not obligated to know, we could not easily have known, and yet we are responsible. For example, unbeknownst to the owner, there is a buried fuel tank on her property; it leaks and contaminates a neighbor’s well. Though the cause was an unknown unknown, the owner may still be held responsible.15
Conversely, the same sources, contexts, and derivative responsibilities apply to the obligation not to know (En). There is one exception: there is no natural, universal obligation not to know. One might argue, for example, that all moral agents have the obligation to know the reasonably expectable consequences of their actions, but there is no parallel argument for content that they must not know. The obligation not to know is specially created; it arises only in constructed contexts. Acting as a blind referee or judge, waiving one’s right to access letters of reference, promising not to read a diary—such situations carry the obligation not to know certain things. Jurors routinely have such a special obligation: they are enjoined not to learn certain information; they may be sequestered to protect and ensure their ignorance.
As an obligation, this negative epistemic imperative is quite odd. In usual cases of obligation, if S has an obligation, its force is directed toward S; it enjoins S to do certain things (and in a minority of cases, not to do certain things). But S’s obligation not to know X not only directs S’s conduct, it also serves to enjoin all others not to inform S about X, directly or indirectly. One who deliberately gives a juror illicit information may be held responsible, even charged with a crime; and whether the juror comes to know by design or by accident, the obligation is violated and juror must disclose that fact to the judge. A breach of this negative epistemic obligation usually disqualifies a juror. My obligation not to know may lead me to refrain from seeking information, to deflect willing informants, and even to take protective steps to avoid accidentally receiving forbidden information. Fortunately, obligatory ignorance is typically directed toward an explicit and rather narrow slice of knowledge.
I referred earlier to a doctor’s obligation to inform a patient of diagnoses and other findings regarding the patient’s health. It is a familiar example of another epistemic obligation: the obligation to report or to inform. Many such disclosure obligations are recognized in professional codes of conduct or institutionalized in laws, and those who keep silent may be prosecuted; but sometimes the moral obligation falls on anyone who knows. Who is responsible for reporting suspected child or domestic abuse or sexual assault? Laws may identify specific persons in certain roles who must report or face penalties; the moral responsibility, however, belongs to anyone who has a reasonable suspicion. A responsibility to report may apply even when the victim pleads to keep the matter secret. The rationale for overriding the victim’s right of privacy seems to involve the judgment that the individual’s autonomy is already compromised by victimhood. Thus, regardless of the victim’s expressed desires, the self-interest of the victim and the public good require the information to be known by authorities.
In some situations, there is no specific person or agency one is obligated to inform, but one nonetheless has a responsibility to reveal, to make public disclosure. That is the obligation felt by whistle-blowers. When these epistemic obligations are distributed across a group, and when the obligation is simply to reveal rather than to inform someone in particular—“someone needs to expose this”—it seems easier to avoid responding. Just as individuals are less likely to help someone if they are among a group of bystanders than if they are the only spectator, so a collective responsibility that falls on everyone who knows seems to fall on no one in particular, not on each one.
One might also be obligated not to inform. Everyone knows there are times when it is wise to keep silent even though it may be difficult. But silence may be legally or morally obligatory. The case of Edward Snowden is seen by many, apparently including President Obama, not as a courageous case of whistle-blowing in which putatively unconstitutional practices of government surveillance are made public, but as a treasonous violation of Snowden’s obligation not to disclose classified information. (It seems now that he did not actually take an oath, but signed a standard form that made nondisclosure a condition of employment.)16 In any event, oaths and pledges, confidentiality and nondisclosure agreements, and gag orders may be used to create obligations not to inform or reveal.
Divulging personal information about a friend may be disloyal; spilling a shared secret may be a betrayal. Breaking a client’s confidences may violate a professional code of ethics. But how much weight one gives to those obligations, legally and morally, may depend on other factors. Acknowledging an obligation does not imply that it may never be overridden by more serious obligations, though clearly the intent of these restrictive instruments is to constrain one’s actions and to exact a penalty if they are breached. A prima facie obligation is real and its violation is not to be taken lightly. Granted, the obligation to keep confidences is voluntary if it derives from a contract. But in practice, the signing of such agreements is often coercive: parties to a suit are told they will receive settlement only if they sign a nondisclosure agreement. In many cases, there is not even coerced agreement—as when a judge issues a gag order that no parties to a case before the court may speak with the media, for example.
All these practices bear a burden of ethical defense because: (1) they restrain freedom of expression; (2) they are frequently motivated by secrecy rather than privacy; and (3) they are instituted without regard for the interests of the greater public and its possible right to know. For example, nondisclosure agreements are used appropriately by corporations to protect proprietary information from being used by competitors. But they are used to protect not only information about products and plans, but also corporate practices and communications that may violate public interest. Their restraining effect is displayed clearly when these agreements intersect with a more open epistemic context. Consider this conflict: an able undergraduate science student pursued a research internship for academic credit, a condition of which was to present the results of the research at a campus colloquium. His internship, however, was with a chemical company that required him to sign a nondisclosure agreement under which he could not discuss the nature of his work with anyone outside the company. Or, to take a more general case: boards of trustees and university presidents have imposed gag orders on employees who are party to information about personnel matters that have become controversial on campus or newsworthy. Apart from the fact that such orders have no legal standing, they conflict with the transparent-and-free-flow-of-information ethos of colleges and universities. In short, it is too easy to use such practices to protect shabby, embarrassing, dangerous, or unethical practices.
Aristotle set the framework for analyzing the relation of ignorance to voluntary action—and hence to moral responsibility.17 He observes that actions that are coerced or produced by external forces are not voluntary. Aristotle also asserts that actions are not voluntary if they are done because of ignorance. Speaking broadly, for an action to be considered voluntary, the agent must know what she is doing. But this claim requires further distinctions, for which I will use a more contemporary situation.
Imagine passengers on a train that is equipped with an emergency cord that will stop the train precipitously. There are three ways to pull the cord. (1) Sarah pulls the cord deliberately, knowing what she is doing. This is a voluntary act for which we might praise or blame her depending on the reason for her act. (2) Philip pulls the cord thinking it is used to call the conductor. Philip acts from ignorance; in a sense, he does not know what he is doing. The action is not voluntary. (3) Michelle is very drunk and pulls the cord; in a different sense, she too does not know what she is doing. She pulls the cord in ignorance, not from ignorance. Oddly, Aristotle gives some classificatory weight to the agent’s feelings post factum: if Michelle regrets her drunken act when she is sober, it was a “nonvoluntary” action (neither voluntary nor involuntary).
From our perspective, the responsibility we bear for our ignorance forms a continuum. Recall Tom and his soup? If a chef unknowingly were to serve Tom soup that contained a poison, he would be acting from ignorance, not voluntarily. In Aristotle’s terms, he was ignorant of the particulars of the case: he did not know there was poison in the soup. We do not blame or condemn the chef. (The chef might feel that he was in some sense responsible, that the action was piacular if not immoral.) But suppose the chef knew there was arsenic in the soup but was ignorant of the commonly known fact that arsenic is poisonous: his ignorance is more culpable, though the act of poisoning is not intentional. But, if the chef was, as Aristotle would say, ignorant of universals and somehow did not know that poisoning was wrong, we are likely either to show little tolerance for his ignorance and hold him fully responsible, or—if he is unable to tell right from wrong in general—treat him as sociopathic or mentally incompetent and in need of therapy.
As I noted earlier, we may rightly be held responsible for our ignorance based on a wide range of factors, but ignorance may also serve as an excuse. Professions of ignorance are often used to deny responsibility. Genuinely exculpatory ignorance that has resulted in a tragic or harmful outcome involves the agent’s unknown unknowns, or things the agent could not have known or had no occasion or obligation to know.18
First, the general statement: the quality of relations in an epistemic community may raise moral concerns. Now more specifically: systematic epistemic bias, disregard for legitimate epistemic authority, patterns of willful ignorance, practices of withholding or distorting vital information—all raise questions of justice. British philosopher Miranda Fricker has offered an incisive critique of two forms of epistemic injustice: “testimonial injustice, in which someone is wronged in their capacity as a giver of knowledge; and hermeneutical injustice, in which someone is wronged in their capacity as a subject of social understanding.”19 The first occurs when a source is routinely given more or less credibility than is deserved. The second occurs when actions or practices are denied the meaning and import they “deserve” epistemically, perhaps because relevant concepts are inchoate or suppressed. If women’s testimony about sexual harassment is regularly discounted, it is epistemic injustice. It required the development of the concept of sexual harassment to bring into focus the import of many kinds of interpersonal actions; it placed a pattern of behavior under a concept carrying ethical weight. If the concept is inchoate, resisted, contested, or suppressed, those interpersonal interactions are not understood. The victim is not understood as victim, rights are not seen as rights, and the harasser is not seen as harasser. As Fricker argues, those who are marginalized experience both forms of injustice; indeed these tendencies contribute to their marginalization.
Ignorance is often the privilege of the powerful. One in power has the luxury of not needing to know. If, for instance, Samuel has no desire to know about the homeless, or perhaps desires not to know about “those people,” he will remain ignorant because he has no need to know. He can ignore them. And insofar as Samuel does cognize homeless people, he is cognitively free to use distorted stereotypes and false knowledge. Whether such ignorance is constructed willfully or inadvertently, it is endemic, and yet complicity in it will likely be denied. It is natural to seek innocence, reject guilt, and deny implication in subordination. Such blasé ignorance serves to protect one’s self-image of goodness, or at least one’s moral luck. In our society of economic inequality, privilege becomes another contributing factor in the culture of ignorance.
According to philosopher José Medina, such privileged ignorance is a form of epistemic injustice that sustains gender and racial oppression. However deeply embedded, it is vincible ignorance, and Medina advocates a “resistant imagination,” a “kaleidoscopic” consciousness of epistemic pluralism and dissonance, as tools for breaking down the barriers.20 Another philosopher, Barbara Applebaum, has argued that an additional technique is required: instead of teaching only the conception of language as communication, we must teach language as discourse, a mode in which language acts, transmits power, and “constitutes subjects as certain types of beings.”21 These analyses suggest a different type of epistemic obligation: not simply to inform, but to disrupt privileged ignorance and the conceptual framework that creates it.
In a survey of moral claims on knowledge and ignorance, we frequently find references to the dispositions of knowers and to the qualities of epistemic communities. Medina speaks commandingly of a “resistant imagination” and a consciousness that is “kaleidoscopic”; Fricker advocates “critical openness”; I have mentioned “epistemic laziness” and “hostility” to the truth. Such dispositions, individual or collective, affect both individual learning and the dynamics of epistemic communities. So, I turn next to a discussion of epistemic virtues and vices, and to the possibility of virtuous ignorance.