Lisa Tessman
When Doing the Right Thing Is Impossible
New York: Oxford University Press, 2017 (ISBN 978-0-190-657581)
Reviewed by Anita M. Superson, 2018

Lisa Tessman argues in this concise book (based on her earlier Moral Failure: On the Impossible Demands of Morality) that morality is risky, messy, hard to live with, and impossibly demanding because it yields conflicting requirements that make acting immorally impossible to avoid. Tessman's argument turns on her view that there are some moral requirements that are immune to critical, reflective scrutiny because part of what is morally required is that it is unthinkable not to do what is morally required. One of her examples is that of a parent who must consider whether to kill her crying child so that the Nazis do not discover and then kill the group of people of which they are a part. Tessman believes that morality requires that the parent not even consider whether to kill her child because there are some acts that are unthinkable. Tessman uses many real and engaging examples throughout this book, which is aimed at a general audience with little or no background in philosophy. I think it meets this goal because she clearly explains her points, perhaps a bit too much: the book is repetitive at times. This book is not overtly a feminist book; rather, it is about normative ethics and its demands. Crucially, it is mainly about what does and should motivate moral action, which of course has implications for feminists, especially given the feminist debates about whether moral action should be motivated by reason or care.

Tessman starts by arguing for the existence of genuine moral dilemmas, which are situations in which there is a moral requirement to do two different acts, the agent cannot do both, and the requirement to do the act the agent does not do remains after acting. Tessman rejects the commonly held view among philosophers that there are no moral dilemmas because of the principle 'ought' implies 'can': one cannot be required to do something unless it is possible to do it. That is, when the agent is faced with a moral conflict between two requirements, either the agent can determine which one overrides the other and thus cancels it, or the agent takes one requirement to be the all-things-considered moral requirement that, when fulfilled, cancels out the other because it then becomes impossible to fulfill. Tessman disagrees with these attempts to escape moral dilemmas because she believes that some moral requirements are nonnegotiable. Which ones are these?

Tessman defines nonnegotiable moral requirements as those that are not eliminated when they are overridden in cases that involve a value that is irreplaceable, and that cannot be replaced with another value or compensated for with some benefit, and that if lost, lead to a cost that no one should have to bear (50, 58). A human life is one such example. Thus a moral requirement involving a loss of a human life is nonnegotiable.

After explaining in the first three chapters the difference between negotiable and nonnegotiable moral requirements and her position on the possibility of moral dilemmas, Tessman turns in chapter 4 to the issue of how we make moral judgments, which lays the groundwork for her view that some acts are unthinkable from a moral point of view. In this chapter, Tessman relies on the work of cognitive psychologists who have found from functional magnetic resonance imaging of the brain that there are two models that we use to make moral judgments. The first model is the reasoning or rationalist model, the model that I think most moral philosophers urge us to use. On this model, we make moral judgments slowly, through a conscious process that is in our control and requires our attention. We use this model when we make inferences and calculations. The second model is the intuitive or intuitionist model. On this model, we make moral judgments quickly, unconsciously, automatically. We use this model when we follow our "gut feelings" (64). Tessman describes the work of social psychologist Jonathan Haidt as concluding that most of our moral judgments are made intuitively. I am skeptical about Haidt's claim that our intuitions are so dominant in our psyche, since morality is complex and cases are different and thus require reasoning, and I am unclear whether brain studies capture these nuances. The focus in the book is on trolley-car cases, but these are "intuition grabbing" cases that do not capture much of morality in my view. For instance, issues such as abortion, euthanasia, gun control, autonomy, rape, privilege, sexism, racism, and lying do not seem to be best settled by intuitions but by reasoning. Moreover, our knee-jerk reactions should not serve as the basis of morality, certainly not without scrutiny about their source. Are your moral intuitions the same as Donald Trump's? What are intuitions, if not reflections of what we are taught? Tessman admits that the biological evidence does not speak to how we should make moral judgments, but only how we do make them (77). But in the rest of the book, she goes on to base her crucial normative points on the empirical claims made here. A significant worry is that what gets lost is the justification of our moral judgments, one of the most intractable issues in moral philosophy. Haidt at least reserves a role for reason in moral judgments, Tessman reports: we use reason after our moral judgments have been made in order to come up with a justification that we can give to others. But Haidt says that only rarely do we use reason alone to arrive at a moral judgment, and even then reason can give rise to another intuition that conflicts with our initial intuition about a case (69-71). I understand Tessman to agree with Haidt, but it would be useful for her to explain the connection between using reason or intuition to arrive at moral judgments, and using reason to justify our moral judgments, the latter of which is the philosophical project

Tessman believes that these different ways of making moral judgments might explain why we take a moral requirement to be negotiable or not. Her explanation straddles chapters 5 through 7. In chapter 5, she defends an evolutionary story about how we come to feel the demands of morality as authoritative. Borrowing from Tamar Gendler, she offers the description of an "alief." An alief is the arational counterpart to a belief and has the following characteristics: 1) it is representational, 2) it is affective, and 3) it is behavioral in the sense that it typically motivates one to action (94). An example of a moral alief is: "wallet not mine/empathic distress of the owner/must return" (96). Tessman notes that we form aliefs automatically and that they motivate action, making them unlike beliefs on both counts. When the link between the representation, affect, and behavior is formed through repeated experience, the behavior is deemed to be morally required. Societies that practice morality as part of cooperation have an evolutionary advantage over those that do not because their members survive and reproduce. Thus, in the wallet example, Tessman says that it pays to be empathetic and to act accordingly. This may be true, but I find the evolutionary story to be uninformative when it comes to difficult moral issues such as abortion. What is the relevant alief? Would it involve empathetic distress with the fetus or with the pregnant woman? And won't aliefs differ across individuals and societies?

Tessman discusses in chapter 6 a major problem with aliefs, which is that we might get things wrong. She offers this alief as an example: "black man/scared of being attacked/shoot fast" (106). The general point is that our aliefs may be informed, for instance, by biases we hold that render them questionable. This brings up the issue of second-order judgments about our first-order moral judgments that on Tessman's explanation come from intuition as just described. She asks whether our second-order judgments should be grounded in reason and serve as a check on our mistaken or biased intuitions. Going against the grain, Tessman says no, because some judgments that are made through a reasoning process would undermine the value that is at stake (108). I take this to be the main thesis of the book, and it leads her to her view that there are some impossible moral requirements, namely, ones that it is impossible to consider doing without thereby transgressing a requirement (119). In this chapter she rejects reason in second-order judgments. Tessman's main examples are actions done out of the sacred value of love. The relevant alief is "love/'I must' do x or not do x/I cannot or should not think about doing or not doing x." In other words, love makes some acts unthinkable to do or not do: it makes a parent not consider not running out into the street to rescue her child or a husband not consider saving a stranger instead of his wife (the well-known example from Bernard Williams) (117). Such acts, according to Tessman, need no justification, and trying to offer one implies that we would consider doing the unthinkable. But I do not find anything wrong with trying to justify even "unthinkable" acts at some meta-level. We should ask why love is so special that it should generate acts that we would find unthinkable not to do, and why we should want love in our lives even though it issues in such actions. Tessman can still make her point about love as a first-order judgment that we intuitively act from, but I would not abandon the philosophical project of justification even for these acts.

Tessman does not return to problematic aliefs until the end of chapter 7, when she admits that intuitive judgments can go wrong when they reflect unconscious biases or are unreliable (133). I should think that, contrary to the view that Tessman has heretofore defended, only reason is going to check these biased intuitions in a first-order judgment. Tessman here surprisingly agrees, stating that in these circumstances, "it is crucial that we do make a second-order reasoned judgment to critically question our first-order intuitive judgments" (133). But this response flies in the face of the view she has been defending all along, that even our second-order judgments are and should be intuitive rather than reason-based. As she says, "when we experience something as unthinkable, it's as if we are making both a first-order intuitive judgment and a second-order intuitive judgment, all at once" (110). In chapter 7, she claims that one might have second-order judgments that our first-order judgment is best made intuitively (the view she has been defending until this point), and other second-order judgments that need to be questioned perhaps from the point of reason, because the alief is suspect or we got the sacred value wrong (136). In the end, Tessman says that we cannot divide cases that require an intuitive judgment and cases that require a reasoned judgment because many moral conflicts are between values best preserved by reasoning and ones best preserved by an intuitive process, and both involve moral failure (136). I did not understand why, once we allow second-order, reason-based judgments to resolve problematic intuition-based, first-order judgments, reason cannot resolve any of our moral conflicts--the best argument wins. My objection is that Tessman has not given us good grounds for sorting out the cases where reason is needed to check on poorly made, intuition-based, first-order judgments and those where even the second-order judgments have to be grounded in intuition because it is unthinkable to do some acts since performing them would violate a sacred value. I am not worried about trolley-car cases ending up in moral dilemmas; I am worried about issues such as abortion ending up as unresolvable moral dilemmas. I, for one, would find it deeply unsatisfying were I to say that morality is messy because we cannot sort out when to use reason and when to use intuition.

Tessman, in chapter 7, considers another sacred value besides love that underlies unthinkable acts: life. She gives the example of Lucas, a boy whose constant drumming annoys his neighbor, Tom, who, wrongly, Tessman believes, has the thought that he would like to shoot Lucas but talks himself out of it by weighing the costs and benefits of doing so. Tessman believes that there is something seriously wrong with Tom for even considering shooting as an option, even if he had the thought in a moment of anger (124). A feeling of horror, not any kind of reasoning, should have ruled it out, she believes. Tom goes wrong because he treats a human life as if it did not have special value (126), albeit in thought, not in action. To reiterate, Tessman's point is that part of the moral requirement not to kill a human is that it is unthinkable to kill a human because human life has special value. Her concern is with the source of this thought: reason or intuition. Tessman compares Tom to Christopher, who intuitively judges that he must intuitively judge against shooting Lucas (126).

But let's change Tessman's example to a more compelling one for the case of reason. The following is based on a true story. Suppose that Lucas is a competent yet evil adult. He cons your neighbor, his uncle, into letting him move into his house purportedly to renovate it. But all he intends to do is to build a deck on top of the garage that is in close proximity to the neighbors' houses, complete with a full wet bar and a huge TV and stereo. His plan is to drag out the project so that he has a place to live all summer and an income from his uncle, since he has no other job. He further cons his uncle into stocking the bar so he can party a lot, which he does, inviting all his friends from the neighborhood bar to whom he owes money. His parties frequently go on into the wee hours of the morning. You call the police, and he retaliates by cranking his radio on the deck steps next to your bedroom windows then walking away, and by chaining up his dog all day under your windows. You learn that he shot his teacher so that he could graduate from high school, and that his father, a police sergeant, got him off the hook. You do not want to move because you have been living in the house near family for 50+ years and you really like the neighborhood. After putting up with his loud parties for half a decade, one night he blasts the music so loud that it can be heard a block away, ruining your state of relaxation and prompting you to have the thought of killing him.

Contrary to Tessman, I think that one should not be faulted for having this thought. Tessman, in the case of Lucas, describes Tom's problem as making a taboo comparison between the sacredness of Lucas's life and Tom's peace and quiet, but I think that a more Hobbesian approach is worth considering. From this perspective, the right value at stake is mutual respect. We have an implicit moral agreement with others that there be mutual obedience to laws and regulations in order for each of us to expect the gains that outweigh the sacrifices morality requires. If a person repeatedly fails to respect others' needs and legitimate interests when they respect him, he puts himself out of the moral sphere in a way. Contra Tessman, on this view, others are not required to overlook his disrespect for them. To put it another way, it becomes a lot less unthinkable that they respond to him in kind. None of this says anything about how we are required to and actually do act toward others. The issue is whether our moral judgments stem from intuition or reasoning. I am not convinced that a person who has the thought of killing this kind of evildoer but reasons herself out of acting on it is morally inferior. I worry also about Tessman's appeal to the sanctity of life, since this value is shared by anti-abortion supporters. In my view, morality is messy because we have to weigh competing values against one another and justify, not merely explain, why and when some values outweigh others.

Anita M. Superson is a professor of philosophy at the University of Kentucky. She works in ethics and feminism. She is the author of The Moral Skeptic (Oxford University Press, 2009), and co-editor of Theorizing Backlash: Philosophical Reflections on the Resistance to Feminism (Rowman & Littlefield, 2002) and of Out from the Shadows: Analytical Feminist Contributions to Traditional Philosophy (Oxford University Press, 2012). She has published articles on practical moral skepticism and feminist moral psychology, and is currently working on a monograph on bodily autonomy.

"Tessman's point is that part of the moral requirement not to kill a human is that it is unthinkable to kill a human because human life has special value. Her concern is with the source of this thought: reason or intuition."