Comments on Andrew Jason Cohen's toleration.
Andrew Jason Cohen begins his book Toleration (1) with the following: "In this book, I aim to provide a clear and lively introduction to the issues surrounding toleration" (p. 1). He delivers on this promise. I hope to return the favor. In my comments here, I aim to provide a clear and lively critique of his book.
Overall, I enjoyed the book for what it is: a brief overview of some of the philosophical issues concerning the concept of toleration. In particular, the conversational tone makes the book especially useful as a springboard for further discussion, for example, in the classroom. I will have some more words of praise to offer below; however, I found Cohen's glossing over certain complications a bit frustrating at times. Cohen's argument would have benefited from his having answered the following three questions: (1) Why is the harm principle the right principle upon which to base a theory of toleration? (2) How is Cohen thinking of the concept of volenti? (3) Is interference (i.e., the abandonment of toleration) ever morally required by the harm principle? Let's consider each question in turn.
(2.) The Harm Principle
Why is the harm principle the right principle upon which to base a theory of toleration? Let's examine Cohen's main claim. Following in John Stuart Mill's footsteps, he says that a violation of the harm principle is the only reason for which one is morally justified in abandoning one's default stance of toleration of others' actions. This "strict version of the harm principle" is the line Mill draws in the sand in his On Liberty, (2) and it has been fodder for philosophical criticism and deep discussion ever since. Indeed, as Cohen says, figuring out what should be tolerated is perhaps the most important question for political morality.
But neither Mill nor Cohen flesh out an argument for their claim that the harm principle is the best or only normative principle for determining which actions fall within the sphere of toleration and which allow for interference. While some have questioned the consistency between Mill's defense of the harm principle and his utilitarianism, drawing a connection between happiness and the harm principle at least points us in the direction of an explanation for why he endorses the harm principle. Mill's case for freedom of thought, speech, character, and action is built on his utilitarian conception about the conditions under which human beings flourish. According to Mill, autonomy is a necessary element in human happiness and autonomy requires an expansive concept of toleration. In this context, the harm principle is a reasonable limit to toleration.
If Cohen holds similar utilitarian commitments, he does not make them explicit in his defense of the harm principle as the normative principle determining when it is morally permissible to abandon toleration. In fact, Cohen explicitly rejects the promotion of autonomy as a reason to uphold the harm principle (pp. 122-23). He does, however, discuss his general orientation when it comes to his brand of liberalism. Perhaps examining this will provide an answer to our question.
When discussing John Rawls, (3) Cohen draws our attention to the distinction between strict political liberalism and comprehensive doctrine liberalism. According to strict political liberalism, no comprehensive doctrines can dictate the laws or other types of interference by the state, while comprehensive doctrine liberalisms of various types allow different concepts of "the good" for human beings to dictate the terms of non-toleration. By this definition, Mill's liberalism falls into the category of a comprehensive doctrine liberalism.
Comparing his own view to that of an autonomist comprehensive liberalism, Cohen says:
I do endorse a comprehensive doctrine liberalism, but it is a "thinner" version--that is, one with fewer commitments. As we will see, it does not require endorsing the promotion of autonomy for all . . . . I take myself to be a comprehensive doctrine liberal, dealing with principles of toleration that are defensible independent of any consensus. (p. 34)
For Cohen, then, the harm principle is defensible independent of any consensus. But rather than providing this defense, Cohen simply takes the harm principle as primary. While there is no denying the intuitive force of harm or a threat of harm as a reason to interfere with the actions of others, this alone doesn't lead to the conclusion that the harm principle is the best or only normative principle we should use to determine when toleration is the right action to take.
Although Cohen spends ample time in Chapter Three defining the harm principle and clarifying the definition, I cannot find an explicit argument for why he endorses this as a principle of toleration. He has a section on "Harm (and Objectivity)," so I expected that this might be part of the argument, but Cohen doesn't express the point in so many words. Perhaps part of my concern here comes from my skepticism about Cohen's "thinner" comprehensive doctrine liberalism. Following Joel Feinberg's lead, (4) Cohen defines harm as a wrongful (i.e., intentional) setback of one's interests. Harms must be objective, according to Cohen, so that the community can reasonably measure when someone's actions harm or threaten harm to another and that person's actions can justifiably be interfered with. In most cases, harms will easily be measurable, but there are, as Cohen acknowledges, complications to determining harm. He discusses one of these complications--baseline considerations--when discussing the case of floating Flo (p. 44). Suppose that I am out for a float in my lifeboat one fine day and come upon floating Flo, who is nearly drowning, through no fault of mine. Am I harming Flo if, instead of rescuing her, I turn the boat around and head in the other direction? As Cohen points out, answering this question depends on determining Flo's baseline situation. Is her baseline the situation you find her in (nearly drowning) or is her baseline her normal situation (being on dry land)?
This is a good illustration of one of the shortcomings of the harm principle. It is intuitively morally wrong for me to leave Flo to drown. Perhaps it's true that we can't say one way or the other whether I have harmed Flo because that would require being able to determine her baseline and Cohen at least seems to think that it is not easy to do in this situation (ever?). I can say, though, based on the Samaritan principle, that I am violating a moral obligation. According to this principle, I have to save others from immediate peril when I am reasonably able and it is not too costly to myself. Without an argument from Cohen for why we should use the harm principle to determine when toleration is called for, it seems to me that we will be required to tolerate behavior that is objectively unacceptable. At the very least, we are not able to determine whether toleration is the right position to take in situations where it would seem crucial for us to make such a decision.
Some political philosophers argue that other principles come into play because of the shortcomings of the harm principle. Kit Wellman, for example, argues that the Samaritan principle comes into play when it comes to questions of whether the state can justifiably coerce us into obeying the law. (5) Wellman, like Cohen, also argues for this principle on objective grounds. The strict harm principle can't explain why the state is justified in coercing us to follow the law in cases where our non-obedience doesn't directly harm anyone. On Wellman's view, were political obligation based only on the harm principle, the state would not be able to provide for our well-being and protect us from the perils of the state of nature. Cohen's discussion of Flo shows that Wellman makes a good point here.
So we return to the question of why Cohen holds that the harm principle is defensible independent of any consensus, while other principles, like the Samaritan principle (similar to what Cohen calls the benefit to others principle [pp. 63-69]), are not defensible in the same way. Why should we accept Cohen's view that interference itself is a setback to one's interests instead of a way of furthering another's interests? Are these not two equally reasonable ways to look at a situation where the state coerces one in order to benefit another? Cohen says that he does not endorse the benefit to others principle because it fails to be mindful of the incentives it sets up and the opportunity costs it imposes; in doing so, it encourages "too much interference" (p. 69). However, this conclusion only follows if Cohen is correct that social ills like poverty are not (or perhaps not always) the result of wrongful behavior by individuals or sets of individuals. In that case, it would seem that any interference is too much because there are in fact no harms that result from impoverishment.
If, on the other hand, we accept the Samaritan principle and we see that allowing poverty to exist where it could be alleviated is wrong in itself, then our conclusion about the benefit to others principle will be the opposite. Cohen doesn't address this concern in his brief discussion of what constitutes wrongfulness. He says that answering the question of what exactly constitutes wrongful behavior would require "in-depth moral reasoning, including work in metaethics and epistemology" (p. 43). Fair enough, but then it's difficult to see why we might easily agree with Cohen that poverty is not harmful--or at least that interfering with actions, like buying new bikes for our children, is "too much interference" (pp. 68-69).
I suspect that were Cohen to flesh out further his "thinner" comprehensive doctrine liberalism, which he calls a middle ground between strict political liberalism and a comprehensive doctrine liberalism, he could clear up at least some of the above confusion. I wonder whether this middle ground in fact exists. Mill can fall back on his idea of happiness, including a strong emphasis on individual autonomy, to explain his defense of preventing harm as the only defensible reason to interfere with the actions of others. Cohen, though, does not want to help himself to other liberal values like autonomy. He wants the harm principle to be the foundation for his liberalism, in which case I think the above question requires an answer.
How is Cohen thinking of the concept of volenti? I want to explore Cohen's claim that "volenti removes any wrongfulness" (p. 58). Volenti means voluntary participation. According to Cohen, whether I choose to ride on the bus when I know ahead of time that some will be engaging in a coprophagic picnic (pp. 58-59) or I decide to commit suicide, as long as I voluntarily participate in these activities, there is no wrongdoing and consequently no harm. In such cases, the right response is toleration.
You would be hard-pressed to find a philosopher friendlier than I am to the idea that consent and voluntary participation absolve others of what would otherwise be considered clear cases of wrongdoing. My intuitions also mostly agree with Cohen's whenever he discusses volenti. However, there are many interesting puzzles in other areas of moral and political philosophy related to the concept of voluntariness. For example, how does the research on implicit association affect our concept of volenti? Is it possible to define rationality in a way that is both true to the social scientific research and philosophically illuminating? Can we use the same measure of voluntariness in both interpersonal situations and when it comes to determining whether state interference is warranted? While I realize that full treatment of these questions may not be appropriate for such a volume, pointing to the questions would be helpful for understanding the depth of Cohen's view about the relationship between the harm principle and consent.
Are there cases where even though a person consents, we should not tolerate the behavior? Cohen sticks to his strict harm principle here. In order to understand how strongly Cohen is committed to the view that volenti removes wrongdoing, let's consider his discussion of the odd case of Armin Meiwes, a German man who published an advertisement looking for a volunteer to be butchered and eaten by Meiwes (p. 83). In this case, Cohen admits that it is likely that the person who ended up volunteering and allowing himself to be slaughtered was not in his right mind, but if he were, then volenti applies and the state should tolerate the action. According to the strict version of the harm principle, the state should tolerate these and similar actions.
Paternalism has been used as a justification for ignoring volenti, but Cohen does not accept paternalism as a principle of toleration. I wonder whether a strong commitment to the harm principle, at least on the part of the state, requires at least some paternalism. If the state's primary goal is to prevent individuals from harming one another, then the state ought to prevent actions that are harmful to most people in most instances. In most cases, if a person wants someone else to slaughter and eat him, that person is not in his right mind. It is wrong to kill and eat a person who has not consented to the act, so the state should prevent such actions.
We could say something similar about other paternalistic prohibitions. In most cases, if someone wants to ride her motorcycle without a helmet, she either does not understand the risks or she understands the risks and is not in her right mind. Although Cohen conceptually denies that it is possible for someone to harm herself, it seems reasonable for the state to prevent self-inflicted "hurts" that come about as a result of ignorance or mental illness. So again, if the state is interested in preventing wrongful setbacks to interests, it ought to prohibit such actions. While paternalism would no doubt set back the interests of a few (those who are in their right minds and still make choices that are highly risky because they enjoy the thrill more than most people), for the most part and given unavoidable epistemological constraints, the state will prevent more harm by prohibiting certain risky behaviors.
There are cases related to the question of volenti that are yet more complicated than the cases above. We might call these mixed or impure cases of consent. Suppose that a child is raised in a cultural group where autonomy is not valued and members of a certain subset are not taught to assert their own choices. For example, I am envisioning a cultural group like certain polygamous sects where women are subservient to their husbands. Still, suppose also that this cultural group allows members to leave the group if they so choose. Suppose that they are explicitly given the choice, say, at age thirty, when they would, under normal circumstances, be considered fully rational and autonomous. Can we really say that those who do not choose to leave or who actively choose to continue to live subserviently do so voluntarily? While it is not obvious that such a culture harms its members (since members who choose to stay believe that they are doing so voluntarily), it is difficult to say whether genuine consent is being exercised in such a case.
Cohen says, "Individuals can certainly be harmed by having their autonomy wrongly set back, but they can also be harmed by having their interest in conforming to a group--that is, an interest in being non-autonomous--wrongly set back" (p. 123). My concern here is that the origin of our interests matters when deciding whether someone has been harmed. A discussion of mixed or impure cases of consent, like the one above, would clarify Cohen's view of volenti and how it removes wrongfulness. While it's right to say that consent takes away wrongdoing, determining what counts as genuine consent is a thorny issue that too often gets short shrift in discussions like this one. While Cohen mentions prostitution, pornography, homosexuality, same-sex marriage, abortion, cloning, stem-cell research, assisted suicide, and euthanasia as examples of actions we should tolerate, he is silent about polygamy. Working through this example would be an interesting test case for sorting out where Cohen comes down on some of the more difficult complications involved in consent and its relationship to toleration.
Is interference (i.e., the abandonment of toleration) ever morally required by the harm principle? Could there ever be a time when interfering with one's harm of (or threat to harm) another is required? Cohen says that answering this question would involve getting deeper into a theory of justice. As he is presenting the principles of toleration here, which is only one part of a theory of justice, he doesn't commit to saying that interference could ever be required. He says, "interferences are, by definition, not acts of toleration and the book is not intended to offer a theory of ways to fail to tolerate" (p. 51). When it comes to interference, his language is always in terms of permissibility or justification. Consider an example, though, where two people's interests conflict and non-interference is simply not an option. It seems that if there is a conflict of interests and there is a possibility that both parties' interests will be set back, if the state does not interfere, the goal of minimizing harm would require the state not to tolerate. Again, if Cohen takes avoidance of harm to be primary in a liberal political theory, then there should be cases where interference will be required. I'm curious about why Cohen doesn't address this question head on.
Cohen suggests that the obligation to interfere is an element of justice rather than toleration (p. 51). However, I wonder what reasons or principles he has for drawing such a conclusion. Cohen endorses the claim that "justice requires that we never interfere where toleration is mandated" (p. 51). Why is it not the other way around? Why couldn't it be the case that justice requires that we interfere where there is wrongdoing and toleration is not mandated? Perhaps this would require accepting the principle of benefiting others who are not necessarily harmed, which Cohen explicitly rejects. Still, some guidance on either why Cohen rejects this concept of justice or why his theory of toleration does not speak to it would be helpful in evaluating Cohen's expansive view of toleration. This comment is more of an invitation for Cohen to expand upon his thoughts in this section of the book.
At times, Cohen seems to be presenting a complete theory of the principles of toleration, but at other times he seems to hold back from such an ambitious project. He says toward the end, "I favor a strict version of the harm principle and rejection of the others [i.e., other proposed principles of toleration] though I would not claim that the arguments I have provided on that score are definitive (at least not all of them)" (p. 150). What are we to make of this?
To my mind, the biggest problem for Cohen, if he does not present an argument for when the harm principle requires us to abandon toleration, is that often one action both sets back the interest of one and furthers the interest of another. Let's consider slavery as an example. Surely, Cohen would argue that slavery is a quintessential case of where toleration should not be tolerated, since slaves' interests are wrongly set back. Even here, though, couldn't we also look at the state choosing not to tolerate this action as an act of benefiting one group and wrongly setting back the interests of another group? Slaves were, at one time, legally property.
How are we ever to justify interfering with one group's interests in order to promote the other's, if there is no consensus about whether some action is wrong? It seems that more should be said about harm being a wrongful setback of interests. We could reasonably put all hard cases in terms of choosing whose interests are set back; without a second orienting principle like utility to weigh in the balance, it's difficult to see who has the rightful claim to non-interference. In his brief discussion of wrongdoing, Cohen points to wrongdoings that are widely agreed to be wrong (p. 43). Of course, consensus won't settle all cases. While consensus settles the case about slavery today, at the time when the U.S. government was deciding whether to interfere, consensus did not settle the issue.
The harm principle can't settle all issues. Does it settle more than other principles? I'm not sure. Perhaps it settles the same number but a different set of disputes than do other principles. It is my hope that Cohen will have more to say in defense of the harm principle as the main principle of toleration in later work fleshing out a complete theory of toleration or justice.
(5.) Praise for Cohen's Toleration
Despite the above questions, there is much to like in this book. Writing an introductory text that is at the same time interesting and worth the time of someone with advanced philosophical training is no easy task, and Cohen does this admirably well. It is impossible for a text like this to address every question; clarificatory questions dressed up as criticisms should not be taken as reasons to reject the ideas within.
I find Cohen's Toleration to be an especially good teaching tool in many respects. He makes clear from the start his overarching goal: to prompt further discussion about toleration, a historically complex philosophical idea. I appreciate that he did not pronounce his argument as the end-all-be-all statement on toleration that readers ought to memorize. At the same time, I appreciate that Cohen manages to avoid another pitfall that introductory books like this too often fall into: presenting the material in such an evenhanded, balanced way that what is philosophically interesting about the topic gets totally whitewashed, leaving the task of finding the value in it entirely up to the reader.
I also appreciate the transparency in the introduction and throughout. I especially appreciate Cohen's statement in the Introduction titled "Expectations and Warnings," where he writes: "[A]s the author, I will not refrain from making judgments. As the reader, you will decide for yourself if my being judgmental is a problem. That is not my concern. I am concerned to help you understand and appreciate the idea and practice of toleration" (p. 2). More generally, I like the message for students who are new to philosophy: don't shy away from passing judgment. Cohen makes the point that this is a necessary and natural thing to do from time to time--at least for beings possessing the ability to think critically. This is a wonderful reminder and invitation to students who are tentative about passing such judgments.
Finally, I like that Cohen takes a clear, firm stance on the question of when toleration is called for and when it is not. There are no mincing words here; the view is quite straightforward. His view about when toleration rightly ends (i.e., when the harm principle has been violated), while not without controversy, certainly creates clear guidelines for action and can be used well for teaching argument reconstruction.
There is a lot to appreciate about this book, both its goals and its method of achieving these goals. I hope my comments presented here lead Cohen and others to think and write further on the topic of toleration as it relates to liberal political philosophy.
(1) Andrew Jason Cohen, Toleration (Cambridge, UK: Polity Press, 2014).
(2) John Stuart Mill, On Liberty (New York: Penguin Books, 1986).
(3) John Rawls, Political Liberalism (New York: Columbia University Press, 1993).
(4) Joel Feinberg, Harm to Others (New York: Oxford University Press, 1984).
(5) Christopher H. Wellman, "Toward a Liberal Theory of Political Obligation," Ethics 111, no. 4 (July 2001), pp. 735-59.
Emily M. Crookston
Coastal Carolina University
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Symposium: Andrew Jason Cohen's Toleration|
|Author:||Crookston, Emily M.|
|Article Type:||Critical essay|
|Date:||Dec 22, 2016|
|Next Article:||Comments on Andrew Jason Cohen's toleration.|