Fifteen Eighty Four

Academic perspectives from Cambridge University Press


Virtual Roundtable: Paternalism

Moderator: Let’s start with the question at the heart of the book to help outline the different facets of this debate: how morally problematic is paternalism?

Danny Scoccia: Paternalistic interference with the choices of adults seems to me always problematic, always prima facie wrong, always in need of justification.  In that respect it is just like deception.  But just as there are circumstances in which deception is all things considered right, there are circumstances in which paternalism is too.  The harder question: is it justified only in cases where the target suffers from mental impairment/lack of competence and/or ignorance about the available options?  Soft paternalism is generally unproblematic; is hard paternalism ever permissible?

Paternalistic interference with the choices of adults seems to me always problematic, always prima facie wrong, always in need of justification.

J.S. Blumenthal-Barby: I think that Danny is right here. One thing I would add is that the extent to which paternalism is morally problematic depends on what precisely one is talking about when they use the term paternalism. Traditionally paternalism meant interference with a person’s liberty in order to make them better off. Now, however, many people describe acts (or even attitudes) as “paternalistic” that do not interfere with freedom at all; they just revolve around beliefs that Person A knows better than Person B what will make B better off.

Lawrence Alexander: I think one might regard as paternalism any restriction on a person’s liberty, any deprivation of goods available to that person, any deprivation of access to information or arguments for that person, any manipulation of that person’s environment, any attempts to deceive that person, and perhaps even the choice of what to donate to that person that the actor believes will benefit the person but that the actor believes are contrary to what the person affected currently desires (or, in the case of information and argument, might come to desire). That’s a mouthful; and not everything described is equally problematic. I think the most difficult question, however, is how to draw the line between soft paternalism—e.g., you stop someone from drinking what he thinks is beer but what you know to be poison—and hard paternalism. After all, in the latter case, you can say that if the person were fully informed, he’d want what you believe he should want.

Douglas Husak: Almost every question about paternalism is problematic—-including the question of whether and to what extent paternalism is morally problematic.  I tend to believe the normative difficulties are generally exaggerated. Paternalistic constraints don’t typically prevent persons from achieving ends they desire.  When drafted correctly, these constraints prevent persons from engaging in conduct that creates unacceptable risks of outcomes they dislike.  No one wants to go through a windshield, and paternalistic constraints such as seat belt regulations help to minimize
the likelihood of such awful occurrences.

Jeremy A. Blumenthal: (A) I find aspects of earlier definitions useful-paternalism is intervening in someone’s decision-making or behavior in order to make that person better off (by SOME definition).  But I don’t think it can be just ANY restriction/deprivation/deception.  (B) Whether the particular intervention is morally problematic depends, inter alia, on incorporating (not substituting) a balance of its benefits and costs into the typical deontological objections.  Benefits might include the increase in welfare; costs might include infringement on autonomy-though empirical research shows people can be less protective of autonomy, less likely to learn from their mistakes, more likely to defer to others, etc., than is traditionally assumed (more on that later, I hope).

Michael Cholbi: I agree that the proponent of a particular act of paternalism bears the burden of showing that it is permissible, rather than the other way around. In terms of how morally problematic paternalism is: Though I’m happy to *define* paternalism in terms of intervening in people’s exercises of liberty, how morally problematic paternalism is depends on the extent to which it interferes with rational agency. For example: Giving someone a monetary disincentive — a tax, say — to avoid fatty foods (motivated by the paternalistic aim of trying to augment their health) is less problematic than spiking the water supply with a drug that makes fatty foods taste bad. The latter is more pervasive interference with the power of rational choice than the former, and so is a more morally problematic instance of paternalism.


Moderator: Libertarian paternalism, or “nudging,” has recently been promoted as an ethical middle ground: the setup of a lunchroom can simply be rearranged so that patrons have better access to and are more likely to make healthy choices, but they are not barred from making unhealthy decisions. Where does this approach fit in the view that some forms of paternalism are more problematic than others? Does it solve the moral problems of interfering with someone’s choice, as some claim, or does it introduce new ones?

J.S. Blumenthal-Barby: This approach makes very salient that some forms of interfering to change another’s behaviors, decisions, or attitudes for their own good (what we might call paternalism) are more problematic than others given that it exposes the wide variety of ways that we might interfere. Some are more restrictive than others, some are more covert than others (which take away an agent’s ability to recognize and respond to reasons, which some think is key for autonomy), and some are more offensive or objectionable than others.

What if healthy choices aren’t your choice?

Lawrence Alexander: Given that things like buffet offerings have to be in some order or another, it doesn’t seem objectionable to set them up so as to encourage the most prudent choice (as those setting things up define the prudent choice). Currently, commercial buffets are set up to encourage patrons to choose the least expensive (to the business) but most filling choices. They have a perfect right to do so. And they would have a perfect right to set things up to encourage the healthiest choices. (Actually, the healthiest and the most profitable choices turn out in this case to be the same—put the salads first and make them look really good.) Where there is no particular order prescribed for offerings, nudges seem to be unproblematic forms of paternalism.

Douglas Husak: (So-called) libertarian paternalism shows (inter alia) how difficult it is to characterize paternalism as involving an “interference” in the choices of agents.  If such practices as reconfiguring the goods in grocery stores interfere with anyone, they interfere with the preferences of owners who seek to maximize their profits.  Items have to be located somewhere; why privilege the placement that makes the most money for grocers over the placement that protects the health of consumers?  Any serious attempt to answer this question takes the debate far away from traditional philosophical controversies about the justifiability of paternalism.

Danny Scoccia: Though “nudges” aren’t coercive/don’t restrict liberty, many if not all are “manipulative” in the sense that they seek to exploit a person’s ignorance about her options and/or defects in rationality (epistemic or “economic”). Nondeceptive manipulation, as well as deception (a kind of manipulation), coercion, and compulsion, seems to me to belong among the possible means by which a paternalist might “interfere” with another’s choices for his own good. To my mind the most important question about nudging paternalism is whether any of it could violate the right to personal autonomy or sovereignty, or whether (as libertarians suppose) only restrictions on liberty can violate that right.

Michael Cholbi: What makes nudges tricky is that they seem to operate sub-rationally: Nudges don’t “argue you” into doing what’s best for you so much as take advantage of certain limitations generated by your cognitive architecture in order to benefit you. Nudgers don’t so much as substitute their judgment for others’ as bypass others’ judgment for those others’ benefit. In cases like the lunch buffet, you will be nudged no matter how the food is presented. I find it harder to grasp what paternalism amounts to in such cases: If I *must* influence you subrationally, and you will benefit if I influence you in a particular way, but harmed if I influence you another way, then this seems paternalistic but the best moral option nonetheless. Paternalism can be all thing considered permissible even if objectionable.

Jeremy Blumenthal: I agree with Michael. And as he implies (or I infer), I think this sort of “nudge” would only be morally problematic under an extreme approach of appealing to autonomy infringement. Indeed, this rearranging-the-lunchroom approach (perhaps obviously) strikes me as a good example of using individuals’ own biases to benefit them—similar to setting default rules to “opt out” of an investment or pension portfolio rather than “opt in” to it, in order to take advantage of people’s tendencies to adhere to a status quo.

Christian Coons: I am especially concerned about “nudges.”  First, nudges may be effective only when its targets are unaware.  Proponents like to claim that we’ll be influenced anyway—e.g. “some food must go first”—but intentional influence is not justified just because another influence would be present otherwise.  A strong wind may influence your movements, but this cannot then justify my—even counter-vailing—”push.” Second, rather than making undesirable activity “costly” in light of ones deeper ends/values, nudging appears to treat these deeper ends themselves as objects to be manipulated. Arguably, this is not a mere failure to respect the “target’s” autonomy, but a denial of her autonomy. “Nudgers”, indeed, often suggest that there’s nothing we “really” prefer anyway—our preferences simply vary depending on context. But saying THAT challenges the possibility that we could reflectively endorse a nudge—a possibility usually required for justifying a paternalistic measure!


Moderator: A major objection to paternalistic practices is voiced by John Stuart Mill, who places value on individual liberty above all. But why should we be awarded so much autonomy when it doesn’t necessarily give us the means to know what is best for ourselves—cigarette taxes and mandatory Social Security contributions may infringe on our free choices, but do a Surgeon General’s fine-printed warning and a semester of high school economics equip us to make serious decisions about our health and fiscal futures?

Douglas Husak: Mill has been roundly criticized for awarding so much prominence to liberty in contexts in which it is doubtful (to say the least) that its exercise leads to utility.  Consistent utilitarians should be far more enthusiastic about paternalism than Mill.  So should deontologists.  If we want to create the conditions under which agents are most likely to become rational, autonomous agents, we should be more receptive to many instances of paternalism than most moral and legal philosophers appear to be.

If we want to create the conditions under which agents are most likely to become rational, autonomous agents, we should be more receptive to many instances of paternalism than most moral and legal philosophers appear to be.

Lawrence Alexander: As I said previously, the problem is where to draw the line. Tyrants can always take refuge behind “I’m only forcing them to be free.” All imprudence—and for some, all immorality—stems from ignorance. And sometimes it takes a lifetime for people to really understand their best interests. So, like Doug, I can endorse soft paternalism, and even some hard paternalism. But all hard paternalism can be reinterpreted as soft paternalism, and I don’t see where the line between them is in principle.

J.S. Blumenthal-Barby: Here it is helpful to distinguish between the instrumental value of autonomy and the intrinsic value of autonomy. Someone might argue that regardless of utility produced (instrumental value), autonomy should be awarded because of its intrinsic value, or its tie to other values such as self-respect, creativity, etc—as Dworkin has noted. Another helpful distinction is global autonomy vs. local autonomy where local autonomy refers to particular decision/behavior instances and global refers to the general capacity to govern oneself and determine one’s values and the course of one’s life. Local autonomy may be enhanced by soft paternalism or libertarian paternalism in certain cases.

Jeremy Blumenthal: Jenny [Blumenthal-Barby] makes useful distinctions, to which I’m sympathetic.  Generally speaking, though, I am (perhaps heretically) somewhat skeptical, or at least curious, about automatic appeals to the intrinsic value of autonomy in that Millian sense—especially given data showing that people in many instances prefer not to make choices, or derive less utility from choices when more options exist, or prefer to have others make their choices. . .


Moderator: Moving to a specific example, Mayor Bloomberg has championed soda-size restriction measures that are about to go into effect: after March 12, no New York City restaurant will be able to serve soft drinks over 16 oz. What are the ethics of this restriction?

Douglas Husak: This case involves a huge fuss over almost nothing.  Consumers can still drink as much as they want and restaurants can still sell as much as they want—although not in a single container.  The restriction is mostly symbolic; it limits liberty to a tiny degree but is likely to accomplish very little.  Opponents should be pressed to describe what measures they would prefer that have a realistic chance of curbing the problem of obesity.  The free-market alternative— “personal responsibility”—has not been an effective strategy.

Michael Cholbi: I agree with Doug that the controversy over this is mostly overblown—though I think it can be salutary when an issue of small substance gets people thinking about larger issues (governmental paternalism, in this case). As I understand it, there’s some nudging going on here, inasmuch as many studies suggest that the quantities in which foods are served shapes how much we eat (a person may eat all the food on a smaller plate but report feeling just as full or satisfied as she would eating all the food from a larger plate, etc.). Given that, limiting the size of soda containers looks like an attempt to nudge us toward healthier eating. The argument made in our earlier thread about nudging applies here: So long as soda is sold at all, it must be sold in a container of some size, and it doesn’t seem terribly objectionable to restrict the sizing so as to benefit consumers. (And some consumers might prefer this sort of paternalistic intervention on their own behalf, a kind of protection against their own tendencies to overindulge, rendering it non-paternalistic.) I don’t see that restaurants have a right to sell it in whatever size container they choose, and as Doug notes, restaurants are still free to sell additional containers and consumers are still free to buy them.

Danny Scoccia: Eliminating the sale of “Big Gulp” sized sodas at fast food restaurants seems to me too a pretty innocuous form of paternalism. Public indignation over the public health measures pushed by Mayor Bloomberg reveals something bizarre about and maybe unique to American political culture— an anti-government populism that the NRA, the Tea Party, etc. are able to exploit. A couple of years ago Michelle Obama tried to launch an educational campaign about the child obesity epidemic, and right wing talk radio went berserk. Glenn Beck (or maybe it was Rush Limbaugh) warned that this was the first step on the slippery slope leading to jail time for adults who want to eat French fries.

J.S. Blumenthal-Barby: I second comments about this being a huge deal over nothing but take it as an occasion to revisit a remark that I made in round one, which is that for some people, paternalism is an attitudinal concept, and I think that part of the fuss here is about a perceived paternalist attitude on the part of regulators (“the government”), which some may be concerned may lead to actual interferences with people’s liberty to make them better off.


Moderator: What do a government-instituted soda-size cap and the resultant controversy indicate about the perception of trust between the government and the governed in our political system?

Michael Cholbi: Some skepticism about government intervention, particularly when it is paternalistically motivated, is healthy, in my opinion. But we’re in a period (in the U.S., at least) where such skepticism and mistrust have become unhealthy. Indeed, my read is that the agenda of one of the two major political parties is to defund and discredit government, thereby ‘proving’ the ineffectiveness even of otherwise desirable government intervention. Being small scale, the soda-size cap controversy plays directly into the anti-government rhetoric and distracts us from the more large scale policy discussions we need to have about retirement savings, climate change, and the like—issues that governments might attempt to address through paternalistic efforts to influence individual behavior for citizens’ benefit.

Lawrence Alexander: I disagree with Cholbi. I think a government that has a debt as large as the GDP, borrows 40 cents of every dollar it spends, and has unfunded liabilities in the $200 trillion range doesn’t merit being trusted. Nor is there any realistic tax fix that anyone has proposed—because none can be proposed. Moreover, the incentives of all the relevant actors are to do nothing about this. Big gulp sodas are a trivial matter—because the government can’t tackle important ones. My faculty is a whole lot smarter than the government, and I wouldn’t trust them to run much of anything. If government wanted to influence really important matters, it would attempt to deal with single parenthood and all the well-documented pathologies associated with being a child without two parents—usually without a father. It would also attempt to deal with the declining birthrate, especially among the most educated. When these problems are set into a world of global competition and mobile capital, the prospects for the U.S. look grim. But hey, we’re making people travel to the fountain twice to get their 16 ounces of soda, so we should be given a pass in not tackling these other problems.

J.S. Blumenthal-Barby: Important here is the issue of transparency and openness (to input of multiple viewpoints/people) of the procedure by which the nudge was designed (its target and its mechanism). One way to increase trust, and arguably to increase the ethical acceptability, is to increase the transparency and openness of this procedure. Thaler and Sunstein mention this point briefly in their book and invoke Rawls for its source. Incidentally, whether such transparency and openness does increase trust is an empirical question, and whether such transparency and openness negates the effect of nudges is also an empirical question (more of a challenge for some nudge cases than others—the soda size cap case not being of of those cases).

Douglas Husak: This question is hard to answer because the campaign against the soda-size cap was heavily financed by the food industry.  Levels of trust would be easier to detect reliably if one side had not invested so much to skew perceptions.

Jeremy Blumenthal: This particular cap shows PROBLEMS in trust—e.g., procedural concerns about who made the relevant decision (NYC Board of Health with Bloomberg-appointed members as opposed to the NYC City Council, etc.).  It also reflects a mistrust that led to knee-jerk appeals to autonomy, with one advocate arguing that “the people of New York can make their own decisions about what they eat or drink” and that in the battle against obesity, “more choice rather than less choice is the way to go.”  As was pointed out earlier, however, it’s not clear that leaving things to the citizenry and to the marketplace—with the health problems that result—is in fact the way to go.


You’ve heard the experts, now it’s time to weigh in! What do you think? Should other people, like the government, be able to make decisions for our own good?

Latest Comments

Have your say!