Sunday, April 29, 2007

Freedom and actual/ideal Preferences

Survey time! I have three questions for you, dear reader:

1) Suppose I am weak-willed, and will regretfully fail to do X even though I judge X to be for the best and indeed would prefer to do it (if only I could find the willpower). Does it violate my freedom when you force me to X for this reason?

2) Suppose I am misinformed, and refuse to do X because I fail to appreciate that it is a necessary means to achieving my heart's desire. If I knew better, I would choose to do X. Does it violate my freedom when you force me to X for this reason?

3) Suppose my values lack coherence. I refuse to do X because it isn't something I think I care about. But further reflection would bring me to care non-instrumentally about X after all (say my other values implicitly commit me to this). Does it violate my freedom when you force me to X for this reason?

Further: in each case, is my idealized self rationally bound to endorse your paternalistic intervention? What does the answer to this tell us about the moral status of intervening?


  1. I'd say that 2 and 3 definitely violate one's freedom. We are not all rational in our desires by nature, so there is no telling ab initio what we would or would not choose if we had all the information. Far far better is information campaigns to improve our understanding of what we hold valuable for ourselves.

    1 is more difficult since here people want to change. I think if someone would be wished to be denied access to drugs most of the time other than when in the midst of cravings, we can take their usual position as permission to restrain them. The problem is of course that then there are two sets of conflicting desires in the person both of which we cannot satisfy. I'm not wholly sure how we can choose one over the other as the person's true preferences.

  2. I would have to say that all of them violate freedom if we define it as autonomy. There are some alternate definitions of freedom that I can think of (mainly Amartya Sen's) that are much more reconcilable in those examples.

  3. Hi Richard

    Interesting set of questions. I would personally think no in each case, but then I am influenced by Pettit's account of freedom as non-domination. Basically on this account you are only unfree when you are dominated. Being dominated is having someone with the (not necessary exercised) ability to arbitrarily interfere with you. Arbitrary interference in this context is interference that doesn't track your long term, rational interests.

  4. I think they all would violate your freedom, but depending on the circumstances, that's not necessarily a bad thing. Assuming X (or your heart's desire) ought to be chosen (in some objective sense), then coercion is justified in some circumstances, it seems to me. Freedom is not the greatest good, even if its moderate exercise is a necessary condition of that good.

  5. I don't think 'freedom' is a univocal notion, so I would say yes or no to each depending on the particular sense of 'freedom' meant and the particular details of the case. Some people regard all freedom as freedom from coercion, which would make all the scenarios clear violations of freedom; on different conceptions of freedom, like that suggested by David Hunter above, none of them would be. And I think one needs this finer-grained sort of analysis to handle questions like these.

  6. 1. It does violate your freedom. But you may be able to enter a contract where you agree to do X and can be compelled if you fail to perform.

    2. Clear violation. I can just provide you with the correct information. If this doesn't violate freedom then hardly anything does.

    3. I can imagine situations where we are entitled to assume someone else's value set is consistent (or at least not inconsistent in specified ways). If you eat in a restaurant you have to pay.

  7. I think as per he comments in the thread we need to determine the definition of freedom we are using and if it just means "good stuff" (i.e. if freedom is by definition only freedom which is good)

    And as Pejar suggests "what is the valid moment of insight (or part of us) that defines what we prefer and what is the weak moment (or part of us) that defines what we do?*


    1) maybe what we need is a request by that person to be restrained (here we introduce contracts as a 'fair' thing that therefore by definition does not violate "good freedom") Otherwise I guess it breaches his freedom - maybe it does anyway...

    3) similarly as Nigel points out 3 could be paying for the food that you agreed to eat. Ie in some cases not a breach (excepting implied contracts from breaches in this case).

    2)however 2 would appear to violate free will. In a sense that is to say "people are allowed to be irrational".

    however I am more in favour of david/Ipse's position as a policy instrument - i.e. that freedom is not the greatest good. (Particularly in that short term freedom might compromise long term freedom – which is part of why we have contracts)

    * interesting here is - do we instinctively declare societies expectations higher desires and hedonism lower desires?


  8. probably no violation, but it might depend on the details of the case.

    yes, a violation

    yes, a violation

  9. Thanks for your responses! I don't have very clear intuitions here, but I'm most inclined towards N/Y/Y, suggesting that "freedom" is a matter of acting as you actually prefer (or "will").

    But in principle, I think I would endorse all three interventions, since attaining my ideal preferences seems to be what's ultimately best for me.

    I still oppose paternalism in practice, however, for the Millian reason that each individual tends to be the "best judge" of their own interests. In other words, paternalism rarely succeeds.

    Tentative conclusion: conventional "freedom" (understood as 'acting as you actually will') is of instrumental value. Nevertheless, there remains a sense in which our "rational autonomy" is foundational, namely: real value must reflect our idealized desires. We cannot be bound by external "ought"-claims that we'd never come to endorse ourselves from an ideal position of full information, etc.

  10. Might be interesting to contextualise each of the scenarios, for example in terms of public health issues or organ donation, since most people will agree that they ought to donate their organs, but not quite get around to signing up (A weakness of will case). Pettit relies fairly heavily on the case of Ulysses asking his crew to tie him to the mast so that he can hear the sirens as an example of something like 2.

  11. Yes, yes and yes. "Freedom" I think is ordinarily understood as getting what one chooses even where choices don't reflect our "true" (i.e. rationally considered, informed) preferences, and even if it's contrary to our interests. The wedge goes between preference (understood in the idealized sense) and choice.

    Whether intervention is morally ok is another thing because it could be that restricting a person's freedom to do a given action results in an overall expansion of his freedom or some other good consequence that on net would leave him better off. If I bully my kid into doing his homework I am certainly restricting his freedom time-slice-wise if you will but I'm (I hope) expanding his overall freedom on net since academic achievement will open a wider range of career options, etc.

    Maybe that's the ambiguity in the questions that confuse us--time-slice (or short term) vs. global.

  12. I say "yes,yes, and yes," and further, your idealized self is bound to oppose the paternalistic interventions. The goal is to achieve both conventional and idealized freedom: to do what you will at the moment and also what you will if you were fully rational and informed because these are one and the same thing. That's virtue. The goal requires learning and practice. Being forced to act by a paternalistic supervisor moves you away from the goal by inhibiting your practice or even, in severe cases, crippling it. Because that is so, the fact that the paternalist made you do what you would have if you were virtuous is irrelevant to whether you should oppose the paternalism.

    Think of being forced for an entire year by a genius neuroscientist to act at every moment as you would if you were fully rational and informed. Don't you recoil in horror at the prospect? And isn't that because you want freedom and virtue but the neuroscientist gives you neither and even inhibits your training yourself to act as you should by your own will power?

    (There are exceptions in cases where my suboptimal action would be disastrous for me, as for example when I am about to race a motorcycle drunk and without a helmet and you hide my keys. I should be glad you did so in that sort of case because major disasters impede my progress along the path to virtue more than the paternalism required to prevent them.)

  13. I guess the idea genius neuroscientist might be so good that he would give you the impression that you had free will (as long as you desire it) so the issue might be to maximize apparent freedom and apparent virtue. Although expect that sort of debate is well explored elsewhere.

    but yeah you brought an enlightening new perspective to the debate. Infinite paternalism certainly sounds creepy at first glance.


  14. I have always supposed that one cannot force another to do something (as opposed to preventing someone from doing something). I may coerce you toward doing x, even deprive you of the ability to do anything until you do X, and I suppose that is a form of forcing one to do X, so, with that caveat ...

    ... Of course all three scenarios violate your freedom, in some senses of the phrase. But do any always "violate my freedom" in all essential senses of the term?

    Consider: "I am very stubborn, and in order for you to force me to do X you have to deprive me of all liberty," vs. "I am very compliant and in order to force me to do X you need only speak sternly to me," and the entire spectrum between.

    There is an irony in one's "choice to be stubborn to degree X," in that the holding of one's own degree of stubbornness is a freedom, the maintenance of which determines the degree of other deprivations.

    Similarly, the notion of violation of freedom refers to a choice (to a certain point). I may choose to be obedient to an authority, and allow that person to "force me to X."

    But the term "violation" must refer to being subjected to some kind of damage, or at least to the possibility of damage. None of the above scenarios, as described, can categorically be regarded as damaging, but the greater the degree of one's stubbornness, the more likely that the coercion might lead to damaging consequences.

    All interventions have multiple "ontologies" -- the ontology of the intervenor and of the intervened, if you will. So in some cases, intervention can be morally suspect in the universe of the intervenor, but have nothing but positive effects in the universe of the intervened. Having said that, it is hard to draw other generalizations.

    -- nss


Visitors: check my comments policy first.
Non-Blogger users: If the comment form isn't working for you, email me your comment and I can post it on your behalf. (If your comment is too long, first try breaking it into two parts.)