Freedom: an Experimental Analysis

Consider the following case:

Tanya lives in a small, newly created country in Eastern Europe. Perhaps the most important issue in the region is the treatment of a disenfranchised minority that lives throughout the country. Tanya truly dislikes the minority and wants to further damage them if she can. While public opinion concerning the minority varies greatly, the government has taken the side of the minority. Consequently, a ban has been placed on any action or public speech that is intended to hurt the disenfranchised minority. In other words, the government has made laws against hurting the minority, but Tanya wishes she could hurt them.

Now ask yourself: 'To what extent do these laws diminish Tanya's freedom?'

Once you have decided on the answer to this question, consider a very similar case with one important difference: Tanya wants to help the disenfranchised minority.

Tanya lives in a small, newly created country in Eastern Europe. Perhaps the most important issue in the region is the treatment of a disenfranchised minority that lives throughout the country. Tanya truly cares about the minority and really wants to help them if she can. While public opinion concerning the minority varies greatly, the government has sided against the minority. Consequently, a ban has been placed on any action or public speech that is intended to help the disenfranchised minority. In other words, the government has made laws against helping the minority, but Tanya wishes she could help them.

Now ask yourself the same question again: 'To what extent do these laws diminish Tanya's freedom?'

During an experiment I conducted in which participants were presented with these two cases, I discovered an very interesting result. Participants thought that Tanya's freedom was much more diminished in the second case than in the first. In other words, subjects thought that people's freedom was much more diminished when they were prevented from doing something morally good than when they were prevented from doing something morally bad. After noticing this interesting result, I conducted two other studies which further confirmed the interesting effect found in the first survey.

I would like to know how other people interested in experimental philosophy might explain this effect.  Specifically, what is it about the folk concept of freedom that it elicits this result? Any and all suggestions are welcome. 

I propose one possible explanation and a survey all the experiments in the full paper, here: Freedom: Morality and Folk Intuitions

Comments

[jonathan weinberg]

I'd worry that there's maybe a confound here, in that harming and helping are not truly parallel in this sort of situation. Presumably in the first case there are plenty of other laws in place forbidding all sorts of ways of hurting members of that minority -- killing one of them would be murder, for example, with or without such a law. So in a case where there were no laws specifically about the minority, Tanya's freedom to hurt is _already_ more limited than her freedom to help. So it may follow that a law further limiting her ability to harm is less of a restriction than a law limiting her ability to help. (My apologies if you cover this in the paper -- I'm just going on the blog post!)

Unrelatedly: I wish that we x-philes could wean ourselves a bit more away from immediately thinking about all of these sorts of results as involving the _concepts_ of the philosophically-relevant issues. It's an atavistic holdover from old-fashioned "conceptual analysis" conception of analytic philosophical method. It came in with the armchair, and I don't see why we should keep it around once we've tossed the armchairs to the curb.

[Aaron Boyden]

There's a long history of philosophers writing about freedom connecting being free to acting rightly. It seems, for example, that on Kant's view you're only really free when you're acting rationally, which means you're also doing the right thing. Admittedly, there's also a long history of mocking such philosophical views as sophistry, but perhaps studies like this show that Kant and his ilk weren't as disconnected from the thoughts of ordinary people as they are sometimes accused of having been.

[Brandon]

Deontologically and ethically, it makes sense to be more disturbed by the hindering of morally "wrong" efforts than morally "right" efforts. An individual's freedom ends where another person's begins, thus, hindering someone's "freedom to hurt others," seems ethically okay to me. I do believe that there is no difference in the quantitative "amount" of freedom being hindered, but the scopes and scales are much different.

[jonathan weinberg]

Brandon's point can dovetail nicely with mine -- I was thinking of explicit legal prohibitions that already constrain the "freedom to harm", but actually it's also not at all crazy for someone to say something like "Tanya was never (morally) free to harm the minority people in the first place", i.e., such actions were not within the scope of morally permissible actions for her, even without the law there.

[Jussi Suikkanen]

Here's a question:

If I asserted that the law that borbids harming the minority diminishes my (or Tanya's) freedoms in some conversational context, what would you think of me as a person?

Answer to this question might help to think what kind of implicatures giving that answer carries irrespective of what the concept of freedom is like.

And, if I wouldn't like to be seen as that kind of person, already the implicatures can explain why anyone would not want to give that answer in the survey.

[John Wehrle]

I wonder how the participants in these studies would have responded if you had included questions about the "right to . . ." in addition to freedom questions. These are concepts that mix "freely" in the zeitgeist but which seem to carry very different implications. If Tanya and Katya have a right to do something then they surely should be free to do it. But, if there is an unarticulated misreading of "freedom to . ." as "a right to . ." then I can easily see how the average response would give a different answer. This would also explain why the morally permissible options scored roughly the same on the freedom question.

Are you open to another survey that would control for a right to X and the freedom to do X?

[Sinduja Ragunathan]

Freedom in basic terms, according to a layman, not educated in refined teachings of philosophy, is the ability to express a spontaneous flow of thought in words or action. Hence, what I view as freedom is largely confined to what kind of actions I think is possible or permissible spontaneously. For eg., If I am a person who has been wearing a purdah from the time I was born due to the customs of my locality and conditioned right from childhood that it is the right thing to do, I would seldom be forced to revolt later. If someone comes and says "Hey, they are curtailing my freedom and not allowing me to wear the purdah! ", my response would be "Of course...what do you mean freedom? It is the right thing to do! "

Coming back to the case in hand, hence, we all have a certain view of what is right and wrong. Hurting a group of people is viewed as wrong by the majority of mankind conditioned in ethics and code of conduct. Hence, I think there exists no right to hurt them and if there exists no right, how can there be a suppression of the right? Hence, we feel the freedom was not curtailed there since we are unable to see how such a freedom can exist

[Sinduja Ragunathan]

However if the same survey is conducted among a large group consisting of people, all of whom dislike the minority, possibly bacause of some negative past experience, then i assume that they would all think the freedom was curtailed in the first case, since they think it is right to suppress a group of people who do not deserve respect due to such and such reasons.

Also, if the survey was conducted with the same group as was done in the original experiment, but with a slight modification that the group consists largely of terrorists who wreck the calm of the nation, the reverse results would have resulted. Hence, freedom is largely an individualistic opinion of what is acceptable or right.

[Jonathan Phillips]

Dr. Weinberg, thank you for your comments. I think your proposal about there being other laws already restricting certain immoral actions is really interesting and definitely could be influencing peoples' judgments. Fortunately, this proposal can be easily tested. If we imagine a case in which something which has no law restricting it, but is still considered morally wrong, then do the judgments of freedom differ from those in the first survey when that action is restricted? I conducted another study, (included in the paper) in which there was law enacted simply because of an irrational fear on the part of a dictator's wife. The law subsequently restricted an action which a woman named Katya was going to perform (in one variant it was a morally wrong one, and in the other it wasn't). In this case, there was no law previously restricting either action, in fact both actions had been going on for a long time and while the dictator knew about them, he didn't care enough to stop them. Participants still judged Katya's freedom to be much less diminished when she was stopped from doing the immoral action. Given that the proposal in this case could not explain the results, I propose another explanation which you mentioned in your second comment (in response to Brandon's). Specifically, participants in the surveys did not consider Tanya or Katya to be free, in the first place, to perform morally wrong actions.

As to the idea of continuing to use the term "concept" I think it is an accurate critique. Do you know of a better term to describe the cognitive processes which determine how we use a particular idea? We often talk about intuitions but isn't there some understanding of the idea of 'freedom' which makes it applicable in certain situations and not in others? What should we call this latent understanding? Thank you again for your comments and suggestions.

[Jonathan Phillips]

Jussi, the concern that people are responding to these surveys in a way that doesn't truly reflect their typical use of 'freedom' is a good one to bring up. While, the surveys were clearly anonymous, there is certainly still the chance that people are responding in the way they think they should, rather than according to how they actually judge the reduction of freedom. Part of this problem seems to be natural to experimental study of intuitions which involves morality. Interestingly though, participants in this survey had no problem saying that the law did in fact diminish Tanya's freedom. Even in the case in which Tanya wanted to hurt the minority but was stopped from doing so, her freedom was reduced just over 3.5, halfway on a 7 point scale (1 was labeled, "not at all" diminished, and seven was labeled "completely" diminished). It was simply the case that when Tanya was stopped from helping the minority her freedom was judged to be much more reduced. Thank you for bringing up the point; it is a good one to discuss especially in experiments involving difficult moral situations.

However, there is a similar, more general, worry that peoples' judgments were influenced by a value judgment of some part of the situation, rather than simply a judgment of freedom. One particular worry was that participants may have thought a law which restricted Tanya from helping a minority was simply a bad law, and therefore diminished freedom more. I conducted another study, (included in the paper and mentioned in response to Dr. Weinberg's comments) in which there was law enacted simply because of an irrational fear on the part of a dictator's wife. Participants were asked to judge whether or not this law was a good law, and then subsequently were given one of two situations, one in which a woman named Katya was restricted from a morally good action and one in which she was restricted from a morally bad action. While participants overwhelmingly judged the law to be a bad law, the difference in terms of the loss of freedom in the two cases was still statically significant. Given that this type of bias could not explain the survey results, I suggest that perhaps it really is something about the morality of the restricted action which is creating the difference in judgments of freedom.

[none]

John, you raise a point which was also touched on by Dr. Andrew Cohen on the public reason blog (http://publicreason.net/2008/10/22/experimental-philosophy-of-freedom/).
I would be open to the idea of running a further study with the added question about rights in an attempt to control for the possible conflation of rights with freedom. However, my response to Dr. Cohen's point about rights was that if people really do have a rights-based concept of freedom (or tend to mix "freedom to" with "a right to") it would seem that our freedom would generally not be lost without a violation of our rights. Yet, I imagine that there are many cases in which a restriction is placed on an individual that does not violate any clear right but still reduces her/his freedom significantly. Perhaps here we might imagine all the gas station owners agreeing to simultaneously raise prices so much that most people are no longer able to drive, as one possible example. Given that there are clearly cases in which we may loose freedom and not have our rights violated, perhaps the concepts are actually pretty distinct. At the very least, the onus would be on the rights-based freedom advocates to prove that participants really were conflating the two.

[Jussi Suikkanen]

Jonathan,

I'm not sure you quite address the point. The idea is that there are many things that affect our use of the terms we have. One of them is the meaning or the terms, i.e., to what things do the terms truthfully apply. Another one of them is what kind of implicatures do the terms generate in actual contexts of use. Now, no-one needs to deny that moral judgments play a role in the use the term freedom. It's just that you assume that this is because the meaning of the term (i.e., its truth-conditions) is in part moral. I think this isn't very plausible unless the pragmatic explanation has been ruled out first because it seems pretty plausible in this case.

The point is that we wouldn't bring up restrictions to our activities as losses of freedoms unless we cared about the activities. This is why if I say 'X restricts freedom' others will assume that I care about X - why else would I bring this up? If this is right, then I wouldn't say of a immoral thing that it restricts my freedom much because others would take me to endorse that bad thing to some extent. This is why people are happier to say that restrictions to helping reduce freedom more than restrictions to harming.

[Jonathan Phillips]

Sinduja, thank you for your comments. You touched on a lot of different things, but I want to address two of the main ones.
First, you followed up John Wehrle's comment on the issue of the relationship between rights and freedom. My position on this is something like this: we understand ourselves as free to do things we don't have the explicit right to do, and thus, it seems that, if we really want to discuss freedom and folk intuitions, it might not be helpful to limit the discussion of freedom to just those things which we have the right to do.
The second point to which I want to respond is about your idea that there is some sort of cultural relativity to this idea of rights and morality. I fully agree with you. I address this in the conclusion of my paper as well. I think one of the problems with folk intuitions about freedom is that they are going to be relative to what is moral for a given individual or culture. In this sense, the folk understanding of freedom may not be very functional if we imagine a situation in which a group of diverse individuals try to come to an agreement about the how much a certain law will diminish citizens' freedom. However, there is also some strength in this understanding of freedom-as-relative in that it allows for a flexibility which many other more rigid theories of freedom have lacked (and suffered for). I think what is most important here, is not to determine whether folk intuitions of freedom do or do not create a tenable/coherent theory of freedom, but instead to understand how the folk actually understand freedom. This understanding may give more weight to one or another philosophical theory of freedom, and at the very least will allow us to recognize when and where we are diverging from folk intuitions in further philosophical discussions about freedom.

[none]

Jussi,
Thank you for replying, I am sorry if I didn't correctly address your point. I think one thing that is different in our approaches is that my research is responding to a set of theories of freedom which really do suggest that restrictions on activities which "we don't care about" actually are restrictions on freedom in the exact same way as restrictions on things "we do care about." So part of my work was intended to suggest that folk intuitions differ from these theories. Jeremy Bentham, Isaiah Berlin, and Thomas Hobbes, to name a few, proposed such theories of freedom.

Beyond this different approach, I am not sure I totally understand what you are suggesting. As I understand it, your proposal is something like this: If we say X reduces freedom by restricting A from Y, then we are implying that we care about Y. If Y is an immoral act, then saying X reduces A's freedom means we care about an immoral act. So people tend to say X reduces A's freedom less when Y is immoral to insure against the implication that they endorse Y, which is immoral. Thus, the morality of the restricted options has nothing to do with the actual amount of freedom lost and participants were simply expressing a negative sentiment toward the restricted immoral option.

I have two thoughts on this type of explanation.
First, if this is a correct description of the cognitive process of participants when taking the survey, then I am unclear why participants generally suggested that this restriction reduced Tanya's freedom so significantly (over halfway on a scale from "Not at All" to "Completely"). It would seem to me that the reduction of freedom would have been much less, unless participants in general endorsed further hurting a disenfranchised minority to some significant extent.

Secondly, I ran another survey, included in the paper, in which a dictator made laws against the creation of either: low-quality soap-operas, well-written newspapers, or violent pornography. Participants were asked both the extent to which freedom was diminished by this restriction and how much they valued that option. If your suggestion is correct, and it is true that in general people don't want to endorse violent pornography or low-quality soap operas, then we should expect significantly lower scores on the amount of freedom lost in these two variants. However, this was not at all the case and there was no significant difference in any of the freedom judgments, though the value judgments varied widely. So either people, in fact, did wish to express an endorsement of violent pornography and low-quality soap operas though they didn't value it (and not express an endorsement of well-written newspapers though they valued it), or their judgments of freedom were not influenced by the endorsement-worry. In the same way, I can imagine many other options which I don't at all care about but the loss of which would result in my judging my freedom to be diminished.

I hope this addressed your point, but if not, I am always happy to continue discussing these really interesting ideas.

[Francesco Guala]

I'm very glad to see this paper, 'cause I always thought that the measurement of freedom was potentially a goldmine for experimental philosophy. The intuitions you report have become more or less orthodox in the debate thanks to Amartya Sen. Others (including myself) believe that these intuitions mix freedom with other values, which in contrast we should try to keep separate in our theories. There are various value-free metrics of freedom around, from very crude (Pattanaik and Xu) to rather sophisticated ones (e.g. Ian Carter, A Measure of Freedom, OUP). I have myself defended a rather complex metric based on people's "potential preferences" that avoids making any commitment to substantial values (Journal of Theoretical Politics, 2005, joint work with Sebastiano Bavetta). I think that the interesting question is whether people under the right circumstances are willing to make distinctions such as the ones we (philosophers) make. In other words: is it possible to separate assessments of freedom from other value-based judgments? As in natural science, experiments can (and should) be used to isolate and measure "pure" phenomena whenever possible (Ian Hacking dixit). At any rate, all this is meant to say that I'm very interested and willing to collaborate.

[John Wehrle]

You point is well taken and I will definitely check out Dr. Cohen's site.

However, if we are talking about folk intuitions and not philosopher intuitions then it seems highly likely to me that ordinary folks would say that your gas station owners were violating their customers rights. There right to what? I'm not sure, to drive like Americans presumably.

You and I might disagree with them, as I tend to do with my students, but they nonetheless consistently express a right to whatever they feel free to do.

If born out in surveys this conflation would lead us to either accept a seemingly flawed folk view of freedom or refine it. Yet, how much refining can experimental philosophy take before it ceases to be experimental and becomes Aristotelian?

Perhaps folk theories must be clarified and broken down into proto-first principles that cohere with each other - an editing process that can be more or less heavy handed, a process more art than method?

So, I suppose my long winded question is this: Is the difference between experimental philosophy and traditional philosophy a difference in kind or degree?

[Jonathan]

You may be familiar with them already, but may I recommend some of the studies done by Paul Robinson and John Darley on criminal punishment and moral intuition? See here for an excellent example - http://ssrn.com/abstract=10195

[John Wehrle]

Very interesting! Thank you. I will add this to my ever growing reading list.

[unmarshal]

My judgment would be based on voluntary association and coercion. If she was trying to hurt the minority, I'd doubt that the minority would welcome this. Therefore, her freedom has not been diminished because there is no freedom to coerce others. If she wanted to help the minority and the minority was willing to take the help then I'd say the government is being coercive and restricting her freedom.

[Marcus]

Isn't this just a variation of "your right to swing your fist ends where my nose begins"?