You Are Never Entitled to Your Opinion

Ever!  You are not even entitled to "I don’t know."   You are entitled to your desires, and sometimes to your choices.   You might own a choice, and if you can choose your preferences, you may have the right to do so.  But your beliefs are not about you; beliefs are about the world.  Your beliefs should be your best available estimate of the way things are; anything else is a lie.

If you ever feel tempted to resist an argument or conclusion by saying "everyone is entitled to their opinion," stop! This is as clear a bias indicator as they come.   It may irritate you to give in, but honesty demands it.    

Yes, you can sometimes reasonably reject a claim that appears well supported by  arguments.   After all, a reliable source might recommend against the claim, but not give its reasons.  Or the arguments might come from a biased source, good at searching selectively for arguments favoring a certain side.  But beware the temptation to use these as blanket excuses to ignore good arguments.

The idea that everyone is entitled to their opinion comes in part from our cult of democracy.   We are proud to live in a society where we all can "have our say" by talking and voting, and those who claim we do not know enough to have a useful say, we suspect are trying to disenfranchise us.   We must therefore all know enough to have a useful say on any public issue.  But that is wrong.   

A related error is the idea there are two kinds of topics: facts and opinions.  On facts you can be wrong, so you just go to experts and then believe them.   On opinions no one can be wrong; the weight of opinion is the weight of power; those who say there are experts are trying to trick you into giving up your power.

My students don’t expect much class discussion when I teach math or physics, but they do expect discussion on topics related to personal choice or public policy.  And they expect any opinion expressed eloquently and with passion to be worth as good a grade.  (At least non-economics students expect these things.)

Jared Diamond writes:

Anybody, scientist or no, feels entitled to spout forth on politics or psychology, and to heap scorn on what scholars in those fields write.  

It is true that some topics give experts stronger mechanisms for resolving disputes.  On other topics our biases and the complexity of the world make it harder to draw strong conclusions.  And topics where most of us can contribute relevant information may be good topics for discussions, and not just lectures. 

But never forget that on any question about the way things are (or should be), and in any information situation, there is always a best estimate.   You are only entitled to your best honest effort to find that best estimate; anything else is a lie.   

Added: M Lafferty points us to this similar article by Jamie Whyte

GD Star Rating
loading...
Tagged as:
Trackback URL:
  • http://pdf23ds.net pdf23ds

    Well, there is one sense in which everyone (in America at least) is entitled to their opinion. The sense is that our government is restricted from dictating to us any of our opinions, or (ideally) preventing us from articulating or disseminating them.

    And whatever conceptual slipping leads to the regular cries of “Censorship!” and “First Amendment!” when people are censored by the owners of private forums, or even sometime when they’re simply strongly disagreed with, is probably the same conceptual slipping behind the sentiment you write about in the post.

    I think the sentiment could be unique to the US. I haven’t heard it elsewhere, and other countries don’t have the same level of legal protection for free speech.

    I’m not sure exactly what “not being entitled to one’s opinion” entails, now. Simply that no one is obligated to give you the time of day simply for expressing it? Or do you mean to go further, and say that you aren’t entitled to ignore evidence against your opinion? Well, people *are* legally entitled to do that. And anyone who’s *trying* to be rational already knows not to do that. What change in behavior or attitude are you trying to exhort in this post?

  • Rue Des Quatre Vents

    Well, there are pragmatic reasons for not wanting to pursue the truth: doctors who lie to terminally ill patients because optimism may help these patients recover, mothers who do not want to believe their son is a murderer…as Nozick said in one of his last interviews, there’s no knock-down argument that in particular cases, we ought to believe the truth. In general it’s good policy, but in certain instances, we may well be better off having our own false opinion.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Pdf and Rue, there can be social conventions about who is entitled to what, different from legal rights, and different from what people would choose without such considerations being present. I would like for there to be a weaker social convention that people are entitled to believe anything they like, and more of a social obligation to believe accurately.

  • William Newman

    I would really like people to be held to have some responsibility for their opinions: at least losing a lot of credibility fast when too stubbornly endorsing something too silly. I’m not sure, though, how you can get that to happen. The best chance that I can imagine is that somehow increased computer power and communications bandwidth might let us cheaply and flexibly tie someone’s silliness to a corresponding slamdunk refutation, and/or help us keep score who turns out to be right. (Possibly through multiple layers of being right: Alice turned out to be right that Bob is a reliable source because Bob said Candice is a reliable source and Candice’s predictions turned out to be useful in that they reduced surprise about the world by 55 relevance-weighted entropy units per day relative to the background consensus…) But I’m not all that optimistic about that. (Even though idea futures are sorta in the same space, and do seem sorta promising.)

    Instead we have a situation where the Aumann (?) disagreement paradoxes you (Robin Hanson) have written about don’t seem very paradoxical to me. Note that it’s a cliche that people’s *behavior* is heavily influenced by other people’s *behavior*. Why then not have people’s expressed opinions be as heavily influenced by other people’s expressed opinions? Or more heavily, since the agreement or disagreement between opinions can easily be a perfect boolean yes or no, while similarity of behavior is more often a messy judgment call? My working guess is that this is very largely because expressed opinions are such cheap talk. Alice judges Bob’s expressed opinions to be very cheap talk and so unconvincing; or even when Alice becomes completely convinced that Bob is correct, she may judge it’s worth pretending other opinions in order to advance some agenda other than truth, and continue to protest that she holds her old opinions. Few people change their expressed opinion when union and their political allies support banning imports from countries with lower labor standards in order to benefit the workers there. Paradox? People being too foolish to use simple Bayesian-ish rules to learn truth about the world from other people? I doubt it. If instead the UAW were in the habit of lobbying in Japan or Europe to cut off US auto imports until the US adopts Japanese or European labor laws, I expect a number of other labor organizations would start paying careful attention and perhaps even consider doing similar things themselves…

    If there were no adaptation to selective pressure at work, then it might be quite clever for organisms to use DNA that works well in similar organisms somewhat indiscriminately. In practice, though, any mechanism for that tends to become a playground for malicious exploitation, viruses and the like. (Similarly, it evidently seemed quite clever to Microsoft engineers to make executable content so convenient…) So in the real world, while it is very valuable to have some protocol for swapping DNA, it is also extremely important to be quite devious and conservative about it, or else you get overwhelmed by all the ways that other actors in the world learn to serve up bogus stuff to manipulate your behavior.

    It seems thoroughly nasty to me when humans serve up known-to-be-wrong information simply in order to manipulate other humans’ behavior. I think it would probably be a good thing to take it very seriously and come down hard on it. Alas, at the moment all cultures that I know of seem to be pretty tolerant of it in many circumstances. Short of the technofantasizing above, I’m not sure how to change that; instead we just adapt to it.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    William, the disagreement issues are regarding your beliefs and their beliefs; if you or they are lying, then of course that is another matter. (And let’s all try to keep the comments from being too long.)

  • http://profile.typekey.com/ocmpoma/ ocmpoma

    “A related error is the idea there are two kinds of topics: facts and opinions.”
    I found that interesting. So, if Bob says that chocolate tastes better than vanilla, and Fred says that vanilla is better, who is correct?

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Ocmpoma, many claims are relative to context; the truth of “it rained” depends on which day I’m talking about. And the truth of “tastes better” can depend on who is doing the tasting. But if chocolate actually tastes better to me than vanilla right now, and I know this, I am not entitled to any other opinion on this subject than that chocolate tastes better.

  • M Lafferty

    Jamie Whyte is eloquent on this point: http://www.timesonline.co.uk/article/0,,1072-1207971,00.html

  • CalcaMutin

    In a discussion one has the right to pull out and deem the discussion not worth pursuing. In such cases I sometime say “Some lies are more cool than other lies. Everyone is entitled to believe the lies of his/her choice.” Isn’t this what freedom is? Denying this option can lead to excesses of (political?) thought correctness.

  • Rob Spear

    > But never forget that on any question about the way things are (or should be), and in any information situation, there is always a best estimate.

    Questions about how things “should be” will always involve some measure of personal choice, and therefore don’t fit into your scheme, even if we were to have faith in the dispassionate Bayesian-reasoning expert which your posting requires.

    e.g. is the Kyoto protocol a good idea? This depends on how you weigh the costs of the possible environmental damage if it is not implemented, versus the possible economic costs if it is. Some people will value these things differently from others.

    The other problem is that when it comes to questions of policy, the “experts” will tend to profit or lose depending on which way the question is solved – i.e. environmental experts on Kyoto. Perhaps there is some way of allowing for this by calculating how much the experts will benefit and reducing their credibility accordingly?

  • http://julesandjames.blogspot.com/ James Annan

    Your claims about Bayesians being unable to honestly disagree rely on them having the same prior. Is that a reasonble assumption?

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Calca, yes, you may have a right to not talk.

    Rob, one need not be Bayesian to think there is one best estimate for any information situation. I had a moral sense of “should” in mind, and moral claims are usually understood as being not a matter of personal choice. My claim applies even to preferences, however, in the sense that you should admit that you want what you want; wishing you wanted something else doesn’t make it so.

    James, I promise I will post on priors soon, and that would be the best place to discuss that issue.

  • Paul Gowder

    Along the lines of Rob’s comment, what reason is there really to believe that there aren’t multiple best estimates of some things? Preferences make an interesting example of this. Let there be two people, A and B, each of whom has a non-harmful preference along dimension X (views of the good life, automobile color, whatever). Each of A and B would make the “best estimate” of their preference merely by accurately stating it. Since there’s no normative content to those preferences, there are objectively equal.

    Now suppose I wish to examine my life, and decide that I need to form a preference on X, where previously it was unformed. The only data I have are the preferences of A and B on X. They both seem pretty good to me. I have to choose one (my car can only be painted one color). It seems like I’m in a place where there are exactly two “best estimates” of the preference I should (or will, after I choose) have.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Paul, a best estimate can be that two things are the same regarding some criteria.

  • Rue Des Quatre Vents

    Robin, fair enough. You want greater adherence and respect for the truth than is current. But the underlying point of my previous post was that, for many things, the density of our desires creates the buoyancy of our beliefs. Unless you can show that everyone really wants there to be greater accuracy or that greater accuracy will procure something else they desire, you might have a hard time convincing them otherwise.

  • http://www.catbirdseat.typepad.com Ray G

    A qualifed opinion is better than “I don’t know,” but this runs up against a problem with unqualified opinions given as fact or simply stated too strongly.

    That is, someone who makes a statement such as – citing the next thread on the board – “How many apples are on the tree?” and someone says with feigned authority “Oh, about 350.”

    I’ve been in the military, I’ve been in small business, and corporate America, and there is a typical trait among many leaders. They feel some kind of pressure to always have an answer, and will give unqualified opinions when in certain cases, “I don’t know, but I’ll find out” would do much better.

    I was under the impression that this “opinion-as-fact” example might not apply until I read the next thread up concerning the Instant Message, and so now I think it fits.

    If I go to someone I respect and trust, and say “I have a very lucrative deal concerning an apple tree I haven’t seen. All I need to know is how many apples are on the tree, approximately anyway.” And the trusted fellow says “350” instead of “Hmm, I don’t know, but it’s such and such month, the tree is currently in X climate. If I were you, I’d google apple trees in that area, and make an educated guess as to how many should be on the tree this time of year.”

    Otherwise, “I don’t know” would much less harm. But on the more basic point of simply having a discussion on general topics, there is a balance involved, and discretion to be considered. Some people aren’t worth the trouble of sharing opinions with.

  • http://www.catbirdseat.typepad.com Ray G

    Also, and this may be reading too much into this, but couldn’t multiple best estimates to get to one truth segue into Poincare’s science/philosophy?

  • randpost

    You are not entitled to have an opinion about one being entitled to his opinion

    • JG

      But that’s your opinion (and so the infinite loop continues)

  • http://profile.typekey.com/sinatraj/ Joseph Sinatra

    Robin,

    Granted that there is a ‘best’ estimate for any information situation and therefore like you say, one is only entitled to their best honest effort to find that best estimate, cannot individuals come to different conclusions as to what the best estimate is, depending on how they value different pieces of that information set?

    Different individuals could come to different ‘best’ judgements about what ‘ought’ to be because even with the ‘same’ information situations, some could think certain aspects of that information is more important than other aspects.

    For example, I’m thinking of different ways of assessing social justice.

    • Dreamvillian

       Best is very subjective, without measure.

  • http://profile.typekey.com/robinhanson/ Robin Hanson

    Joseph, in this post I’ve said nothing about disagreement. Disagreement has been, and will continue to be, discussed in other posts here at Overcoming Bias.

  • http://lonelygenius.wordpress.com LG

    I resolved the belief/opinion problem by more specifically defining them as objective and subjective respectively; meaning, one is a question of fact that is either the case or is not the case, the other is a question of preference that has no correctness state.

    When people retort a statement of fact I make with “That’s just your opinion,” I correct them by saying that it’s not an opinion at all — it’s a belief, and our beliefs (as is often the case is exchanges like these) cannot both be correct at the same time.

    That’s why I can like the color blue and never be wrong, but I can be wrong if I’m a [insert pet hated religion or movement practitioner], since said movement makes claims of fact about reality.

    You are correct though, Robin, in that once I hold an opinion about the color blue, there is a claim of fact to be made about what my favorite color truly is, but I think that’s kind of pedantic since I can honestly and truly change my favorite at will, since it’s a property of me not of the color.

  • Chris

    Robin’s post looks somewhat provocative and lightweight. He hasn’t defined ‘entitled’, he hasn’t defended his remark on not being entitled to ‘I don’t know’. He hasn’t considered that an n-dimensioned reality is subject to n! different reductional representations, each of which is ‘true’. Above all, he hasn’t considered that the utility of continuing a discussion with someone whose arguments are confused and whose aims are obscure may be slight.

  • J Thomas

    But never forget that on any question about the way things are (or should be), and in any information situation, there is always a best estimate. You are only entitled to your best honest effort to find that best estimate; anything else is a lie.

    I must disagree. When I have finite resources I must estimate how many of those resources are worth devoting to my answer for any particular question.

    The best estimate for how many resources to devote to finding a particular best estimate, depends on everything else that’s worth my attention. I won’t find out how much attention each topic is worth until I have thoroughly explored it in the context of everything else I know.

    This is not a trivial problem to solve. I say that an ethic which demands I must correctly solve a problem I cannot solve, is a bad ethic.

  • http://hanson.gmu.edu Robin Hanson

    LG, I’m not so sure you can change your favorite color easily at will.

    Chris, you lost me.

    J, I never claimed you should make a large effort. But given any level of effort, there is your best estimate given that effort, and you have no right to believe anything else.

  • Lightwave

    Given a moderate amount of effort, your best estimate might obviously be wrong to someone who’s put in much more resources and effort into solving a particular problem. And what’s the practical difference between “the best estimate” and “opinion”, anyway?

  • http://www.yohami.com yohami

    I just wrote a post similar to this, I found yours checking for similar ideas

    in my experience this “im entitled”, I get it mostly from girls?

    http://yohami.com/blog/2010/12/14/you-are-not-entitled-to-have-an-opinion/

    • Arkady Rose

      Possibly because you feel more entitled to challenge the opinions of women than men. Or perhaps you feel women’s opinions are less valuable? Either way, I’d say it speaks more about your attitude than theirs.

  • amazingtroll

    You are nopt entitled to not know? that is the most stupid opinion ever. If you have no information you don’t know. Why should I pull something from my ass instead of admitting that I don’t know?.

  • Allison G

    A former teacher of mine recently emailed this link to some of her former students. She agreed with the majority of the post, but held the position that people are entitled to say “I don’t know” because it’s true sometimes.

    But it isn’t true–you will always have a best estimate. Anything that can be related to prior experience will allow you to place a degree of certainty on an answer. You always know what the possibilities are, and you can certainly decide which is most probable. It seems the phrase “I don’t know” is really an excuse that means “I don’t think my information, and hence my belief, is more accurate than yours.”

  • Pingback: Confusions in Artificial Intelligence | Accelerating Future

  • Sev

    Er. Producing a crap estimate from minimal resources might bias your lines of thinking on the topic even after you receive sufficient information to make a non-crap estimate. Which might be counterproductive.

    If you know enough to expect that any estimate you made is very unlikely to be true, mightn’t it be wiser to refuse to estimate?

  • John Doe

    I think I heard someone once say during a certain event that opinions are about things and not people. He or she said something about having the same information. I couldn’t agree more because people can be unpredictable and always subject to change. What’s the point of having an opinion when people defy it anyway?

    Not everyone acts the same, so what’s the point of generalizing people into these groups except to justify prejudice? Not everyone believes in the divine or morality, so what’s the point of having it around other than to justify your own life? Not every person adheres to certain standards, so what’s the point of having them except for yourself and yourself alone?

    What is the point?