= 27d567bc2d48d9f40e9ccc20ea370b15)
fubarobfusco comments on [Link] Admitting to Bias - Less Wrong Discussion
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (129)
Bear in mind that the labels "liberal" and "conservative" are not timeless or neutral. There is no reason to expect that there would be equal proportions of "conservatives" and "liberals" in any given intellectual movement, phyg, profession, or other group, without first asking what those labels are used to mean, and whether those meanings are in accord with, or opposed to, that group's goals or interests. It is not reasonable to proclaim that unequal representation of political labels in a group is indicative of bias on the part of that group's members — at least not without considering and rejecting other hypotheses for that inequality.
The specific views associated with these labels in the mainstream have shifted and changed around over time. In the U.S. today, the term "conservatism" is heavily associated with (among other things) opposition to secularism, science, and sexual equality. If you call yourself a "conservative", people may expect that you support creationism, school prayer, and the persecution of gays — just for instance. Thus, one would expect that fewer biologists, atheists, or gay people would call themselves "conservative" than in the general population — not because biologists, atheists, and gays are biased against conservatism, but because the word "conservative" means, in part, opposition to biology, atheism, and gays.
This is not because of something inherent about "conservatism" (which is, after all, a label and has nothing inherent about it at all). It has to do with the specific movements which have claimed that label. (Again, these labels shift — few would expect of a "progressive" today that they favored eugenics, despite that having been once a position labeled "progressive".) People may have good reasons to decline to associate themselves with a label, even if they may agree with some other views that have sometimes been labeled "conservative" in the past — and this is not bias, at least, not in the sense that we generally use that term here.
I find it interesting that this is never pointed out when occasional casual speculation based on reasoning similar to my own (except without actual examples) that our large libertarian minority might be biasing us towards libertarian ideas?
Same goes for those talking about Feminist biases btw.
Our large cryonicist population is probably biasing us towards cryonicist ideas. Our large computer scientist population is almost certainly biasing us towards explanations of consciousness/intelligence which rely on computational metaphors. Our large, outspoken atheist contingent probably biases us away from conveying positive attitudes towards religion, lest we be publicly ridiculed. Are these biases problematic too?
Social groups form based on shared ideas. Of course they will be biased! I think most of us would be unhappy in a totally unbiased LessWrong, because we'd be so busy debating things like theism vs. atheism that we'd have no time to delve into the finer-grained implications of our philosophical worldview.
I feel like you're conflating "unbiased" with "zero-knowledge". Flat-earthism is not an accepted viewpoint here, but that's not because we're biased against flat-earthism, it's just that flat-earthism is stupid. If it turned out that flat-earthers on average were happier and better at getting things done than other people, and if we failed to acknowledge that because flat-earthism is stupid, then that would indicate a harmful bias against flat-earthism.
If the reason we don't debate theism vs. atheism is mere bias, then we're all fooling ourselves terribly. If the reason is that we've thought sensibly about it and come to the conclusion that atheism is correct and the expected benefit of further debate is less than the time and energy costs, then that's okay.
I agree with you; I am abusing terminology.
For the record, though, I do think we have a bias against theism in the sense that you've described. I understand why LessWrong might choose to consider the "does God exist?" question settled, but we go further than that. We frequently discuss how terrible religion is, and applaud efforts to promote atheism, despite the benefits to happiness etc. that religion provides.
I think it's understandable that we have such a bias. On LessWrong, we value truth and truth-seeking; this goes far enough that it's almost a moral value. When other groups actively discourage truth-seeking, we oppose them.
I don't know anything about social psychology, but my experience with other social sciences suggests that liberal, humanist values are deeply ingrained in their system. What I mean is, the social sciences are not just truth-seeking engines looking for facts about humanity. They have their own moral values attached (e.g. reducing discrimination).
Just as we on LessWrong are reluctant to engage in discussions with people who oppose truth-seeking, social scientists may be reluctant to engage in discussions with conservatives, because conservatives tend to hold moral values that are actively opposed to the social sciences' agenda.
This article suggests implicitly that the social sciences should be about truth-seeking, not about promoting some political/moral agenda. The willingness of social scientists to admit their bias against conservatives suggest that they feel otherwise.
If the social sciences aren't about truth seeking the same way physics or biology are I think we shouldn't be calling them sciences.
Those benefits are primarily based on the communities. If we build similar secular communities, like the link I gave suggests, then there wouldn't be a special benefit to religion.
Religions can make the community-building easier.
At least they provide additional reasons for: "Why should I bother joining your community?".
You misunderstand. I'm saying that the argument that we are biased towards say libertarianism because of our demographics (and don't forget libertarians are another minority on LW) is brought up, but the counterargument that fubarobfusco has described against this sort of reasoning is not brought up.
(@_@)
This was not about low representation being an argument for discrimination, this was about people in a field out and out admitting in huge numbers that they would blatantly discriminate against people hurting their careers because of political affiliation!
Worse they would hinder papers and research that carried ideas they don't like for political reasons? Don't you see how that disfigures our map of reality and turns out to be a mockery of what the scientific process should be?
And even if that would have been that argument, like it partially seems to be in say this NYT article describing the work of Dr. Haidt:
sigh
Naturally I do agree with you and in the commentary I did invoke an argument based partially on demographics that we may be similarly biased. I don't expect there to be many conservatives on this site. I also don't expect there being many women or poor or non-white people here either. And no fubarobfusco "non-white", "woman" and "poor" are also not timeless or neutral.
What I do expect of people here is to work against biases against ideas that aren't comfortable to groups they identify with!
I'm going to move my own argument out of the OP and into a separate post so people don't confuse it with the point of the article and the primary takeaway that current socially psychology research is probably on average compromised by political bias leftward.
My argument was that similar demographics and status games here probably recreate the same problem. This compounds the progressive bias we inherit from academia.
If you want to get a job providing safety equipment for workplaces, you should probably not proclaim that you believe that workplaces are too safe. If you want to get a job as a doctor, you should probably not announce yourself as a believer in Christian Science and faith-healing. If you want to get a job as a Friendly AI researcher, you should probably not declare that you believe any AI that has been blessed by the Pope is assuredly friendly.
The common use of "conservatism" today proudly includes positions that are anti-science; as such, it is unsurprising that scientists might consider affiliation with that label to be evidence of incompetence, unseriousness, or opposition to their field. I do not see a need to introduce the hypothesis of "political bias leftward" when it is quite possible that t he scientists doing this so-called "bias" are merely taking the claimed beliefs of conservatives seriously.
I could make the same case about "liberalism". The difference being that since liberals have more influence in academia, they can actually force most scientists who disagree with them to keep quiet.
I want to see you do this, but unlike many things that people here are overly concerned about, this idea seems right at the heart of the mind-killing part of politics, so it would probably be best to do it outside of LessWrong.
Look at the discussion below this comment. Notice that the people trying to argue the liberal side aren't even bothering to argue that their position is "true", but merely that people should pretend it is for the sake of making people feel good. Call it what you want, it's definitely anti-science.
You may note that in the thread you linked, there is a current unresolved disagreement LITERALLY BETWEEN YOU AND ME. This is clearly a problematic example when it comes to my own learning, and is the exact reason that I was interested in that example being moved outside of LessWrong; perhaps I should have additionally specified that the example would ideally come from outside of LessWrong.
Yes, and do you disagree with the characterization of the argument I gave above?
At least in part.
My primary objection remains that it will be vastly more difficult for me to learn to identify the bias that you are indicating exists if your only examples of it appear to be personal attacks. I don't mean to say that this is what your example is, obviously I only posted one or two comments out of dozens, but my silly little ape brain isn't letting me consider your proposal objectively so I would like it if you could provide a different example.
Given that you seem to be a liberal, what makes you think you won't be equally mind-killed by the other examples?
It looks like here you have inadvertently provided a good argument for the opposite of what you wanted. Namely, what you write applies even if your belief that workplaces are too safe is correct. (Workplaces can certainly be too safe by any reasonable metric, at least in principle. Imagine if office workers were forced to wear helmets and knee pads just in case they might trip over while walking between the cubes. Then imagine a thriving industry of office helmets, an ever expanding bureaucracy for regulating and inspecting them -- and august academic experts getting grants to study them and issue recommendations for their use.)
If your stated beliefs are misaligned with the institutional incentives in the business or bureaucracy in which you work, it will indeed be very bad for your career. And what reason do you have to believe that the institutional incentives in the contemporary academia are aligned with the truth on all (or even on most) ideologically charged matters?
That's odd, it looks to me as if you're taking a rather loose analogy in a direction somewhere away from the topic. Getting back on topic:
My point was that "conservatism" isn't a thing — it's a label, and that people's responses to that label have to do with what they take it as referring to.
It's been noted elsethread that the survey has serious problems. One of them is that it doesn't ask what the surveyed psychologists think they are talking about when they say "conservative". If you ask someone, "What do you think about conservatives?" you will get different answers based not only on what that person's values are, but what they think "conservative" means.
If scientists use "conservative" to mean "a person who values religious doctrine over scientific results", then you are ill-advised to represent yourself as "conservative" when trying to get a job from a scientist. Especially if you don't mean that when you say "conservative"!
Note the difference between "scientists use 'conservative' to mean 'a person who values religious doctrine over science'" and "scientists think that conservatives value religious doctrine over science". The latter implies that scientists are referring to an objective class of "conservatives" whereas the former considers that scientists may not be referring to the same set of people when they say "conservative" that someone else refers to by that word.
I think we have a problem of sneaking in connotations here.