Scientific Studies of Racism?


I’m fairly swamped today, so I want to crowd-source this: Can we get some links to scientific studies of racism?

In my recent post, “Playing the Victim”: Oppression and a Catch-22, I pointed out a conundrum. I said that if people speak out about oppression and marginalization and bigotry, they get accused of “playing the victim card,” and the marginalization becomes invisible. But if people don’t speak out about it… then, obviously, it stays invisible. And I posed this question to people making this “victimhood” accusation: How, exactly, would you like marginalized people to proceed? Is there any possible way we can make oppression and marginalization and bigotry visible, which will meet with your approval?

I thought it was pretty obvious that this was a rhetorical question. Especially given the follow-up: “And why, precisely, do you think your approval matters? Why do you get to be the ones who decide which forms of oppression and marginalization and bigotry are important… and which ones are not? Why do you think that decision should be up to you?”

But I got this reply, from Emil Karlsson, who apparently took the question non-rhetorically:

I wrote about the general problem with the approach Greta Christina and others make in my post The Plural of Anecdote is not Scientific Evidence, although it is much more general.

The simple answer to the question is “more scientific evidence, less anecdotes”. The problem with anecdotes is that (1) there is no independent corroboration and (2) it is not possible to know how representative they are (i.e. the extent and scope of the problem).

Someone who does this incredibly, spectacularly and fantastically well is e. g. zoebrain (comment 14). Zoebrain provides us with a scientific report on the extent of the discrimination of trans people. This is how things are suppose to be done, not providing a few testimonials.

Why can’t Greta Christina and others do this more often? Why can’t you discuss scientific studies looking at racism and transphobia more? Or better yet, contribute to carry them out in e. g. the skeptical or atheist community?

The gold standard for claims is scientific evidence and just as we can reasonably ask for it in cases involving medical treatment and claims about the world at large, so too should be be able to ask for it when it comes to the extent of discrimination without having to get claims like “you are dismissing the experience of group X” or anecdotal testimonials thrown in our face.

In my opinion, this is a ridiculous distinction. We need anecdotes and scientific evidence on racism. We need scientific evidence to demonstrate the hard reality of racism; we need anecdotes to humanize the issue, to make the terrible reality of it clear, and to get people to freaking well care about it.

What’s more, if the question on the table is, “Does racism happen in the atheist community?” — then anecdotes of blatant unquestionable racism are, in fact, a pertinent response to the question. They still leave open questions such as “how often does this happen”… but if the question on the table is, “Does this happen?”, and a whole lot of people say, “Yes, here are examples of it,” then unless you think those people are lying, that settles the question. The existence of one black swan disproves the hypothesis that all swans are white… and the existence of loads of examples of racism in the atheist community definitely disproves the hypothesis that racism never happens in the atheist community.

But it’s also a ridiculous question because the reality of racism is extremely well-established, with study after study after study. Charging into a conversation about racism and saying, “Give me scientific evidence that it exists!” is about as absurd as charging into a conversation about vaccinations and saying, “Give me scientific evidence that vaccinations even work!” It’s one of the reasons that, in the Race and Inclusivity — A To-Do List post I put on my blog, “Get your “Race and Racism 101″ on Google or at the library. Don’t expect people of color who come to your group or event to bring you up to speed” was on the list. It’s incredibly frustrating to have to re-introduce ground-level concepts to people who are jumping into the conversation but haven’t bothered to do their homework. (And since atheists are a subset of our society at large, it would be an extraordinary claim indeed to assume that atheists are somehow miraculously free of this racism.)

However, I’m swamped today, and I really don’t have time to do Google-Fu, and email all my friends and colleagues who have sociology and psychology studies at their fingertips, and otherwise spend the entire day lining up links to the countless studies demonstrating the reality of racism. So I’m going to crowd-source it. People here who do have sociology and psychology studies at your fingertips… can you please provide links to scientific studies on racism? Thank you.

Comments

  1. Alverant says

    All I know of is the episode of “Through the Wormhole” when they asked “Is there a Master Race”. One person did talk about her research that there were IQ differences between the races and that different races were better at different things. However the people who designed the IQ test made it bias in favor of things they were good at and left out the others. I can’t site a source apart from the TV show but I’m sure it’s available online somehow and you could get her name there. But … that would probably take more time than looking up the reports on google.

    So this post is almost a complete waste of time. My bad.

  2. cortex says

    Here’s a nice review of implicit racial bias by Dovidio. I have lots more (grad student in social psych), but would have to go through syllabi and stuff, which I don’t have time for today. This article does a pretty good job of addressing how people of different races can interpret the same event very differently, among other things.

    http://happy.cs.vt.edu/~manuel/courses/diversity-F10/readings/Diversity/2002-dovidio-why%20cant%20we%20jut%20get%20along.pdf

  3. says

    Hey, Emil, congratulations on your newfound interest in racism. Since you’ve had no interest up to this point, you may be surprised to know that we have a blogger here at FtB who makes scientific studies demonstrating racism a part of his blogging beat. He’s really outstanding a presenting the material in a fashion that’s accessible to people without much grounding in the topic.

    He has a category on his blog that contains numerous studies you’ll probably want to catch up on. Have fun reading. Report back when you’re done.

  4. says

    I know of a few interesting studies, but most of my sources are secondary rather than primary… Hopefully they are helpful nonetheless (if skewed slightly towards Canadian data).

    This one shows the surprising problem of passive racism starkly, despite being a small study (“People… think they’re egalitarian, they think they’re fair, they think they’ll react in negative ways towards racism, but that might not actually be the case”):
    http://www.cbc.ca/news/technology/story/2009/01/08/racism-tolerate-study.html

    The Toronto star reporting on general feelings of alienation, based on a recent large scale survey (“Crunching thousands of numbers from 41,666 people interviewed… found skin colour – not religion, not income – was the biggest barrier to immigrants feeling they belonged…”):
    http://www.thestar.com/News/GTA/article/634117
    http://www5.statcan.gc.ca/bsolc/olc-cel/olc-cel?catno=89-593-XIE&lang=eng

    Here we see Canadian universities doing just what large numbers of other people seem to be doing (“Simply put, members of senior administration denied that racism existed in their institutions, and would respond defensively to cases of racism.”):
    http://studentseyeview.wordpress.com/2011/09/11/canada-wide-study-of-anti-racism-policies-points-finger-at-university-administration/

    I’m inclined to think the first of these is the most important to address. It’s easy not to recognize internal biases…

  5. Azkyroth, Former Growing Toaster Oven says

    First response to that entire line of argument:

    Why the FUCK should people’s ability to participate equally in society and be secure in their persons and lives be held hostage to the willingness of (the highly non-representative demographic group of) credentialed scientists to formally study their plight?!

  6. Azkyroth, Former Growing Toaster Oven says

    Second response: where the FUCK did people get the idea that “skepticism” means “I’ll keep believing what I already believe and WANT to believe until I’m dragged kicking and screaming out of it?”

  7. says

    I just searched Web of Science for the keyword “racism” and refined the search to categories with the word “psychology” in them. There were 1504 results

  8. says

    Greta Christina:

    In my opinion, this is a ridiculous distinction. We need anecdotes and scientific evidence on racism. We need scientific evidence to demonstrate the hard reality of racism; we need anecdotes to humanize the issue, to make the terrible reality of it clear, and to get people to freaking well care about it.

    I agree that anecdotes are valuable for e. g. emotional impact, but the point I was trying to make was that only scientific evidence, not anecdotes, provides you with rational reasons to conclude anything about the size of the problem and that it is the question of size, rather than existence, that is the relevant question.

    What’s more, if the question on the table is, “Does racism happen in the atheist community?” — then anecdotes of blatant unquestionable racism are, in fact, a pertinent response to the question. They still leave open questions such as “how often does this happen”… but if the question on the table is, “Does this happen?”, and a whole lot of people say, “Yes, here are examples of it,” then unless you think those people are lying, that settles the question.

    Obviously racism exists (i.e. there have been more than zero incidents of racism) in the atheist community. But this is so obvious that almost no evidence is required. I also have to ask: have any serious person ever put forward the hypothesis that there is no racism in the atheist community whatsoever? That no atheist in principle could possibly make any racist statement? I find this hard to believe.

    As an initial estimation, racism is probably about as prevalent in the atheist community as in the general population. If we have no other evidence, then all we have to go on are base rates. If this is reasonable, we also have some idea about what sort of solutions could be reasonable.

    Note that I am not, in any shape or form, questioning the existence of racism. I am not even questioning that it has a high prevalence. What I am questioning is the emphasis you and others are putting on anecdotes and the question of existence, when the focus clearly needs to be on prevalence and solutions questions (and possibly emotional impact of stories).

    With that said, I am very happy that you seem to understand the value of scientific evidence over anecdotes in this context. I look forward to more discussions about the hard evidence in the future.

    Charging into a conversation about racism and saying, “Give me scientific evidence that it exists!” is about as absurd as charging into a conversation about vaccinations and saying, “Give me scientific evidence that vaccinations even work!”

    I never asked for scientific evidence that racism exists. My point was that it is much more productive to talk about the size of the problem than if it merely exists or not and that anecdotes tell you nothing about the size.

    Even if we substitute in my actual argument, the analogy is still flawed. The actually comparison should be between testimonials that a vaccinated person did not become infected with the pathogen that he or she was vaccinated against versus the results of efficacy studies. Clearly, testimonials like that would tell you nothing about efficacy (i.e. how prevalent protection is). To know something about efficacy, you would need scientific studies.

    Stephanie Zvan:

    Hey, Emil, congratulations on your newfound interest in racism. Since you’ve had no interest up to this point, you may be surprised to know that we have a blogger here at FtB who makes scientific studies demonstrating racism a part of his blogging beat

    Actually, I have actually debunked racists on my blog for quite a while. See for instance “Problems with Racial Realism and Race-Based Rhetoric” and “Mailbag: The Absurdity of Race Realism”. I have also spent what is probably thousands of posts debunking racists at various forums (such as Rational Skepticism and Forum för Vetenskap och Folkbildning).

  9. says

    Azkyroth, Former Growing Toaster Oven:

    First response to that entire line of argument:

    Why the FUCK should people’s ability to participate equally in society and be secure in their persons and lives be held hostage to the willingness of (the highly non-representative demographic group of) credentialed scientists to formally study their plight?!

    Where did I claim that?

    My point was, and still is, that you need scientific studies, not anecdotes, to get an accurate view of the size of the problem which is the more relevant question in the situation. This is because a consideration of size will lead into appropriate solutions in a way that getting stuck on arguing about existence does not. For instance, if it is difficult to find testimonials (because of, say, social stigma), then that may leads you to incorrectly underestimate the size of the problem, which will undermine efforts at finding appropriate solutions.

  10. says

    “We study race in the labor market by sending fictitious resumes to help-wanted ads in
    Boston and Chicago newspapers. To manipulate perceived race, resumes are randomly assigned
    African American or White sounding names. White names receive 50 percent more callbacks
    for interviews. Callbacks are also more responsive to resume quality for White names than for
    African American ones. The racial gap is uniform across occupation, industry, and employer size.
    We also find little evidence that employers are inferring social class from the names. Differential
    treatment by race still appears to still be prominent in the U.S. labor market”
    http://www.economics.harvard.edu/faculty/mullainathan/files/emilygreg.pdf

    “A random survey of nonstudent adult residents was undertaken to determine whether exposure to network news has a demonstrable effect on racial attitudes and perceptions of African Americans. After controlling for a number of factors, results revealed that exposure to network news depressed estimates of African American income. In addition, network news exposure increased the endorsement of African American stereotypes, particularly the view that African Americans were poor and intimidating, and was positively associated with higher racism scores.”
    http://onlinelibrary.wiley.com/doi/10.1111/j.1460-2466.2008.00387.x/abstract

    “Ostensibly color-blind, the US drug war has been and continues to be
    waged overwhelmingly against black Americans. Although white Americans
    constitute the large majority of drug offenders, African American communities
    continue as the principal “fronts” in this unjust effort.”
    http://www.hrw.org/sites/default/files/reports/us0508_1.pdf

    “Racism continues to be a significant cause of wrongful convictions. While 29% of people in prison for rape are black, 64% of the people who were wrongfully convicted of rape (and then exonerated through DNA) are black. Moreover, most sexual assaults nationwide are among perpetrators and victims of the same race (the federal government says just 12% of sexual assaults are cross racial), but two-thirds of all black men exonerated through DNA evidence were wrongfully convicted of raping white people.”
    http://www.innocenceproject.org/Content/In_200th_DNA_Exoneration_Nationwide_Jerry_Miller_in_Chicago_Is_Proven_Innocent_25_Years_After_Wrongful_Conviction.php

  11. says

    Sam Sommers at Tufts has conducted many psych studies on the effects of racism on judgements/decision-making.

    For example, in a 2000 paper, Sommers and Ellsworth gave a mock-jury task to white and black participants:

    “the defendant’s race did not influence Whites when racial issues were salient. But in the non-race-salient version of the same interracial case, White mock jurors rated the Black defendant more guilty, aggressive, and violent than the White defendant. Black mock jurors demonstrated same-race leniency in both versions of the trial,
    suggesting that racial issues are generally salient in the minds of
    Black jurors in interracial cases with Black defendants.”

    They conclude:

    It is also worth emphasizing that the present studies suggest that White jurors (and by extension extension White police officers, White judges, White lawyers, etc.) still demonstrate bias in cases where racial issues are not emphasized”

    The paper is available (along with other interesting papers) at the bottom of this page:
    http://www.ase.tufts.edu/psychology/sommerslab/researchPublications/raceRealWorld.htm

  12. says

    My point was that it is much more productive to talk about the size of the problem than if it merely exists or not and that anecdotes tell you nothing about the size.

    @Emil – so you’re looking for scientific studies of what proportion of the atheist community exhibits racist ideologies and behaviour? What makes you think that something like that would exist? There is no scientific evidence suggesting the prevalence of atheism in the atheist community. Shall we start being hyperskeptical about that too?

    If your position is that it exists and is bad, then you can stop talking because everyone already gets that. Some of us are trying to do something about it, and we don’t actually need to know anything about the size to know it’s a problem and how to deal with it.

    It is interesting that you’ve raised pretty much the identical argument that is used to dismiss (or otherwise minimize) the existence of misogyny in the community. For the same reason that those arguments are fatuous and empty, yours is.

  13. says

    There is some interesting research on health disparities (they disproportionately effect Latina and African American women)over at guttmacher.org Been a while since I’ve been on there. If I recall they do mention discrimination, lack of opportunities as a contributing factor in many instances etc. So, here is a link to the search with several papers that address this issue.
    http://www.guttmacher.org/search/index.jsp

  14. says

    Crommunist says:

    @Emil – so you’re looking for scientific studies of what proportion of the atheist community exhibits racist ideologies and behaviour? What makes you think that something like that would exist? There is no scientific evidence suggesting the prevalence of atheism in the atheist community. Shall we start being hyperskeptical about that too?

    Please don’t put words in my mouth.

    My point boils down to this: posting testimonials tells you nothing about the size of a problem. Therefore, when critics ask for evidence we should response by providing scientific evidence rather than anecdotal testimonials.

    There is nothing “hyperskeptical” about this and we all do it regularly in any other topic, such as medical claims. Can a proponent of a new treatment respond that we are being “hyperskeptical” if we ask for evidence?

    I also explained that in the absence of evidence concerning the prevalence of racism in the atheist/skeptical community one is justified in taking the base rate prevalence in the overall population as an anchor. Even if the atheist/skeptical community is not completely representative of the overall population, the margin of error is probably not that large.

    If your position is that it exists and is bad, then you can stop talking because everyone already gets that.

    I correct flawed arguments made by people (i.e. providing testimonials and focus on the exist question instead of evidence and focus on the size question), even if those people happen to agree with my general position (i.e. racism exists, is bad and is frequent).

    Some of us are trying to do something about it, and we don’t actually need to know anything about the size to know it’s a problem and how to deal with it.

    Of course you do not need to know the size of e. g. racism to know that racism exists (i.e. has occurred at least once), but that is not the point here. The point is that you actually do need to know (or should know) something about the size of a problem before dealing with it.

    That is because reasonable solutions must reflect the size of the problem. If you are on a boat that is filling with water, your solution to that problem will depend on whether it is just raining a bit outside, if waves crashing over onto the boat, if water is leaking into a one particular area that is already tightly sealed off, or whether the boat has been torpedoed by a destroyer.

    For instance, what if you have underestimated the size of the problem? What if your boat has been torpedoed by a destroyer and you just think that you are going trough rough seas? In a similar fashion, if you have underestimated the size problem of racism, then the deployed solutions are not strong enough.

    It is interesting that you’ve raised pretty much the identical argument that is used to dismiss (or otherwise minimize) the existence of misogyny in the community. For the same reason that those arguments are fatuous and empty, yours is.

    On the contrary, an appreciation of the actual size of the problem of misogyny, that is, presenting and discussing the evidence that the problem is practically significant is the only way to rationally disprove people who don’t think that misogyny is a problem in practice in the skeptical community.

  15. says

    Oh I just remember this one I saw last year.
    “Even when the researchers made statistical adjustments to ensure they were comparing apples to apples — that is, scientists at similar institutions with similar academic track records — the disparity persisted. A black scientist was one-third less likely than a white counterpart to get a research project financed, the study found.”
    http://www.nytimes.com/2011/08/19/science/19nih.html

  16. Greta Christina says

    I agree that anecdotes are valuable for e. g. emotional impact, but the point I was trying to make was that only scientific evidence, not anecdotes, provides you with rational reasons to conclude anything about the size of the problem and that it is the question of size, rather than existence, that is the relevant question.

    Emil Karlsson @ #9: I am going to say this as calmly as I can: It is not up to you to decide what the relevant question is.

    I also have to ask: have any serious person ever put forward the hypothesis that there is no racism in the atheist community whatsoever?

    I don’t know how you’re defining “serious person”… but the whole reason I was collecting stories of instances of racism in the atheist community is that I was in an argument on Twitter with more than one person claiming that examples of racism in the atheist community were outliers or anomalies, and that racism in the atheist community was a trivial problem that could be ignored.

    Obviously racism exists (i.e. there have been more than zero incidents of racism) in the atheist community. But this is so obvious that almost no evidence is required.

    Then why were you insisting on seeing that evidence — and being so dismissive of the collection of personal anecdotes? And in any case: I wish it were obvious. To far too many people, it’s not obvious at all.

    As an initial estimation, racism is probably about as prevalent in the atheist community as in the general population.

    I completely agree. This is almost certainly true. And given that racism in the general population is pretty freaking prevalent, that means that it’s almost certainly pretty freaking prevalent in the atheist community as well — and we therefore should be taking action about it.

    What I am questioning is the emphasis you and others are putting on anecdotes and the question of existence…

    One blog post collecting anecdotes, in direct response to people denying that these sorts of incident take place, is too much emphasis?

    My point was that it is much more productive to talk about the size of the problem than if it merely exists or not…

    If people are denying that racism exists at all, it is productive to point out that it does, in fact exist, and to point to examples of it. Why are you so dismissive of this? Why can’t we discuss scientific studies of racism, examining the prevalence of the problem and its causes and the most effective ways to alleviate it… and also tell stories about our personal experiences with it?

    Again: You keep talking about an excessive emphasis on anecdotes over data. But I wrote one post — one — collecting anecdotes about people’s experiences of racism in the atheist community. Is one blog post collecting people’s stories really too many?

  17. says

    @Emil

    You’re a big fan of that “medical science” analogy, and I can understand why – if it were in any way relevant to what you’re saying, it would be a knockout. However, when someone says “I have a causal mechanism for some effect”, and we express skepticism, that is in no way the same as someone saying “X exists” and you saying “where’s your evidence?”

    If someone says “homeopathy exists”, and I say “show it to me” and you pull out a vial and say “look, here it is”, I’d look like a complete idiot if I said “anecdotes aren’t data!” It’s there. Right in front of me. Anecdotes are not data when evaluating causative claims – the existence (or prevalence) of racism in the skeptical community is not a causative claim. It’s an existence claim.

    That whole bit about boats and torpedoes was complete nonsense. I’m not sure what your point is – that racism might be so much of a problem that we need… artillery? If I were far more cynical, I would guess that your position is that racism is being overestimated and you don’t think it’s worth addressing until you see some kind of census of people who are willing to admit that they are openly and notoriously racist (which will never happen for reasons that I am sure are obvious to you). What ‘evidence’ would qualify for you to demonstrate the size of the problem?

  18. says

    @Emil:

    My point boils down to this: posting testimonials tells you nothing about the size of a problem. Therefore, when critics ask for evidence we should response by providing scientific evidence rather than anecdotal testimonials.

    Uh, those are two different things.

    If you want to know the “size” (what do you mean by that? The prevalence? Do you think it is some constant rate that never changes and is not context-dependent?), what does that have to do with “evidence” that these problems exist and are experienced by people?

    The point is that you actually do need to know (or should know) something about the size of a problem before dealing with it.

    Even if it happens to only a handful of people, it’s still wrong and needs to be addressed. Your focus on “size” is bizarre. We can deal with it whether it’s only happening to a couple of people or whether it’s happening to thousands of people. We do that by speaking out about our experiences, not necessarily by seeking out a quantitative study and sharing that.

    We can address these problems without trying to come to some understanding of an exact “size.” In fact, it’s probably not possible to find some sort of exact number or even a rough estimate, because racism (and sexism and homophobia and so on) happen in different proportions depending upon contexts and people experience events differently. We know that it happens, we know that it’s common (things you yourself admit to), and I’m curious why that’s not good enough for you and why you feel that without knowing the size that it’s not worth talking about?

    As an aside, whoever first said the plural of anecdote is not evidence should be slapped. You cannot determine what will serve as “evidence” unless you know what research questions and/or what theoretical framing you’re going to take. To say that anecdotes are not evidence is bullshit because if my research question is “Do people of color racism within the skeptical community?”, then anecdotes are most definitely “evidence” because I’m not asking for a quantity.

    So, basically what you are doing is setting up “scientific evidence” to be equal to “quantitative data”, and that’s simply not the case. It seems like what you want is some quantified number because I guess in your mind that makes the issue valid? Like, if we were somehow to quantify it and we found only one single person experienced racism in our community, that’s a non-issue for you? Or you don’t think that we should pay attention to racism as much within our community because only one person has experienced it? Further, how in the hell do you even know if racism is occurring if you’re not asking for people’s experiences–I’m sorry, “anecdotes”?

  19. says

    Greta Christina says:

    It is not up to you to decide what the relevant question is.

    But I never stated that the relevant question is the size because I decide that it is. I provided what I thought were reasonable arguments for why we should consider size the relevant question. Those arguments may be wrong of course, but never did I make the assertion that relevance is up to be personally.

    I don’t know how you’re defining “serious person”… but the whole reason I was collecting stories of instances of racism in the atheist community is that I was in an argument on Twitter with more than one person claiming that examples of racism in the atheist community were outliers or anomalies, and that racism in the atheist community was a trivial problem that could be ignored.

    But is it not the case that these individuals acknowledge the existence of racism, but incorrectly believe that it is so rare that it is not a practical problem?

    If so, I think the reasonable counterargument to that is to discuss evidence that the size of the problem is large, thereby defeating the flawed notion that it is a trivial problem? Providing testimonials in such a situation, would seem counterproductive because those people are (1) not actually denying existence and (2) they may come to falsely believe that because no scientific evidence is presented, no scientific evidence exist.

    Then why were you insisting on seeing that evidence — and being so dismissive of the collection of personal anecdotes?

    I am insisting that the discussion should be framed around the size of the problem, not merely it’s existence. I accept that racism exists, that it is morally vile and that it is a big problem. But for those that do not, evidence provided should be in the form of quantitative estimations of size, not primarily testimonials. Call it a dispute about what the most successful strategy is for combating those individuals.

    I completely agree. This is almost certainly true. And given that racism in the general population is pretty freaking prevalent, that means that it’s almost certainly pretty freaking prevalent in the atheist community as well — and we therefore should be taking action about it.

    I wholeheartedly agree.

    But against those pesky nay-sayers, we should launch our attacks with scientific studies, not primarily with testimonials (although we keep in mind that testimonials can have a beneficial emotional impact).

    You keep talking about an excessive emphasis on anecdotes over data. But I wrote one post — one — collecting anecdotes about people’s experiences of racism in the atheist community. Is one blog post collecting people’s stories really too many?

    I suspect that you are partially right. Of course one post is not excessive. However, I think that the strategy of collecting anecdotes has been a prevalent trend on many social justice issues (e. g. misogyny) by many skeptical bloggers. It was not my intent to pick just on you.

  20. says

    Crommunist says:

    You’re a big fan of that “medical science” analogy, and I can understand why – if it were in any way relevant to what you’re saying, it would be a knockout. However, when someone says “I have a causal mechanism for some effect”, and we express skepticism, that is in no way the same as someone saying “X exists” and you saying “where’s your evidence?”

    The analogy I made was between X is a successful treatment, not that there is a causal mechanism for some effect. Imagine that the person presents you with testimonials from people. Alright, but how do we know that it is an accurate estimation of the size of the effect of that treatment. What if it successfully treats just 1 out of 1 million? For that we need scientific studies, not testimonials. Testimonials says nothing about the size of the effect (or problem).

    That whole bit about boats and torpedoes was complete nonsense. I’m not sure what your point is – that racism might be so much of a problem that we need… artillery?

    Exactly. You can think of that as increased government or legal intervention, if you’d like. The general point is that rational solutions need to be proportional to the size of the problem.

    If I were far more cynical, I would guess that your position is that racism is being overestimated and you don’t think it’s worth addressing until you see some kind of census of people who are willing to admit that they are openly and notoriously racist (which will never happen for reasons that I am sure are obvious to you). What ‘evidence’ would qualify for you to demonstrate the size of the problem?

    Now you are putting words in my mouth again, despite the fact that I have repeated many times that racism exists, is large and is morally vile.

    My general point has been that it is much more productive to talk about the size of the problem than if it merely exists or not and that anecdotes tell you nothing about the size.

  21. says

    Imagine that the person presents you with testimonials from people. Alright, but how do we know that it is an accurate estimation of the size of the effect of that treatment?

    Oh, okay. I see what’s happening. You don’t understand why anecdotes aren’t evidence, you just know that anecdotes aren’t evidence. The question of how many people it helps as a proportion of the population is actually not a question that anyone in medical science asks ever. The question they (we) ask is ‘does it work better than X’ where X is either the next best treatment or a placebo. In that case, saying “it worked for soandso” is not sufficient because you have not controlled for other explanations, not because soandso represent an unknown proportion of people suffering from whatever.

    it is much more productive to talk about the size of the problem than if it merely exists or not and that anecdotes tell you nothing about the size

    Asserted without evidence. You might think it’s more “productive” (ye gahds how I hate that word in this context) but that assumes that everyone’s priorities are the same as yours. You may be shocked to learn that social scientists who study racism are interested in causative mechanisms and demonstrating effect size (which is a very different thing from prevalence). The proportion of the population affected by it is not important, because the methods to deal with racism (or any -ism) are orthogonal to that statistic.

    You still haven’t answered my question of what kind of “scientific studies” you would consider sufficient to demonstrate that it’s a big enough problem to care about. Or have we already done this on Twitter where you’re the guy who claims to care about the problem, but repeatedly demands evidence for all those other irrational people who won’t act on anything that doesn’t affect them personally unless it meets some kind of minimum standard for population size?

  22. albiefarinas says

    We all witness racism, homophobia and misogynism in our every day lives… A white Christian woman being a racist towards an African American Misogynist, who’s bulling a Mexican transsexual, that’s abusing child, that’s abusing a dog or a cat…. Or any combination thereof…. Please, get over it….!

    Act in accordance to your conscience….!

  23. Greta Christina says

    I am insisting that the discussion should be framed around the size of the problem, not merely it’s existence. I accept that racism exists, that it is morally vile and that it is a big problem. But for those that do not, evidence provided should be in the form of quantitative estimations of size, not primarily testimonials. Call it a dispute about what the most successful strategy is for combating those individuals.

    Emil Karlsson @ #24: And yet again, I say: Why are you so up in arms about one blog post — one — collecting stories of personal experiences with racism? Why do you think one blog post means this is what we’re “primarily” talking about?

    Stories of personal experiences have power… especially if you’re trying to make a case that these kinds of incidents actually do happen, and to convey their real-life impact, to people who are fervently denying this. The fact that you, personally, accept that racism exists and is bad does not mean that everyone else does.

    Nowhere have I said, “We should totally ignore scientific statistics on racism.” I am saying, “Stories matter.” If you’re not disputing that — then stop arguing. Do you really not understand the power of marginalized people telling their/our stories? And do you really not know anything about the history and the context of people from a privileged group telling people from a marginalized group to stop telling their stories, or saying that their stories aren’t important?

    Oh, and as for this:

    However, I think that the strategy of collecting anecdotes has been a prevalent trend on many social justice issues (e. g. misogyny) by many skeptical bloggers.

    What scientific evidence do you have for this? This seems to be the heart of your claim: that social justice bloggers place a disproportionate emphasis on stories instead of data. Do you have any data to back up this claim?

  24. Greta Christina says

    BTW, Crommunist has put up a post in his blog, compiling several posts discussing specific pieces of scientific research on racism:

    I can’t believe I actually have to answer this question

    He has also sent me this list of links to studies about racism:

    http://www.theroot.com/buzz/bad-news-those-blackest-names
    http://campaignstops.blogs.nytimes.com/2012/06/09/how-racist-are-we-ask-google/?hp
    http://www.racialicious.com/2012/05/31/new-study-backs-up-what-weve-been-talking-about-all-this-time/
    http://thinkprogress.org/justice/2012/05/10/481589/nypd-stop-and-frisk-young-black-men/
    http://today.duke.edu/2012/04/jurystudy
    http://www.nytimes.com/2012/03/06/education/black-students-face-more-harsh-discipline-data-shows.html?_r=2&hpw
    http://www.google.com/hostednews/afp/article/ALeqM5ieIKEf6GvJAwc1iBJZ1itH-HGbyA?docId=CNG.1732b21b28ee34447047f9aa12dd08c5.b31
    http://www.jahonline.org/article/S1054-139X(08)00416-3/abstract
    http://abcnews.go.com/Health/Wellness/black-children-pain-meds-er/story?id=16231146#.UBrNhbSufW-
    http://www.psychologytoday.com/blog/between-the-lines/201204/studies-unconscious-bias-suggest-racism-not-necessarily-perpetrated-ra
    http://www.washingtonpost.com/business/economy/report-finds-racial-discrepancies-in-upkeep-of-foreclosed-properties/2012/04/04/gIQAB7W8vS_story.html
    http://www.npr.org/templates/story/story.php?storyId=120985060
    http://www.good.is/post/debunking-the-myth-that-minorities-receive-more-college-scholarships/
    http://www.allgov.com/Top_Stories/ViewNews/Black_Americans_Given_Longer_Sentences_than_White_Americans_for_Same_Crimes_120204#.TzAlfkFg6eY.facebook
    http://www.nytimes.com/2012/01/21/business/blacks-face-bias-in-bankruptcy-study-suggests.html?_r=1&pagewanted=all?src=tp
    http://www.washingtoncitypaper.com/articles/39580/dc-leads-nation-in-per-capita-marijuana-arrests-crime-stats
    http://www.bbc.co.uk/news/world-us-canada-14584860
    http://www.pewsocialtrends.org/2011/07/26/wealth-gaps-rise-to-record-highs-between-whites-blacks-hispanics/
    http://thechoice.blogs.nytimes.com/2011/06/20/college-board-findings-on-minority-men/

    Is anyone still in doubt about this?

  25. says

    Crommunist says:

    Oh, okay. I see what’s happening. You don’t understand why anecdotes aren’t evidence, you just know that anecdotes aren’t evidence.

    I think I made the reason why anecdotes is not evidence clear in the very comment that Greta cites in this blog post. Anecdotes are not evidence because there is no independent corroboration and it is not clear that anecdotes are representative.

    The question of how many people it helps as a proportion of the population is actually not a question that anyone in medical science asks ever. The question they (we) ask is ‘does it work better than X’ where X is either the next best treatment or a placebo.

    When I say “work”, it is implied that I mean the actual pharmacological effect i.e. “how much better than placebo”.

    Asserted without evidence. You might think it’s more “productive” (ye gahds how I hate that word in this context) but that assumes that everyone’s priorities are the same as yours.

    No, I provided reasons why size is relevant. It is the same reason as in medicine. If you know that treatment X is better than treatment placebo, that does not tell you anything useful on its own. You want to know how much better treatment X is. If the difference is so small that it doesn’t make any relevance in practice, then we understand the relevance of the size question.

    I am not asserting my personal priorities over others, but rather present why size considerations are objectively more important.

    You may be shocked to learn that social scientists who study racism are interested in causative mechanisms and demonstrating effect size (which is a very different thing from prevalence).

    Actually, effect size is often just a general term for an estimate of quantity from a sample. So prevalence can be an effect size. So can the age of the earth. Or the number of people who like vanilla ice cream over papaya. Or any quantity that can be estimated from a sample. For our purposes, you can think “effect size” when I use the term “size”.

    You still haven’t answered my question of what kind of “scientific studies” you would consider sufficient to demonstrate that it’s a big enough problem to care about

    That is because I already accept that racism is a big problem. My quibbles is about the most rational way to argue against those who disagree (evidence vs. testimonials).

  26. says

    If the difference is so small that it doesn’t make any relevance in practice, then we understand the relevance of the size question.

    Okay, now bring this point home, because you’ve lost me. What does this have to do with the existence of racism in the skeptical community? Is your position that people (not you, clearly, but some unnamed others) think that racism exists but think the prevalence is not large enough for them to care about? And if we could say “look, 36% of minority atheists have experienced racist abuse” that they will say “this is above my arbitrarily-determined threshold for things I care about, therefore I will jump right in!”?

  27. Greta Christina says

    I think I made the reason why anecdotes is not evidence clear in the very comment that Greta cites in this blog post. Anecdotes are not evidence because there is no independent corroboration and it is not clear that anecdotes are representative.

    Emil Karlsson @ #29: Translation: “Anecdotes are an appropriate response to Question A, but not to Question B. I think Question B is more important than Question A. Therefore, I’m going to discount the importance of anecdotes… even if what is being discussed is Question A.”

  28. says

    I don’t mean to intrude; I see you are all having a nice back and forth, and excuse me if I have misinterpreted something. Anecdotes cannot tell us about the size of a problem, I agree.

    However, if the same anecdotes of behavior we all know is relatively common in society at large keep popping up in the atheist community, maybe they can give us some idea of how to address particular problems, even if they only are experienced by a few people. I don’t see the point in waiting for a comprehensive survey of atheist racial minorities when we can learn to identify possible problematic behavior and correct it now.

    I have had my American citizenship questioned quite a few times, for instance. (Frankly, I don’t think that’s anyone’s business.) I know this is anecdotal, but wouldn’t that be enough to suggest that people in the community not question someone’s citizenship based on preconceived notions about that person’s group affiliation? Will it really hurt their feelings that much? Even if they have never done it before, wouldn’t it be beneficial to bring that into someone’s consciousness, so they can store that tidbit for later? “I’m thinking of questioning that girl’s citizenship right now…oh wait, I remember reading about how that was upsetting to some folks, maybe I should ask her what music she likes instead.”

    I mean even if I’m the only one who benefits, I would greatly appreciate not being questioned in that way. :3 Is it really that awful to say, “a few people say they have experienced this, we don’t know how extensive this behavior is, but can everyone make sure not do that? Thanks.”

  29. 'Tis Himself says

    I am insisting that the discussion should be framed around the size of the problem, not merely it’s existence. I accept that racism exists, that it is morally vile and that it is a big problem. But for those that do not, evidence provided should be in the form of quantitative estimations of size, not primarily testimonials. Call it a dispute about what the most successful strategy is for combating those individuals.

    I fail to understand why the “size” of the problem is more important than the existence of the problem. You accept that racism exists. You appear fixated on the quantitive aspects of racism rather than the qualitative aspects. Is there some threshold number where racism ceases to be a minor problem and becomes a major problem?

    Hefflefinger et al (2010) say that overt racism is present in 18.34% of atheists. Since Buttmunch’s seminal 1999 paper shows that overt racism can be ignored until it reaches 19.26%, at which time it becomes “noticeable” then we can merely make tut-tut noises about atheistic racism.

  30. says

    Greta Christina says:

    .Nowhere have I said, “We should totally ignore scientific statistics on racism.” I am saying, “Stories matter.” If you’re not disputing that — then stop arguing.

    Actions speak louder than words. If you personally focus more on anecdotes than evidence, that means that you think that the former is more relevant than the latter.

    You did not respond to my central counterargument, which was:

    But is it not the case that these individuals acknowledge the existence of racism, but incorrectly believe that it is so rare that it is not a practical problem?

    If so, I think the reasonable counterargument to that is to discuss evidence that the size of the problem is large, thereby defeating the flawed notion that it is a trivial problem? Providing testimonials in such a situation, would seem counterproductive because those people are (1) not actually denying existence and (2) they may come to falsely believe that because no scientific evidence is presented, no scientific evidence exist.

    I would really like to have a direct comment on this from you Greta. Does it make sense? If so, I think I have demonstrated the reasonableness of my argument.

    We both agree that anecdotes have the power to move people, but that is different from the purely consideration of what counts as a rational counterargument.

    What scientific evidence do you have for this? This seems to be the heart of your claim: that social justice bloggers place a disproportionate emphasis on stories instead of data. Do you have any data to back up this claim?

    The heart of my claim is that it is unreasonable to argue from testimonials instead of evidence and that the strategy of collecting anecdotes has been a prevalent trend on many social justice issues (e. g. misogyny) by many skeptical bloggers. I am not entirely sure it is disproportional because what exactly would the correct anecdote to evidence ratio be? In my opinion 0:1, but of course that is not something I can decide.

    There are several places where this has been the norm such as your blog post about racism, your post “Why “Yes, But” Is the Wrong Response to Misogyny”, Watson’s “Reddit Makes Me Hate Atheists”, Zvan’s “Would It Be Immoral to Rape My Friends?”, McCreight’s “Re: Reddit Makes Me Hate Atheists”, The skeptical teacher’s “Misogyny in the Skeptical Movement: “Don’t Feed the Trolls” Panel from SkepchickCon 2012″ etc.

    I am not saying this is a big problem of course, but even small errors needs to be corrected. But I see what you are trying to do. Your next response will probably ignore the central point I have been trying to make all along and tell me that I am providing nothing but anecdotes, and can therefore be dismissed. It would completely misunderstand what I am saying, but I guess I am used to that happening here.

  31. Brownian says

    But for those that do not, evidence provided should be in the form of quantitative estimations of size, not primarily testimonials. Call it a dispute about what the most successful strategy is for combating those individuals.

    I can’t believe I have to ask for this, but what is Emil Karlsson’s evidence that quantitative estimations of size are more effective in convincing people that racism exists, that it is morally vile and that it is a big problem?

    Assuming ‘effectiveness in changing people’s mind’ is what he means when he says ‘should’.

  32. Brownian says

    Ah, I see it has been asked.

    What scientific evidence do you have for this? This seems to be the heart of your claim: that social justice bloggers place a disproportionate emphasis on stories instead of data. Do you have any data to back up this claim?

    The heart of my claim is that it is unreasonable to argue from testimonials instead of evidence and that the strategy of collecting anecdotes has been a prevalent trend on many social justice issues (e. g. misogyny) by many skeptical bloggers

    So, none.

    That’s some fucked up.

  33. says

    Crommunist says:

    So you’re just not going to answer that question?

    Why should I answer a question that assumes that I accept that racism is not a large problem? I do accept that racism is a large problem, therefore it makes no sense to ask me what evidence would be required for me to start accepting that racism is a large problem. I already accept it.

    Okay, now bring this point home, because you’ve lost me. What does this have to do with the existence of racism in the skeptical community?

    Here is how I understand the situation.

    1. A bunch of racists engage in a Twitter war with Greta.
    2. These racists believe that racism exists, but that it is so trivially low that it does not warrant attention.
    3. Greta responds by providing testimonials e. g. examples of racism in the skeptical community.
    4. I reply that it is more reasonable to provide scientific evidence about the size of the problem of racism because i. it is a more rational reply since they are not directly denying existence and ii. an understanding of the size of the problem is more relevant for deploying appropriate solutions than testimonials
    5. As a hypothetical example, I considered the possibility that the problem of racism is being underestimated (i.e. one way in which solutions and size of problem can be disconnected). If so, then that would lead to the deployment of solutions that are not strong enough. Therefore, it makes sense that knowledge of the size of the problem is relevant for deploying solutions.

    Is your position that people (not you, clearly, but some unnamed others) think that racism exists but think the prevalence is not large enough for them to care about? And if we could say “look, 36% of minority atheists have experienced racist abuse” that they will say “this is above my arbitrarily-determined threshold for things I care about, therefore I will jump right in!”?

    When talking about the practical significance of a result, we must abandon the dichotomous thinking of “significant / not significant” and think in terms of estimation. Therefore, there is cannot exist such an arbitrary cut-off. The figure has to be interpreted in the sociological context by researchers who work in the field.

    Consider medicine again: we test treatment X against placebo and we find that it does better. That is one thing. But we still have to decide if it works better enough to justify recommending it to doctors and patients. The only rational way to estimate this is to interpret the obtained effect size in the biological context by a professional in the field. Say the different is 10%. To me, 10% seems a lot. But what if the obtained effect size of the treatment is within weekly fluctuations? In other words, there are clinical factors in the biological context that will determine if something is practically significant or not.

    In other words, we make arguments that show that the prevalence of racism in the skeptical community, in its sociological context, is practically large. Then there is nowhere left to run for racists from a rational standpoint.

  34. says

    Greta Christina says:

    Emil Karlsson @ #29: Translation: “Anecdotes are an appropriate response to Question A, but not to Question B. I think Question B is more important than Question A. Therefore, I’m going to discount the importance of anecdotes… even if what is being discussed is Question A.”

    I made an argument in an earlier comment (23) that the question that was actually discussed by the Twitter racists, per your retelling, was B (size), not A (existence).

  35. says

    Brownian says:

    I can’t believe I have to ask for this, but what is Emil Karlsson’s evidence that quantitative estimations of size are more effective in convincing people that racism exists, that it is morally vile and that it is a big problem?

    Assuming ‘effectiveness in changing people’s mind’ is what he means when he says ‘should’.

    No, I mean rationally refuting in a logical sense (like how evidence of X would refute the claim ~X), not effectiveness in changing people’s mind. I do not think that evidence or anecdotes are effective in changing the minds of entrenched racists.

  36. Brownian says

    No, I mean rationally refuting in a logical sense (like how evidence of X would refute the claim ~X), not effectiveness in changing people’s mind.

    Wait, what? So the argument is that you feel that there’s a right (as in correct, proper) way to argue that racism is important to fight, but that’s divorced from its effectiveness?

    I do not think that evidence or anecdotes are effective in changing the minds of entrenched racists.

    Did I use the term ‘entrenched racists’ (I notice earlier you used the term ‘critics’) or is that a bit of a goalpost shift?

  37. Brownian says

    Like what does this:

    I do not think that evidence or anecdotes are effective in changing the minds of entrenched racists.

    have to do with this:

    I reply that it is more reasonable to provide scientific evidence about the size of the problem of racism because i. it is a more rational reply since they are not directly denying existence and ii. an understanding of the size of the problem is more relevant for deploying appropriate solutions than testimonials

    ?

  38. Greta Christina says

    Actions speak louder than words. If you personally focus more on anecdotes than evidence, that means that you think that the former is more relevant than the latter.

    Emil Karlsson @ #35: And for the zillionth time, I say: I wrote one blog post — one — in which I focused on anecdotes about people’s experiences of racism. From this, you are claiming that my focus on anecdotes instead of data is disproportionate. Even though this very post, the one we’re commenting on, is focusing on data. ???

    But is it not the case that these individuals acknowledge the existence of racism, but incorrectly believe that it is so rare that it is not a practical problem?

    If so, I think the reasonable counterargument to that is to discuss evidence that the size of the problem is large, thereby defeating the flawed notion that it is a trivial problem? Providing testimonials in such a situation, would seem counterproductive because those people are (1) not actually denying existence and (2) they may come to falsely believe that because no scientific evidence is presented, no scientific evidence exist.

    This is a straw man. I am not arguing that we should focus on anecdotes instead of data. I am arguing that we need both.

    The heart of my claim is that it is unreasonable to argue from testimonials instead of evidence …

    Same straw man. I never said we should argue from testimonials instead of evidence. I said we should argue with testimonials and evidence.

    And I say yet again: Do you really not know anything about the history and the context of people from a privileged group telling people from a marginalized group to stop telling their stories, or saying that their stories aren’t important?

    There are several places where this has been the norm such as your blog post about racism, your post “Why “Yes, But” Is the Wrong Response to Misogyny”, Watson’s “Reddit Makes Me Hate Atheists”, Zvan’s “Would It Be Immoral to Rape My Friends?”, McCreight’s “Re: Reddit Makes Me Hate Atheists”, The skeptical teacher’s “Misogyny in the Skeptical Movement: “Don’t Feed the Trolls” Panel from SkepchickCon 2012″ etc.

    But that’s just a list of examples of incidents you had a problem with! Where’s your data that for how often these incidents occur? And are you really not seeing the irony here?

    I am not entirely sure it is disproportional

    Then stop making the argument that it is.

    because what exactly would the correct anecdote to evidence ratio be? In my opinion 0:1…

    In other words, you think there should be zero anecdotes. In other words, everything you said about anecdotes having their place was malarkey. Why should I continue to debate with you? Why should I assume that you’re debating in good faith?

  39. Kevin Schelley says

    Ok Emil, if you think such studies are so necessary, are you willing to fund one? Gathering anecdotes is cheap and easy. While it is not a perfect solution to the problem, I don’t see anyone rushing to give funding to study the problem of racism in the atheist community.

    It’s like you’re complaining someone is collecting stones to fix their foundation when really they should be hiring a contractor to be doing it with concrete and rebar. One of these things takes a lot more money, and if you’re not willing to help fund the thing you find preferable, then stop complaining about someone else’s choice.

    If a scientific study of this is so important to you, then do the study yourself.

  40. josh says

    Crommunist and Greta need to learn how to be wrong. This is embarrassing:

    Crommunist-“The question of how many people it helps as a proportion of the population is actually not a question that anyone in medical science asks ever.”

    Of course they do, it’s the only way you can decide if a medical treatment is worth it in terms of investment and risks. Remember the flap about revising the recommended breast cancer screenings? Or recall the ‘results not typical’ caveat skeptics should always point out for advertised dietary supplements and weight loss miracle cures.

    Greta-“I was in an argument on Twitter with more than one person claiming that examples of racism in the atheist community were outliers or anomalies…”

    In other words, you were arguing about size, which is why Emil contends that vague testimonies aren’t going to be very persuasive.

  41. says

    This is a straw man. I am not arguing that we should focus on anecdotes instead of data. I am arguing that we need both.

    But do you agree with my reasoning in that particular section? Those twitter racists questioned did not question the existence of racism, but accepted existence and incorrectly believed that the problem was trivially small?

    If so, It seems that you have conceded the point. Providing testimonials do not rationally disprove their position. Providing scientific evidence regarding the size of the problem does.

    Then stop making the argument that it is.

    You started using the phrase “disproportional”. I was talking about how errors should be criticized even if they are not frequent.

    In other words, you think there should be zero anecdotes. In other words, everything you said about anecdotes having their place was malarkey.

    No, no. That is not what I mean.

    We have to separate two different contexts here:

    (1) the value of anecdotes in rationally refuting the claim that the problem of racism is trivially small.

    (2) the value of anecdotes in actually convincing people that racism is not trivial.

    In my opinion, anecdotes are almost completely irrelevant for (1). But as you and I agreed earlier, anecdotes are incredibly good at influencing people emotionally (2).

    So when I talk about a 0:1 ratio of anecdotes to evidence, I only mean in the context of (1). You are absolutely, positively, 100% correct that anecdotes are extremely valuable for emotional impact and thus have a pragmatic justification, but that was not my beef.

    Sorry for expressing myself so muddled.

  42. Greta Christina says

    Greta-”I was in an argument on Twitter with more than one person claiming that examples of racism in the atheist community were outliers or anomalies…”

    In other words, you were arguing about size, which is why Emil contends that vague testimonies aren’t going to be very persuasive.

    josh @ #45: No. Let me clarify.

    The people I was debating on Twitter were specifically saying, “Give me examples of this racism in the atheist community you keep talking about.” (Yes, they were also saying that any examples they’d seen were outliers or anomalies. Yes, their argument was self-contradictory and incoherent.)

    So I collected examples. I did, in fact, also refer to hard data: I referred over and over again to the many studies that had been done on racism in society in general, and argued that it would be an extraordinary claim to think atheists were immune. But they kept saying, “You haven’t shown me any examples of this happening in atheism!” So, given that there are no scientific studies I know of on regarding racism among atheists specifically, and given that they kept hounding me for examples, I did the best I could in a short time, and collected a large number of stories.

    And I now see that this is a Catch-22. If you point to general data about racism, people say, “But I don’t see any examples of it — point to some!” And then, if you point to examples, someone else will say, “You’re focusing too much on anecdotes and not on data!”

  43. Greta Christina says

    But do you agree with my reasoning in that particular section? Those twitter racists questioned did not question the existence of racism, but accepted existence and incorrectly believed that the problem was trivially small?

    If so, It seems that you have conceded the point. Providing testimonials do not rationally disprove their position. Providing scientific evidence regarding the size of the problem does.

    Addressed in my reply to Josh, in #47. In the debate in question, I was providing data on the general prevalence of racism which was being dismissed, and I was specifically being asked for anecdotes. And they were asking for data which, as far I know, does not exist — namely, data on the prevalence of racism in the atheist community specifically. In the absence of that data, I hoped that a large number of stories about racism among atheists might persuade them that this was a serious problem.

    We have to separate two different contexts here:

    (1) the value of anecdotes in rationally refuting the claim that the problem of racism is trivially small.

    (2) the value of anecdotes in actually convincing people that racism is not trivial.

    In my opinion, anecdotes are almost completely irrelevant for (1). But as you and I agreed earlier, anecdotes are incredibly good at influencing people emotionally (2).

    So when I talk about a 0:1 ratio of anecdotes to evidence, I only mean in the context of (1). You are absolutely, positively, 100% correct that anecdotes are extremely valuable for emotional impact and thus have a pragmatic justification, but that was not my beef.

    Emil Karlsson @ #46: So what you’re saying is that it’s irrational to pay any attention to which arguments will actually be persuasive? It’s irrational to pay attention to what kind of communication is actually effective?

  44. says

    Alright, I will rephrase:

    For those tweets that accepted the existence of racism in the skeptical/atheist community, but wrongly thought it was trivial, would be rationally refuted by scientific studies showing that racism has a high prevalence together with the base rate argument (i.e. probably not much difference between skeptical community and overall population).

    For those tweets that did not even accept the existence of racism, then examples of racisms in the form of anecdotes would rationally refute their position.

    Finally, anecdotes are highly valuable for influencing people emotionally, so they have a pragmatic justification for being used even in cases where their rational force is low.

    How does this sound?

  45. Greta Christina says

    For those tweets that accepted the existence of racism in the skeptical/atheist community, but wrongly thought it was trivial, would be rationally refuted by scientific studies showing that racism has a high prevalence together with the base rate argument (i.e. probably not much difference between skeptical community and overall population).

    Emil Karlsson @ #49: Rationally, yes: these people should have accepted these studies as refuting their point. But they didn’t. And they kept insisting on examples of racism in the atheist community. Are you going to continue chiding me for providing them?

    And you haven’t yet answered a question I have now asked more than once: Do you really not know anything about the history and the context of people from a privileged group telling people from a marginalized group to stop telling their stories, or saying that their stories aren’t important?

  46. says

    Emil Karlsson @ #46: So what you’re saying is that it’s irrational to pay any attention to which arguments will actually be persuasive? It’s irrational to pay attention to what kind of communication is actually effective?

    No, I would not go so far.

    Take the medical analogy again. A company that has created a certain treatment could potentially convince more people about efficacy if they use more emotional arguments than scientific studies. But as a society, this is not really how we would want things to be done. Surely, we want decisions about medication to be based on more on evidence than emotional arguments, irregardless of how effective emotional arguments are for convincing people.

    Even if something has a large power to convince people, it would not necessarily mean that it is a reasonable method of going about things.

    A victory based on mostly emotional arguments is a hollow victory, but I now understand that you think both evidence and anecdotes are important.

  47. 'Tis Himself says

    Emil is asking for a study on the incidence of racism in the atheistic community. I doubt such a study has ever been conducted. If Emil will provide us with the funding, we can accumulate the data and analyze it.

    Ball’s in your court, Emil.

  48. qdbp says

    @Emil: You still haven’t really addressed the fact that because racism is so prevalent, it’s fairly absurd for someone to claim ignorance on the “scientific evidence” of racism.

    “Give me scientific evidence that it exists!” is about as absurd as charging into a conversation about vaccinations and saying, “Give me scientific evidence that vaccinations even work!”

    Whether or not that applies to you, the fact is it still applies.

  49. says

    Greta Christina says:

    And you haven’t yet answered a question I have now asked more than once: Do you really not know anything about the history and the context of people from a privileged group telling people from a marginalized group to stop telling their stories, or saying that their stories aren’t important?

    It was not a factor in my evaluation of what arguments are most rational or practically effective.

    Emil Karlsson @ #49: Rationally, yes: these people should have accepted these studies as refuting their point. But they didn’t. And they kept insisting on examples of racism in the atheist community. Are you going to continue chiding me for providing them?

    If you agree with the propositions in 49, I have no further disagreement with you.

  50. says

    ‘Tis Himself says:

    Emil is asking for a study on the incidence of racism in the atheistic community. I doubt such a study has ever been conducted. If Emil will provide us with the funding, we can accumulate the data and analyze it.

    Ball’s in your court, Emil.

    No, I made the argument that data from the population at large plus the base rate argument should be enough to rationally refute the position of racists.

  51. Greta Christina says

    Greta Christina says:

    And you haven’t yet answered a question I have now asked more than once: Do you really not know anything about the history and the context of people from a privileged group telling people from a marginalized group to stop telling their stories, or saying that their stories aren’t important?

    It was not a factor in my evaluation of what arguments are most rational or practically effective.

    Emil Karlsson @ #54: Then shame on you. I urge you to find out about that history, and to pay attention to it the next time you patronizingly tell marginalized people to not tell stories about their marginalization because it’s an irrational form of argument.

  52. Greta Christina says

    No, I made the argument that data from the population at large plus the base rate argument should be enough to rationally refute the position of racists.

    Emil Karlsson @ #55: And despite having been asked several times, you haven’t yet said what people should do when this is not enough, and when people irrationally reject this argument.

  53. says

    Greta Christina says:

    Then shame on you. I urge you to find out about that history, and to pay attention to it the next time you patronizingly tell marginalized people to not tell stories about their marginalization because it’s an irrational form of argument.

    It may be emotionally and socially insensitive of course and yes, maybe I need to work on that, but the origin and history of a claim has no relevance for its rationality and calling an irrational argument irrational can hardly be patronizing.

    In the end, your reply is yet another straw man. I have never said that marginalized people should not tell their stories. On the contrary, they should be free to share their stories if they so choose. My argument was that such stories cannot be considered scientific evidence, does not rationally refute false claims about the problem being trivial and says nothing about the actual size of a problem.

  54. says

    And despite having been asked several times, you haven’t yet said what people should do when this is not enough, and when people irrationally reject this argument.

    Then we conclude that those people cannot be rationally persuaded and move on to others that can be?

    I mean, the problem of people being irrational and rejecting reasonable things is going to exist whether we use scientific evidence, anecdotes, both, neither, plum pudding or a cosmic death ray.

    It is not a unique problem to my approach and if you think your objection is forceful enough to undermine it, the objection will also undermine your own (and any other plausible approach).

  55. Brownian says

    A victory based on mostly emotional arguments is a hollow victory, but I now understand that you think both evidence and anecdotes are important.

    What? A hollow victory? For the rights of groups that are oppressed?

  56. Greta Christina says

    I have never said that marginalized people should not tell their stories.

    Emil Karlsson @ #58: Actually… yeah, you kind of did. In response to the question, “If you’re accusing people of playing the victim card, is there any possible way we can make oppression and marginalization and bigotry visible, which will meet with your approval?”, you responded, “more scientific evidence, less anecdotes.” Given that I have created exactly one blog post asking people to tell their stories about racism, asking for less anecdotes amounts to asking for no anecdotes.

    Also, you keep insisting that anecdotes aren’t a rational form of argument… even though you concede that they can be effective, and are therefore rational by the effectiveness standard. Again, given the ugly, nasty, patronizing, silencing history of marginalized people being told to stop telling their stories — and in particular, being told to stop telling their stories because they’re being “irrational,” and being told that their subjective experience isn’t relevant and that others will tell them the right way to talk about their marginalization — this is something you might seriously reconsider.

    It may be emotionally and socially insensitive of course and yes, maybe I need to work on that…

    Gee, ya think?

  57. says

    What? A hollow victory? For the rights of groups that are oppressed?

    No, we are still talking in terms of victory in the question of the size of the problem (i.e. convincing people that the problem is large with anecdotes), not victory in terms of socially beneficial outcomes of civil rights / social justice movements.

    It would only be hollow in the sense that anecdotes would not rationally refute the false claim that the problem of racism is low.

    Clearly, a victory for civil rights and social justice would not be hollow.

  58. reneerp says

    I have been reading this and trying to put my finger on what’s bothering me about Emil’s insistence on data to verify the prevalence of racism in the atheist community (and out for that matter). There are two things than seem to be true.

    1, He’s been given links to both existence of racism as it affects people’s perceptions of people of color, but has not acknowledged those links.
    2. He hasn’t got a good idea about how people change their beliefs or opinions.

    1 is poor manners, but 2 is worth some discussion. For one thing, facts don’t work on most people. For a second, people who have made huge shifts (say from being neo-nazis) have made them through personal experience, not some data-driven revelation. Narrative is a powerful way to bring people to understanding they could not have experienced otherwise.

    Changing minds is not like finding an efficient medicine. It’s not like debunking homeopaths. It’s about communication and data are only a part of any story.

  59. Greta Christina says

    It would only be hollow in the sense that anecdotes would not rationally refute the false claim that the problem of racism is low.

    Emil Karlsson @ #62: Translation: It would only be hollow in this definition of “hollow victory” that I just made up, in which whatever I want to talk about — i.e., my own incredibly narrow definition of what constitutes a rational argument — is more important than the rights of marginalized and oppressed people.

  60. CPS says

    Personal anecdotes are not only useful rhetorically or in demonstrating the existence of discrimination and harassment within the atheist community, they also tell us a great deal about the forms harassment and discrimination take. This information is invaluable to those who are actually trying to address these problems. The individual stories help give us a picture of exactly what sorts of harassment and discrimination are in play and so gives us an idea of what measures need be taken to prevent similar incidents from occurring in the future. With so many people having recently argued that sexual harassment policies would be ineffectual, it is important to be able to have these stories on hand in order to point out how certain policies might have addressed certain problems on particular occasions. How do you evaluate the efficacy of or defend a particular policy when ‘harassment’ is just some abstract, academic concept? Without concrete, real life examples this is nearly impossible.

  61. josh says

    Greta @ 47:
    Thanks for the clarification. If they asked for examples then it’s appropriate to ask your readers to post examples, but, unfortunately, that doesn’t really solve the problem of how typical or not those examples are. Also, I note that almost none of the replies to your examples post gave specifics, links, quotes or named names. I still think you’re trying too hard to make Emil wrong rather than understand and consider what s/he is saying, but I have a better understanding of what prompted your initial post.

  62. says

    Emil Karlsson @ #58: Actually… yeah, you kind of did. In response to the question, “If you’re accusing people of playing the victim card, is there any possible way we can make oppression and marginalization and bigotry visible, which will meet with your approval?”, you responded, “more scientific evidence, less anecdotes.”.

    No, that is something you infer from context not something I said. The reason I made my initial comment was against the use of anecdotes in questions about size, which I argued was the relevant question from a scientific standpoint. I did not agree with the sentiment you criticized in the blog post, namely people who accuse others of playing the victim card.

    I do not think that people who tell their stories of marginalization are playing the victim card. Even the very concept of “victim card” seem absurd to me.

    I responded to the strategy you employed in isolation.

    Sorry for the confusion.

    Given that I have created exactly one blog post asking people to tell their stories about racism, asking for less anecdotes amounts to asking for no anecdotes.

    I gave examples where you and others in the skeptical community employed such a strategy on other topics.

    Remember here that I am referring to less anecdotes in the particular context of trying to convince people that a problem is not trivial in size, not less anecdotes for intentionally using stories with emotional impact. That is something else and I have no problem with that.

    Also, you keep insisting that anecdotes aren’t a rational form of argument… even though you concede that they can be effective, and are therefore rational by the effectiveness standard

    We are using two different types of rationality here. I do not know if there are any words for the distinction in English, but I think they are originally from German. I am using Wikipedia definitions.

    Wertrationalität – Rationality in accordance with moral demands instead of organizational and technical demands.

    Zweckrationalität – Technical rationality; rationality in accordance with organizational demands instead of moral demands.

    I forget which one is which, but rational in one sense means doing things that effectively attains your goal. The other one is that rational means that which follows principles of reasoning.

    Clearly, there are circumstances where these are not identical. The best way to obtain a goal might be entirely against principles of reasoning and vice versa.

    So I guess it boils down to what type of rationality you prefer. The question one has to ask oneself is this: “Is the goal important enough to compromise with principles of reasoning?”

    But this seems like a moot point since we seem to agree that both evidence and anecdotes are valuable, although in different ways.

  63. says

    Greta Christina says:

    Emil Karlsson @ #62: Translation: It would only be hollow in this definition of “hollow victory” that I just made up, in which whatever I want to talk about — i.e., my own incredibly narrow definition of what constitutes a rational argument — is more important than the rights of marginalized and oppressed people.

    So you’re saying that you have no problem compromising with principles of reasoning as long as it attains your goal? See, I can also make straw man arguments. :)

    No, I did not make it up. The entire conversation has been about what methods are most rational in winning the argument against racists. I have not discussed solutions to the problem in any particular detail and so this has not been about the political victory for social advocates.

  64. ischemgeek says

    Emil, you’re trying to argue that because a screwdriver isn’t as versatile as a hammer, we shouldn’t use screwdrivers at all. As I said in Greta’s Catch-22 post, you’re missing the poin of what anecdotes are used for. You’re right that anecdotes do not provide information on scope of a problem. They are not a statistically significant sample size, even if some of them do come with supporting evidence, contrary to your assertation that they’re by nature unsupported (consider Ashley Mills’ anecdote about sexual harrassment, which was backed up by other witnesses). Here’s the thing, though: That’s not what anecdotes are for.

    Anecdotes provide something similar to a case study: An account of “this happened, here is how it happened, this is what made me figure out what was going on, and here is what I did about it.” A case study cannot on its own prove anything. What it can do is show that there is cause for concern or study in an area.

    A case study can also disprove an absolute statement: Joe claims that sheep are white. I found a black sheep. Therefore, I can conclude that not all sheep are white. I cannot say how many sheep are black, but that doesn’t mean that knowing that not all sheep are white is a useless bit of knowledge. Further, the more black sheep I’m able to find, the more I’ll be able to gain a rough idea of what sample size I need to collect to get a reliable idea of how often black sheep occur in a flock. In this case, anecdotes do serve to disprove the idea that racism is not a problem in the athiest/skeptic community. That’s an absolute statement, and therefore even one example to the contrary can disprove it. Does it detail the depth of the problem? No. But it does tell us that there is a problem and it gives some qualitative insight on how to recognize the problem and the subjective experience of victims.

    As a final note: I’ve noticed that a lot of people here seem to glorify the quantitative and ignore the value of qualitative data. If I use pH paper, the data it gives me isn’t quantitative like an analytical series of pH titrations would be. It’s qualitative. But it still serves a purpose: First of all, if I need quantitative information about the pH, it helps to know whether I should be mixing up a solution of standardized base or standardized acid. Titrating like with like is useless to me. Qualitative data gives you insight to the nature of the problem that helps you design your quantitative experiments.

    Second, sometimes qualitative data is all you really need: if the pH of the pool water I’m testing is within the ‘safe’ color zone of the pH paper, that’s good enough. I don’t need to know that [H+] is exactly 10-7.4213 +/- 0.001% in order to know that people aren’t going to get acid or base burns off it, that I’m not going to have problems with bacteria or algae, and that I won’t have scaling. The qualitative ‘safe’ vs ‘not safe’ is good enough. Anecdotes can be a social pH paper to give us that qualitative ‘safe’ vs ‘not safe’ judgement.

    If a number of people come forward and say “yeah, I’ve had that happen. It’s happened more than once and I’m not really that thrown by it because it happens so much. Other people in my group have relayed similar experiences to me, too,” in regards to racism, that gives us the qualitative ‘not safe’ and we can assume that as in the greater population, racism is a significant problem in the atheist community in absence of quantitative evidence to the contrary. Anecdotes like this would also give us a qualitative impression of the severity of the problem: they are inconsistent with a small-scale problem that is mainly unusual outliers (note that these are the dominant form of anecdotes in a lot of social justice discussions).

    If instead there’s a very small number of people who come forward at all and those who do tell stories along the lines of, “I had that happen once, and I was really surprised when it did because normally it’s not like that at all. It was really wierd. Other people reacted quite well, and overall I was happy with how it was dealt with. Nobody else in my group has ever experienced anything like that. It was strange,” that gives a qualitative ‘safe’. These anecdotes are inconsistent with a widespread problem and suggest the problem is more likely small, isolated outliers – qualitatively, people relaying the anecdotes are on average saying that it is not a common problem for them and that it is not common among their circle.

    So don’t dismiss anecdotes out of hand because they’re not quantitative. Qualitative data has value, too.

  65. Greta Christina says

    So you’re saying that you have no problem compromising with principles of reasoning as long as it attains your goal?

    Emil Karlsson @ # 68: No. I am saying that, when illustrating the seriousness of racism, both anecdotes and data are valid and important, that both serve different rational functions, and that using both does not compromise my principles of reasoning.

    The entire conversation has been about what methods are most rational in winning the argument against racists.

    No. That’s just what you derailed the conversation into.

  66. says

    Emil Karlsson @ # 68: No. I am saying that, when illustrating the seriousness of racism, both anecdotes and data are valid and important, that both serve different rational functions, and that using both does not compromise my principles of reasoning.

    That’s fine with me, as long as we are being clear with what sort of rationality (Wertrationalität or Zweckrationalität) we are talking about.

    No. That’s just what you derailed the conversation into.

    The point remains: what I was referring to was not political victories for social justice movements, but a rational victory (in the sense of adhering to principles of reasoning kind of rationality, rather than attaining the goal kind of rationality) against racism on the question of the size of the problem of racism.

    Also, since this post appears to be devoted to my views (or rather what you think were my views), I can hardly derail it by explaining why those beliefs about my views are wrong.

  67. Greta Christina says

    Emil Karlsson @ #72: Actually, the original topic of conversation was people who accuse marginalized people of “playing the victim card,” and the Catch-22 in which speaking about marginalization is seen as “playing the victim card” and is therefore rendered invisible — but not speaking out about marginalization also renders it invisible. In that post, I rhetorically asked these people, “Is there any possible way we can make oppression and marginalization and bigotry visible, which will meet with your approval? And you responded, “Yes! You could tell fewer stories and show more data!”

    This makes two mistakes. It assumes that people who accuse people of “playing the victim card” will be persuaded if they just see better data. And it places the blame on the victim. It treats this “playing the victim card” Catch-22 as if it were a rational argument that could be overturned by rational evidence — if only marginalized people would be more rational in how they present that evidence.

    And again, it does this in an ugly historical context of marginalized people repeatedly being told, “Don’t tell your stories, your stories don’t matter, we will tell you how to talk about your marginalization.”

    If you really don’t see how and why this is problematic, I don’t think I’m going to bother with this any longer.

  68. smhll says

    Threadrupt. Leaving my house in 5 min.

    I think we are looking at the nature of racism in the skeptical community (not the size). We don’t need to measure the number of these incidents. We pursuing the approach that of listening to people talk about their alienating experiences, and then looking for ways to stop doing that to them. The desired measurable outcome of our efforts is having greater minority representation in our community. That’s a pretty easy number to estimate in real life settings by counting heads. (I don’t see a need for a great degree of precision, just a trend line.)

    (If someone wants to improve my ideas and use the right names for the variables, please do so. I was striving to be comprehensible without resorting to raising my voice.)

  69. Greta Christina says

    Oh, on the actual topic of this post — namely, scientific studies of racism — someone made a comment on my Facebook page that I thought was worth sharing:

    ***

    I wrote the following as a response to a comment claiming that if anyone goes through the computer game of life on the hard setting, rather than the easy setting, it is white men. Some of it might be helpful: I would be very interested to see where you are getting your statistics. Generally speaking, blacks are much more likely to be killed by other people than whites are. For instance, in 1997, the rate of homicide victims per 100,000 was 6.7 for white males, and 47.1 for black males. http://www.allcountries.org/uscensus/139_homicide_rates_by_race_sex_and.html In 2005, the rate per 100,000 was 3.3 for white people and 20.6 for black people. http://bjs.ojp.usdoj.gov/content/homicide/tables/vracetab.cfm More generally speaking, the life expectancy of a black American man is, as of 2010, 71.8 years while white male life expectancy is 76.5.

    Simply not dying is not the only way to measure life though.

    Within the age category of 15 to 64 years, the prevalence of disability was low for Asians and Pacific Islanders (9.6 percent) and high for American Indians, Eskimos, and Aleuts (26.9 percent). The rate was 17.7 percent among Whites, 20.8 percent among Blacks. http://www.census.gov/population/www/pop-profile/disabil.html

    A black person is less likely to get a job – see http://money.cnn.com/2011/09/02/news/economy/black_unemployment_rate/index.htm – in part because black people are perceived as less competent, even when the black people in question are mythical, in that black sounding names were appended to white people’s resumes and given to people for review as part of a lab study. See See http://sites.nas.edu/wocconference/resources/exploring-bias/ and http://www1.eeoc.gov//eeoc/meetings/2-15-12/benard.cfm?renderforprint=1 and http://neuropoly.com/tag/black/

    38.2% of black children lived in poverty in 2010, compared to 12.4% of white non-hispanics. http://www.npc.umich.edu/poverty/

    Blacks are more likely to get cancer, and more likely to die when they do. http://seer.cancer.gov/faststats/selections.php#Output

    A black male is more likely to be pulled over by the police than a white one. If pulled over, he’s more likely to be searched. If searched, he is slightly *less* likely than a white man to have drugs on him, but he is more likely to be arrested for those drugs. If arrested, he is more likely to be prosecuted. If he is a minor, he is more likely to be prosecuted as an adult. If prosecuted, he is less likely to be released without bail pending trial. He is less likely to have bail money if bail is required. A side effect of this is that a black man is more likely to take a plea bargain and plead guilty, irregardless of guilt, because the other option is more likely to involve staying in jail for the 6 months to 2 years it can easily take to get to trial. When he goes to trial, he is more likely to be convicted. If he is convicted he will generally receive a longer/harsher sentence than a white man – he is more likely to go to jail than a white man, even when convicted of the same crime. If he is convicted of murder, he is more likely to get the death sentence. It’s not much better for black women. See, among other things, http://www.newjimcrow.com/

    If he gets out of jail and applies for a job, in at least one study while 17% of white job applicants with criminal records received call backs from employers, only 5% of black job applicants with criminal records received call backs. http://colorlines.com/archives/2011/03/snoop_from_the_wire_claims_innocent_for_drug_arrest.html

    So, actually, I think the “easy setting” analogy is pretty good, if somewhat simplified.

  70. Brad says

    English appears to not be Emil’s first language. This is causing you to talk past each other to some extent. (sorry about the idiom)

  71. Erista (aka Eris) says

    Oi. This discussion is vexing me very much, so I will try to be brief.

    Dear Emil,

    You do not get to tell other people how much racism must be present and how much racism must be impacting them before they can do something about it. If someone has 100% anticdotal

  72. Erista (aka Eris) says

    Oops, sorry, my last post went through unfinished.

    Let’s try again.

    This discussion is vexing me very much, so I will try to be brief.

    Dear Emil,

    You do not get to tell other people how much racism must be present and how much racism must be impacting them before they can do something about it. If someone has 100% anecdotal evidence indicating that racism is a problem TO THEM, they can focus on it. If it turned out that racism only happened to one person ever (ha ha), then that person and their friends would be MORE than justified in fighting racism. There is no threshold at which you get to tell people that they can’t deal with issues that are impacting those people or the individuals who are important to those people.

    I don’t know what the fuck you think your point is. Is there some magic percent that you will start caring? What if 5% of people experienced being treated in a racist manner? 10%? 30%? Do you really think there is some goddamned number where you get to decide what other should spend their time on? You say you want data, but I don’t know what you want to do with it.

  73. Mark Erickson says

    If there is such a thing as an Internet character witness, I’ll be one for Emil. I stumbled upon his blog over a year ago and he posts good stuff. But that’s just background and a basis for my suggestion for those in this large audience to give his blog a chance and a read.

    I agree with brad that you are talking past one another, but probably for more than just language reasons. Apologies to Emil if I get this wrong, I’m just trying to help. I think Emil is a type of hyper-rationalist and/or lacking in empathy to see things through another’s eyes. The plural of anecdote is not data, all agree. However, Emil doesn’t “get” that’s not the crucial issue here and how the history of marginalization impacts the debate. It’s personal. He might not be able to “get” that. And he’s very stubborn with lots of practice at dispassionately evaluating and forming arguments.

    So while this can’t really have an agree to disagree amicable ending, I hope that – well I don’t know what to hope for. Just had to put myself into the middle I guess. May no good deed go unpunished.

  74. CPS says

    Uh, Mark…isn’t “I think Emil is a type of hyper-rationalist and/or lacking in empathy to see things through another’s eyes.” just another way of saying “The guy is an ass-hat”? And isn’t “However, Emil doesn’t ‘get’ that’s not the crucial issue here and how the history of marginalization impacts the debate. It’s personal. He might not be able to ‘get’ that.” more or less just accusing him of being kind of stunned? I know you mean well, but as a character witness, I’m not sure you are doing him any favours.

  75. Brownian says

    Mark, I don’t particularly care about Emil’s character. I accept that he’s probably a good person, with benevolent intent.

    That’s irrelevant to the fact that I don’t think he has any idea of what he’s talking about, other than some weird notion of honorable skepticism that trumps efficacy.

  76. Mark Erickson says

    Cps, they are actually two different things. My character witness is for his body of work. Probably the wrong expression. I only know him through his blog. The second paragraph is about neurotypicalness. It might be a factor here. (again, apologies to Emil for this, I could be completely wrong) My son is autistic. If he came up to you and said something socially inappropriate, you would be wrong to call him an asshat. So I’m just saying give that interpretation a chance. Be generous. It doesn’t usually lead you wrong.

  77. Mark Erickson says

    Brownian, honorable skepticism is exactly right. That is exactly what he’s going for. I’m not sure how that opposes efficacy in this case, I must admit I didn’t read every word of the back and forth. I’m not arguing who’s right and who’s wrong.

  78. Brownian says

    Brownian, honorable skepticism is exactly right.

    it does not trump actually reducing the real effects of discrimination.

  79. Greta Christina says

    I think Emil is a type of hyper-rationalist and/or lacking in empathy to see things through another’s eyes.

    Mark Erickson @ #79: You say that like you think it’s a defense.

  80. TwoPiDeltaIJ says

    I do not think it is correct to say that Emil is going for honorable skepticism or hyper-skepticism, I think instead, just skepticism based on rational argument (which was explained to mean logically rational and not rational in the sense of effective choices). I think he has been very clear about that. In that framework, he seems to be entirely correct (as far as I can tell) in his argument. That is not to say that Greta et al. are wrong, I think that Emil is correct they were using different uses of ‘rational’ in their respective arguments.

    From a rational (logical construction of argument) point of view:
    1) It is true that the source of a claim (or its history) have no bearing on its truth.
    2) It is true that anecdotes are not relevant arguments to anything other than questions of existence.
    3) Once you have presented a rational (in this sense) argument to someone who can not refute it and continues to assert the counter position, you are done having a rational argument with them. You are very unlikely to convince someone who refuses to argue reasonably with reason (which has been pointed out several times in this thread by both sides).

    Emil also points out there is a flaw in using the other form of rational argument because it can be understood as ‘the ends justify the means.’ The response to this was something like ‘that is not what I am arguing for.’

  81. Ciaphas says

    Near as I can tell, Emil’s argument could be summed up as “It’s better to fail to convince people rationally than to successfully convince people emotionally” .

    That is not a rational position to hold when talking about civil rights.

  82. Erista (aka Eris) says

    @Mark

    I don’t see how any of what you said would indicate that we should be any more accepting of what Emil said than we are already being. Your argument seems to boil down to something equivalent to, “Yes, he did just punch you in the face and take your wallet, but you need to understand that he just doesn’t care about your well being.”

    Emil doesn’t have to “get” why any of this is important to anyone. That is not the issue. The issue is that Emil is arguing that people have no right to be upset about racism; he’s arguing that because the significance of racism hasn’t been proven TO HIM, that other people shouldn’t spend their time and energy That is a serious issue.

    I don’t know your son, but I do know what it’s like to be somewhat neural atypical, and I do know that when I do something inappropriate, something that I hadn’t expected to hurt someone and don’t understand why what I said was hurtful, I am sorry for what I did; I do not insist, “Well, you have to provide me with empirical data supporting your right to be upset.” I understand that even if I do not “get” where the hurt comes from, that hurt is very real, and it doesn’t work via empirical data.

  83. ischemgeek says

    @Eris (Erista)

    Yes. All of that. I get really annoyed by those who try to excuse bad behavior by bringing up ASDs. I have several relatives and friends who are on the spectrum. Yes, you would be wrong to call someone with autism an asshat for doing something inappropriate. However, there is a big difference between the behavior pattern generally exhibited by those with ASDs and the behavior patterns of neurotypical people who are just insensitive.

    Someone with an ASD generally acts similar to someone neurotypical accidentally made a social gaffe: They are usually surprised that people were hurt, and may ask questions about the source of the hurt, but they will in general accept that the hurt exists, apologize and try to make amends. The difference is that while the neurotypical person in face-to-face interaction will usually realize the problem and start trying to correct it on their own (by picking up on the hurt or offense in the other person’s body language and facial expression), the person with an ASD will more often need the problem explicitly brought to their attention by someone else. People with ASDs, in my experience, tend to ask a lot more questions about the source of the hurt and will often mentally dissect the encounter afterward, looking for where they went wrong. This isn’t a 100% always happens this way thing, and yes, people with ASDs can be jerks sometimes, but in general, this is how I’ve found they react.

    Someone who’s just being insensitive may be equally surprised that people are professing hurt in the case where they honestly did not intend to offend, but will then either judge the hurt (“there’s no reason to get all upset about it!” and variations on it), deny the hurt (variations on inappropriate ‘playing the victim’ accusations and ‘pics or it didn’t happen’), or downplay the hurt (variations on “even if it did happen, it’s isolated and not a big deal” or “I didn’t mean to do that so you shouldn’t be that upset. It’s not like I intended to be an asshole…”).

    And then there’s people who are just assholes: Those who have been warned that a certain behavior causes hurt and who do it anyway because they either don’t care or they’re sadistic and take pleasure in others’ hurt.

    Neither of the last two cases should be excused by bringing up ASDs. Trying to excuse such behavior by bringing up autism is ableist by:
    1) implying that those with autism-spectrum disorders are lacking in empathy (they’re not; they have difficulty reading social cues and expressing themselves. There’s a difference),
    2) denying that those who are neurotypical can make asinine social gaffes and act like jerks sometimes. We can.
    3) implying that such severe social gaffes are diagnostic of autism-spectrum disorders – that is, implying that people with autism-spectrum disorders consistently make social errors of that severity, to the point that if you see such a social gaffe, you can assume the person in question has an autism spectrum disorder much like seeing someone using an inhaler can let you safely assume they have a respiratory disorder.

    Finally, please note that I think there’s a difference between “being insensitive” and “making a social screw-up” even among people who are neurotypical. IMO, the difference between a whoops and a wrong is how you react to realizing you did it/being called on it.

  84. Maggie says

    I was trying your blog for the first time. You totally lost me when you called the distinction between anecdotes and scientific evidence “ridiculous.”

    Last time I read something like this, I was mauled by a tiger. I don’t want to get mauled again, so I encourage everyone to be wary of this post.

  85. says

    “if the question on the table is, “Does this happen?”, and a whole lot of people say, “Yes, here are examples of it,” then unless you think those people are lying, that settles the question.”

    Didn’t CS Lewis say something similar? Here are some anecdotes about a guy who says he’s God, and the anecdotes say he believed it, so he must be God, right? Or wrong? Isn’t it quite possible that some of these anecdotes have been fabricated, and others have grown in the telling, and some are subjective evaluations by people whom you would not normally trust to accurately report events?

    And isn’t this why we need objective scientific evidence, to determine how much to trust the anecdotes which have drawn our attention to what appears to be a problem? Because as objective investigators with a strong regard for truth and accuracy, we want to make sure that our actions are based on the best possible information we can get.

    Don’t we? Or do we just trust CS Lewis’s anecdote about the anecdotes about the anecdotes told by Jesus?

  86. says

    @Jon Jeremey, #92: Do you really not see any difference between claims of experiencing racism and claims of experiencing God? The existence of God is an extraordinary claim, whereas the existence of racism is so beyond ordinary that it is systemic. Apply some critical thought to this, for the love of dog.

  87. Erista (aka Eris) says

    @Jon Jeremy

    No one is saying that scientific evidence isn’t important; the only thing anyone is saying is personal stories are ALSO important. To say something along the lines of “only scientific studies have any bearing on how our community can/should/does deal with racism” isn’t reasonable. Humans are not creatures that are ruled solely by peer reviewed sources; we are also made up of our own personal experiences and the experiences of others. In fact, people are more influenced by personal experiences than peer reviewed papers if for no other reason than many people rarely if ever even glance at a peer reviewed paper. On the other hand, no one gets through life withing experiencing the stories of life. For example, many people decide to spend significant portions of their lives raising money for cancer research after they or someone they care about experiences cancer, and many an anti-gay bigot has been converted when they come to know and love someone ho is gay or realize that someone they already know and love is gay. These people were not convinced by some peer reviewed article; they were convinced by stories of every day life.

    This is why creating a distinction whereby scientific evidence is allowed but personal anecdotes are not is incredibly strange and completely counterproductive. We are not emotionless beings who are influenced only by data. To act like we are is to ignore a truly massive chunk of what it means to be human.

  88. Greta Christina says

    You totally lost me when you called the distinction between anecdotes and scientific evidence “ridiculous.”

    Maggie @ #91: [facepalm]

    Did you bother to read the very next sentence? The one where I explained the reason I thought this distinction was ridiculous? The one where I made it clear that of course there’s an enormous distinction between anecdotes and scientific evidence — and that both are valid in an argument? The one where I made it clear that it’s important to have both, exactly because they’re so different? The one where I made it clear that the “distinction” I was talking about was not a distinction between anecdotes and scientific evidence… but an artificial distinction between using just one or just the other in an argument, rather than both?

    Reading for comprehension. Try it sometime.

  89. abenaafreeka says

    “Remembering and telling the truth about terrible events are both prerequisites for restoring social order and healing individual victims” (J. Herman M.D.,Trauma and Recovery:From Domestic Abuse to Political Terror, 1994)

    Having had personal, educational and professional experience with working a program of recovering from some terrible events, “keeping it simple” and “flip the script” are tools used to assist VIEWING situations differently or from an alternative perspective and I apply that thinking to those presenting the smoke screen of “playing the victim” when I ‘tell my story’.

  90. Steve says

    There is no “link” for this one because it was published as a book by Grove Press in 1965: “Sex and Racism in America,” by Calvin C. Hernton.

  91. says

    I had always thought sharing one’s experiences would help foster empathy and humanize folks. I have some family members who were anti LGBT, but they changed their minds about that because they realized they knew people who were LGBT. These people shared their stories, their trials, etc, and that made my family members more empathetic to them. Now they are very pro LGBT.

    I think this, in tandem with scientific studies that shows discrimination against group x is a wide spread problem, is a good method for raising awareness and promoting change. Essentially what this doe is humanize people. Too many folks, unfortunately, think people from group X are merely debate topics, not people endowed with the ability to feel and hold unique insights directly relevant to their experiences.

  92. says

    One more thing, I experience discrimination sometimes, and it hurts me a lot. Most of it is outside of the atheist/skeptical community. However, being that many people in the community don’t look like me, and aren’t treated like me for the most part, sometimes they have a hard time understanding I go through stuff. If I try to share, for instance, how hurt I felt when I experience x, it really hurts to be told “are you sure? that’s never happened to me? Maybe you were acting in a way that caused this reaction?

    Being told that I should not share my experiences, the things that happen in my life, is really insulting. And the thing is, I can share all day long the things I have experienced as an atheist, and no one questions whether or not someone told me “you’re going to hell.” I’m pretty sure it’s because they experience similar treatment. I don’t think I want to be a part of a community that tells me, implicitly or explicitly, that the ongoings in my life have little or no value because my experiences are different from theirs and because of that I need to pad each account with comprehensive data because I could be a liar or just looking for excused to be offended or hurt. But experiences that are similar to their experiences, will be accepted at face value and you will be treated with empathy.

  93. Terry R says

    A little late however I believe that the following links may be helpful:

    Project Implicit – http://projectimplicit.net/index.html

    “Project Implicit is the product of a team of scientists whose research produced new ways of understanding attitudes, stereotypes and other hidden biases that influence perception, judgment, and action.”

    An interview with the developers of Project Implicit outlining the research.

    http://www.edge.org/3rd_culture/banaji_greenwald08/banaji_greenwald08_index.html

  94. ben says

    Greta@61:

    you keep insisting that anecdotes aren’t a rational form of argument… even though you concede that they can be effective, and are therefore rational by the effectiveness standard.

    Perhaps it’s a hollow victory because it legitimises irrational arguments, which grants more power to the weaker side. I guess that any time you appeal to an emotion rather than making a rational argument you are (1) suggesting to rational people that your own position is too weak to be supported by facts; (2) condoning emotional “arguments”; and (3) making the arguee less likely to rationally resist similar emotional “arguments” from the weaker side. It seems that the best solution for all concerned is to try to educate people on the power of rationality, rather than just stooping to their level.

    What if that fails and you find yourself wanting to redefine “rationality”? I like Machiavelli as much as the next aspiring evil overlord, but if your measure of rationality is “We won”, then you should consider switching from blogs to guns. But I still think that you’d like to be in a position to recognise when you’re wrong by some other means than that someone else has a bigger gun.

    Why is effect size important? Because it lets you pick your battles. What if you could prove that racism were universal in the atheist community (or wherever) _only_ in that darker skin led to longer waits for service at the bar, and that the difference between white and black was 1.3 +- 0.004 seconds? That would be strong evidence for omnipresent racism, for sure, and it should be addressed IFF all the bigger problems already had been. But we have limited time and energy and we can’t solve all the problems, and emotion without data is a poor way to decide on the importance of a problem. Otherwise you get situations like the USA’s reaction to terrorism: your chances of dying due to obesity-related illnesses are about 1:3, by car crash 1:80, by terrorism 1:9000000. Would it make sense to channel a million times more effort and money and time into solving the obesity crisis, a hundred thousand more times the effort into reducing car addiction, etc., than into fighting terrorism? 9/11 showed us that there is a problem with terrorism; further study showed that it’s less urgent and less devastating than some others. Of course any good person would like to solve all the problems, but what if that’s impossible? If racism were causing a few million times less hurt than obesity-inducing conference food, then I would propose focusing more energy on the menu than on racism. And vice versa. In any case, anecdote is a pretty poor measure of problem size, and tends to lead to poor allocation of resources.

    What anecdotes and emotional pleas _can_ do is the following:

    * Convince stupid people–at least until they hear a more compelling emotional plea from some other side. If you can convince enough people all at once to reach “critical mass”, then the existence of other positions will be challenging (whether or not they’re more rational (hollow victory etc.)).
    * Make people aware that the problem is prevalent enough to warrant further study!

    I don’t know why I didn’t notice any discussion of the second point until #70 (ischemgeek), who hit this quite eloquently. But I’m not sure the right people listened:

    Greta @ 73:

    This makes two mistakes. It assumes that people who accuse people of “playing the victim card” will be persuaded if they just see better data. And it places the blame on the victim. It treats this “playing the victim card” Catch-22 as if it were a rational argument that could be overturned by rational evidence — if only marginalized people would be more rational in how they present that evidence.

    The first “mistake” is not necessarily a mistake. There are nontrivial numbers of people who can be persuaded by data (but they’re not receiving the bulk of “personal fact” research funding or publicity right now). It usually doesn’t happen immediately, and often enough it doesn’t happen at all (varies depending on sense of identity in connection with their error), but it happens. I have anecdote to support that, but of course the fact that society makes progress is decent evidence. The second “mistake” is rubbish: I didn’t see Emil asking minorities to provide the studies, but rather asking opinionmakers and decisionmakers to do so. It seems quite reasonable that influential people might want to advocate study, but maybe I misinterpreted Emil?

    Erista (Eris) @ 78:

    If it turned out that racism only happened to one person ever (ha ha), then that person and their friends would be MORE than justified in fighting racism. There is no threshold at which you get to tell people that they can’t deal with issues that are impacting those people or the individuals who are important to those people.

    MORE than justified, to address a single event that happened in the past? Sure, people can fight for whatever they want. But they should not expect that anyone else will take very much interest in fights against small problems when large problems exist. And I’d argue that using emotional pleas to try to distract people from bigger issues is selfish and antisocial. So we’re back to: in what order do we want to tackle all these issues?

    On the other hand, anecdote gives us _some_slight_ idea of the size of a problem. Thinking that you know more about the size of a problem than you actually know is foolish, but the opposite is true too.

    I’ve now given two examples that trivialise racism. I feel I need to spell out for some people here that I know quite well that racism exists and is a big problem. I’m just trying to illustrate why asking “How big is the problem?” might not be irrelevant, and how quantitative data might be kind of a big deal.

    So it’s not a question of “Racism is at 18.3% and our threshold for action is 19%”. It’s a question of “Given that we have these problems and this much time and these possible options for addressing the problems, is racism causing enough hurt that we should move some resources away from combating sexual harassment?”

    Emil: your clarity and patience are astounding. Keep it up!

  95. says

    I’m diving into this extremely late, but…

    Emil, it strikes me that you’re insisting on first-order rationality to the point of second-order irrationality. Rationality is about observing the world and taking actions based on how the world currently is, not what you think the world ought to be. In the real world, people are not rational: force-feeding them rational evidence and hoping that they will magically sprout an appreciation for rationality in the process is not going to cause them to accept your evidence, much less actually become more rational. They will simply discard everything you have to say: Confirmation Bias, itself a powerful effect, will be the very least of your problems on an emotionally charged topic like racism.

    If you were to do a controlled study comparing ways to convince people to change their mind on long-held beliefs… and I recall that various such studies have been done in the Heuristics and Biases literature (though I’m too lazy to look them up and cite them right now)… then you would find that quoting prevalence statistics and efficacy results at people has such a small effect that you could use that as the placebo control for more effective forms of persuasion, such as emotional anecdotes or even slippery slope arguments.

    In effect, you’re chastising Greta for being rational about the problem she faces. I suggest reading Rationality is Systematized Winning on Less Wrong.

Trackbacks

Leave a Reply