The bias started out with a good intention - "all people should have the same rights and opportunities".
But the official, carefully constructed non-offensive party line reasoning behind it in the US seems to be something like "all people of all types ever are the same", which is just not true. Any one deviating from the party line is automatically labelled bigot, and anyone defending these people is lumped into the same bucket.
Its NOT wrong to make generalizations about people. Our brains are pattern finding machines - we generalize all the time. It IS wrong to use them as justification to treat certain people differently - because a statistic computed on a population of millions doesn't really tell you anything about a specific person. So it makes sense to discard the generalization prior - i.e be unbiased and not view someone through stereotypes.
Its a hard problem - tolerating honesty but not bigotry, someones right to be offended vs. the offenders right to speak their mind etc etc. But right now US culture is very biased towards not being offensive, at the cost of truth - at the least, this kind of attitude does not belong in research. Research needs to be about truth.
The politics are addressing racism is hard. White people don't want to take responsibility for something that happened before they were born; minorities won't accept their current problems are their fault.
Neither group is wrong for feeling the way they do, but the hard truth is accepting these positions is the only starting point to undoing hundreds of years of assholery.
Slavery was not a white problem, it was a humanity problem of multiple skin colours, and in some places, still is.
No one should be held accountable for their ancestors actions. I certainly don't look at anyone Who is Japanese and think "Pearl Harbour", it's just not sensible.
Agreed. In super-liberal circles, i see people sounding alarms at the mere mention of race, even in a completely non-offensive way.
You can't tackle a problem you aren't even willing to talk straight about. The current system is something like attempting to make an ostrich walk while its head is forced in the sand.
We categorize problems endemic to particular racial groups all the time.
For instance, blacks, hispanics, and poor whites are largely over-represented in the prison population. The reasons for this are many, but are largely endemic to the system, rather than being endemic to these groups. Inequality of education and opportunity is, in my opinion, the primary factor; followed closely by the War on Drugs. But other factors include inequality in how law enforcement treats suspects in these groups and inequality in legal defense.
Solutions to some of this are fairly straightforward. Educational opportunities should be improved for the underserved. Real legal defense should be provided for those who cannot afford it.* Law enforcement needs to be overhauled and demilitarized.
The problem is there is no political will among people who matter to make these changes.
I think maybe you are talking about cultural issues of particular groups that contribute to problems. There is the concept of "acting white" which does hold down certain ethnic groups in some areas, particularly education. I do think those problems should be addressed, but I also think they are addressed, just not at city, state, or national level. I see these problems addressed largely in support groups within and without communities. My college, and many universities have programs tailored to provide help for underserved groups and addressing these kind of issues falls under their aegis. (A matter of pride, the program developed at my college served as the template for all such programs in colleges and universities across California.)
But even so, I think addressing the larger issues will not only solve many of the issues that are specific to particular groups, it will also provide the moral grounds to avoid accusations of racism when addressing those issues directly.
*In my county, the public defenders are private. They are paid per case and are therefore incentivized to take on as many cases as possible. The quickest way to deal with a case is to encourage their client to take a plea deal regardless of the merits of their defense. In other counties, the public defenders are holding public offices and are incentivized, therefore, to win cases to further their careers. However, often these defenders are given case loads so large they can literally only spend seconds or minutes on any individual case.
I mostly agree with you, especially all the factors you mentioned. But like you said, I am "talking about cultural issues of particular groups that contribute to problems".
I think they play a bigger role in perpetuating these problems than you may think. Or at least, they offer a good way for us to attack these problems.
You're right. I think the problem is that white people are bad. Now it will be easier to solve the problem, and white people will benefit from this knowledge.
Btw, in areas where it's OK to apply math (like medicine or spam filtering) we have a science which tells us how to determine posterior probability given prior and new information.
The more academic paper I linked to (by the author being discussed in this article) suggests that people use stereotypes a lot like a Bayesian prior. Once they have knowledge about a specific individual, that knowledge usually drowns out the stereotype.
The part that I found particularly interesting is towards the end of this article. That using stereotypes [as priors] makes people more accurate. And ignoring stereotypes makes people less accurate.
This is another thing that makes stereotypes tricky - it takes a little bit of statistical insight to see how they can be both right in some sense but completely wrong in the other, but statistics is very unintuitive, and math doesn't grab headlines. So that kind of argument is difficult to use with non-nerds, but ultimately i think its the only accurate one.
That brings up another good point in that its really a perfect opportunity for media outlets that like to turn controversy into clicks. Its very controversial and very broadly applicable.
I do my best to avoid generalisations about people, unless they consciously choose to label themselves (eg. KKK is racist, feminist is sexist, etc).
There is a difference between looking for patterns in people and stereotypes. For example, walking down a dark lane at night in Kings Cross (Australia) is dangerous with any person you meet. Seeing a black person in Pitt St is unlikely to be dangerous. A person wearing tracksuit pants and a backward cap walking in an agitated manner may represent a threat. A woman in a dress might or might not be a CEO. Now, how does a stereotype of skin colour, gender, sexuality, etc help anyone?
I agree that the US is particularly biased and that truth is the victim. I went to code.org with my young son who asked "are boys not welcome?" The link on the title page said "why girls should be in IT". Political Correctness - another word for prejudice. The idiot who approved it didn't give a crap about the message to young boys, they're just a casualty of war.
I agree. This entire issue is rooted in different axioms about what should be "equal". I hear suggestions and implications that many believe equality of outcome is the goal for which we should be striving - the basic assumption here being that all people are inherently equal, and if equal outcomes are not achieved, then it must be a systemic issue causing the inequality of outcome.
The problem is the axiom of "all people of all types ever are the same", as you pointed out. It's false. Our inherent advantages and disadvantages are as varied multi-dimensional as human individuality itself. A proper axiom from which to base a philosophy upon would be that every individual is inherently different, with advantages and disadvantages determined by nature. Given this truism, it can only be asked of a human society that all these different people be provided an equal playing field. This playing field does not guarantee equal outcome and does not provide preference - It does not aid the inherently disadvantaged, nor does it cripple the inherently advantaged.
The trick, of course, is in implementing a "level playing field" given some of the generational feedback loops which exist. That the wealthy tend to stay wealthy and the poor tend to stay poor, generation after generation, shows that there is some "unlevel" part of the American playing field. What we should expect in an ideal "level playing field" would be that children born in any socioeconomic situation would have an equal likelihood of ending up in any other socioeconomic situation as their outcome. Perhaps slightly bias this towards moving "up" the economic ladder to keep pace with the ideal of incentive and capitalism, but the point remains.
Many of the liberal bias coming out of the soft sciences would be sharply curtailed if we started a discussion about the fundamental axioms. A fair society is one that acknowledges the inherent differences in humanity and still provides equal rights for all.
With this said, I really don't think the US has done a bad job at creating something close.
One of the more surprising findings is that ethnic and gender stereotypes are a lot more accurate than social psychology papers (see table 10.4). I haven't double checked the meta-paper that derived the first column, but if true, this is a scathing indictment of social psychology.
I suspect that the idea that stereotypes are inaccurate might be a noble lie to counteract how bad we are at statistics.
Let's say we have a stereotype "blue people are quick to anger." Even if there is truth in it, what it really says is that the median for the phenotypic trait "quick to anger" is somewhat higher for blue people.
That does not mean that a given blue person is quick to anger since the whole population is not the median. It also doesn't mean that white or black or red people are not quick to anger, etc.
People tend to take statistical generalizations about wholes and apply them like facts to the parts. In the case of people it results in great injustice. Given the above imaginary stereotype, does it follow that blue people should be barred from becoming police or caring for children?
So I wonder if we haven't made the whole thing taboo and constructed a kind of elaborate ruse to pretend group differences don't exist to correct for this.
We have lots of hacks to correct for irrationality. Inflationary currency for instance is a hack to correct for our psychological association of bigger nominal prices with actual progress. Religion is full of hacks to short circuit undesirable social behaviors too. I have also suspected for years that God might be a hack to create a "virtual alpha" to placate our need for an alpha male and allow the emergence of the rule of law.
Stereotypes can be accurate but also become false over time. Continued belief in them can also make them self-fulfilling. It wasn't until relatively recently that "women are bad at math" was examined and found to be a product of various cultural mores. Let's not confuse current accuracy of stereotypes with some permanent classification of behaviors innate in a group.
One point that is often overlooked is that, even if a stereotype tends to be accurate, there's, in many cases, no need to assume that it applies to a particular person.
That is, rather than making a judgement, you can, and often should, be agnostic until some definitive evidence comes to hand.
That is completely contradictory. If stereotypes tend to be accurate, then that is a reason to believe it applies to a person of that group.
As the article states, people (rightly) abandon these when dealing with a specific person, but mathematically speaking, a Bayesian approach would assume the stereotype until evidence to the contrary (with varying degrees of conviction depending on the prior). As long as you do not have a ridiculously strong or weighted prior and don't act on this presumed belief before confirmation, it is a useful thing.
Edit: this is an attempt to paraphrase the article's reasoning for moving from stereotypes to their researcher as a subject.
Because people will shut down interesting research which doesn't conform to the stated position of the social justice movement. This happens even if the research is high-quality and relatively unbiased. The research on stereotyping is a good example because the research doesn't attempt to cast stereotypes in a bad light, therefore it must be bad research.
That's exactly his point, and honestly, I agree with it.
An addition that I would make is a large point that he made by mentioning the paper with the faulty statistical data-- there are a lot of research papers being published that are conforming to the social justice movement, but aren't well backed, yet still accepted because of their motivation. The field is being diluted by this bad research that's trying to "change the world," instead of searching for objective truth.
“But out of 1145 participants, only ten agreed that the moon landing was a hoax!” he said. “Of the study’s participants, 97.8% who thought that climate science was a hoax, did not think that the moon landing also a hoax.”
So what? This doesn't necessarily invalidate the argument.
Curiously though, it contributes to falsely exaggerated stereotyping, in the sense that given the title and abstract I might expect a far greater percentage (>2.2%) of moon landing disbelievers among climate change disbelievers until until I take time to read through column 8 of table 3 on a potentially publicly inaccessible pdf. Will the journalists report these details? Maybe. Maybe not. Which plays into a larger possible tendency in social psychology: It's a poor choice to publish negative sterotype about groups supported by your social justice allies (most of the field, apparently), but it may be more warmly received to sterotype the political opposition, easiest with the low hanging fruit of conspiratorial right wingers.
Nobody ever claimed it did. The "what" is that it shows the evidence presented in the paper is bad for that argument. That says nothing about the truth of the argument.
My important take from the article:
"people apply stereotypes when they have no other information about a person, but switch them off when they do" - so stereotypes are not as bad as previously thought :)
The problem is that people often have important interactions, in some form, with those that they don't really know at all. Two politically relevant examples would be police officers arresting someone, or recruiters/interviewers reading resumes and interviewing people.
I for one will be downvote this link into oblivion, because it's scientifically proven that information like this can have a negative effect on women and minorities in tech.
I realize picking up on sarcasm on the internet is tricky, but it's also a bad sign for society when a ridiculous circular statement is close enough to many people's genuine beliefs that it's taken seriously.
This sort of thing further reinforces the typical disdain that "hard" scientists have for "soft" sciences. And it's a shame because these are important things to understand with rigor.
But the official, carefully constructed non-offensive party line reasoning behind it in the US seems to be something like "all people of all types ever are the same", which is just not true. Any one deviating from the party line is automatically labelled bigot, and anyone defending these people is lumped into the same bucket.
Its NOT wrong to make generalizations about people. Our brains are pattern finding machines - we generalize all the time. It IS wrong to use them as justification to treat certain people differently - because a statistic computed on a population of millions doesn't really tell you anything about a specific person. So it makes sense to discard the generalization prior - i.e be unbiased and not view someone through stereotypes.
Its a hard problem - tolerating honesty but not bigotry, someones right to be offended vs. the offenders right to speak their mind etc etc. But right now US culture is very biased towards not being offensive, at the cost of truth - at the least, this kind of attitude does not belong in research. Research needs to be about truth.