Hacker News Reader

Comment by Kenji | Hacker News Reader

I don't think the other people are at fault here. I think it's reasonable to assume that someone with a stereotypically black name is black, just like it's reasonable to assume that a random IT professor is male based on the fact that 90%* actually are. As long as you owe up to your error when you notice it (and e.g. apologize). It's a reasonable optimization that works out more often than not.

The problem only starts when the treatment of the person depends on the category they belong to and if that difference in treatment is not based on reason and evidence.

*example, may not be accurate

View on HN

We all make decisions and assumptions based on our past experiences. It's perfectly natural that if we keep seeing the same input leading to the same output over and over, we begin to associate that input with that output. Sometimes the input doesn't lead to the output, which helps us inform our future actions, but there's nothing inherently wrong with making assumptions and taking actions based on our experiences.

We all have to make decisions based on incomplete information. This means the decisions you make are approximations; their accuracy is contingent upon a lack of good information. It's not about assumptions per se: whatever you assume will be partially wrong.

With that said, if you want a high degree of accuracy in matters like this, you're going to have to give up many of those highly efficient heuristics we've evolved. That may not be a good idea. Instead we develop more heuristics that tell us when to apply simplifications and assumptions, and those will often fail.

Using name is a reasonable way to optimize figuring out someone's race. The question is: why are we trying to figure out someone's race? If you're talking about reducing effort, isn't the easiest thing to just not ponder the question of someone's race? People wouldn't bother to try and answer that question if they didn't care about what the answer was.

Now, the fact that people care is not in and of itself evidence of racism. To some extent, people just want to put a face and a narrative to a name. The question is, what's in that narrative?

That's roughly right in a moral sense, but in a Bayesian sense I don't think it's unreasonable. If race (or background or country of origin, etc.) and some attribute (education, wealth, etc.) are correlated, and names and race are correlated, it's intuitive that names and that attribute are likely correlated.

It's a huge problem when you reverse the process and say, "Despite evidence of having X attribute, I'm going to treat you as though you don't because you belong to a set where X attribute is less common." So overlooking someone's resume because of their race/class/background/etc. is absolutely racist and unreasonable. Having a biased expectation a priori, however, is not necessarily unreasonable. However it may be morally wrong because having and acting on those expectations reduces the opportunities given to members of those groups who are outside the norm, which in turn reduces the group's ability to change those correlations and expectations.

Nobody meets you and says: "While it's reasonable, in a Bayesian sense, to conclude that someone with a 'black' name like mine grew up in a low-income, low-education household, in fact both my parents have graduate degrees and are high-income professionals."

A lot of those assumptions that people make, while statistically justified, will never be rebutted in individual instances because it's socially impractical to do so.

Yes. That's what I was trying to say with the "morally wrong because having and acting on those expectations reduces the opportunities given to members of those groups" bit. Creating uphill battles doesn't allow the situation to improve.

Well, in the one example given, a waiter used the name to discern that the debit card would probably belong to a black person. That seems like a perfectly innocent use.

If a waiter isn't certain who owns a particular credit card, I do not think handing it out to someone they think matches the name, is all that sensible a course of action.

Especially when they are literally holding a card with the name of the person they are looking for on it.

There are fairly well established social methods for dealing with that conundrum and none of the good ones involve the waiter handing the card to the person they think is most likely.

Yes and no. Surely the people handing his credit card to his black friends after receiving it from him have reason to know better.

Besides that, I think the point is not just that people assume he's black, which would just be a curiosity, but that that assumption shapes their attitudes toward him.

Our bureaucratic preoccupation with racial and gender categories is much like those who segmented bloodlines into Untouchables, or royalty. (As if it was ever impressive to be related to some glorified dictator called a "king".) Consider that, depending on your racial category, it may be ok for the US state to choke you to death in public.

Even just to speak normal English, you have to basically surveil into people's genders. (He/she. One consequence is you have to cram many people into a false category, like a bureaucrat only having yes/no on the form.)

"Consider that, depending on your racial category, it may be ok for the US state to choke you to death in public."

It is never OK, ever.

I think they clearly don't mean "really ok" but "will be treated as ok by the government".

Your jab against monarchy is unwarranted. A hereditary monarch who is recognized as legitimate has better incentives than a garden variety military dictator.

Jabs against monarchy are always warranted.

>it may be ok for the US state to choke you to death in public.

For a certain value of 'ok' that accepts that most Americans find such behavior repellant, and that it's not actually legal (much less mandated by the state) for cops to murder people for being of the 'wrong race' (despite it being, apparently, easy for them to get away with.)

And the Soviet Union also had a wonderful constitution that enshrined legal protections for bodily integrity, religious practice, freedom of speech and thought...

"No, I'm not Japanese, you idiot; 100% white European descent, in fact!"

"Oh, I'm sorry Mr. Shigeru Kobayashi, I just ass ... "

"Yes, like other members of the rubble, you just assumed, thereby making an ALL out of U and ME."

'I just ass'

It is absolutely unreasonable. When you say it "works out more often than not" I would ask "For whom?" Many snap judgments based on race lead to the death of unarmed black men and women.

Studies have shown that resumes with black-sounding names get fewer callbacks and we treat black men like white men with a criminal record.

Are Emily and Greg More Employable than Lakisha and Jamal? http://www.nber.org/papers/w9873

Norton Pscyhology: Devah Pager and Dalton Conley discuss racism and the stigma of criminal record https://www.youtube.com/watch?v=CofLE3q3Qh0

You are conflating the observation that a person who has a name predominantly used by african-americans is most likely african-american with further, discriminating, acts based on that observation.

Much more succinct than how I said it.

"We are blind to our blindness." Where do the generalizations stop?

As others have mentioned, we generalize all the time. We have to.

No single stone is like the other, and yet we we talk about stones and pebbles like they were some uniform set of platonic ideals and some sharp boundary separating one from the other.

We lack both knowledge about each individual and the capacity to hold all detailed information of all individuals in our heads to not shove them into categories.

Of course once you know someone better you slice them into finer categories ("polite person", "work-a-holic", ...) and thus add facets and tidits to your model of that other person. But that in itself is a continuum from broad categories to unique bits of information about a person.

If you're walking down a street at night in a city with a high crime rate and a group of tattooed, muscular youths comes around a corner then you're likely not going to treat them as individuals.

You will at least consider crossing the street because you categorize them as potential threat. Even if that particular group of youths is no threat, it's still a matter of probabilities they might be and avoiding them seems a reasonable risk-avoidance strategy because it costs you little.

You simply don't have the time or necessary information to treat them as individuals.

So as long as your generalizations are statistically sound (instead of outdated stereotypes) they are useful and should only stop once better information becomes available.

> So as long as your generalizations are statistically sound (instead of outdated stereotypes) they are useful and should only stop once better information becomes available.

I agree with that. However, in certain cases, it's called discrimination and it's illegal.

>Are Emily and Greg More Employable than Lakisha and Jamal?

That's impossible to say on an individual level, but it is very likely. If you phrase the question differently, there's a pretty clear answer:

>On average, are two white people more employable than two black people?

Without knowing any further info about the possible candidates - Yes. Looking at education statistics in the US, it becomes clear pretty quickly, that the average white person has a higher education level, than the average black person for example. That's a pretty important factor to look for as an employer. I'm sure statistics on work experience will look similar.

And of course this is quite racist, but as long as different races exist, that really just means taking reality into account. Pretending there are no differences between different population groups is insane and counterproductive.

> On average, are two white people more employable than two black people?

Instead of above, the question you need to ask is "On average, are two white people who are applying to your job more employable than two black people who are applying to your job ? I don't believe the answer is yes. A person who has overcome the statistics you quote has the same ability and employability regardless of race.

> Instead of above, the question you need to ask is [...]

Yes, excellent point. (And we can go further still.)

I've been calling this the "fallacy of ignored instantiation". Personally, this is something I've been ruminating on for a while and have meant to do a writeup on it for a couple years now. Hopefully, though, that name is descriptive enough that a certain crowd would, upon hearing it alone, understand almost exactly where I would go with it.

(Alternatively, if anyone knows of any good treatments that already exist on this topic, please let me know.)

The thing is that when you're looking at job candidates, you typically have far more reliable indicators of employability than race. So it is not justifiable to make race a significant factor in your decision. These sorts of biases tend to show up even when race is absolutely demonstrably non-informative. (E.g., when reviewing journal articles.)

I think you missed the point that the studies actually found that black-sounding names got more callbacks despite having the same credentials, e.g., degree, GPA, etc.

Exactly wrong. They get fewer callbacks.

Thanks. I meant to say fewer callbacks.

Two white people are definitely more employable on average, even if they're felons and the black people have college degrees!

We perpetuate those statistics on work experience, etc., because the playing field is uneven to begin with. We have a nice little system in which further discrimination is "taking reality into account" once we've ensured a substandard elementary education and reduced chances at a good first job. Read the study: it, of course, controls for educational and work experience.

Education statistics cut both ways.

Ask the question this way.

Are Brittany McKnight and John Wilson more employable than Lukwesa Nakazwe and Hijani Lungo?

Yes and it is obviously about race. Why? On average, Africans immigrants are more highly educated than native born Caucasians.

It's not about education. It's not about statistics. It's about racism. There have been studies where researchers submitted resumes with identical education and employment backgrounds but substituted "white sounding" names for "black sounding" names and the ones with the "black sounding" names were half as likely to be contacted by prospective employers.

With the same education and the same work experience, "Je'Marcus Johnson" is half as likely to even be interviewed as "Mark Johnson".

Survivorship bias. I went to a happy hour a week ago right after the Baltimore riots when a real estate exec starting going on and on about how blacks just needed to work harder - like many different immigrants.

This is survivor ship bias - comparing immigrants who left hell for the American Dream vs people living an American Nightmare. This is akin to comparing Olympic gold medal winners to members in a high school band. A better comparison would be comparing people of a similar socio-economic level in different countries.

I thoroughly enjoy those kinds of situations.

I use them to either get the person to re-evaluate their thinking or admit in public that they're just a racist.

One of my fiance's family friends works for the FBI and we were discussing racial profiling. He didn't understand why innocent people object to it. I explained to him that I'm not a criminal and it's always be a negative experience if someone treats me like I am one.

He asked if it doesn't make me feel safer to know that law enforcement is doing its job. I explained that if their only reason to suspect me of anything is the color of my skin, that it's racist and that I already know that I'm not a criminal. I wasn't a criminal yesterday. I'm not likely to become a criminal tomorrow and if law enforcement is wasting time, both theirs and mine, by bothering me; they're not out there actually catching criminals. I saw the expression on his face change. That thought had never occurred to him. The discussion ended with him conceding that he had never considered that before.

Using a name that is statistically far more likely to be a black male is what is "working out". I think we can agree that inferring someone's race and gender from their name is not the same as being racist or gender biased.

The second half of the parent comment states the problem is if you treat people differently based on those assumptions. Guessing that someone called Jamaal is probably black is like guessing that someone called Sue is probably female.

I was going to say "Johnny Cash kind of destroyed that last one", but on reflection, he really didn't. Sue is still probably female, despite Johnny Cash...

Okay, then we need to remove all stereotypes: republicans are rich old men that want to hurt the poor people, religious people don't like gays, anyone that didn't vote for the current president MUST be racist, if something it's written in PHP..it must be terrible because the language is terrible, the police are all racist and want to kill unarmed black men. I could go on.

There are plenty of stereotypes and generalizations that continue to go on even here on HN..without anybody seeming to care.

"Many snap judgments based on race lead to the death of unarmed black men and women"

A small percentage of this is going to happen no matter what, and it needs to be prevented. The problem I have is that society is not allowed to use statistics or facts any longer because it's not politically correct.

If I'm a police officer going into a neighborhood that has had multiple crimes, shootings, and deaths, and the perpetrators were black, I'm going to be extra careful if I see someone that is black doing anything suspicious. This is just being smart.

"Are Emily and Greg More Employable than Lakisha and Jamal?"

The resumes they used were nowhere to be found on the site, so I don't even know if I can believe this study.

"Norton Pscyhology: Devah Pager and Dalton Conley discuss racism and the stigma of criminal record"

Of course there is a stigma of a criminal record. There should be, because it's not a good thing. If there are two applicants, 1 with a criminal record and one with slightly worse experience and no criminal record..I'm going to chose the latter every time. Sorry, but if you don't want to be judged on your criminal record, don't commit crimes.

No, the studies showed that black males with NO criminal record were called back at the same rate as white males WITH a criminal record.

> If I'm a police officer going into a neighborhood that has had multiple crimes, shootings, and deaths, and the perpetrators were black, I'm going to be extra careful if I see someone that is black doing anything suspicious. This is just being smart.

OK, but note that this has a side effect of innocent blacks getting stopped and questioned with a higher probability than innocent whites (where by innocent I mean not doing whatever activity it was you suspected them of when you stopped them). Ans since you are going in extra cautious about blacks, if you find something that is borderline between worth charging them for or letting them off with a warning, you'll probably be more likely to charge a black and let a white off. Result: among blacks and whites doing the exact same things, blacks are more likely to acquire a criminal record...

...

> Of course there is a stigma of a criminal record. There should be, because it's not a good thing. If there are two applicants, 1 with a criminal record and one with slightly worse experience and no criminal record..I'm going to chose the latter every time. Sorry, but if you don't want to be judged on your criminal record, don't commit crimes.

...which you want to use to avoid hiring them.

> It is absolutely unreasonable. When you say it "works out more often than not" I would ask "For whom?" Many snap judgments based on race lead to the death of unarmed black men and women.

So your solution is to rewire the brain so that it doesn't generalize? Is this what contemporary discussions of prejudice attempt? Guilting people for aspects of cognition?

No, I think it's better to at least admit that this sort of subconscious bias is a problem and that we don't know now to fix it yet.

I think you are absolutely right. The other point that can be made is that P(person is black | name=Jamaal, person is US citizen) is much higher than P(person is black | name=Jamaal), since that is a common name in Arabic and in Muslim countries http://en.wikipedia.org/wiki/Jamal. One can argue that the extra "a" in the spelling of Jamaal as opposed to the more common transliteration Jamal is a marker but I don't think regular people notice this.

Note that the other weirdness is that the common US usage of "black" to commonly refer to African-Americans is very problematic: An arab may look dark skinned but most would never consider themselves "African" or there are African-Americans that have lighter skin colors than, say, people from Sri Lanka. Similarly for the label "white". For example, when we say "white men's advantage" there is an implicit assumption, not always stated or well-understood by users, that this refers to white men in the US, or perhaps more broadly to countries where there's a large population of "non-whites" that have been historically repressed.

Personally, when faced with a form that requires me to fill in race, I mark "other" or "does not want to disclose" rather than "white", since I don't feel I fully understand what the connotations of that label are for the purposes of that form.

I would push back on that view and say other people are absolutely at fault here. if thats not something we can't or won't own up to as a society, then we will have to continue to live with the consequences of stereotyping which, good and bad, harm us all in many subtle and not-so-subtle ways. difficult, but necessary.

Are generalizations really not allowed? Do we have to handicap our brains?

I actually worked for the progressives and saw campaign analytics first-hand. It was one of the most racist things I've seen.

And it works.

Ok. So why is it legal for insurance companies to increase price for young drivers, but not for men* (*at least in some countries)? Both groups cause more accidents than average,but it's legal to "stereotype" against one group but not the other? To anyone shouting that it's not "stereotyping" but true statistic, then I would reply that it's also true that Jamaal is mostly used by black men, therefore the probability of someone with that name being black is very high, and not a stereotype, but merely an observation.

It's asking much to not generalize, generally. Like, are you going to ask why you go on green and stop on red every time you come to a light? Not all red lights mean stop.

I think we can artificially not think of ethnicity when hearing names, but simultaneously realize people can choose non ethnic names or ethnic, not withstanding new immigrants. And, it's not clear that when people do internalize ethnicity and name associations that it results in negative outcomes. There was a recent study which contradicted previous studies when they controlled for other factors.

In any event, people make all kinds of assumptions and are often wrong, but we don't call them out because we have not become 'conscious' of them, till someone points them out.

We can make excuses for this behavior, but we shouldn't make it look OK by using the word "reasonable". It's certainly understandable why people make those assumptions, but those assumptions are neither based on reason nor fair.

I think it's reasonable to assume that someone with a stereotypically black name is black

It's easy to assume that. It really is a very rough guide though, so if you stick to it, don't be surprised if you end up looking like a complete idiot on several occasions throughout your life.

At the same time, I can kind of understand the point he's getting at.

Similar issue: I worked with someone via email for several months thinking they were female because their name is Kim before a coworker mentioned "uh, pretty sure Kim's a dude." You run into similar issues with e.g., women named Chris. I don't think being thrown by situations like that makes anyone look like an idiot.

Looking at it from an ethnic perspective, I'd be similarly thrown by finding out a guy named Santosh was a European white male.

Ideal world, sure, we wouldn't make assumptions about identity based on names. But it happens. I'm not sure that there is an answer that'd make everyone happy, to be honest.

> Ideal world, sure, we wouldn't make assumptions about identity based on names. But it happens. I'm not sure that there is an answer that'd make everyone happy, to be honest.

Of course we would, unless in the ideal world names would be uniformly distributed among population in a completely random way. But they're not.

The only problem here is seeing reasonable inference from a known probability distribution as ethically dubious.

Wasn't meaning ethics, just the risk of looking silly over time.

If you tend to make a lot of assumptions about someone just from their name before encountering them, then you are pretty much guaranteeing that you are going to get burnt a lot over time, unless you don't meet many people, as it simply isn't that good a guide.

I agree. Names just ain't that good of a heuristic.

I guess the point I disagree on is that I don't see it as getting burned. If you treat people differently based solely on their name, sure, that can cause problems.

For me when I'm mistaken, it's just a matter of making note of it and moving on. Not a big deal.

>Of course we would, unless in the ideal world names would be uniformly distributed among population in a completely random way. But they're not.

That's essentially the same point I was getting at.

This is what amazes me about white America:

1: 14% of the population is black,

2: 90% of white America have never spoken to a black person let alone have a black friend.

3: So where do their stereotypes come from, rap?, news?, (with regards to height) NBA?

4: They don't like to talk about this and this will be flagged off HN.

It amazes me with so little to no interaction with a group, white America can infer a black persons character and be so confident in doing so.

EDIT: "In fact, PRRI's data show that a full 75 percent of whites have "entirely white social networks without any minority presence." The same holds true for slightly less than two-thirds of black Americans."

Look at it another way 66% of black people say they do not have a white person in their social network. So roughly 34% of 14% of the population have a person in their social network from the white population. So how the hell can all these white people claim to have all these black friends? How can all these white people know so much about people they don't have meaningful interactions with?

>90% of white America have never spoken to a black person let alone have a black friend.

Are you being serious? Citation needed.

So, your overall question is a good one, but #2 is a howler. The reported statistic I've seen is that 40% of Americans say they don't have a close friend of a different race. But I'm confident that 99% of white Americans have spoken to a black person.

Except for a few states like Vermont, most states and most cities have significant black populations. See, for instance: http://fivethirtyeight.com/features/the-most-diverse-cities-.... Even in spite of that segregation, people do talk to each other (they may not live on the same block, but they do eventually meet each other).

Do you live in the US? If you don't, please do try to get basic facts about us right. Our country has a lot of problems, but you can't understand them without having a basic factual grasp of our demographics.

"The reported statistic I've seen is that 40% of Americans say they don't have a close friend of a different race."

Firstly that says of another race, could be arab,persian, Italian, asian, etc... Secondly, no way in hell 60% of america can have a friend from a pool of 14% of America. This is impossible looking at how race groups tend to cluster.

More evidence you're confused about the US. In the United States, white people don't consider Italians a different race. They're white.

You're right about the precise claim of what I said. But people who've never talked to a black person are disproportionately likely to have not talked to an Asian, Arab, etc. States that have high proportions of non-Black racial minorities tend to have higher proportions of Blacks.

Bullshit, I've heard white ppl talk shit about Jersey Shore and use racial slurs against Italians. Secondly look at the Silicon Valley numbers for diversity, how many Asian vs Black? So your point no black interaction == no asian interaction is incorrect.

Regarding minority populations: it's not equivalent. But it is often the case. Think Texas, New York City, Los Angeles, or Chicago.

People talk shit or joke about the French, Italians, Germans too, but French and German people are considered white. Same with Poles and Italians.

Those stats you gave did not consider people of Italian or German descent as non-White.

Hm. My wife (Italian descent) reports differently. During her southwest upbringing she was treated as anything but white. I guess it depends upon your environment.

Perhaps I'm too quick to assume my experience is universal on this point. At school growing up, the white people were fairly mixed in actual ethnic background. Statistically, though, any researcher would consider Italian white if they used "white" as a category.

Would it be too personal to ask what she meant? Was she taken for Hispanic, or did people use anti-Italian stereotypes?

The 2nd. She was not welcome everywhere in town; she was viewed with suspicion in stores; her family's social circle was several notches lower than expected from her fathers' position (scientist at Los Alamos Natl Labs).

I'd have supposed Los Alamos, what with the laboratories, to be more cosmopolitan than Denver; yet so far as I know, that didn't happen in Denver. To be sure, I didn't have a particularly elevated social circle, and maybe the people at the Denver Country Club had some issues. But as far as I know, most people in Denver sorted ethnicities into Black/Chicano/Anglo.

Strangely, once Hispanic folks 'from the valley' were allowed on base (to clean and drive etc), Italians moved up. The Hispanics were the new bottom of the heap.

I'm sorry to hear that.

At the moment, throughout the U.S., Italians are considered "white". In the recent past, in certain areas of the country, they (Southern Italians, at least) were considered "black". They were lynched and not allowed to marry "whites". Racial categories are ever-shifting social constructs that serve temporary political and social purposes.

http://www.africaresource.com/rasta/sesostris-the-great-the-...

You're absolutely right, and I'm am only meaning to talk about the present.

> But I'm confident that 99% of white Americans have spoken to a black person.

I'm no more confident of that than in the 90% haven't made up statistic that this was offered in response to.

The truth is probably somewhere in between (and may be closer to the 90% haven't side, especially if, as suggested in another comment, one excludes the minimal interactions around things like retail commercial transactions and the like, and only includes substantive conversations.)

> Except for a few states like Vermont, most states and most cities have significant black populations.

That doesn't mean that the black population interacts with the White population all that much.

Eh, OP made a damn extreme claim. If you exclude extremely glancing interactions, I'd walk back the claim about 99%.

But I will say that any place I've ever lived or visited, the majority of White people would have had conversations with Black people. I just can't imagine someone in Pittsburgh, New York City or Raleigh not having done that.

No I did not!

"In fact, PRRI's data show that a full 75 percent of whites have "entirely white social networks without any minority presence." The same holds true for slightly less than two-thirds of black Americans." Now isolate the 14% black population from all minorities, take into account how minorities cluster and the 75% number is for black friendship is probably even higher than 90%

Your stats are wrong, read below and thats covering all minority groups. now isolate the 14% black group from those minorities. White ppl just don't have black friends, deal with it.

""In fact, PRRI's data show that a full 75 percent of whites have "entirely white social networks without any minority presence." The same holds true for slightly less than two-thirds of black Americans.""

Source?

I'm asking because I find it hard to believe (almost impossible) that 90% of white people have never spoken to a black person.

How many white people do you know that have spoken to a black person, not the lady at Mcdonalds but an actual interaction?

Every single person I worked with in Rocky Mount had spoken with Black people and worked with Black people. They lived on the same streets and their kids went to the same schools. And many of those white people who interacted with Black people every day were stunningly racist (and I should say many weren't). Maybe they categorized "different types" of black people. But they could hold a civil conversation, then turn around and stereotype, or drop the n-word repeatedly.

Your vision of what people have been exposed to is a cartoon.

This is the number I've seen: http://www.reuters.com/article/2013/08/08/us-usa-poll-race-i....

Since i cannot respong to you below here is the updated numbers: http://www.washingtonpost.com/blogs/wonkblog/wp/2014/08/25/t...

I think those are not updated, but using a more restrictive methodology. And while it's not invalid, I think the statistic I cite is more relevant. For your initial question, who have you talked to, the acquaintances at work are more important than confidants.

Like I said in my original post, you're asking a good question. As far as I understand it, most people's stereotypes are not based on significant interaction with other people. They're picked up from other people who held those stereotypes.

It's just that I think you made a major exaggeration going along with it.

The sad truth I've heard people who were racist as hell say "they're just blaming that boy because he's black." They were capable of seeing someone as human in a particular instant, then turning around and uttering the most hateful things imaginable. They could talk to a black man civilly then turn around and talk about what "the brothers" do. Not having talked to black people wasn't their problem.

All of them.

This has to be a troll.

All of them. Seriously, where do you live?

At a guess, 80% of the white people I know. I live in Washington, DC, where you would have to go out of your way not to have black co-workers. Yet I had black teachers and classmates in high school in Denver long ago.

All the white people I know (since knowing people takes "actual interaction") have spoken to a black (or, at least, a multiracial person with black enough features to be generally considered "black" by observers) person enough to meet that standard, pretty much by definition.

But, while I think your 90% number is a substantial exaggeration, you do make a good point that this is probably far less common than most people think it is.

>> 90% of white America have never spoken to a black person let alone have a black friend.

Where in the world did this statistic come from?

Please source statistic #2.

You have, seemingly without irony, just stereotyped "white america"... as someone that stereotypes "black america".

You deserve to be voted down. But not for the reason you think you do.

Reply on news.ycombinator.com