With all the discussion of false rape convictions on this blog lately, I thought it could be worth pointing out that one form of false rape conviction doesn’t involve false accusations at all. Many false rape convictions now being discovered due to DNA analysis involve genuine rape victims who make good-faith, but mistaken, IDs.
One disturbing thing is the apparent connection between false IDs and race:
The racial mix of those exonerated, in general, mirrored that of the prison population, and the mix of those exonerated of murder mirrored the mix of those convicted of murder. But while 29 percent of those in prison for rape are black, 65 percent of those exonerated of the crime are.
Interracial rapes are, moreover, uncommon. Rapes of white women by black men, for instance, represent less than 10 percent of all rapes, according to the Justice Department. But in half of the rape exonerations where racial data was available, black men were falsely convicted of raping white women.
“The most obvious explanation for this racial disparity is probably also the most powerful,” the study says. “White Americans are much more likely to mistake one black person for another than to do the same for members of their own race.”
On the other hand, the study found that the leading causes of wrongful convictions for murder were false confessions and perjury by co- defendants, informants, police officers or forensic scientists.
One thing that may reduce false IDs is changing police proceedures for IDs by victims; the methods used by police too often send unspoken messages to the victim that “this guy here! He’s the one!” The Innocence Blog has an interesting post about methods of making the ID less subject to police bias.
“On the other hand, the study found that the leading causes of wrongful convictions for murder were false confessions and perjury by co- defendants, informants, police officers or forensic scientists.”
All the false convictions I’ve read about involved this. Since I’m fond of forensic science and used to read a lot of stuff on that, most of them involved forensic scientists who either lied or were so incompetant they should have lost their jobs. They would tell the jury stuff like only ten percent of the men in the country could have raped the woman, and that the accused was in that ten percent, but in reality, because of the woman’s chemistry, less of half of all men could be elminated.
If I were on a jury, I would have gone for guilty in a lot of these cases no question, because it wasn’t just the woman’s testimony; it was her testimony backed by “the experts.” In a purely “he said/she said” sort of case I think I would still have reasonable doubt and would find it difficult to convict (which may be why the cops falsify evidence – they’re sick of all these guys getting off).
I suspect that while a man can rape a women all by his little selfie, it’s tough for a woman to frame a guy and get a conviction for rape without help. What shocked me about one case is that the woman recanted and the forensic evidence was very skewed, but it was still a fight to get the guy out of jail.
I guess some people have way more faith in the judicial process than I do.
Sheryl
I don’t want to commit myself to the position that the difficulty of cross-racial identification has nothing to do with racism, but I don’t believe it necessarily has all that much. I lived in Samoa for a couple of years, and even people I knew fairly well consistently mixed me up with any other white woman within four inches in either direction of my height. Samoans (in Samoa, I don’t have any experience with a Samoan community in the US) really don’t seem to have much anti-white racism going on; my strong impression was that there was some psychological effect producing the ‘all white women look alike’ result that was separate from racial animus, and it seems reasonable that the same is true with whites and African-Americans in the US.
But generally, you’re absolutely right that a lot of work has to be done on making certain that eyewiness IDs of strangers, particularly cross-racial IDs, are reliable before anyone is convicted on that basis alone.
Interesting point, LizardBreath. It may be that the failure to be able to ID faces is, in some case, a form of racism that has nothing to do with racial animus.
When we lived in Korea, our landlady once confused my husband (tall, dark haired) with another tenant (not nearly so tall, red hair and freckles). I thought that was pretty much the ultimate in the “they look all alike” category. Koreans are racist in the sense of being very aware of race (and whether you’re Korean or not), but most Koreans in her generation like Americans pretty well.
There was a recent article in, I believe, New Scientist that discussed a “study” in which several of England’s top fingerprinting experts were gathered in a conference. They were told, “We’re going to give you a copy of the fingerprints on file for Brandon Mayfield — the Oregon lawyer who was falsely linked to the Spain terrorist bombings — and the fingerprints found on the bag in Spain. Please, you experts, show us how we can avoid making the same mistake here in England.”
In fact, rather than give them the Mayfield fingerprints, they actually gave them fingerprints from a recent case in which they had served as an expert witness. Only 20% came back to say, “No, these really are the same prints.” The rest either “proved” that the fingerprints they had previously claimed were identical were in fact from different people, or else came back and said that they weren’t sure one way or the other.
In the classes I teach, I try to learn student’s names from their ID card photos. I find it quite difficult–the static photo misses a lot of features that help me identify people, like the way they move and hold themselves, not to mention height. (Unfortunately, most of my students are white, so it’s not a question of race here.) I guess that the photos in police lineups are probably better, but has anyone done a study of how good people actually are at identifying faces from photos?
“Racism” sounds like criticism. What we might be talking about here is something more innocent, unfamiliarity. Members of any unfamiliar group tend to look like one another, as these stories illustrate, not because there is any moral flaw here, but just from lack of experience. In the same way and for the same reason most arabic letters look alike to me.
It’s even more complicated — there’s more context. Back when Bennett was being lampooned for his comments about Freakonomics, I wrote about crime statistics, noting that one of the things Levy said repeatedly is that crime/race stats are notoriously problematic.
Why? Because there’s a little matter of the difference between crime rates and clearance rates. To address the drawbacks with the Uniform Crime Report data, they use other measures, such as the National Crime Victims Survey. To quote myself:
Crim 101
>>I don’t want to commit myself to the position that the difficulty of cross-racial identification has nothing to do with racism, but I don’t believe it necessarily has all that much. I lived in Samoa for a couple of years, and even people I knew fairly well consistently mixed me up with any other white woman within four inches in either direction of my height. Samoans (in Samoa, I don’t have any experience with a Samoan community in the US) really don’t seem to have much anti-white racism going on; my strong impression was that there was some psychological effect producing the ‘all white women look alike’ result that was separate from racial animus, and it seems reasonable that the same is true with whites and African-Americans in the US.>>
You could argue, though, that institutional racism and resulting segregation have made black people so foreign to their local white counterparts that they might as well be white women in Samoa.
I really noticed the same “they all look the same” effect when I went from my small, rural town with a negligible Asian population to a college town with a much higher asian population. I had a lot of difficulty telling Asians of the same sex apart from one another. I still find it very hard to believe that, by some objective measure, they’re not more homogeneous than other American populations, though I’m likely wrong about that.
While I haven’t had a huge amount of contact with blacks, I’ve never had the problem to the same degree with them.
One theory I’ve heard about this is that people remember faces by remembering any notable aspects of the face, anything that deviates from the mean. If they haven’t seen a lot of faces from a certain group, all of those faces will differ from the mean in very similar ways, even though the intra-group differences are just as large.
I’ve heard of a really interesting set of research, which I’ve never been able to follow up, which basically says we don’t look at people’s whole face when we identify them. When we’re laying down the ‘recognising people’s faces’ bits of our brains we only look at the bits that tend to be different, and that people whose ancestors come from different areas tend to have different areas of their face that tend to be different.
But piny is right – all this shows is how segregated our society is. If it was the explanation it wouldn’t only be black men who were wrongfully being convicted, but many cross-race identification.
I think everyone is aware of how unreliable eyewitness identification can be even for simple bystanders and witnesses, forget people who have been traumatized. I don’t doubt that it’s even more unreliable when we’re talking about cross racial identification. There’s no doubt that this needs to be addressed within the criminal justice system. Still, I think using a word like racism can be a little bit problematic in this context. I certainly understand why it’s being used, but it does imply malicious intent, and considering that fact that I’ve seen more than one comment around the web about “even if it’s a false ID, it’s still her fault, who made the ID?” and there’s this whole mythology about women being vindictive and out to get men, it makes me uncomfortable. I agree 100% with your point about reforming police proceedure.
A discussion about identifying people in one’s own group vs. an “out-group” came up in one of my social psychology classes, and I seem to recall a statement in my textbook that most people have a harder time telling apart people in the “out-group”, and that it’s essentially because of lack of familiarity. I’ll see if I can dig out the cite.
Okay, my textbook doesn’t appear to specifically deal with race, although I could have sworn that came up in the class discussion. But my notes have vanished into thin air, apparently.
However: “In-group members tend to perceive those in the out-group as being more similar to each other (homogeneous) than they really are, as well as more homogeneous than the in-group members… [This has been studied using rival universities in a variety of domains, from athletics to music preferences, by having participants from each group make predictions about the in-group and out-group members.] … When the target person was an out-group member, the participants believed his choice was more predictive of what his peers would choose than when he was an in-group member… In other words, if you know something about one out-group member, you are more likely to feel you know something about all of them.” (“Social Psychology”; Aaronson, Wilson, Akert, & Fehr; page 482-483)
That sounds to me a lot like what commenters have been saying with regards to race, here: misperceiving more similarity in people of other races than one’s own than really exists.
Part of the problem with even discussing these matters is that the word “false” has two different meanings. It can mean “not true” as in “it is false that 2 + 2 = 5”, or it can mean “wrongful”. I tend to take it to mean “wrongful” in the context of a conviction, but “not true” in the case of a report. But even that’s problematic. A report will consist of a large number of claims, some innocuous, some innocuous in isolation, but incriminating when taken together with other claims. Even a false report will contain many true statements, and a substantively true one may contain intentional or unintentional false claims.
Suppose someone is genuinely raped, but distorts the events in her report? Suppose that as a result of that distortion the wrong person is identified as her rapist? She’s then culpable, isn’t she, for the forseeable consequences of her wrongful action, without being any the less the victim of the rape.
There is also a problem with using the word ‘accuser’ to describe the complainant, which frames the discussion in a way which places sole responsibility upon her. To give an extreme example, there was a case in the UK some years ago in which the rape victim (it is undisputed that she was raped) testified that the defendant (who was known to her) was not her rapist. He was nevertheless convicted on DNA evidence which later proved to be faulty. It is hard to see how she could reasonably be called his “accuser”, let alone a false accuser, given that her testimony was, in fact, true.
I think reddecca is right – people of different races tend to vary in different areas (European Caucasians, for example, tend to have a lot more variation in hair color than, for example, Koreans or black Africans). So someone without a lot of experience with people of a different race might have a hard time distinguishing people of that race because he is not used to noticing their intraracial variations.
There was a discussion of this at Gene Expression a while back.
But piny is right – all this shows is how segregated our society is.
That’s true. The issue isn’t even so much that one is perceiving more similarity than exists in a race other than his own, but that one is perceiving more similarity than exists in a race with which he is unfamiliar. So if a person has trouble distinguishing black men from one another it does suggest a lack of interaction with black people.
However, even white people can get “confused” with subcultures of white people under certain circumstances. Even having lived in an area with many Amish for 15 years, I’m ashamed to say that a lot of them look alike to me. True, they dress the same and have the same hair styles. But that’s really no excuse.
There could be much more sinister racism going on here too, I wonder if they can tell from those DNA samples if the semen was even from a black guy? Often when people want to report a false mugging or a false auto theft, they’ll “witness” a black guy, figuring it’ll be easier to make that claim credibly. Not sure why a woman would need to claim that she had been raped when she hadn’t been, but if women ever do that, chances are they would follow the same pattern as people that report false muggings.
Also, the fact that it is false convictions that are mostly victimizing black men might point to disparities in lawyers and police departments more than the person who ID’s the suspect in a line-up.
I guess that the photos in police lineups are probably better, but has anyone done a study of how good people actually are at identifying faces from photos?
a few years back the discovery channel did a piece of false ID. they had a man run into a college classroom, steal the professor’s bag, and then run out. the man literally paused for a moment and looked directly at the students. if i recall correctly, not a single person actually picked the right guy out of a line-up.
therein is the problem with eyewitness testimony. line-ups give the perception that one has to pick from the group. while who is chosen still lies in the hands of the chooser, there could be better methods of presenting the material. once a person has been chosen, even if he is innocent, it is not very likely anyone will catch it before it actually goes to trial. and once he is convicted, there is nothing but an uphill battle.
Even having lived in an area with many Amish for 15 years, I’m ashamed to say that a lot of them look alike to me. True, they dress the same and have the same hair styles. But that’s really no excuse.
It seems to me that a lot of that has to do with the common characteristics of the people that you grow up around. For example, the way I tend to distinguish/recognize people is by their hair. That was the simplest way to differentiate the people around me when I was a child. (I don’t think that I really knew what my parents faces looked like until I was 30)
That method of recognition fails when dealing with people who have similar hair. I’m not ashamed of it, it is the method that worked best for a long time. I simply have not completely adjusted to my new(er) world, yet.
This is interesting, if not immediately on topic: It seems that if a black person is involved in a crime any whites who witness the crime will focus on him/her/them and therefore the witnesses are less able to identify any whites also involved. I’m not sure what that means, but it can’t be good.
Personally, by the way, I’m lousy at identifying people. If I meet someone outside of the expected context, even people I’ve known for years, I have trouble recognizing them. And I have a hard time remembering what people look like. Once when I was in school I went to a professor’s office to look for her. She wasn’t there and as I was leaving a man, also looking for the same professor, wandered in. I said something like “she’s not here”, he asked if I knew where she was or when she’d be back, I said no, we both went on our way. A little later I ran into the prof in the hall. After talking for a few minutes, I mentioned that a man had been looking of her. She asked what he looked like. I was totally stumped. I couldn’t even remember with any degree of certainty basic things like his race, age, height, etc. This was someone I had seen for several minutes and talked to. Totally gone. So if you’re ever on a jury and I ever act as a witness to ID a suspect who is a stranger to me, just blow my testimony off: it’s probably no better than random. (On the other hand, I can remember license plates and similar strings of numbers/letters. I once called the police after being hit (though not hurt) by a hit and run driver. I could give them the license plate but not the make or color of the car. Go figure.)
If [difficulty of making cross-racial identifications] was the explanation it wouldn’t only be black men who were wrongfully being convicted, but many cross-race identification.
Three thoughts:
(1) An important question here is what is the breakdown of interracial rapes? In other words, what percentage of interracial rapes involve a black perpetrator vs. a white, Asian, Latino, etc. perp?
(2) Also, how likely are interracial rapes reported? It strikes me that a white woman raped by a black man would be more likely to come forward than a black woman raped by a white man.
If most rape charges where the complainant and the alleged rapist are different races have black defendants (either because of different crime rates between the races or because of different rates of reporting a crime), then it it is likely that there would be more false convictions simply because all else beng equal, a higher number of convictions would mean more false convictions. To really see what is at play, we would need to look at false convictions of black men as a percentage of all convictions of black men, and then total false convictions vs. total convictions of all men.
(3) Whites are the vast majority of the population in the U.S., so it is harder for minorities not to be exposed to whites than for whites not to be exposed to minorities. Put another way, minorities would be more familiar with whites than whites with minorities. Therefore, it seems to me that the percentage of whites whp would have difficulty making cross-racial identifications of minorities would be larger than the percentage of minorities who would have such difficulty with whites.
A few points….I think those who made the comments about racial segregation are right. We should be comparing cross cultural comparisions with cross racial comparisions. This is about people of different races who live in the same culture, but have very little contact, especially Whites, which brings me to my second point.
I am fairly certain that on the whole most ethnic/racial groups are more accurate when identifying people of their own group. However, I am confident that Blacks do a better job identifying Whites than vice versa. Whiteness is really pervasive in this culture, and Black folks are exposed to White people and White standards whether they want to be or not–on TV, at their jobs, on the street and so on. It reminds of the phenomenon that happens to many Black folks when people mistake them for celebrities who they look nothing like. In fact, I have a Black acquaintance who used to get Whites to buy him drinks because they thought he was an NBA player, just because he was slightly tal (6’3″) and Black; he could never convince any Black person that he was a famous athlete. LOL!
I think about 98% of rapes by whites have a white victim, while about 40% of rapes by blacks have a white victim. (I have to be vague, because it’s been a couple of years since I looked at this, and I haven’t been able to find the data in a few minutes of searching. You’d probably need to look somewhere around here.)
Since these discussions are often framed in terms of whether there is targetting of white women by black rapists, I looked at that, and found no valid statistical justification for the claim. Rather, there was a lot of abuse of statistics by those with an agenda.
Dianne:
Is it possible that you have Asperger’s syndrome? I do (self-diagnosed, and subsequently confirmed by an expert). I can memorise an eleven-digit phone number instantly, but cannot remember what I’m supposed to be doing in two days or where I’m supposed to be.
John Howard:
Kanin’s notorious study is apposite. Here‘s a Microsoft Word file of the unedited article. (There’s at least one version floating around which is heavily edited and annotated, and not worth bothering with.) Here‘s Google’s cache of the same.
Not only possible, but likely. I’ve never had the diagnosis confirmed, though.
JohnHoward,
Interesting question about determining race from DNA testing. The answer–no, with the tests they use, you couldn’t exactly determine race, although you could guess.
What they do is specifically test loci (the plural form of locus, which is basically a specific physical place on a chromosome) in non-coding regions, that is, parts of the genome that are not made into proteins. These loci that they test must be 1) independent from eachother and 2) have a fair amount of polymorphism, meaning that they must be different in different people.
When it comes to the different alleles (different DNA sequences at the same loci) alleles that are common in one “racial group” are common in all the rest, alleles that are rare in one group are rare in all of the rest. But the frequencies are slightly different, i.e., one allele might have a prevalence of 5% in the European-American population, but a prevalence of 15% in the African-American population. So it is more likely that a person with that allele is African-American, but a European-American is by no means excluded.
Even in coding regions of the genome, we are remarkably similar. If they got into the genetics of hair and eye and skin color and height, maybe they could start making predictions about what the person looked like. However, this would be more difficult than traditional DNA testing (which looks only at length) because you would have to look at sequence, and look at several genes at once. (Plus having a gene doesn’t mean that you express the gene. Yes, teachers lie.)
Yeah, the topic posted here is a serious problem.
In some cases, rape victims can not make accurate identifications for different reasons. It doesn’ t mean they aren’t being truthful. They aren’t out to get the wrong person, but the rapist. They trust the justice system, which isn’t always doing right by the victim or any potential misidentified suspects. Sometimes it makes mistakes. Sometimes the agencies have their own agendas. And even when they realize they’ve made a mistake, they can be loathe to correct it.
All LE agencies have ideas about Black men that are racist and sexist. And about White women that are racist and sexist. And often these beliefs taint investigations of interracial rape, just like sexist and/or racist beliefs taint intraracial rapes, i.e. women of color can’t be raped, period or women of all races deserve it if they dress a certain way, or drink alcohol.
It’s very difficult to believe that these belief systems don’t impact some rape investigations, unless you are in denial. And those belief systems also impact prosecutory agencies as well, which too often do not look too closely at what LE agencies are doing in their rape investigations that they forward to prosecutory agencies.
In cases where mistakes are made, you have people who are suspects who shouldn’t be. You have rape victims who might believe that the police have caught the real perpetrator b/c they trust the police. In reality, the rapist is still out there, and may have already or may soon rape other women, as well. So nailing the wrong person for a crime doesn’t do anything but cause more damage to all parties involved including women who might be victimized by a free rapist.
I’ve talked with Black men who have been charged with rapes then cleared through DNA tests, and almost always, the victim is White. Almost always, the victim either can’t positively identify them, or they say that it’s not that person. In a couple of cases, the victim said the rape suspect was not even the right race(i.e. the suspect is Black, the rapist is Latino) but even with their refusal to ID the suspect, the case still pushes forward, unless a DNA test comes back negative. But DNA sample testing isn’t possible in all cases, and DNA testing is only as good as the agencies(LE, laboratories, etc.) conducting the test.
And, with any rape statistics, it comes down to whether they involve rapes reported to LE agencies, or rapes reported by victims as part of a survey. A lot of rapes do not get reported, so they don’t get counted in statistics including annual UCRs which skews the numbers quite a bit. And it could skew the numbers of reported rapes by victims within different races, differently.
I think for women of color especially. The justice system and its componants tends to treat them as if they are invisible or can’t be victims. LE agencies for example tend to rely on racist stereotypes just like they do sexist ones too often.
Even with a DNA test in their favor, it can still take months to free a wrongly charged individual. That’s not uncommon at all. I knew a guy who was in jail for six months after a negative DNA test came back. The rape victim had already refused to ID him(the rapist was light skinned Hispanic or African-American, he was very dark-skinned African-American) and he was arrested after a canine dog(with a history of mistakes) hit on the apartment complex he lived at and police arrested him b/c he lived there.
Yup. And juries as well. Also, just because a person may plead out to a rape or a lesser charge, doesn’t necessarily always mean they are guilty, b/c many times, Public Defenders at least where I’m from, well tell defendants especially Black men to plead out on most crimes b/c a jury will never believe them at trial.
And as for “confessions”? Well, they might be coerced by detectives. Hundreds of cases involving crimes in Cook County were tossed out b/c the confessions were coerced. There was also the Central Park Rape case, where the confessions were tossed out after some of the defendants had served prison time.
Not that all confessions are coerced, but there’s enough of them that have been to scrutinize these confessions much more closely.
You know, Paul Craig Roberts (the architect of the Reagan tax cuts, now an outspoken paleoconservative critic of the Bush administration, published on such diverse sites as VDARE (right-conservative), LewRockwell.com (right-libertarian), and Counterpunch (traditional left)) has written often about the “convict at any price” mentality of a lot of people within the judicial system. Convicting a person whom the victim insists was not the rapist would probably fit right in with the idea that it’s not about the accuser or the accused, it’s about the prosecutor’s chance of winning that sweet promotion.
Paul Craig Robert’s latest piece on this is available at Counterpunch, LewRockwell, and VDARE (I list all three so you can choose which site you would rather patronize).
I’m writing an opinion piece about some posters that went up in Vancouver, written on them is “Just because you regret having a one night stand, does not mean it wasn’t consensual. Don’t be that girl”. The message seems to be about discouraging false accusations of rape among young women… Although that’s besides the point. I’m looking for information on the subject… and none of the links to your sources work. Would you mind fixing them or sending me links to the information? I’m having a hard time finding them on Google.
Both of the links in my post here, are still working. Can you tell me specifically what link it is you’re talking about?