Viral load

Image copyright 2012 Paul Pierson/The Enliven Project

Brother Seth and I exchanged words.  Not blows--this exchange occurred on Facebook and Seth is a gentleman, which means he's unlikely to resort to his fists; and I, meanwhile, am an attorney, which means I'm likely to respond to any show of force with wretched, craven shrieks of, "Oh God, not in the face, it's my moneymaker!" and  "I'll sue!"

We came to words over a miserable illustration, and it didn't even depict anyone in a compromising position involving farm animals and fifty gallons of baby oil.  No, the picture in question is an illustration by Paul Pierson for The Enliven Project, which was picked up by Dylan Matthews of The Washington Post and has apparently gone viral since.

Which is unfortunate.  The image has problems--many of which are pointed out by Amanda Marcotte at Slate.  Marcotte's response is devastating, but only scratches the surface of self-evident problems with The Enliven Project's statistical methodology.

Here's the methodology Enlighten used:

For those of you who have asked, here is the background on the stats we used:

• Some reports suggest that only 5-25% of rapes are reported to authorities.  Other suggest that close to half are reported.  We assumed 10%, which is dramatic, but possible.
• Of the rapes that are reported, approximately 9 are prosecuted.  
  
• Of the prosecuted, 5 result in felony convictions.  This is across the board for all felony prosecutions, not just rape.
  
• Assuming that 2% of reported rapes are false and a 10% reporting rate, the graphic assumes that 2 of 1000 rapes are falsely reported (assuming a rape can’t be falsely reported unless it’s reported in the first place)

A few things to notice.  The 5%-25% number for reported rapes comes from a British report published in 2007, the fifty percent comes from an American Department Of Justice clearinghouse that includes data accumulated between 1973 and 2011, and the 10% comes from... well, they made that up.  The "9 are prosecuted" (they presumably mean nine percent) comes from a survey of crimes reported in the United States from 1950-1998.  I'm having a helluva time figuring out what the "5 result in felony convictions" is based on, but the source linked to is a DOJ bulletin titled "Felony Defendants in Large Urban Counties, 2006".  [EDIT: As best I can tell from another quick survey of "Felony Defendants", the number here seems to be a bizarre reading of a line on page 3: "Murder defendants comprised 3% of the defendants charged with a violent felony, while rape defendants accounted for 5%. (See Methodology for the specific crimes included in each offense category.)"  Enliven's interpretation is contradicted four pages later: "Regardless of adjudication method, a majority (72%) of convicted defendants were convicted of the felony offense with which they were originally charged (figure 7). More than three-fourths of defendants convicted of driving-related, weapons, or murder offenses were convicted of their original arrest charges. In comparison, about half of convicted rape defendants were convicted of their original arrest charges."  Good grief.]   The "2% of reported rapes are false" comes from a pamphlet or bulletin offering prosecutors advice for dealing with "False Reports: Moving Beyond the Issue to Successfully Investigate and Prosecute Non-Stranger Sexual Assault" (emphasis added).

This is not nitpicking.  The graphic Enliven created is presumably intended to have an effect, to convey some kind of information that inspires one to action.  But what, exactly, is it conveying?  If we look to the title of the page it appears on and ignore the captioning on the graphic itself, it seems to be confronting us with "The truth about false accusation"--but the truth of false accusation where and what and when and how?  In Great Britain in 2007?  In the United States in 1962?  In urban areas as opposed to rural counties?  Stranger rape as opposed to non-stranger rape?  And if we consider the graphic itself--as presented, say f'r'instance, in the context of the Washington Post's blog page, where it makes no pretense to truth about false accusations of rape somewhere and somehow but, rather, is billed as "The saddest graph you’ll see today", we're further divorced from any context or meaning.  Is the saddest truth that so many rapes are unreported, or that so few are prosecuted?

We have no idea.

Marcotte's points at Slate are worth visiting in brief, though you should read her critique.  The most notable one to call attention to is the fact that the graphic's author confusingly conflates rapes and rapists, when it's indeed probable that many rapists commit multiple rapes; this is especially important to notice when you're looking at the upper-right-hand corner of the chart and realize that a serial rapist may only be jailed one time.  One also has to point out that there is a difference between whether a rape occurred and whether someone was falsely accused of the rape: consider, as an example, the Ronald Cotton case, in which an innocent man was mistakenly accused despite a victim's heroic efforts to bring her rapist to justice; she nevertheless misidentified Cotton as her attacker and Cotton spent ten years in prison while another man bragged about getting away with the crime until DNA exonerated Cotton and brought justice round to the right man.

The thing here is: this issue is too important and complicated to be bullshitted about like this.  I'm not pointing out these methodological and presentational flaws to say that rape should be minimized as a social or legal issue.  What I'm saying is, well-intentioned but completely lazy and divorced from reality visuals intended to rouse gut reactions do not help--in fact, they hurt.

One of the things that needs to be emphasized on that particular score is that the methodological and presentational flaws in the Enliven graphic may in fact understate whatever problem it is we ought to be talking about.  That is, Enliven makes assumptions about the frequency of rape occurrences and reportage that may understate the number of rapes that occur and aren't reported.  Meanwhile, the fact that the image is so direly wrong about whatever it's trying to say certainly makes it possible for people who want to dismiss issues involving sexual violence for whatever reason to do so by pointing out that individuals concerned about sexual violence are (for lack of a better word lacking in sexist connotations) "hysterical".  Or at least ill-informed and lazy with their use of facts.

What we need are adult conversations.

For instance, we need to make sure we know what we're talking about.  By this, I don't mean making some bogus and ignorant Todd Akinish "distinction between "legitimate" rape and (by implication) "fake" rape.  What I mean is that we use the word "rape" within the legal system in ways that are both maddeningly specific and unconscionably vague, to describe wide ranges of misbehavior that may have different causes and may not merit one-size-fits-all solutions.  And outside the legal sphere, we're prone to use the word in so many ways and in so many contexts that it risks losing meaning altogether.

For instance, if we're talking about "unreported rapes", it might be helpful to know whether we're talking about unreported non-consensual sex between strangers, unreported non-consensual sex between non-strangers (note the emphasis I made above relevant to the "False Reports" pamphlet), and unreported statutory offenses of any kind.  Not because there's some difference in "legitimacy" or some such nonsense, but because the approaches we might take to encouraging women to report rapes committed by strangers might be different from the strategies we might take to encourage women to report date rapes or rapes committed by spouses or boyfriends, and none of the approaches we take to encouraging women to report are likely to have much bearing on how we encourage victims and families to report child molestation (i.e. statutory rapes, where consent is irrelevant because the victim's status makes consent impossible to begin with).

Perhaps more critically, we might observe that from the perspective of deterring rapists, deterring non-stranger non-consensual sexual assaults may be as (relatively) simple as the kind of educational programs that became very popular on college campuses in the 1990s, where young men are told that "no means no" and that it's simply wrong to have sex with anyone if there's any doubt whatsoever of their consent.  Whereas stranger rapists are likely suffering some psychological pathology and certainly aren't educable in any case (I sincerely doubt anyone willing to grab and assault a complete stranger doesn't know what they're doing is wrong unless they are suffering a psychiatric disorder that renders them incapable of making the distinction at all).  And pedophiles are yet another category requiring another program altogether.

The tendency in the law and in society, however, is to lump all of these things together.  (The Enliven graphic certainly does--they have to, at least in part because reports like "Felony Defendants in Large Urban Counties, 2006", although informative and useful, aren't so fine-grained as to distinguish between defendants who plead to one kind of rape or another.  [Correction: "Felony Defendants" does offer a definition of rape on page 16: "forcible intercourse, sodomy, or penetration with a foreign object. Does not include statutory rape or non-forcible acts with a minor or someone unable to give legal consent, nonviolent sexual offenses, or commercialized sex offenses."  N.b. the definition used specifically excludes non-forcible rapes, e.g. the stereotyped date rape scenario in which someone has sex with an incapacitated victim without the use of force.  Which, y'know, some of us would call a form of rape.])  Again, lest someone mistake the point here: distinctions are not between "legitimacy" and whatever the opposite of "legitimacy" implies; the distinctions we're drawing are between how we, as a society, need to deal with different kinds of crimes and categories of behavior.  We might well agree date rape and child molestation are equally immoral and we might decide they ought to be punished equally by a court of law, while also readily noticing that one might be prevented by general education of boys and girls as soon as they're ready to learn about sex and the other only by identifying specific individuals with deviant sexual urges and offering them intense therapy (for instance).

We need to figure out whether or not we should even be talking about these things at the same time at all,  because they may only look like the same issues.  For instance, there's an unfortunate possibility that pedophilia is a sexual orientation; child rape can not and should not be treated as anything other than an ugly, serious criminal offense, but (and I've mentioned this above) if it's a neurological imperative that must be restrained, it's in a different remedial category from other kinds of sexual assault.  Maybe we shouldn't be talking about it at the same time we're talking about other kinds of rapes--maybe we even need new terminology, going back to what I wrote earlier about agreeing upon what we're talking about in the first place.

The Enliven image generates a lot of heat but no light.  It's meant to shock, but there's no point in our being shocked for the sake of being shocked.  I'm doubtful of the point of being shocked at all, actually--not because rape (of whatever sort) doesn't offend the moral conscience, but because I think this is a situation where anger ought to be portioned off to its proper place and a sober look needs to be had as to how we reconcile our moral values and our laws, how to adapt our customs without losing sight of our traditions.  How we solve our problems instead of alternately ignoring them or screaming incoherently about them.

I feel like I need to mention, you know, that we are talking about complicated problems.  We have an adversarial system that gives those accused of crimes a sacrosanct right to face their accusers in public; there's a great deal of value in that principle, but it's obvious that's a setting in which someone who has already been violated in the most personal way conceivable might face humiliation and further trauma.  I don't have a real answer to the point I'm raising there, only that I wish people would look at it as a complicated piece of work and not just a matter of whether little meeples on a chart ought to be beige or red, or if "enough" of them are, or only a mere two are black and that somehow says something about the other 2,498.

There's also one more thing I'd like to point out by way of a comparison or contrast.  Slate ran a critique of the Enliven graphic on the same day they ran another graphic:

I mention it because the Tuskegee Institute did exactly what I think Enliven failed to do: they took a shocking, ugly crime, and they came up with a simple, stark visual representation that effectively communicates something to the viewer.  And they did that in part by being precise: 

In 1959, Tuskegee defined its parameters for pronouncing a murder a "lynching": "There must be legal evidence that a person was killed. That person must have met death illegally. A group of three or more persons must have participated in the killing. The group must have acted under the pretext of service to justice, race or tradition."

Obviously, Tuskegee offered the chart as a historical item, though lynchings were a contemporary problem in 1959; but what they offered here was content and context for dealing with something appalling in a manner that I think incites or encourages action.  There is a wrong here, and we know where it is and where it was and what it is: we have a place to focus our attention.  If the Tuskegee graphic is a form of propaganda, it at least has the virtue of being intellectually honest.  It comes by its power to shock through the truth, without forcing numbers through a strainer and throwing them at a wall to see what sticks in the audience's mind.

Well, I wish someone at Enliven had bothered to do the same.