Sexual Assault Nurse Examiners

June 16, 2010 10:53 AM

Sexual Assault Nurse Examiners (SANEs) are registered nurses who have special training and experience in forensic evidence collection (conducting “rape kits”) and survivor-centered care. They also increasingly conduct forensic examinations on suspects accused of sexual assault(PDF). There is evidence that, through this work, SANE programs increase the prosecution rates of sexual assault cases. However, as a result of a recent Supreme Court Case, Crawford v. Washington, the role of SANEs is being increasingly curtailed(behind a pay wall).

One important role of SANEs is to testify as expert witnesses. When a survivor is unwilling or unable to testify (e.g., in cases of rape/homicide, when the survivor dies before trial, when the survivor is very young, mentally ill, or disabled, when the survivor was intoxicated and thus cannot remember what she said to the SANE), a SANE may read the statements that were recorded in the patient’s medical records. Since the SANE must ask about details of the assault to determine where to search for evidence, and has an emotionally supportive role, statements made in the presence of a SANE are often about critical aspects of the case.

While statements made outside of court and used in court to prove the truth of the case are heresay and thus typically not permitted in court, statements made to medical providers are exempt from such rules, since it is generally recognized in the law that a patient will not lie to their doctor. However, Crawford v. Washington says that if the statements made to someone usually exempt from heresay are "testimonial" (i.e., the person making them knows they can be used in criminal proceedings), then the person making the statements may be more likely to lie, and thus the statements are not admissible in court.

SANEs may be able to present themselves in a clear medical role with survivors. However, this ruling clearly threatens spoken evidence obtained from examinations conducted on suspects. It also threatens important relationships between SANEs and law enforcement, since if they appear to be too close to a survivor, statements made are more clearly testimonial. Such relationships are a critical part of an effective systems response to sexual assault.

posted by emilyd22222 (57 comments total)

Also germane but not linked above because I was involved in this research (linked here with jessamyn's permission): recent evidence(pdf) indicates that SANE programs contribute to higher prosecution and conviction rates. The relationship between law enforcement and SANEs was identified as an integral part of the increase in prosecution/conviction rates.
posted by emilyd22222 at 10:56 AM on June 16, 2010

statements made to medical providers are exempt from such rules, since it is generally recognized in the law that a patient will not lie to their doctor

A simple google search on "patients lie to doctors" destroys that assumption.
posted by Ardiril at 11:16 AM on June 16, 2010

Duh! ... yeah ... and your problem is? Hearsay is hearsay. Makes sense to me. A party (the victim) can always give direct evidence if it is that important to them ... which is why we have a rule against hearsay. A SANE's job is not to take verbal evidence.
posted by jannw at 11:20 AM on June 16, 2010

My employer helped establish SANE nurses in my community at one particular medical hospital years ago. This post makes me sad, because starting January 2010, our Catholic affiliated hospital added a recruitment of trained SANE nurses. I thought progress was in the works. Damn caselaw. The criminal attorneys will be all over this. Progress needs to continue (DESPERATELY!), because most civil attorneys in this town still refer to their own clients as "playing the victim" because the attorneys refuse to understand the complexity of DV and Sexual violence and the laws therein. ^groan^
posted by psylosyren at 11:26 AM on June 16, 2010

I'm not tracking the problem here, and my job revolves in part around dealing with evidentiary issues and sexual assault cases. If the suspect makes a statement, that's NON-hearsay when used by a party opponent (e.g. prosecution) by definition under the hearsay rules, and can be used to prosecute assuming there are no problems with things like Miranda rights being required. If the alleged victim makes a statement for medical purposes, that is generally not viewed as testimonial under cases construing Crawford . . . because statements to the doctor/nurse that are made for medical treatment actually aren't expected to be used for purposes of prosecution.

But of course if the SANE or anyone else tells the person they are asking questions to assist in prosecution, then what is said is going to be testimonial and excluded under Crawford if the alleged victim doesn't testify. The same is true of statements made to professional child interviewers, for the same reason.

Does this really matter? It is pretty unheard of to pursue a sexual assault prosecution without the alleged victim testifying, because it is so difficult to prove such a case.
posted by bearwife at 11:30 AM on June 16, 2010

That's not quite the correct definition of testimonial as defined in Crawford (or more specifically, in Davis v. Washington). I wrote a paper on the last of the Crawford trilogy, Giles, which also factors in here, specifically regarding the forfeiture doctrine. I'm not going to self-link to my own writings, but the whole thing is a mess. For non-attorneys, pretty much any succinct statement you know about hearsay is inadequate to describe the federal rules of evidence and there are a lot of very complex issues here. For attorneys who don't specialize in Evidence or trial practice, you probably remember learning all the exceptions - it's even worse now.

For those who wonder why this is a big deal, here's a sample scenario, which is sadly too common.

Let's say a woman, D, leaves her abusive husband. She goes to the police to file a report against him in order to receive a restraining order. The husband, P, is so infuriated that he hunts down D and rapes her. D goes to the hospital and tells the SANE that P raped her and said he was going to kill her. That night, P kills D because he's mad that she left him.

Current law states that anything P told the police is inadmissible as hearsay, and that given that D likely contemplated that her statement to the SANE would be used in a trial, anything she told the SANE is inadmissible as well. Up until Crawford, the police report would have been seen as non-testimonial and up until Giles, the murder of the witness would have invoked the forfeiture doctrine.

If a prosecutor wants to try P, he has to do so without introducing the most relevant evidence, something that many domestic violence activists believe will lead to an increase in the acquittal rate of men tried for killing the partners (which was the case in Giles itself). So this is a pretty serious deal.
posted by allen.spaulding at 11:33 AM on June 16, 2010 [1 favorite]

A party (the victim) can always give direct evidence if it is that important to them ...

Clearly you have never been, and never known, a rape victim. Grow some sensitivity.
posted by dirtynumbangelboy at 11:37 AM on June 16, 2010 [5 favorites]

recent evidence(pdf) indicates that SANE programs contribute to higher prosecution and conviction rates.

According to emilyd22222's self-link, police refer only 67% of complaints for prosecution, even with SANE nurses, and only 29% lead to conviction or guilty plea. That seems disturbingly low to me.

A serious question: what is the ideal prosecution and conviction rate? Is the correct rate of referral and conviction 100%?
posted by anotherpanacea at 11:59 AM on June 16, 2010

Not to quarrel, allen.spaulding, but current law where? In my jurisdiction statements made for purposes of medical diagnosis and treatment (versus for purposes of prosecution) are indeed considered non testimonial under Crawford.

I'd add that in most murder cases, which is what your hypothetical posits, it is extremely rare to need to show exactly why the defendant murdered the victim to prove the case. Nor is it likely the government would try to prove a rape charge too, given the inability to call the dead victim to testify that the sex was not consensual -- and given that murder carries a very long prison term anyway.

I'm no fan of Crawford, but in practice I'm not seeing an impact on sexual assault cases. There is some impact on domestic violence cases, but much less than was feared when the case was handed down.
posted by bearwife at 12:04 PM on June 16, 2010

Is the correct rate of referral and conviction 100%?

At the simplest level. 100% prosecution and conviction is ideal whenever the punishment is uniformly applied and exactly correct to deter socially costly misbehavior without generating other social costs in the form of overdeterrence of certain desirable behaviors. This ignores the cost of enforcement, various types of opportunity / scarcity of police (hah) or prosecutorial or judicial resources costs, false convictions or acquittals, etc.
posted by Inspector.Gadget at 12:07 PM on June 16, 2010

Is the correct rate of referral and conviction 100%?

Well ... duh! yes ... of course. Due to the historical oppression of women, and the massive gender power imbalance (and their unique special snowflake position), we must right this societal wrong by ensuring that all complaints of sexual assault made by women against men result in conviction and the maximum penalty (which should be increased to include castration). Rules of evidence which inconveniently get in the way of this should be ignored in these cases.

Didn't you get the memo?
posted by jannw at 12:09 PM on June 16, 2010

Ugh. I'm sorry I asked. jannw, I get that you think you're taking some sort of fundamental 6th amendment stand here, but there are plenty of hearsay exceptions and it's not quite the best place to dig in.

I do think it's a serious question: my gut says that 29% is way too low, but I'm not sure what the "right" number is. 80%? I've read that DWI cases have a 91% conviction rate, for instance.
posted by anotherpanacea at 12:16 PM on June 16, 2010

since it is generally recognized in the law that a patient will not lie to their doctor.

What? That seems like an odd assumption? Should statements you make to your doctor be able to be used against you in court?
posted by delmoi at 12:18 PM on June 16, 2010

I do think it's a serious question: my gut says that 29% is way too low, but I'm not sure what the "right" number is. 80%? I've read that DWI cases have a 91% conviction rate, for instance.

I think that instead of shooting for a percentage, we should shoot for the most fair-minded system possible. Justice is more important than a superficial statistic.
posted by esprit de l'escalier at 12:20 PM on June 16, 2010 [1 favorite]

At the simplest level. 100% prosecution and conviction is ideal whenever the punishment is uniformly applied and exactly correct to deter socially costly misbehavior

Even in the case when the charges are false? Odd view.
posted by delmoi at 12:22 PM on June 16, 2010 [1 favorite]

I've read that DWI cases have a 91% conviction rate, for instance.

Well, I would imagine that the conviction rate for crimes where you are caught by the police doing them and an objective measure of guilt is possible the conviction rates would be pretty high.
posted by delmoi at 12:24 PM on June 16, 2010 [1 favorite]

[few comments removed - rape is a difficult topic, please be respectful to other commenters or go to MetaTalk, thank you.]
posted by jessamyn at 12:25 PM on June 16, 2010

Even in the case when the charges are false? Odd view.

You'll notice I reserved the cost of false conviction, because the fallibility of the justice system is something one should think about immediately after figuring out what punishment is likely to deter a guilty party who will certainly be caught. They are basically two separate cost-benefit questions that you have to assemble (with others) in finally answering the question.
posted by Inspector.Gadget at 12:32 PM on June 16, 2010

I get that you think you're taking some sort of fundamental 6th amendment stand here, but there are plenty of hearsay exceptions

Well, most hearsay law isn't constitutional. Only a small fraction of hearsay issues raise Confrontation Clause (6th Amendment) concerns. It makes sense to treat those cases differently from garden-variety hearsay. So "there are plenty of hearsay exceptions" is a bit too cavalier for me. We shouldn't suddenly lose all concern for defendants' rights when the charged offense is rape. Of course rape is a horrible thing ... and so are many crimes, which is why we heighten the level of procedural protections for those accused of crimes.
posted by Jaltcoh at 12:50 PM on June 16, 2010

I think this is all well and good as long as the victim's pay for their own damn rape kits.
posted by Mental Wimp at 12:51 PM on June 16, 2010

Some survivors are uncomfortable speaking with the police because of drug use, undocumented immigrant status, privacy, or simple distrust of police.

These are all pretty valid concerns because the legal system is stacked against victims. We've had cases where a victim reported a rape and requested the police don't tell their parents because they weren't comfortable. The police officers actually followed her from the hospital to her home and told her parents. Ideally, victims would know to contact advocates, get professional medical attention from SANEs, and convict their rapist. In reality, victims rarely contact advocates, most hospitals do not have SANEs (or sometimes, it's a gamble depending on who's on duty when you reach the hospital), and the effective prosecution and conviction rate for rape is about 6%.
posted by yaymukund at 1:02 PM on June 16, 2010 [1 favorite]

Only a small fraction of hearsay issues raise Confrontation Clause (6th Amendment) concerns.

Could you clarify this? Are you pointing at the ways hearsay exceptions are used outside of criminal prosecutions, or are you saying that there are exceptions to the hearsay rule in criminal contexts that don't raise Confrontation Clause issues. Because I'm not sure how I understand how the latter point could be true.
posted by anotherpanacea at 1:09 PM on June 16, 2010

Only a small fraction of hearsay issues raise Confrontation Clause (6th Amendment) concerns.

Could you clarify this?

Sure. The hearsay rule generally excludes statements made outside of court that are presented for the truth of what the statement asserted. There are a lot of evidentiary exceptions to this rule.

The confrontation issues only arise when 1) the person who made the statement in the first place doesn't testify and hence isn't available for cross examination and 2) the statement was made in a context that is testimonial, i.e. that indicates the speaker intended the statement to be used in court.

So, for example, there isn't a Confrontation Clause problem -- although there may be hearsay rule problems -- if an alleged victim testifies at trial and is available to be cross examined (confronted) about prior out of court statements.
posted by bearwife at 1:14 PM on June 16, 2010 [1 favorite]

I think that instead of shooting for a percentage, we should shoot for the most fair-minded system possible. Justice is more important than a superficial statistic.

How do you measure fair-mindedness, and how do you know we haven't already achieved it? Doesn't one use statistics in a case like this, at least in part, in order to get a sense of where we are and where we might have to go?
posted by OmieWise at 1:34 PM on June 16, 2010

There's a MetaTalk thread about the viability of this post given that crucial content is behind a paywall.
posted by OmieWise at 1:48 PM on June 16, 2010

Only a small fraction of hearsay issues raise Confrontation Clause (6th Amendment) concerns.

Could you clarify this?

Sure. ...

To add to bearwife's comment, I'll just note that the hearsay rule itself is incredibly old. It's older than the United States. (Wikipedia traces its origins to the 17th century and says it started taking its modern form in the early 19th century.) But hearsay never even raised an issue with the Confrontation Clause until the Supreme Court decided the ground-breaking case of Crawford v. Washington in 2004. So, historically, the percentage of all hearsay issues that have been Confrontation Clause (Crawford v. Washington) issues has been tiny.

And here's an example of hearsay that doesn't raise Confrontation Clause issues: any hearsay evidence in any civil case! Only criminal defendants have confrontation rights, so the Confrontation Clause is never at issue in a civil lawsuit.
posted by Jaltcoh at 1:50 PM on June 16, 2010

I think that instead of shooting for a percentage, we should shoot for the most fair-minded system possible. Justice is more important than a superficial statistic.

How do you measure fair-mindedness, and how do you know we haven't already achieved it? Doesn't one use statistics in a case like this, at least in part, in order to get a sense of where we are and where we might have to go?

You need much more than the statistic of "conviction rate" to decide "where we are and where we might have to go." What is happening when someone quotes that single statistic is that the reader fills in a world of unknowns with his prejudice and individual experience. One needs much more information.

The quotation of this statistic also sets up the false ideal that each conviction is a positive development and each acquittal a negative one. Isn't the better ideal that when there is sufficient, reliable evidence, we expect a conviction and vice versa? In cases where there is a poverty of evidence, the positive development must be acquittal -- regardless of actual guilt -- even though it's hard to stomach.
posted by esprit de l'escalier at 1:57 PM on June 16, 2010

Jaltcoh is right except for one small legalistic detail, which is that pre-Crawford cases did consider the Confrontation Clause as well as the evidence rules when dealing with many criminal cases involving hearsay evidence -- but used a much different analysis than Crawford does. Crawford greatly narrowed the field of admissible hearsay evidence in criminal cases that would also get past a Confrontation Clause challenge. Crawford's biggest impact has been on child sexual assault cases, which used to rely much more on the use of out of court statements made by very young children, and on domestic violence cases, which often involve unavailable victims.

Having said all that, I still don't see a problem with using SANE nurses in sexual assault cases -- that happens routinely in my jurisdiction, post Crawford.
posted by bearwife at 2:17 PM on June 16, 2010

Thanks for the correction, bearwife.
posted by Jaltcoh at 2:51 PM on June 16, 2010

You'll notice I reserved the cost of false conviction, because the fallibility of the justice system is something one should think about immediately after figuring out what punishment is likely to deter a guilty party who will certainly be caught. They are basically two separate cost-benefit questions that you have to assemble (with others) in finally answering the question.

So we should have 100% conviction for people who are definitely guilty? That's... not very informative.
posted by delmoi at 2:59 PM on June 16, 2010 [1 favorite]

I do think it's a serious question: my gut says that 29% is way too low, but I'm not sure what the "right" number is. 80%? I've read that DWI cases have a 91% conviction rate, for instance.

The ideal number of reported rapes that should result in conviction is the number of reported rapes which did in fact happen, assuming no one is falsely convicted.
posted by furiousxgeorge at 3:08 PM on June 16, 2010

Damn it delmoi.
posted by furiousxgeorge at 3:09 PM on June 16, 2010

Jaltcoh: when I look at the list of hearsay exceptions I helped my wife memorize for the bar exam, quite a number of them seem like potential confrontation clause issues. But my understanding is that that this is one of those places where you have layers of common law crafting, codification, Supreme Court parsing, and now law and econ style evaluations all wringing the last vestiges of a principle exception from the hearsay rule and leaving a confusing maze that, well, requires a ton of memorization. If the gist of your point is simply that most cases are civil cases, and civil cases don't raise Confrontation Clause issues, it seems like you might have initially overstated it.

(bearwife, I too was thinking of the line of child sexual assault cases that started with Maryland v. Craig and which was supposed to be overturned by Crawford. But this also comes up, for instance, in the question of whether forensic analysts must be cross-examined before their results are admitted.)
posted by anotherpanacea at 3:24 PM on June 16, 2010

Isn't the better ideal that when there is sufficient, reliable evidence, we expect a conviction and vice versa? In cases where there is a poverty of evidence, the positive development must be acquittal -- regardless of actual guilt -- even though it's hard to stomach.

Well, sure, which is why this is a question at all rather than just an assertion that a 100% conviction rate for all reported rapes is what we should shoot for. You seem to be misunderstanding the import of a statistic like conviction rates. The rate contains some of the nuances you suggest are important. We don't know from the rate whether the low conviction rate is because prosecutors aren't developing good evidence, cops are failing to pursue cases effectively, juries are being lenient...but the rate gives us something to suggest that we should take a closer look. I like justice as much as the next person, but what I take to be your reaction against data leaves us without anything but aspirations with which to craft our argument for more justice.
posted by OmieWise at 3:43 PM on June 16, 2010

anotherpanacea, you are right, but frankly so far in the courthouse where I hang out, that has been an inconvenience, not a barrier, for criminal prosecution. I.e., defense attorneys understandably don't stipulate to forensic expert conclusions, so the expert has to show up and testify and be subject to being cross examined -- but of course the experts cooperate and do that.
posted by bearwife at 4:14 PM on June 16, 2010

FYI, all, Emilyd22222 was kind enough to send me the article behind the paywall. Here's what the prosecutor author says about the role SANEs should maintain with alleged victims:

If the SANE acts independent from law enforcement, maintains a caring
but objective stance toward the facts before her, and
conducts herself as a professional focused only on her
patient, then most courts would agree that her elicitation
of information from her patient was indeed solely
for the purposes of diagnosis and treatment. A hospital based,
independent SANE program with objective examiners
should have little trouble presenting themselves to
the court as medical professionals and not individuals affiliated
with law enforcement. In turn, the prosecutor
should be able to argue that statements made to medical
professionals such as these are non-testimonial ones.

I.e., as the prosecutor says, statements to medical personnel for medical purposes are normally not viewed as testimonial and hence as barred in criminal cases by Crawford.
posted by bearwife at 4:29 PM on June 16, 2010

bearwife, does it give any evidence of SANE "curtailment"?
posted by anotherpanacea at 4:38 PM on June 16, 2010

Hearsay Exception Song: do not rely on this for legal advice.
posted by anotherpanacea at 4:48 PM on June 16, 2010

anotherpanacea: No. The next thing the prosecutor author says is:

However, if the SANE presents herself as an adjunct
of law enforcement or someone whose primary job is
to secure convictions in rape cases for the prosecution,
she will very likely put her status as an objective medical
professional in jeopardy. A skilled defense attorney
will argue to the judge, under Crawford or one of the
many state cases that have referenced it, that because
of the SANE’s close identification with the law enforcement
function, she cannot be regarded as solely a medical
provider. Statements made to her must be considered testimonial.
This clear bias toward one side of the case will
hurt her ability to testify in a compelling manner about
both fact and expert matters. It may also close off the possibility
of using statements made to her in the event that
the patient becomes unavailable to testify.

And I completely agree: a SANE who views their role as that of a law enforcement adjunct isn't going to be able to testify about the statements they got from alleged victims who think what they say will be used in court.

What the article says nothing about is the impact of SANE interaction with suspects. But I'll point out that people who are in custody and are being interrogated by law enforcement need to be advised of their Miranda rights, or their statements are inadmissible under the 5th amendment. So SANEs who act as adjunct law enforcement to question suspects without Miranda probably won't be able to talk about what the suspects said either.

I don't think emilyd22222 understood the article. (In her defense, she isn't a lawyer.)
posted by bearwife at 5:14 PM on June 16, 2010

So we should have 100% conviction for people who are definitely guilty? That's... not very informative.

No. You start with figuring out what level of punishment is ideally necessary to deter all guilty parties and no innocent parties for a given legal standard, and then you alter the whole calculus based on projections about false convictions and acquittals and the social costs thereof, as well as trying to predict side effects and taking into account enforcement costs and similar. It gets difficult when you have a lot of information to work with, but at bottom it's straightforward: you adjust the punishment upward as the likelihood of a guilty party being caught and convicted decreases in order to force guilty parties as a class to internalize the social costs of their behavior.
posted by Inspector.Gadget at 5:23 PM on June 16, 2010

Jaltcoh: when I look at the list of hearsay exceptions I helped my wife memorize for the bar exam, quite a number of them seem like potential confrontation clause issues. But my understanding is that that this is one of those places where you have layers of common law crafting, codification, Supreme Court parsing, and now law and econ style evaluations all wringing the last vestiges of a principle exception from the hearsay rule and leaving a confusing maze that, well, requires a ton of memorization. If the gist of your point is simply that most cases are civil cases, and civil cases don't raise Confrontation Clause issues, it seems like you might have initially overstated it.

No, I just gave civil cases as one particularly clear example. Many hearsay issues in criminal cases also don't involve the Confrontation Clause. For instance, Crawford v. Washington doesn't apply to a statement that isn't "testimonial" (whatever that means). But a nontestimonial statement could still be inadmissible in a criminal case under the general hearsay rule (not the Constitution).

The reason the Confrontation Clause + hearsay comes up a lot in studying for the bar exam is that it's a particularly intricate, ruled-based area of law that bar examiners have decided is very testable. Also, it's pretty new, so the bar review courses might have more concern about whether people have adequately learned it in law school. The frequency of subjects on the bar exam generally doesn't bear any connection to how often they occur in real life. For instance, the Rule of Perpetuities is an ancient and notoriously counterintuitive law that has generally been replaced across the country by more rational statutes. But the bar exam will go ahead and test you on the crazy 16th century rule because everyone learns it and it's easy to write difficult exam questions about it. (And surely you don't think the world has many fertile octogenarians?)
posted by Jaltcoh at 5:43 PM on June 16, 2010

No. You start with figuring out what level of punishment is ideally necessary to deter all guilty parties and no innocent parties for a given legal standard, and then you alter the whole calculus based on projections about false convictions...

Okay, well now you're just saying that the conviction rate should be set at the appropriate level. Which is quite different from your original statement that "At the simplest level. 100% prosecution and conviction is ideal"

Anyway, I find the idea that we should simply seek to set the conviction rate at some sort of socially optimal level without regard for the actual guilt or innocence of any individual suspect somewhat creepy.
posted by delmoi at 5:49 PM on June 16, 2010 [1 favorite]

Right, I'm talking about the simplest conceptual level. I certainly don't believe that 100% arrest->conviction rate is desirable in the real world. What I'm getting at is that there is a discernable optimal punishment for a given crime, subject to problems arising from minor variations in trial procedures as well as the ones that I listed above. But if you were explaining to someone asking in the abstract how to mathematically determine the ideal deterrence situation (and it occurs to me now that might be a misreading of the original question), you'd start like this:

Imagine that the social cost of parking beyond metered time is $10, because all spaces in our hypothetical downtown are equally valuable and demand for spaces is constant for the length of the thought experiment. Given perfect traffic enforcement, every overparker should get a $10 ticket, which eliminates any incentive for people who are indifferent - or value the extra time spent doing whatever rather than moving their car - to overpark. What's left is only the overparkers who value the time at more than $10, and will fully compensate society for the cost while generating some additional value. Now imagine you have very flawed enforcement - say, 1 in 10 overparkers is caught. Then, to return to this optimal level of deterrence, every caught overparker must be fined $100. Provided that this new policy is announced, signage is clear, etc., the risk of ensnaring an unsuspecting, indifferent person is nil.

Now when you add false convictions and false acquittals, a real time/money/manpower budget, etc. you begin to add layers of complexity that you need to factor in everywhere you can't perform a purely procedural fix (which has its own costs, but they aren't generally paid out of the same budget). In rape cases, for instance, the consensus is that we want to overdeter - so we punish convicted rapists severely, because we don't want people to even toe the legal line. We don't consider "borderline consensual" sexual practices to have much if any value. We jack up the penalties even further because many rapists are never caught. We don't (consciously) add in much weight for false convictions because both false convictions and false acquittals are both so costly to society and drawing the legal standard is already difficult. So, getting back to the main point, it's possible to think through to a real-world standard in terms of simple situations and endless refinements - and in fact this is the outline of the approach that many economists take in writing about the penal system. It's much easier to do this in a small town with traffic tickets with a lot of known quantities, and with complex crimes for the moment I think we can't do much better than a heavy thumb on the scale.

Anyway, I find the idea that we should simply seek to set the conviction rate at some sort of socially optimal level without regard for the actual guilt or innocence of any individual suspect somewhat creepy.

There, I was writing from the perspective of assuming perfect punishment and a perfect legal standard to point out that in that idealized situation 100% conviction is optimal, but to point out that it was just a model and that a host of complications make it unworkable if applied hastily.
posted by Inspector.Gadget at 6:14 PM on June 16, 2010

We don't (consciously) add in much weight for false convictions because both false convictions and false acquittals are both so costly to society and drawing the legal standard is already difficult.

Oops, that should be

We don't (consciously) add in much weight for false acquittals because both false convictions and false acquittals are both so costly to society and drawing the legal standard is already difficult.
posted by Inspector.Gadget at 6:15 PM on June 16, 2010

Inspector.Gadget, I think the above assumes we're designing a deterrence system rather than both an investigation system. It also assumes that rapists are rational choosers. For that matter, it assumes the same thing about democratic societies. This is not, in any case, the kind of 'ideal' I had in mind.

I guess the closest approximation to what I meant would be something like: what would be the rate of prosecution and conviction be in a non-misogynistic, non-patriarchal society? Not the Kaldor-Hicks optimal society, just one that doesn't hate women.
posted by anotherpanacea at 8:09 PM on June 16, 2010

But this also comes up, for instance, in the question of whether forensic analysts must be cross-examined before their results are admitted

This was the crux of Melendez-Diaz v. Massachusetts. I'm sorry I'm too busy to really contribute here, this is one area where I've written a few articles and can contribute. Crawford shouldn't control the issue of SANEs, Davis should. It would seem, perhaps, that a jurisdiction that allowed this as non-testimonial might be getting Davis wrong. Information given to a SANE would seem to be closer to a statement made for a restraining order than it is a 911 call. An individual at a hospital certainly would know that testimony given could be used in future prosecutions, as opposed to resolving an ongoing emergency. I could look more closely, but I suspect that a court interpreting Davis would not allow this, especially given the most recent Confrontation Clause rulings.
posted by allen.spaulding at 8:28 PM on June 16, 2010

Inspector.Gadget: "We don't (consciously) add in much weight for false acquittals because both false convictions and false acquittals are both so costly to society and drawing the legal standard is already difficult."

Isn't it conventional legal philosophy that a false acquittal is much preferable to a false conviction? Ah, yes, Blackstone's formulation, "better that ten guilty persons escape than that one innocent suffer".

Rape is horrible, but it seems that all regard for the defendant's rights go out the window when it enters into the picture. It's horribly unpopular to even question an allegation of rape, but surely we should apply the same skeptical examination to rape allegation as to any other allegation of wrongdoing?
posted by Joakim Ziegler at 8:57 PM on June 16, 2010 [2 favorites]

Anyway, I find the idea that we should simply seek to set the conviction rate at some sort of socially optimal level without regard for the actual guilt or innocence of any individual suspect somewhat creepy.

Concern about a low conviction rate is not equivalent to an attempt to arbitrarily raise it by just convicting some random extra suspects.

Instead, you can look at the rate as an indicator of a potential problem, which can be either explained by some rational means or understood as an indicator that justice is lacking for some disproportionate number of victims in these cases. And then you can investigate why.

Say, for instance, that the rape conviction rate per alleged victim were .0001%. Would you just say, well, let's ignore that, because any concern about that percentage is inherently an attempt to raise it through arbitrary convictions? Would it be a meaningless statistic, and as long as the forms were followed, we should be satisfied that we're doing fine?
posted by palliser at 9:22 PM on June 16, 2010

Rape is horrible, but it seems that all regard for the defendant's rights go out the window when it enters into the picture. It's horribly unpopular to even question an allegation of rape, but surely we should apply the same skeptical examination to rape allegation as to any other allegation of wrongdoing?

Right. I'm talking about whether the pair of (i) the elements of the crime of rape and (ii) the punishment meted out as they together impact deterrence, not about general procedural and evidentiary rules. The former two have to basically move in tandem to make deterrence work. In considering the rule to be applied, the false acquittal / false conviction problem seems to be more directly controlled by defining the elements of rape.

I like strong procedural protections; my default stance is against any sort of alteration of standards for particular classes of even heinous crimes.
posted by Inspector.Gadget at 9:32 PM on June 16, 2010

I notice that in Hammon v. Indiana, there's still an issue of batterers using spousal privilege to prevent their victims from testifying. WHAT THE FUCK?
posted by anotherpanacea at 6:51 AM on June 17, 2010

Crawford was also about spousal privilege. Crawford stabbed a man he thought was trying to rape his wife. She disagreed. Crawford claimed the stabbing was in self-defense but the wife thought it was not. She was barred from testifying under marital privilege laws. In Washington, marital privilege does not extend to out-of-court statements admissible under a hearsay exception, so the prosecution attempted to admit the wife's tape-recorded statements to the policy.

None of this would have happened if Washington recognized the backwardness of this application of marital privilege - based as it is on the assumption that husband and wife always share the same interests is demonstrably not true. There's a reason why these cases all involve domestic violence and spousal privilege reform would also be an important step.
posted by allen.spaulding at 7:19 AM on June 17, 2010 [1 favorite]

I'm not a lawyer. But as I understand it, in Crawford it's a standard spousal privilege, i.e. to protect marital intimacy or whatever I have to be able to tell my wife about all my horrible crimes without fearing that she'll be forced to testify. But in these cases, the immunity is supposed to be waveable by the testifying spouse. The privilege supposedly inheres in the witness, not the defendant.

In Hammond we're literally talking about a husband beating his wife with his fists and then preventing her from testifying. On the one hand, I'm glad that counts as waving your Confrontation Clause rights, because... duh! But on the other hand, how can any state have left the domestic battery/marital rape loophole unclosed in our era? WTF Indiana?
posted by anotherpanacea at 7:41 AM on June 17, 2010

The privilege supposedly inheres in the witness, not the defendant.

This is not true. While marital privilege varies from jurisdiction to jurisdiction, in many states, including Washington, the defendant can forcibly prevent a spouse from testifying, even if the testifying spouse wishes to.
posted by allen.spaulding at 7:50 AM on June 17, 2010

OK, allen.spaulding, do you practice in Washington? Because that is my jurisdiction. And let me fill you in on what the law actually is here.

First, a spouse or domestic partner is "incompetent" to testify against another spouse/domestic partner in Washington EXCEPT in civil cases between them, criminal action or proceeding involving a crime by one against the other, in a criminal action that was pending when the marriage or partnership began, or a criminal action involving an alleged crime by the spouse/domestic partner against a child, or proceedings involving treatment for substance abuse, mental illness, crisis response, or sexually violent predators.

I.e., in most cases, and certainly in domestic violence, child abuse and rape chases, a spouse/partner is competent to testify against the other.

Similarly, the spouse/partner privilege in Washington only protects confidential communications between the spouses/partners, and again there are a lot of exceptions, including testimony in civil cases between them, testimony concerning a crime by one spouse/partner against the other, testimony concerning a crime by one spouse/partner against a child, testimony in a proceeding for non support or desertion, testimony about efforts to induce a spouse/partner not to testify (witness tampering), and testimony proceedings involving substance abuse, crisis response, mental illness, and sexually violent predators.

So again, the privilege doesn't allow a defendant to forcibly or otherwise prevent a spouse/partner from testifying in most cases, and certainly not in domestic violence, rape or child abuse cases.

I'd add that I heartily disagree with your analogy that statements to medical personnel are like statements for protection orders (which are clearly testimonial as they are made to a court!), and I challenge you to cite a case anywhere that adopts it. I also would ask if you can cite a single appellate case in any jurisdiction that precludes statements in a medical setting to medical personnel as "testimonial" under Crawford.
posted by bearwife at 10:42 AM on June 17, 2010

bearwife - Crawford itself is the example you want. A husband forcibly prevented his wife from testifying when he tried to kill her lover (or rapist, depending on perspective). I mean, this is what we've been discussing from the beginning, no? Cite the exceptions you want, it's demonstrably true that marital privilege does not inhere in the witness in Washington and that the spouse can use the privilege to silence the voluntary testimony of his spouse.

And I keep citing Davis, which is also a case from your jurisdiction. In that case, a 911 call was deemed non-testimonial. In defining what counts as testimonial, this is what the court had to say:

Statements are nontestimonial when made in the course of police interrogation under circumstances objectively indicating that the primary purpose of the interrogation is to enable police assis-tance to meet an ongoing emergency. They are testimonial when the circumstances objectively indicate that there is no such ongoing emergency, and that the primary purpose of the interrogation is to establish or prove past events potentially relevant to later criminal prosecution"

So, this is the test. Does a statement made to a SANE after the "ongoing emergency" has ended count as testimonial? Davis would seem to indicate that it would. This is not to defend the law, just google my name and you'll see my written critique of this nonsense. But Scalia's view of the Sixth Amendment is openly hostile to women, as many have written. As written, if the statement is made to someone who could potentially testify in later prosecution once the emergency is over, well, it's almost certainly testimonial. I hate it, but that's the recent jurisprudence.
posted by allen.spaulding at 8:56 PM on June 17, 2010

A husband forcibly prevented his wife from testifying when he tried to kill her lover (or rapist, depending on perspective).

Now that is an interesting twist -- he can't prevent her from testifying if he hurts her, or if he abuses a child and she is a witness (going by bearwife's information), but he can hurt someone else she cares about and prevent her from testifying. Not so uncommon, actually, for abusers to harm mothers/fathers/sisters/brothers of their immediate victims.
posted by palliser at 6:05 AM on June 18, 2010

Well, allen.spaulding, we have lots of areas of agreement. I think Scalia is hostile to women too. (And children -- or maybe just worried about all those poor priests facing trials based on what molested children have told investigators.) Also, while you have read these cases and written about them, I too have read these cases, and have taught about them and their impact on domestic violence prosecutions. Moreover, I work in a court, and I see plenty of criminal cases raising hearsay and confrontation clause issues.

Here is where we differ. I know of no case, not in Washington and not elsewhere, precluding the admission of statements made to medical personnel in the context of receiving medical treatment, under the authority of Crawford or Davis or Giles. Do you?

Palliser, you are right -- but there tend to be other available witnesses, including the victim of the violence, for domestic violence cases involving crimes against family members of spouses/partners.
posted by bearwife at 3:32 PM on June 18, 2010