They're always guilty ... or are they?

The deceptively simple case of Diaz v. United States

The least popular post in the history of this Substack is Juror Egos and Juror Ids. In that post, I opined that Evidence is the most interesting law school class and then proceeded to analyze an obscure Confrontation Clause case called Samia v. United States. As it turns out, readers are more interested in abortion and Trump than evidence law. Who knew?

Having failed to learn my lesson, I will once again address an obscure case about evidence law: Diaz v. United States, which will be argued in the Supreme Court on March 19. Diaz is an intriguing case because, on the surface, it looks extremely simple. The facts are easily comprehensible to a lay person. The legal provision at issue is two sentences long. And best of all, there is no relevant case law to weigh us down. You’d think the case would be resolved easily—either there’s a straightforward right answer, or there’s a single dimension of disagreement in which the Court can pick a position without too much fuss.

Unfortunately not. The case abounds with disagreements. Textual disagreements, philosophical disagreements, empirical disagreements. It’s like a fractal—if you zoom in on any one sub-issue, the case does not get any simpler, but instead you get a bunch of disagreements about that sub-issue. Why that is—and what that says about legal reasoning—is the subject of today’s post.

The simplest case ever

(Couldn’t settle on a single art style this week, so there will be a bunch.)

For a Supreme Court case, Diaz is remarkably easy to understand.

Ms. Diaz drove from Mexico into the United States. Border patrol officers searched her car. Alas, 28 kilograms of methamphetamine were found hidden in the door panels.

The government charged her with drug trafficking. Her defense at trial was that she didn’t know the drugs were in the car. She must have been a blind mule, she claimed.

At trial, the government called a “Homeland Security Investigations Special Agent” to testify as an expert. The expert had no personal knowledge of Ms. Diaz’s case, but nonetheless offered the following testimony:

Q. Agent Flood, based on your training and experience, are large quantities of drugs entrusted to drivers that are unaware of those drugs?

A. THE WITNESS: No. In extreme circumstances —actually, in most circumstances, the driver knows they are hired. It’s a business. They are hired to take the drugs from point A to point B.

Q. And why aren’t—why don’t they use unknowing couriers, generally?

A. THE WITNESS: Generally, it’s a risk of your— your cargo not making it to the new market; not knowing where it’s going; not being able to retrieve it at the ending point, at your point B. So there’s a risk of not delivering your product and, therefore, you’re not going to make any money.

On cross-examination, the agent testified that drug traffickers occasionally use blind mules, but in circumstances very different from Ms. Diaz’s case.

The jury convicted Ms. Diaz.

The question before the Supreme Court is whether the expert witness’s testimony was admissible under Federal Rule of Evidence 704(b). Rule 704(b) provides: “In a criminal case, an expert witness must not state an opinion about whether the defendant did or did not have a mental state or condition that constitutes an element of the crime charged or of a defense. Those matters are for the trier of fact alone.”

Ms. Diaz claims that the expert’s testimony was inadmissible because it was “about” whether she knew about the drugs, i.e., had “a mental state … that constitutes an element of the crime charged.” The expert testified that drug couriers usually know that they’re transporting drugs. The purpose of this testimony was to persuade the jury that Ms. Diaz knew she was transporting drugs. Therefore, she argues, his testimony was “about” whether she knew about the drugs.

The government responds that the expert’s testimony wasn’t “about” Ms. Diaz’s mental state because the expert never mentioned her. According to the government, Rule 704(b) prohibits an expert from opining on the defendant’s mental state. Testimony about the general practices of drug cartels isn’t “about” Ms. Diaz’s mental state, even if a jury might rely on that testimony to draw an inference about her mental state.

That’s it. That’s the entire case! Praise the Lord, there’s no case law on point.

So who’s right?

Whatever, it’s semantic

At the highest level, the case boils down to one of those philosophical questions of life: what does it mean for something to be “about” something else?

Is marriage “about” love? Are law firms “about” money?

Is Waiting for Godot “about” Godot, even though Godot never appears? Was Agent Flood’s testimony, which didn’t specifically mention Ms. Diaz’s mental state, “about” Ms. Diaz’s mental state?

IDK. I guess it depends on what “about” means. Perhaps we should just introspect deeply about how we typically use the word “about” until an answer pops into our heads?

Seems like a dissatisfying way of resolving the case, though. Perhaps, instead, we should learn about the history and purpose of the rule before deciding what “about” means.

Unfortunately, analyzing history and purpose makes the case harder, not easier.

Hotel California

Before 2011, Rule 704(b) provided: “No expert witness testifying with respect to the mental state or condition of a defendant in a criminal case may state an opinion or inference as to whether the defendant did or did not have the mental state or condition constituting an element of the crime charged or a defense thereto.” In 2011, Rule 704(b) was amended into its current form, which bars an expert from “stat[ing] an opinion about whether the defendant did or did not have” a mental state.

Note the difference: the former version of the rule appeared to assume that an expert could testify “with respect to the mental state or condition of a defendant in a criminal case” and merely barred the expert from testifying “as to whether” the defendant had the “mental state or condition constituting an element of the crime charged or a defense thereto.” By contrast, the current version of the rule bars an expert from “stat[ing] an opinion about whether the defendant did or did not have a mental state or condition that constitutes an element of the crime charged or of a defense.” According to the Advisory Committee’s notes,¹ this edit was intended to be “stylistic only” with no change in substance.

The government makes the following argument:

• Under the now-repealed version of the rule, the government would have won, because the rule implicitly authorized an expert to testify “with respect to the mental state or condition of a defendant in a criminal case” so long as the expert did not specifically refer to the defendant.

• The Advisory Committee’s notes say that the current rule is supposed to mean the same thing as the prior rule.

• Therefore, under the current version of the rule, the government still wins.

I don’t like this argument. It’s not clear to me that the government would have won under the old version of the rule.² But even if it would, the language in the old version of Rule 704(b) should be given zero weight.  It should be wiped out of everyone’s memories, like in Eternal Sunshine of the Spotless Mind. 

Otherwise, you are stuck with the old, ambiguous rule forever. Any effort to “clarify” the rule will automatically backfire, because courts will be forced to apply some unholy combination of the new, clearer rule and the old, less clear rule. It’s like Hotel California: you can amend the rule any time you like, but you can never leave.

It’s also not clear to me that we should care about the Advisory Committee’s explanation for the change. Justice Scalia, wise as usual, wrote a concurrence saying that Advisory Committee notes on the Rules of Evidence shouldn’t be given authoritative weight because … who cares what the Advisory Committee thinks? The rule is the rule. The Advisory Committee’s explanation doesn’t even make much sense. Is the new rule really so much more stylish than the old rule?

Attempting to draw inferences from now-repealed versions of rules makes life too complicated. We have enough rules that do exist. Why should we care about rules that no longer exist? Good riddance!

But evidently, the government doesn’t agree with me. The government’s argument injects two difficult questions into the case:

1. To what extent should we look at a now-repealed version of a rule in order to assess the purpose of enacting a new rule?

2. To what extent should we trust the Advisory Committee’s assertion that the amendment was “stylistic”?  Should we treat this as more trustworthy than traditional legislative history?  Less trustworthy?

The case suddenly seems more complicated than a mere dispute about what “about” means.

Against telepathy

Why does Rule 704(b) exist anyway? Fortuitously, there’s a clear historical explanation for Rule 704(b). Not-so-fortuitously, that historical explanation does not make the case any easier.

In 1981, John Hinckley shot Ronald Reagan in an effort to impress Jodie Foster.  He was charged with attempted murder and found not guilty by reason of insanity.  This led to a public outcry. 

Congress responded by amending the Rules of Evidence.  At Hinckley’s trial, experts on both sides testified as to whether they thought Hinckley was insane. The government’s experts said he wasn’t, Hinckley’s experts said he was.  This seemed hinky, so Congress enacted Rule 704(b), which banned experts from testifying “as to”—now, “about”—the defendant’s mental state.

Well, this intriguing historical information seems like it should be relevant somehow to Diaz. Right?  Bueller?

Let’s present the arguments on this issue in the form of a dialogue between the defense lawyer and the prosecutor.

• Prosecutor: This history shows that we should win.  The purpose of Rule 704(b) was to prevent psychologists from testifying about whether defendants were insane.  That’s totally different from law enforcement experts testifying as to the typical modus operandi of drug cartels.

• Defense lawyer: Rule 704(b) doesn’t say anything about “insanity.”  It says “mental state or condition that constitutes an element of the crime charged or of a defense.”  “Knowing there are drugs in the car” is a mental state that constitutes an element of the crime charged.  So it’s irrelevant that the testimony wasn’t about insanity; the rule is broader than the original justification for the rule.

• Prosecutor: You’re missing my point.  The original justification for the rule was that “mental health experts” have no special insight into the content of a criminal defendant’s mind.  Just because you’re a psychiatrist or psychologist, doesn’t mean you’re telepathic.  But the police officer in this case wasn’t trying to be telepathic.  He was just talking about how drug cartels usually operate—they hire and pay couriers as opposed to using blind mules.  Rule 704(b) wasn’t intended to foreclose such testimony.

• Defense lawyer: You’re missing my point.  Suppose the expert said something like: “In my expert opinion, it is highly likely that Ms. Diaz knew about the drugs; after all, drug couriers almost always know there are drugs in the car.”  That testimony would be barred by Rule 704(b): it is obviously “an opinion about whether the defendant did or did not have a mental state that constitutes an element of the crime charged.”  That remains true even though the witness would be relying on his law enforcement expertise rather than his psychology expertise.  There’s no practical difference between that hypothetical testimony and the testimony that “drug couriers almost always know about drugs in the car”: the obvious implication of saying “drug couriers almost always know about drugs” is that Ms. Diaz knew about the drugs.  So it would be arbitrary to say that Rule 704(b) bars the former testimony but permits the latter testimony.

• Prosecutor: It makes perfect sense to distinguish the two.  There’s no way to read Rule 704(b) to permit the former testimony.  But there is a way to read Rule 704(b) to permit the latter testimony.  And if we can read Rule 704(b) to permit the latter testimony, and permitting the latter testimony would be consistent with Rule 704(b)’s purpose, why not read it that way?   Fine, maybe it seems a bit arbitrary to exclude the former testimony but permit the latter testimony, but general rules always produce arbitrary distinctions at the margin.  Better an arbitrary distinction than an atextual expansion of a rule.

This type of dispute—arbitrary distinction versus atextual expansion—pops up constantly in the law.  In almost every hard case, you will find it lurking somewhere. 

Take, for instance, Trump v. Anderson.  Section 3 of the Fourteenth Amendment applies only to people who previously took an oath “as a member of Congress, or as an officer of the United States, or as a member of any state legislature, or as an executive or judicial officer of any state.” One hotly disputed issue was whether Trump so qualified.  You could generate a closely similar dialogue:

• Anti-Disqualification Lawyer:  Trump never took an oath “as a member of Congress,” a “member of any state legislature,” or “an executive or judicial officer of any state,” so that leaves “Officer of the United States.”  But “Officer of the United States”—by its literal text—encompasses only appointed officers, not elected Presidents.  And the Fourteenth Amendment wasn’t intended to prevent insurrectionist Presidents from holding office again.  So, under both the text and purpose of the Fourteenth Amendment, Trump can be President again.

• Pro-Disqualification Lawyer:  But this would produce an arbitrary distinction.  Prior Presidents would have been subject to Section 3 based on a job they had before they were President.  For example, Joe Biden and Barack Obama both took oaths as “member[s] of Congress.”  Isn’t it arbitrary that the Constitution leaves Trump—unlike every other President since the Fourteenth Amendment’s enactment—free to commit insurrection and run again for President merely because Trump didn’t previously hold a high-level political job?

• Anti-Disqualification Lawyer: It makes perfect sense to draw this distinction.  The Fourteenth Amendment wasn’t intended to cover Presidents. Regardless of its intended scope, however, it unambiguously covers Presidents who previously held high-level political jobs; there’s no way to read it otherwise. But there is a way to read it so that it doesn’t cover Presidents who didn’t previously hold high-level political jobs.  So why not read it that way? Fine, maybe it seems arbitrary for Section 3 to distinguish between Trump and other Presidents, but general rules always produce arbitrary distinctions at the margin.  Better an arbitrary distinction than an atextual expansion of a rule.

OK so … who’s right about this?  Well, which is worse, an arbitrary distinction, or an atextual expansion of a rule?

Answer: There is no answer. These options are incommensurate. No neutral principle can decide which is worse—much like no neutral principle can decide the abstract question of, “what is the platonic meaning of the word ‘about’”?  

Sorry, I don’t have a good answer on this. All I know is that the more you dig, the harder the case gets.

Profiling evidence seems kinda sus

To me, the agent’s testimony just … feels … wrong.  Essentially, the agent was testifying: “I’ve been a cop for years; people like Ms. Diaz are usually guilty; so she’s probably guilty.”  I have a strong intuition that this type of testimony is out of place at a criminal trial.

Why?  A few reasons, which aren’t entirely consistent with each other.

The testimony is too persuasive.  If in fact the police officer is right that he’s seen a zillion drug cases and the drug lords never use blind mules, the jury might treat that testimony as dispositive and ignore the evidence specifically pertaining to the defendant.  Maybe it’s rational for the jury to treat the expert testimony as dispositive, but criminal cases are supposed to focus on the specific facts of the defendant’s case.  The expert’s testimony is a bit like propensity evidence—i.e., evidence that the defendant previously committed the same crime—which is generally inadmissible (subject to some rather porous exceptions).  Whether someone previously committed the same crime is undoubtedly relevant evidence of guilt, but it’s nonetheless excluded because of the risk that the jury will treat it as so relevant that it will ignore the specific facts of the defendant’s case.

The testimony isn’t persuasive enough.  The agent’s general recollections about the operation of drug cartels say little about whether Ms. Diaz herself is guilty.  What experience is the agent relying on, exactly?  Is the agent relying on his recollections of cases in which it was obvious that the couriers knew about the drugs in the car (e.g., the courier admitted to his knowledge or the drugs were in plain sight)?  If so, those cases shed little insight on whether a courier knows about drugs when she denies knowing and the drugs are hidden in the car.  Or maybe the agent is relying on his recollections of cases in which drivers claimed not to know about the drugs in the car and the agent didn’t believe them?  Well, of course the agent didn’t believe them.  He’s an agent!  And those guys usually plead guilty anyway.  Conditional on the defendant denying knowledge and taking her case to trial, does the expert really have any insight on how likely it is that the defendant’s guilty?  

It’s impossible to tell how persuasive the testimony is.  Let’s say the expert testifies that drug couriers are usually guilty.  The defense lawyer cross-examines him and gets him to admit that he doesn’t have specific knowledge of this particular drug courier, the drug couriers he’s familiar with usually plead guilty, and he hasn’t done any studies.  What exactly is the jury supposed to do with that?  Does the expert’s testimony get the prosecutor 10% of the way to “beyond a reasonable doubt”?  50%?  How should the jury balance that testimony with, e.g., testimony that the driver looked surprised when drugs were found in the car?

Each of these issues raise difficult questions:

• Is there even such a thing as evidence that is too persuasive?  Is there some intrinsic value to juries focusing on defendant-specific facts as opposed to an expert’s general life experience? Or is this intuition mere sentimentalism about due process?

• Is there even such a thing as evidence that isn’t persuasive enough?  Juries are supposed to be better at weighing evidence than judges.  If evidence is unpersuasive, shouldn’t we trust the jury to treat it as unpersuasive?  What are we afraid of?

• Balancing contradictory evidence and weighing inferences is something that juries do all the time.  Is there any empirical reason to believe that juries are better at weighing fact testimony (“she looked scared”) than expert testimony (“couriers are usually guilty”)?  Any philosophical reason to distinguish between those questions?

And even setting those issues aside … should we care about any of these issues in interpreting Rule 704(b)?  Rule 704(b) was enacted to prevent the unseemly spectacle of psychologists pretending to read someone’s mind.  Policy considerations about police officers profiling drug couriers have nothing to do with that. This raises a ubiquitous question in the law … When, if ever, should a court rely on policy concerns in interpreting a rule that aren’t the policy concerns that led to the enactment of the rule? 

On second thought, there is some relationship between concerns about “I’ve been a cop for years” evidence and concerns about psychologists testifying about what’s in a defendant’s head.  In both cases, the jury is being encouraged to discern the defendant’s mental state based on an expert rather than its own common sense.  Is that relationship close enough to justify using policy concerns about “I’ve been a cop for years” evidence to inform the proper interpretation of Rule 704(b)?

So who wins?

What’s the takeaway here?

If you want my opinion on how Diaz should be decided, or even on how the many sub-issues raised in this post should be decided, the answer is that I don’t know.  Sorry. 

I’m trying to make a different point:  Without much effort, we have transformed a seemingly simple case into a case packed with intractable questions.  What should this tell us?

1. Judges have lots of discretion.  Each of the sub-questions in Diaz could reasonably be decided in either direction, which means you could have a plausible judicial opinion going either way that says almost anything.  That’s not true in every Supreme Court case, but it’s true in many. 

2. It’s easy to be results-oriented if you want to be.  If you are the type of judge who consistently desires progressive or conservative outcomes, there’s a way to get there while still purporting to act like a principled, process-oriented judge.  Simply subdivide a case into a web of intractable methodological questions that themselves do not have a left/right valence, exercise discretion to decide them in the manner that will lead to the desired result, and presto—you’ve applied what looks like neutral principles, but you’ve still achieved your desired outcome.

3. You can also be non-results-oriented if you want to be.  The trick is to precommit to particular methodologies and stick to them.  For example, prior to reviewing the facts of a particular case, a judge could decide: “I will not consider the potential for arbitrariness at the margin as a basis for enlarging the scope of a rule” or “I will not consider policy arguments unless they are identical, or nearly identical, to the policy arguments that led to the enactment of the rule.” In the real world, it is unrealistic for judges to be perfectly consistent on these issues across cases, but they can at least try to set up methodological guardrails that constrain their subsequent exercise of discretion.

4. One of the promises of AI-assisted judging is to help judges consistently apply particular methodologies without engaging in motivated reasoning.  The AI could break a case apart as I have above, and then say, “in prior case X you applied the principle that a court should disregard a policy concern because it diverges from the policy concern that led to the rule’s enactment, here’s how the same principle should apply here.”  Hopefully, the AI would be braver about pointing out methodological inconsistency across cases than the judge’s law clerks.  Today’s judges would likely ignore such advice, but tomorrow’s judges might come to rely on it.

1. Not all amendments to the Rules of Evidence come from Congress. Sometimes, an Advisory Committee consisting of a mix of judges and lawyers makes recommendations to the Supreme Court to amend the Rules. If the Supreme Court accepts the recommendations, the Rules change. That’s what happened in 2011.

2. The government says that the old version of the rule plainly authorized Agent Flood’s testimony, but I don’t see it. The old version might have authorized testimony allowing a jury to infer mental state; that doesn’t mean it authorized testimony of the form “yeah, people like in Ms. Diaz’s position pretty much always know about the drugs.”