A brief history of the Confrontation Clause

And more fun with Claude.

Jun 26, 2024

Thanks to everyone who reached out to me following my prior post, either in the comments or via private email. I love to talk about AI (or almost anything law-related, really), so I appreciate hearing from you.

Also, in response to a couple of people who asked, I’m not a law professor, although I appreciate the compliment. I’m a practicing lawyer; this Substack is merely a hobby. Indeed, I’m always looking to take on interesting new cases, so if you have one, let me know.

Let’s turn to this week’s topic: the Confrontation Clause of the Sixth Amendment, which guarantees criminal defendants the right to cross-examine witnesses. The Supreme Court decided its most recent Confrontation Clause case, Smith v. Arizona, on June 21. Smith is the fourth in a line of cases addressing a defendant’s constitutional right to cross-examine scientific analysts, which explains this week’s artwork. Here is the Supreme Court’s holding: “When an expert conveys an absent analyst’s statements in support of the expert’s opinion, and the statements provide that support only if true, then the statements come into evidence for their truth.” The Court remanded for consideration of whether the expert’s statements were “testimonial.”

To the non-lawyers in the audience, these statements will seem like gibberish. For good reason: The Confrontation Clause regulates criminal trial procedure and hence does not garner much interest outside the legal profession. That is too bad, because the Confrontation Clause is a wonderful subject, allowing for deep jurisprudential debates without the shackles of controversial subject matter (guns, Trump) that may distort legal judgment.

First, Claude

You will not be surprised to learn that Claude is more insightful about the Confrontation Clause than any mortal.

First, I am now using Claude 3.5 Sonnet, which was released after I wrote my previous post. I can report that it not only produces better answers than Claude 3 Opus, but it’s at least 3 times faster. So, my statement in my prior post that Claude is at least 5,000 times faster than humans is now obsolete; it’s actually at least 15,000 times faster.

Before Smith came out, I asked Claude to predict how the case would be resolved, and Claude correctly predicted a reversal—but we already knew it could do that. In Smith, Justice Kagan wrote the majority opinion, and Justices Alito, Gorsuch, and Thomas wrote separately. After the opinion came out, I decided to check whether Claude could discern the author of each opinion. Thus, I inputted each opinion into Claude, while stripping out identifying information—for example, when I inputted the majority opinion, I deleted “JUSTICE KAGAN delivered the opinion of the Court,” as well as “post, at 8 (ALITO, J., concurring in judgment).” I also asked Claude to explain how it reached its conclusions.

Claude went 4 for 4. Here was Claude’s analysis of the majority opinion (going forward, to ensure compatibility with screen readers, I am going to use block quotes of Claude’s outputs rather than pictures):

Q: Here is the actual majority opinion in the Smith case. I’ve deleted the author of the majority opinion. Based on the style and content, and your knowledge of the Justices’ writing style and substantive views, who do you think the author was? Explain your reasoning.
A: Based on the style and content of this opinion, I believe the likely author is Justice Elena Kagan. Here’s my reasoning:
Writing style: The opinion has a clear, direct, and occasionally conversational tone that is characteristic of Justice Kagan’s writing. For example, phrases like “Where the parties disagree, of course, is...” and “If that explanation seems a bit abstract, then take this case as its almost-too-perfect illustration” are typical of her accessible style.
Use of examples and hypotheticals: The opinion frequently uses concrete examples and hypotheticals to illustrate points, which is a common technique in Kagan’s opinions.
Emphasis on practical consequences: The opinion considers the practical implications of its ruling, which aligns with Kagan’s pragmatic approach.
Confrontation Clause jurisprudence: The opinion takes a strong stance on protecting defendants’ Confrontation Clause rights, which is consistent with Kagan’s previous opinions and dissents in this area (e.g., her dissent in Williams v. Illinois).
Reference to previous dissents: The opinion favorably cites positions from dissents in previous Confrontation Clause cases, including the Williams dissent that Kagan herself authored.
Engagement with textualism and originalism: While not primarily focused on these methodologies, the opinion does engage with text and history in a way that suggests an author trying to appeal to more conservative colleagues.
Critical treatment of evidentiary rules vs. constitutional rights: The opinion’s skepticism of using evidentiary rules to define the scope of constitutional rights aligns with positions Kagan has taken in other contexts.
Focus on jury perspective: The opinion emphasizes how a jury would understand and use the evidence, which is a recurring theme in Kagan’s Confrontation Clause jurisprudence.
While it’s possible another justice could have authored this opinion, the combination of writing style, doctrinal approach, and consistency with her previous positions makes Justice Kagan the most likely author in my assessment.

Alito opinion:

Q. Here is a separate concurrence in the judgment in the Smith case. I’ve deleted the author of this opinion. Based on the style and content, and your knowledge of the Justices’ writing style and substantive views, who do you think the author was? Explain your reasoning.
A. Based on the style and content of this concurring opinion, I believe the likely author is Justice Samuel Alito. Here’s my reasoning:
Defense of traditional evidence rules: The opinion strongly defends established evidence law and criticizes the majority for “inflict[ing] a needless, unwarranted, and crippling wound on modern evidence law.” This aligns with Alito’s tendency to defend traditional legal practices.
Detailed historical analysis: The opinion provides an extensive historical review of expert testimony practices, which is characteristic of Alito’s writing style. He often includes thorough historical context in his opinions.
Pragmatic concerns: The author emphasizes practical considerations and the potential negative consequences of the majority’s ruling, which is consistent with Alito’s pragmatic approach to judicial decision-making.
Citation of precedent: The opinion extensively cites and relies on previous Court decisions, a common feature in Alito’s writings.
Critique of the majority’s reasoning: The author directly challenges the majority’s logic, stating “This is just plain wrong,” which aligns with Alito’s often forceful disagreements with majority opinions he opposes.
Focus on jury instructions: The opinion places significant emphasis on the effectiveness of jury instructions, which is a topic Alito has highlighted in other opinions.
Concern about undermining Federal Rules of Evidence: Alito has previously shown concern about decisions that he believes undermine established legal frameworks, like the Federal Rules of Evidence.
Writing style: The opinion is clear, direct, and occasionally employs rhetorical questions, which are all characteristics of Alito’s writing.
Alternative resolution: The opinion offers an alternative way to resolve the case that is more limited in scope than the majority’s approach, which is consistent with Alito’s tendency to prefer narrower rulings when possible.
Conservative jurisprudence: While agreeing with the judgment, the opinion takes a more conservative approach to changing established legal practices, which aligns with Alito’s judicial philosophy.
While it’s possible another justice could have authored this opinion, the combination of writing style, historical analysis, pragmatic concerns, and jurisprudential approach makes Justice Alito the most likely author in my assessment.

Gorsuch opinion:

Q. Same question for this concurrence.
A. Based on the style and content of this concurring opinion, I believe the likely author is Justice Neil Gorsuch. Here’s my reasoning:
Textualist approach: The opinion focuses heavily on the text of the Sixth Amendment and explores possible interpretations based on the original meaning of terms like “witness” and “testimony.” This aligns with Gorsuch’s strong textualist approach.
Historical analysis: The author cites founding-era dictionaries and considers the historical understanding of key terms, which is characteristic of Gorsuch’s originalist methodology.
Skepticism of judge-created tests: The opinion expresses concern about the “primary purpose” test being “a limitation of our own creation on the confrontation right,” which reflects Gorsuch’s skepticism of tests not grounded in the Constitution’s text.
Concern for practical application: The author raises questions about how the “primary purpose” test would be applied in practice, which is consistent with Gorsuch’s tendency to consider the real-world implications of legal rules.
Cautious approach: The opinion advocates for not addressing issues unnecessary to the case at hand, which aligns with Gorsuch’s judicial restraint in some contexts.
Writing style: The opinion is clear, direct, and uses rhetorical questions to make points, which is characteristic of Gorsuch’s writing.
Citation of lesser-known historical sources: The reference to Chief Justice Marshall’s statement in United States v. Burr is the kind of deep historical dive that Gorsuch often includes in his opinions.
Focus on individual rights: The opinion shows a strong concern for protecting the rights of the accused, which is consistent with Gorsuch’s libertarian leanings in criminal justice matters.
Agreeing with the majority result but not all reasoning: This approach is common for Gorsuch, who often writes separate opinions to express his own legal reasoning even when he agrees with the outcome.
Emphasis on potential confusion: The author’s concern about the confusion that might result from the majority’s approach is typical of Gorsuch’s pragmatic considerations in legal interpretation.
While it’s possible another justice could have authored this opinion, the combination of textualist approach, historical analysis, skepticism of judge-made tests, and concern for practical application makes Justice Gorsuch the most likely author in my assessment.

Thomas opinion:

Q. And for this one.
A. Based on the style and content of this concurring opinion, I believe the likely author is Justice Clarence Thomas. Here’s my reasoning:
Consistent view on testimonial statements: The author explicitly states that they “continue to adhere to my view” about what qualifies as a testimonial statement under the Confrontation Clause. This view, focusing on “formalized testimonial materials,” has been consistently expressed by Justice Thomas in numerous opinions.
Extensive self-citation: The opinion cites many previous concurrences and separate opinions by Justice Thomas on this issue, spanning from 1992 to 2015. This level of self-citation is characteristic of Thomas’s writing.
Focus on formality and solemnity: The author emphasizes that testimonial statements should be those with sufficient “formality and solemnity,” which is a hallmark of Thomas’s approach to Confrontation Clause issues.
Historical analysis: The opinion references the historical context of the Confrontation Clause, particularly practices during the reign of Queen Mary, which aligns with Thomas’s originalist approach to constitutional interpretation.
Rejection of the “primary purpose” test: The author explicitly disagrees with the majority’s suggestion to use a “primary purpose” test, which is consistent with Thomas’s longstanding position on this issue.
Writing style: The opinion is clear, direct, and relies heavily on precedent and historical context, which is characteristic of Thomas’s judicial writing.
Specific legal theory: The focus on “formalized testimonial materials, such as affidavits, depositions, prior testimony, or confessions” as the key to Confrontation Clause analysis is a theory that Thomas has developed and advocated for over many years.
Agreement with result but not reasoning: Thomas often agrees with the Court’s judgments while providing his own distinct legal reasoning, which is the case here.
Emphasis on text and original meaning: The opinion starts by quoting the text of the Sixth Amendment and focuses on the original understanding of terms like “witnesses” and “testimony,” which is consistent with Thomas’s textualist and originalist approach.
Suggestion for lower court: The author suggests that the lower court should consider on remand whether the statements have the “requisite formality and solemnity to qualify as testimonial,” which reflects Thomas’s consistent approach to this issue.
Given these factors, particularly the consistent articulation of a specific legal theory on the Confrontation Clause that Thomas has developed over decades, I am confident that Justice Clarence Thomas is the author of this concurring opinion.

For those curious, I tried the same exercise with United States v. Rahimi. Claude accurately concluded that the Chief Justice wrote the majority opinion and correctly identified the authors of the separate opinions by Justices Gorsuch, Barrett, Kavanaugh, Jackson, and Thomas. It offered detailed explanations for each Justice, which were uniformly perfect, as far as I can tell. With respect to the concurrence by Justice Sotomayor, which was joined by Justice Kagan, Claude thought the “most likely” author was Justice Kagan, but stated that “it’s possible this could be authored by Justice Sotomayor.” I guess there’s always room for improvement! I won’t copy-paste all of Claude’s answers, but here’s a PDF if you’re interested.

Rahimi Claude Authorship Predictions

69.1KB ∙ PDF file

Download

It would be fun to input all of the opinions from this Term into Claude and see how many authors Claude could identify, but I do not have the patience. Law professors, this is why God gave you RAs.

This is impressive, but not that impressive. A sophisticated human law clerk could probably have done the same thing, although not in the same level of detail and not within 15 seconds. To understand where Claude was really impressive, you have to understand a little bit about the case.

And so, my essay about the case will come next. You will have to wait until the end of the post for more Claude. Of course, Claude can write an essay about Smith just as well as I can (probably better), but I do not want to cede complete control of my own newsletter.

Justice is following the rules

Here is the Confrontation Clause in all its glory: “In all criminal prosecutions, the accused shall enjoy the right . . . to be confronted with the witnesses against him.”

Most Confrontation Clause cases have to do with the admissibility of hearsay. As Bar/Bri books are fond of saying, hearsay is an out-of-court statement being offered to prove the truth of the matter asserted. Suppose a police officer interviews a bystander who says she witnessed the defendant committing a crime. The bystander doesn’t show up to trial, so the police officer tells the jury what the bystander said, and the prosecutor uses the bystander’s statement as evidence of the defendant’s guilt. That’s classic hearsay. It raises a Confrontation Clause issue because the defendant can’t confront—and cross-examine—the bystander.

In Ohio v. Roberts, 448 U.S. 56 (1980), the Supreme Court held that hearsay statements were admissible under the Confrontation Clause if they had “adequate indicia of reliability.” Essentially, if the trial judge concluded that the statement seemed trustworthy enough, the statement could be used against the defendant, even if the defendant never had the chance to cross-examine the statement’s utterer.

Roberts was decided by a 6-3 vote, with the three most progressive Justices (Brennan, Marshall, and Stevens) dissenting. And you can see why a certain kind of judicial conservative would support Roberts’ test. Conservatives are supposed to be tough on crime. If the testimony seems reliable enough, why let the defendant, who is probably guilty, get off on a technicality?

Justice Scalia joined the Supreme Court in 1986, and he didn’t see things that way. He proved to be an enthusiastic proponent of vigorous enforcement of the Confrontation Clause—indeed, the most enthusiastic member of the Court. And, to put it mildly, he didn’t agree with Roberts.

Why? Well, from Justice Scalia’s perspective, the judge’s role was not to keep bad guys behind bars. It was to follow the Constitution as written. If the Constitution says the defendant gets to confront witnesses, then the defendant gets to confront witnesses, irrespective of whether the judge thinks the testimony is reliable or not.

Sometimes it’s hard to follow the Constitution as written. Determining what “due process” or “equal protection” means isn’t easy. But determining what the Confrontation Clause means, at least in its core applications, is easy. The defendant gets to confront witnesses. No exceptions! Judges do not have the discretion to decide, on a case-by-case basis, whether to follow the Constitution.

Maybe, in a particular case, excluding seemingly reliable hearsay would result in a guilty defendant going free. But from Justice Scalia’s perspective, a judge’s job was not to do justice in a cosmic sense, but instead to follow the rules. Or a more felicitous way of putting it is: in a society governed by the rule of law, following the rules is justice. And the Sixth Amendment contains a clear rule: the defendant gets to cross-examine the witness.

Roberts contradicted Justice Scalia’s world view in other respects. For instance, the Roberts test required judges to apply an indeterminate multi-factor test to determine whether a statement was “reliable,” which was anathema to Justice Scalia. Also, the Constitution assigns to juries the task of determining whether statements are reliable. The Founders didn’t trust judges, and Justice Scalia didn’t trust them either. Determining whether a statement is “reliable” is precisely the type of determination that judges aren’t competent to make.

But ultimately, these rationales are icing on the cake. At the end of the day, Justice Scalia supported vigorous application of the Confrontation Clause for the same reason that people climb Mount Everest: because it is there. The Sixth Amendment says that the defendant gets to cross-examine witnesses and so the defendant gets to cross-examine witnesses, regardless of whether the judge thinks it’s really necessary.

In Crawford v. Washington, 541 U.S. 36 (2004), the Supreme Court, per Justice Scalia, overruled Roberts. The facts of Crawford were stark. The defendant, Michael Crawford, stabbed a man. A police officer interrogated his wife, Sylvia Crawford, and she said that Michael committed the stabbing. Michael was charged with attempted murder and claimed self-defense. At trial, Sylvia refused to testify, citing the marital privilege. So the government attempted to introduce into evidence Sylvia’s tape-recorded statements to the police officer. The trial judge allowed the statements into evidence, finding them sufficiently reliable under Roberts, and Michael was convicted.

The Supreme Court reversed Michael’s conviction, holding that Michael’s Confrontation Clause right was violated. Regardless of whether Sylvia’s testimony was reliable, Michael had the constitutional right to cross-examine her. He couldn’t do that, so her testimony should have been excluded.

There’s an important detail I’ve left out. The Confrontation Clause guarantees a defendant the right “to be confronted with the witnesses against him.” What does “witnesses against him” mean? Well, in 1603, Walter Raleigh was tried for treason. He was convicted and sentenced to death based on his accomplice’s statements to the Privy Council, even though he couldn’t cross-examine the accomplice. The Crawford Court explained that Raleigh’s trial “has long been thought a paradigmatic confrontation violation.” From this and other historical evidence, the Court concluded that a person is a “witness” under the Confrontation Clause if he is providing a testimonial statement. And, without defining with precision what “testimonial” means, the Court held that statements in response to a police interrogation are a classic form of “testimony.”

Some people love originalism and some people hate it. But, love it or hate it, you should understand it. Reading Crawford is a good place to start. Crawford is a masterpiece. It’s a magnificent blend of history and pure logic. It’s one of the greatest originalist opinions of all time.

Interpolation

Next up to bat was Davis v. Washington, 547 U.S. 813 (2006), in which the Court decided what “testimonial” means. The Court considered two cases in which trial courts admitted hearsay statements and the defendants challenged their convictions under the Confrontation Clause. Both cases involved victims of domestic violence who refused to testify against their abusers, a sadly common fact pattern in Confrontation Clause cases. In one, the statement was a victim’s 911 call, describing an ongoing emergency. In the other, the statement was a victim’s response to police questioning. The questioning took place at the scene of the crime, but there was no ongoing emergency; the crime had already occurred.

The Court, per Justice Scalia, announced the following test: “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 assistance 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, applying that test, the statement during the 911 call was nontestimonial, while the statement during the police questioning was testimonial.

Justice Thomas dissented, arguing that both statements were nontestimonial. In his view, “testimony” historically required a degree of formality, and a victim’s statements to a cop in her living room weren’t sufficiently formal to qualify as “testimony.” The majority responded: “Restricting the Confrontation Clause to the precise forms against which it was originally directed is a recipe for its extinction.”

In my view, the dueling opinions are both reasonable interpretations of the historical evidence. Davis is an excellent illustration of the difficulty of interpolating based on a limited set of historical data points and generating a rule that will resolve intermediate cases. Some people think that cases like Davis prove that originalism is useless. Maybe?

Scissors

Then came Melendez-Diaz v. Massachusetts, 557 U.S. 305 (2009), which is one of my favorite cases of all time.

Some of you may be familiar with the concept of a “scissor statement”—that is, a statement that seems obviously true to half the people who hear it and obviously false to the other half. Melendez-Diaz is a scissor case. To almost everyone who hears the facts, the case is blindingly obvious—except some people think the criminal defendant obviously wins and some people think the criminal defendant obviously loses.

If you want to know whether you’re a formalist or a functionalist at heart, read Melendez-Diaz. If you think the defendant should win, you are a formalist. If you think he should lose, you are a functionalist. It’s a perfect litmus test. Almost no one is in the middle.

OK, here are the facts. Mr. Melendez-Diaz was found with a baggie of white powder. The powder was sent to the lab. A technician tested it and found that it was cocaine. He prepared a “certificate of analysis,” sworn to before a notary public, stating that it was cocaine. At trial, the prosecutor attempted to introduce the certificate of analysis into evidence. The defendant objected, arguing that he had a right to cross-examine the lab technician under the Confrontation Clause.

To formalists, it will seem perfectly obvious that the lab technician must testify live. The “certificate of analysis” is obviously testimonial. It’s a formal, sworn declaration. It contains the key evidence necessary to convict the defendant—that he was found with cocaine. What’s the contrary argument? That the guy providing the crucial evidence isn’t a “witness” because he’s a lab technician as opposed to a police officer? Seriously? Or are we going to let this evidence in because we think it’s reliable and cross-examination isn’t needed? Isn’t that exactly what Crawford repudiated?

To functionalists, it will seem ridiculous that the lab technician must testify live. It’s such a waste of time. He’s just going to repeat what’s on the form. “Yes, I did the test, and I found that it was cocaine.” What’s cross-examination going to accomplish? Even if the guy was lying about doing the test and just made everything up, he’s not going to get mad and admit it on the stand like Jack Nicholson in A Few Good Men. At best, the defense lawyer is going to ask some confusing question like “isn’t your machine unreliable,” the lab technician will be like “durr, I’m just doing my job,” and the jury won’t understand anything. Meanwhile, if the lab technician can’t show up to trial for whatever reason, a drug dealer will randomly go free for no reason whatsoever. As for the formalists’ insistence that they are just following the Constitution wherever it leads, do you really think there is no difference whatsoever between Walter Raleigh’s treason conviction and the failure to call a lab technician to confirm he filled out a form saying that the substance he tested was cocaine?

The formalists won. The Supreme Court, per Justice Scalia, ruled in the defendant’s favor by a 5-4 vote. There’s a follow-up case called Bullcoming v. New Mexico, 564 U.S. 647 (2011), another case involving a lab test, in which the prosecutor called a guy who worked at the lab but not the actual guy who did the lab test and filled out the form. Not good enough, said the Supreme Court in another 5-4 decision: the actual guy who filled out the form must show up at trial.

Trust the experts

Five being more than four, Melendez-Diaz and Bullcoming became the law of the land. But storm clouds were looming. Something weird was going to happen. Why?

Because the rule of decision in these cases was highly formal and didn’t have much to do with fairness. Also, there were four Justices who really hated these decisions and wanted to limit their impact. Given that the decisions themselves didn’t draw a line based on fairness, the dissenters might be willing to distinguish them based on a rationale that really really didn’t have anything to do with fairness.

Enter Williams v. Illinois, 567 U.S. 50 (2012). Mr. Williams was charged with rape. Vaginal swabs were taken from the victim and sent to a lab called Cellmark. Cellmark created a DNA profile from semen found in the vaginal swabs. It then sent the DNA profile to a forensic specialist, Lambatos. Lambatos conducted a search and concluded that the Cellmark profile matched the defendant’s DNA profile, which was already in the state’s database due to an unrelated arrest.

Lambatos testified at trial as the state’s expert witness. She explained that she received the Cellmark profile, which (according to Cellmark) was derived from semen found in the vaginal swabs, and matched it to the defendant’s DNA profile. Therefore, according to Lambatos’s expert opinion, the defendant’s DNA was found on the victim.

Williams was convicted. In the Supreme Court, Williams challenged his conviction under the Confrontation Clause.

The Confrontation Clause problem is obvious under Melendez-Diaz and Bullcoming. The crucial piece of evidence supporting Williams’ guilt is that the DNA profile was, in fact, created from the semen found in the vaginal swabs. And Lambatos didn’t know that. She merely testified that Cellmark told her that. That’s hearsay in exactly the same way that the lab technician’s statements in Melendez-Diaz and Bullcoming were hearsay. Williams never got the chance to cross-examine the Cellmark lab tech to check whether he was lying.

But Williams lost. A four-Justice plurality—consisting of the four dissenters in Melendez-Diaz and Bullcoming—found two creative ways of distinguishing Melendez-Diaz and Bullcoming.

The first one is hard to explain, but I will give it a shot.

According to the plurality, Lambatos wasn’t actually testifying as to the truth of the proposition that Cellmark did the lab test properly. Instead, Lambatos was merely testifying as to the basis of her opinion that Williams’ DNA was found on the victim. Of course, if Cellmark didn’t do the lab test properly, then the premise undergirding that opinion falls apart. But even setting aside Cellmark’s hearsay statement, there was circumstantial evidence that Cellmark did the lab test properly. There was evidence that the vaginal swab was shipped to Cellmark, and if Cellmark did the test improperly, what are the odds that the resultant DNA profile would just happen to be a match to Williams?

Even for an experienced lawyer, it is hard to wrap one’s head around this reasoning. But I will promise you this: if you’re an ultra-formalist, it’s correct. There are no logical errors. Strictly speaking, if the evidence wasn’t admitted for the truth of the matter asserted, there’s no Confrontation Clause problem.

The problem is that no fact-finder is ever going to construe this evidence in this insane way. The fact-finder is going to hear Lambatos repeat Cellmark’s statement, and is going to believe Cellmark’s statement. It’s a kind of comic parody of formalism to think that the fact-finder—a judge in Williams, but typically a jury—will understand these incomprehensible distinctions. But well, live by the formalist sword, die by the formalist sword, right? It’s poetic justice: the functionalist dissenters from Melendez-Diaz and Bullcoming had out-formalized the formalists in order to restore their preferred rule of functionalism.

In fairness to the plurality, the Rules of Evidence permit expert witnesses to rely on hearsay, and this happens all the time. An expert will say “well, I talked to the CEO, and he said that if the patent wasn’t infringed then the company would have sold 100 million units at $10 dollars per unit, so using my financial wizardry I will calculate a damages figure of one billion dollars!” When the expert tells the jury what the CEO said, juries aren’t supposed to treat the CEO’s hearsay statements as evidence of the truth of the matter asserted. So if you say “no fact-finder is ever going to construe this evidence in this insane way,” you are saying that fact-finders always misunderstand expert testimony that relies on hearsay.

Yeah. They probably do. Maybe this is OK when you’re talking about patent trials, but it’s not OK when you’re talking about the Constitution.

The plurality’s second argument was that Cellmark’s statement wasn’t “testimonial” because, at the time of the test, Williams wasn’t a suspect yet. So the lab technician couldn’t have been trying to incriminate anyone in particular.

This is the sort of argument that sounds good when you say it fast. But a moment’s thought reveals how bizarre and random it is. The test was clearly done for purposes of a criminal case. Who cares whether Williams was a suspect or not? What does this have anything whatsoever to do with whether or not Williams should be able to cross-examine the lab technician?

But from the plurality’s perspective, the Melendez-Diaz and Bullcoming decisions were equally bizarre and random. If the Melendez-Diaz and Bullcoming majorities are going to draw a constitutional line that has nothing to do with fairness, can’t they do it too? This is a nice illustration of how constitutional law gets weird—Justices who think a prior decision doesn’t make sense distinguish it on a ground that also doesn’t make sense, yielding a rule that no one likes.

The plurality opinion was joined by only four Justices. But Justice Thomas thought the Cellmark report didn’t satisfy his preferred “formality” test for testimonial statements. Because four plus one equals five, Williams lost, but without any majority opinion.

Finally!

That finally brings us to Smith v. Arizona—the case the Supreme Court decided last week.

The facts were similar to Williams, but more favorable to the defense. It was kind of a ridiculous fact pattern, actually. To make a long story short, Scientist #1, Rast, did lab tests, took notes, and prepared a report concluding that the defendant was possessing drugs. Scientist #2, Longoni, who had no personal knowledge of the lab tests, essentially read Rast’s report and notes to the jury.

The Arizona courts concluded that there was no Confrontation Clause violation because when Longoni informed the jury of Rast’s statements, those statements weren’t being introduced for the truth of the matter asserted, but instead merely to convey why Longoni believed what he believed. Of course, Longoni’s beliefs were irrelevant unless the underlying assumptions—i.e., Rast’s statements—were accurate, and there was really nothing in the record substantiating Rast’s statements. But Arizona argued this wasn’t a Confrontation Clause problem, but instead a problem of insufficient evidence or perhaps the defendant’s failure to request a jury instruction that the jury would have had no chance of understanding.

This hyper-technical distinction was too much for any Justice to stomach. Even the Chief Justice and Justice Alito, who dissented in Melendez-Diaz and Bullcoming, have their limits. The majority opinion held that when an expert relies on testimonial hearsay, the hearsay is being admitted for its truth, and the Constitution is violated. As noted above, this holding is inconsistent with the basic premise of expert testimony, which is that when experts rely on hearsay, juries shouldn’t believe the hearsay just because the expert repeated it. Essentially, the majority opinion held: “Juries are never capable of correctly interpreting expert testimony that relies on hearsay, but it’s only a constitutional problem in criminal cases.” Harsh but probably true. The Chief Justice and Justice Alito thought that the theory behind expert testimony still makes sense sometimes, just not this time.

Arizona had one other theory—that even if Rast’s report was “testimonial,” Rast’s informal notes weren’t “testimonial” because maybe they weren’t specifically created with the trial in mind. The Court didn’t resolve this issue, instead sending it back to the lower court. The argument seems profoundly weird, though. Why does it make sense that the expert can’t rely on the formal report, but can rely on the informal notes, which are rougher and presumably less reliable? OK, Crawford rejects a reliability test, but why adopt a kind of anti-reliability test, in which less reliable hearsay is admitted but more reliable hearsay is excluded?

Justice Scalia might say: “The Constitution doesn’t solve all problems. The Sixth Amendment addresses one problem: cross-examination of ‘testimony’ intended for use at a trial. It doesn’t address the problem of nontestimonial hearsay. So what? It doesn’t address climate change, either. If you want to promote justice on a topic the Constitution doesn’t address, run for office.” If you believe that, you really are a formalist.

Back to Claude

That wraps up my analysis of Smith. Can Claude do any better?

Of course it can. Above I asked: why does it make sense that the expert’s notes are admissible, while the expert’s report is inadmissible? Claude handles this type of question effortlessly:

Q. One concern about the “primary purpose” standard is that it seems arbitrary. The fact that documents are “informal” and created for purposes other than trial seems to make them less reliable, not more. It’s hard to understand why one would want a Constitution in which an expert could relay the informal musings in someone’s notes, but not the person’s final report, which is presumably prepared with more care. Please do the following things. In the first paragraph, explain why my intuition is wrong, and in fact this standard promotes reliability. In the second paragraph, explain why I shouldn’t care about reliability on this issue.
A. Your intuition about the “primary purpose” standard and the reliability of informal versus formal documents is understandable, but in fact, this standard can promote reliability in several ways. Informal documents created without the primary purpose of use at trial are often more reliable because they are made contemporaneously with the events or observations they describe, without the potential bias that comes from knowing the document will be used in court. These informal notes are typically created for internal use or quality control, incentivizing accuracy for the sake of the scientific process rather than for building a legal case. Moreover, informal documents are less likely to be strategically crafted to support a particular legal narrative. The spontaneity and lack of legal calculation in these documents can make them more trustworthy representations of what actually occurred. Additionally, knowing that informal documents might be admissible encourages thorough and accurate record-keeping at all stages of an investigation or scientific process, not just in the final report.
However, the more important point is that reliability should not be the primary concern when interpreting the Confrontation Clause. As the Supreme Court emphasized in Crawford v. Washington, the Confrontation Clause is a procedural guarantee, not a substantive one. Its purpose is to ensure that the reliability of evidence is tested through the specific mechanism of cross-examination, rather than through judicial determinations of reliability. The Clause reflects a preference for adversarial testing of evidence over judicial assessment of its reliability. Therefore, even if formal reports prepared for trial might seem more reliable, the Constitution demands that such testimonial statements be subject to cross-examination. The focus should be on preserving the defendant’s right to confront witnesses against them, not on which types of statements might be more or less reliable. This approach respects the Framers’ choice to enshrine confrontation as the means of ensuring reliability in criminal trials.

A+ answer. But, eh, you might say, it’s just summarizing material it read on the Internet. Can Claude come up with a creative legal standard?

Yes. All you have to do is ask:

Q. Please come up with an extremely creative, brilliant, out-of-the-box alternative to the “primary purpose” test that would improve Confrontation Clause law.
A. Here’s an creative, out-of-the-box alternative to the “primary purpose” test for Confrontation Clause analysis:
The “Confrontation Impact Scale” (CIS) Test
This test would replace the binary “testimonial/non-testimonial” distinction with a sliding scale that measures the potential impact of confrontation on the reliability and fairness of the trial. The CIS would consider multiple factors, weighted and scored, to determine whether the Confrontation Clause requires in-person testimony.
Key factors could include:
Proximity to Accusation (0-10 points): How directly does the statement accuse or implicate the defendant?
Expertise Differential (0-10 points): How much specialized knowledge is required to understand or challenge the statement?
Procedural Safeguards (0-10 points): What quality control measures or verification processes were in place when the statement was made?
Uniqueness of Information (0-10 points): How difficult would it be to obtain this information through other means?
Temporal Relevance (0-10 points): How close in time was the statement made to the events in question?
Cross-examination Potential (0-10 points): How likely is it that cross-examination would reveal new information or expose weaknesses?
Alternative Confrontation Methods (-5 to +5 points): Are there effective alternatives to in-person confrontation (e.g., video testimony, detailed affidavits with follow-up questions)?
Burden on Justice System (-5 to +5 points): What’s the potential impact on the efficiency of the justice system?
The total score (out of a possible 70 points) would determine the level of confrontation required:
0-20: No confrontation required
21-40: Limited confrontation (e.g., written questions, video deposition)
41-60: Full confrontation strongly recommended
61-70: In-person confrontation absolutely required
This approach would:
Provide more nuanced outcomes than the current binary system
Allow courts to balance various interests more explicitly
Adapt to technological changes in how information is created and shared
Encourage more detailed analysis of each piece of evidence
Potentially reduce arbitrary distinctions between similar types of evidence
While complex, this system could be refined over time through judicial application, potentially leading to more consistent and fair outcomes in Confrontation Clause cases.

This is an exceptionally thoughtful proposal that proposes a sliding scale both for whether a statement is testimonial or not, and for what type of cross-examination is required.

Can Claude do it again? It can.

Q. Now, do it again, but a standard completely different from what you just suggested.
A. Here’s a completely different, creative alternative to the “primary purpose” test for Confrontation Clause analysis:
The “Defendant-Centric Confrontation” (DCC) Test
This approach shifts the focus entirely to the defendant’s perspective and needs, rather than the nature of the statement or the intent of its creator. The core idea is that the right to confrontation should be most robustly protected when it matters most to the defendant’s ability to mount an effective defense.
Key elements of the DCC test:
Defense Theory Relevance: The defendant must articulate how confrontation of the statement’s author is relevant to their specific defense theory. The more central the statement is to challenging the prosecution’s case or supporting the defense’s narrative, the stronger the confrontation right becomes.
Impossibility of Alternative Discovery: The defendant must demonstrate that the information they seek to gain through confrontation cannot be reasonably obtained through other means (e.g., expert witnesses, documentary evidence, other testimony).
Confrontation Simulation: Before deciding on full confrontation rights, courts would allow a “confrontation simulation” where defense counsel can submit written questions to the statement’s author. If this process reveals significant discrepancies or areas of concern, it strengthens the case for full confrontation rights.
Prejudice Potential: Courts would assess the potential prejudice to the defendant if confrontation is denied. This includes considering how likely jurors are to give undue weight to the unconfronted statement.
Confrontation Readiness: The defendant must demonstrate “confrontation readiness” by providing a detailed plan of their proposed cross-examination, showing that it would be substantive and not merely procedural.
Dynamic Reassessment: The right to confrontation would be dynamically reassessed as the trial progresses. A statement that initially seemed unimportant might become crucial as the defense strategy evolves.
Implementation:
Pre-trial: Defendants would file a “Confrontation Necessity Brief” outlining their need for confrontation based on the above factors.
During trial: Judges would have the flexibility to grant confrontation rights as the need becomes apparent, potentially pausing proceedings to secure a witness’s presence.
Post-trial: Appellate courts would review confrontation decisions based on how the lack of confrontation actually impacted the trial outcome, not just on the nature of the statement itself.
This approach would:
Tailor confrontation rights to each case’s specific needs
Encourage more strategic and thoughtful use of the confrontation right
Potentially reduce unnecessary testimony while preserving the right where it matters most
Shift the focus from formalistic categories to practical impact on the defense
Allow for more flexible trial management while still protecting core Sixth Amendment rights
While this system would require significant changes in trial procedure, it could lead to a more meaningful application of the Confrontation Clause that truly serves its purpose of allowing defendants to challenge the case against them.

Another carefully-thought out proposal that takes a completely different view from current law, focusing on the defendant’s needs rather than the statement’s nature.

One more?

Q. One more. Last time.
A. Here’s a third, entirely different approach to replace the “primary purpose” test for Confrontation Clause analysis:
The “Confrontation Lottery” System
This unconventional approach introduces an element of randomness to Confrontation Clause decisions, aiming to balance fairness, efficiency, and unpredictability in criminal trials.
Key elements:
Confrontation Pool: At the start of each trial, all potential pieces of testimonial evidence are placed into a “Confrontation Pool.”
Random Selection: A set number of items (based on the case’s complexity and total evidence volume) are randomly selected from this pool for mandatory in-person confrontation.
Challenge Tokens: Both prosecution and defense receive a limited number of “Challenge Tokens” they can use to either:
Force confrontation for an unselected item
Waive confrontation for a selected item (subject to opponent’s veto)
Blind Selection: The random selection process occurs before either side knows which specific pieces of evidence are chosen, encouraging thorough preparation for all potential confrontations.
Tiered Evidence Classification: Evidence is pre-classified into tiers based on its potential impact:
Tier 1 (High Impact): Higher chance of selection, more Challenge Tokens required to alter status
Tier 2 (Medium Impact): Standard selection chance
Tier 3 (Low Impact): Lower chance of selection, fewer Challenge Tokens needed to alter status
Dynamic Rebalancing: As the trial progresses, the judge can add or remove items from the Confrontation Pool based on evolving case dynamics.
Confrontation Alternatives: For unselected items, alternative forms of scrutiny are available (e.g., written interrogatories, video depositions) to ensure some level of examination.
Implementation:
Pre-trial: Parties submit and classify potential evidence, with judicial review of classifications.
Trial Start: The lottery drawing occurs, setting the initial confrontation landscape.
During Trial: Parties strategically use Challenge Tokens as the case unfolds.
Post-trial: Appeals can address the fairness of the overall process, not individual confrontation decisions.
Benefits:
Encourages thorough preparation for all potential confrontations.
Reduces strategic manipulation of evidence presentation.
Balances the need for confrontation with court efficiency.
Provides a degree of confrontation rights even for unselected evidence.
Forces both sides to prioritize their most crucial confrontations.
Introduces an element of unpredictability that may encourage more plea bargains.
Potentially reduces bias in confrontation decisions.
Challenges:
Requires significant changes to trial procedures.
May face constitutional challenges.
Could lead to seemingly arbitrary outcomes in some cases.
Might be perceived as treating serious matters too lightly.
This system radically reimagines how we approach the Confrontation Clause, turning it into a more dynamic, game-like process. While highly unconventional, it aims to balance various competing interests in a novel way, potentially leading to more efficient trials while still preserving core confrontation rights.

I’m speechless.

To those of you who think that all AI can do is summarize things, and that it can’t measure up to the ineffable wonder of human judgment, well, I prefer to believe my lying eyes.

Adam's Legal Newsletter

Discussion about this post