The book, given to Rothstein years ago by an old friend, was a prescient gift: In January, Rothstein received the John Henry Wigmore Lifetime Achievement Award from the Association of American Law Schools (AALS) Section on Evidence. Evidence, Rothstein explains, is the study of “how we prove things and what we use to prove them” — in other words, the use of testimony, documents, photographs and other information to establish facts in court. “An understanding of what is good evidence and what is bad evidence could help people generally in the world — not just courts — to sort fact from fiction in this present critical time when the two are increasingly getting confused,” he says.

In addition to teaching courses in evidence and torts at Georgetown Law, where he has been on the faculty since 1971, Rothstein has advised both houses of Congress on the Federal Rules of Evidence and is credited with writing the first book on the Rules, which went into effect in 1975. He has also authored a leading evidence casebook and more than 100 articles – some of which have inspired robust debate among evidence-law scholars.

Rothstein spoke recently about his current projects, the technologies shaping the future of evidence law and the link between legal scholarship and practice.

You’re well known for your work on character evidence, or evidence related to a person’s personality or general character, which is typically inadmissible in criminal cases. Are you doing anything in that realm at the moment?

I’m working on a forthcoming North Carolina Law Review article with my former student RJ Coleman, now an adjunct professor at the Law Center, that examines whether, when and how past acts that exhibit racism, so-called “racist character evidence,” may be admissible in hate crimes cases. This type of evidence can be vital to doing justice in these cases, but courts often struggle to admit it — and when they do let it in, they don’t have a solid legal explanation as to why. In our paper, we look at seven “gateways,” or paths to admissibility, for this type of evidence. For example, we argue that evidence related to a specific propensity — that is, the tendency to do something very specific, such as commit a particular violent act — is distinct from “character” and therefore not subject to the Federal Rules of Evidence’s ban on character evidence. Our hope is to help guide courts and attorneys involved in hate-crime cases and related proceedings.

I’m also working with Edward Imwinkelried, professor of law emeritus at the University of California at Davis School of Law, on a forthcoming Criminal Law Bulletin article. In the Federal Rules of Evidence, “character” is described in two different ways. Our paper asks: Is it fair to assume that the drafters of the rules meant two different things by these different descriptions? And how does this play out in light of textualism, the dominant judicial philosophy when it comes to statutory interpretation today?

Last month, you presented a paper at the University of Michigan Journal of Law Reform’s 2024 Symposium on “Crawford at 20: Reforming the Confrontation Clause.” What is the Confrontation Clause, and what does your article look at?

The Confrontation Clause of the U.S. Constitution requires that witnesses against the defendant in criminal cases come to the witness stand and be cross-examined. The general idea is that you can’t use second-hand hearsay evidence against the criminally accused. In a 2004 case called Crawford v. Washington, the Supreme Court revolutionized this requirement by ruling that only if a person makes an out-of-court statement with testimonial intent — for example, a statement by a witness to the police describing the defendant’s conduct after an investigation has begun — must they also make that statement in court. The symposium article, also co-authored with RJ Coleman, addresses the questions that Crawford raised that remain unsolved and need solving. There’s a group of about 15 of them.

What’s one example? Why is it important to answer those questions?

Currently, there is controversy surrounding the forensic reporting that labs provide for criminal evidence, such as DNA and blood-alcohol reports. The Supreme Court’s cases in this area say that the analyst or analysts who prepared the report must also come to the witness stand to be cross-examined; you can’t just submit the report on its own. In a current case, called Smith v. Arizona, the Court may reexamine this standard, which some view as too cumbersome because witnesses end up called to the stand unnecessarily, making trials difficult and expensive.

We may want a system that is more efficient — meaning one that is less expensive and cumbersome for the prosecution to try — without sacrificing defendants’ rights. Finding that balance makes it easier to uncover and punish crime without sacrificing the right to a fair defense.

What do you want to work on next?

In the future, I want to explore artificial intelligence (AI) and burgeoning technologies such as facial recognition systems that can be used by police against criminal defendants. I think they pose a problem that is analogous to the Confrontation Clause problem, because we don’t always know what biases or hidden assumptions are built into software by human programmers. Frequently, the police and prosecutors can’t even access this information because it’s considered a trade secret.

To fairly defend themselves, criminal defendants ought to have some kind of access into how these programs were made or written. If there are people who are making evidence against you by way of AI systems or software, they should be present for examination in court.

Much of your work is philosophical in nature. How do you view the link between legal scholarship and practice?

The law is practical philosophy and practical logic applied to real cases. My work has a real effect on cases because that’s the way lawyers work, and that’s what convinces judges – logical chains of analysis leading to desirable results.