The Baldy Center Podcast

The Baldy Center for Law and Social Policy at the University at Buffalo
Podcast Season 4, Episode 26
Podcast recording date: January 20, 2022
Host-producer: Edgar Girtain
Speaker: Rachael Hinkle
Contact information: baldycenter@buffalo.edu

Podcast transcript begins.

Edgar: Hello and welcome to the Baldy Center for Law and Social Policy podcast. I'm your host, Edgar Girtain, and today it is my pleasure to present you Rachael K. Hinkle, Associate Professor in the Department of Political Science here in the University at Buffalo. Professor Hinkle's research agenda focuses on judicial politics, with particular attention to gleaning insights into legal development from the content of judicial opinions through the use of computational text analytic techniques. And if all that sounds a bit intimidating to you, don't worry, you're in good company. But as you're about to hear, Dr. Hinkle is a passionate speaker, a rigorous academic, and a thoroughly articulate speaker. I hope you enjoy this fascinating conversation, and now with any - and now without any further ado, I present Dr. Rachael K. Hinkle.

All right, Rachael, thank you very much for coming on the program today.

Rachael: Thank you for the invitation. I'm happy to be here.

Edgar: So, while I was preparing for this interview, your background actually stood out quite a bit to me. In the University at Buffalo, and, on this podcast, I think we get a lot of legal scholars around who come from, I guess, the practical professional field, you know. They're practicing lawyers, you know, they're clinical professors, but you're a little bit - it seems a little bit different to me. I get the sense that you're coming into this from a deeply academic perspective and that you're very interested in how political and especially social, even personal relationships, these kinds of processes interact with the legal field. Is that an accurate assessment?

Rachael: Absolutely. And it's interesting. I wouldn't have thought of my background as entirely academic, but, that's fair, it is. However, it is also informed by observations of, just kind of, humanity and its natural habitat. So, I did go to law school, I never practiced law, but I clerked for three years. So, a lot of my motivation comes from just remembering stories from that time. And you know, many people who have been former clerks know, like, you don't necessarily tell all the stories in public, but they're there, they percolate in the back of your mind and they inform the kind of work that you do. You think, “gee, that one time that thing happened, I wonder if that's part of a greater pattern.” Which is where my political science training came in.

So, after clerking I decided I was more interested in teaching than in practicing law, so I went back and got a PhD in political science, which brings in academic, big data, lots of statistics, but really caring about broad patterns of behavior. But a lot of times those patterns are informed by the anecdotes you get from being in the weeds, and I have a co-author, Morgan Hazelton, who practiced law for several years, and she brings a lot of that very practical, hardcore legal element to the work we do together as well.

Edgar: So, I hear you're working on a new book.

Rachael: Yes, with Morgan. It's due to come out next fall. The book is focused on the impact of briefs at the supreme court level, so, there's a lot of research looking at how oral arguments affect justices, but we're actually going back to that written level. Like, what do those briefs say that are submitted to the court in writing? Because, of course, the court might hear all arguments from the two sides, but they will read briefs from potentially dozens - hundreds of different entities who file amicus brief. So we look at amicus briefs and litigant briefs, and we look at a variety of features of how that information is created and then how it impacts the court, both in terms of votes and in terms of how opinions are crafted.

We can't really tell causation, but we kind of look for links, like, where does language from briefs show up in opinions? So, we can kind of get a more nuanced sense of that by having the actual text of the written briefs and comparing that to the text of the written opinions.

Edgar: Slow down. What is a brief?

Rachael: Okay. A brief is a written submission to the court, so it's arguments to the judges in writing. By the time oral arguments - happen we're all a little more familiar with oral arguments. I mean, a lot of citizens aren't that familiar with the appellate process, right? We think of court, we think of Law and Order, what we see on tv, the lawyers questioning witnesses, a jury. At the appellate process, the witnesses, the litigants don't have much to do with it. It's pretty much the lawyers and the judges talking about law. And deciding issues of law.

In most appellate courts, including the Supreme Court, the written arguments are really a very large part of the consideration of legal issues. Written arguments are read by the judges and their clerks and their staff. And then, of course, their opinions are produced in a written form. So, the oral arguments are interesting and kind of snazzy and fun to watch and have fun sound bites, but a lot of the actual legal work that's happening is these very carefully constructed briefs that present legal arguments to the court.

Edgar: There were two technical terms that you had mentioned: an amicus brief and a litigant brief, you said?

Rachael: Yes, and a litigant brief, right. So, we looked at these two kinds of briefs, and lawyers like to use Latin just because. So, an amicus brief - and amicus is a friend, right? So, you're a friend of the court, an amicus curiae. Some people say uh-mee-kuss, ah-mih-kuss, I don't know what the correct pronunciation is. If I’m talking to someone, I usually just adopt the one they're using because that seems like the most lawyer-y thing to do. So, an amicus brief as a friend of the court. Essentially anybody can file a brief to the court; they don't have to be a party to the case, right. So, the parties to the case the ones who are suing each other. The Supreme Court is resolving their dispute. So, if you're looking at oral arguments you're focusing on their attorneys because they're the ones making the arguments, the parties to the case. But when you look at briefs you can look at this whole chunk of amicus briefs. Which is, in fact, is more than half of the briefs in our data set. We have a huge data set of briefs that we're looking at.

There's some rules about permission. Really, the Supreme Court doesn't care. If you file a brief, they'll almost always accept it. And even if they don't accept it, they have to read it to say no, which, pretty much seems as good as if they read it and said yes, because now they've read your arguments. So, to get a sense of what that can look like in a given case, like how many people are interested in cases before the Supreme Court, the answer is a lot. And it's not cheap to hire attorneys to write these briefs. They can run into the tens of thousands of dollars pretty quickly. But to get a sense of the scope, so in Obergefell, in the marriage equality case, the briefs submitted to the Supreme Court, if you took them all together and put them in a stack, would be twice the length of War and Peace. [both laughing]

That's one case, right? So, that gives you a sense of the scope of what's - and not every case, that's a particularly large number of briefs, but there is a lot of information being submitted to the Supreme Court in these written arguments by all manner of people; by the people litigating the case, it can be an individual person who wakes up one day and says, “I want to tell the Supreme Court what I think.” It can be the ACLU. Right, it can be anywhere in between. It could be the Solicitor General of the United States, who is the lawyer in charge of telling the Supreme Court what the executive branch thinks about things. And that's a - they're a very powerful institutional actor, so, it can be a wide range of people. It can be a party filing a brief and they've never been in the Supreme Court before. It can be the ACLU filing their, you know, 200th brief in the past five years.

The parties and the filers of amicus briefs, they hire attorneys. You could have an attorney writing a brief who's never written a brief to the Supreme Court before. Or you could have an attorney writing a brief to the Supreme Court who's written hundreds of them or dozens of them. So there's variation in, kind of, experience in that forum, and those are some of the things that we track in the book.

Edgar: And you said that the court is obligated to read these?

Rachael: Um, so, obligated - that's interesting. So, really cool part of the book is, Morgan did some interviews with former Supreme Court clerks, lawyers who write these briefs in their daily practice, and there's a lot of overlap there, right? Folks who were Supreme Court clerks and are writing the briefs. So, she talked to these folks. There’s some really interesting texture throughout the book from those interviews. And one of the things that she asks about is, especially the folks who clerked, was, do the justices read all of these? And the picture that emerges is that the clerks will do some sort of filtering. So, the clerks will read them all and determine which ones they, kind of, pass along to their justice as the most important ones. And one of the key themes we pick up in our research is, how does repetition work, right?

So, those two volumes of War and Peace that go to the Supreme Court in the Obergefell case, that's not all new information, right? It's not like a page turner where you get something new every page. There are a lot of the same arguments being made over and over again. So, one of the things that clerks can do in that filtering process is kind of say to their justice, like, “well yeah here are these ten briefs but they all kind of make the same argument. The one on top does it the best,” right? So, their justice can then read through - I mean, I would guess there's variation in how much actually gets read. It's a tremendous reading load for any human person to get through, you know? Not unlike trying to get through ten thousand petitions to even decide which cases the Supreme Court will hear. So, they have a tremendous reading load. So, my guess is no they don't read all of it, but the clerks do, right? And then there's this filtering process.

Edgar: I'm very curious about what gets through in the end, you know, and how it gets through. I mean, is there any kind of, um, I don't know, like, AI? Or any kind of, like, processing that happens to the data?

Rachael: We can talk to clerks and say, okay, what's the process look like, and get that kind of deep, qualitative dive into what's happening. Empirically, what we actually - what we have access to, we don't have access to those clerk memos saying, “don't read these five briefs, they're dumb,” right? We don't have that information. But what we do have is, we have the text of the briefs that are submitted, and we have the text of the ultimate majority opinion. So, there are some kind of computational AI-adjacent techniques that we use to measure, “oh, what's the similarity between the text in the brief that's submitted and the ultimate majority opinion?” So, it's a complicated process, but here's how it breaks down.

If you want to do a Google search you type in a sentence, right, or a phrase, right, Game of Thrones. What Google is doing underneath is it's running a computation, this is - how similar is this text that you've given me to all the websites I have, and then I'm just going to tell you which ones are most similar. Well, it's not going to use the word ‘of,’ right? Lots of things are going to have ‘of’ that aren't going to have anything to do with Game of Thrones. It's going to look for things that have all of those words, particularly if ‘game’ and ‘thrones’ occur together a lot, it's going to pull that website and say, hey this is really similar. So what we're doing is essentially the same thing. Instead of comparing a snippet to all the websites on the internet, we're comparing a longer text, right, an entire brief, to a majority opinion.

It generates a number between zero and one. If it's close to one, they're pretty similar, if it's close to zero, they're not very similar. So it has that kind of weighting aspect of, like, Scrabble tiles, right? If two opinions, if they both talk about Game of Thrones, they're going to be more similar. If they both use the word ‘of,’ or ‘law,’ or ‘justice,’ that's not really helpful.

Edgar: Wow, fascinating. This actually dovetails nicely into the next question I had. I read that as a source for this book, you have over 25,000 amicus briefs. How does one even begin to analyze that much information and, let alone, make any sense of it?

Rachael: You know, so, it takes a little bit of creativity, a little bit of digging into other literature. So, one example of a challenge we faced was, there's this sense in the legal community, or this concept of the Brandeis brief, right? A certain kind of brief - the first one was originally written in Muller versus Oregon by later Justice Brandeis - but the idea is, it's a brief that advises the court about future ramifications of a decision. It talks about, kind of, social scientific research and facts that the court might not be aware of. So, folks like me, right? Telling the court about what we do. Well, how do you tell if you have 26,000 briefs? We didn't have time to sit there and read all of them and say which one is a Brandeis brief and which one isn't a Brandeis brief.

So we got creative. And we did some - we said, well what is it about a Brandeis brief? Well, one of the things is it presents the court with, kind of, technical information. “If you do x, y and z are going to happen with whatever percent probability,” right? It's technical. It doesn't read like legal language, it reads, like, in more technical ways. So, we dove into some associated literatures about how do you measure technical language. And there's some analogous fields that want to do that for various reasons. They might want to make sure their reports are understandable by a, like, audience, right? So, there's different kinds of reasons.

So, we get to this research and we found out, well, parentheses. If you use parenthetical explanations a lot, that's an evidence of technical language, and there's a couple others. If you talk about causation, that's a thing. And there's um, there's another one in there that I can't think of at the moment. But there are these three elements that we pull out that, using existing software, we were able to count objectively in all of our documents. So, we went back to the original Brandeis brief and we said, well does this idea work? And it does. In the original Brandeis brief, in all these measures, they pop out a lot higher than the opposing brief in that same case.

So we’re like, okay. That seems we have some facial validity here. Is that measure perfect? No. Is there a lot of noise? Absolutely. But there's some signal there. Now, yeah, we're just looking at, you know, parenthetical explanations, and causal, you know, words that indicate causation, like ‘because.’ But it's a measure that works, so, a little bit of creativity, a little bit of thinking outside the box. But we're excited about that because we do that a few different ways, a few different places in the book. Some of it is building on-existing work. That one we kind of came up with from scratch, but then there's some other things like that. But, so, a little bit of creativity and a - and a little bit of being good at explaining to people why the thing you did makes sense.

Edgar: Is there anything else that you'd like to say or add that we didn't touch on?

Rachael: Um, uh, very briefly, just because we talked about the book so much, would you mind if I just kind of gave you a couple of, like, short findings from the book? Like, kind of, the takeaway, what we find out after doing all that data analysis and fancy AI stuff? Like, what are some of our, kind of, cool, fun takeaways?

Edgar: Oh, yeah! Yes, please, please.

Rachael: So, the book is kind of in two parts. In the first part we think about the causes of information. So, what impacts how briefs are written. So, the filer is the entity submitting the brief. It could be the litigant, it could be the amicus. So, one of the things we look at, or that we find out, is that more experienced filers and attorneys present very specific looking amicus briefs. They have more overall information, they refer to the future more frequently, and, this is kind of interesting, they're a little more strategic. They include more citations to opinions authored by the current median justice, right? So, there's some speculation that some briefs are written as, kind of, like, the traditional love letters to Kennedy, right? You find the median of the court and you really try to target them. So by experience we mean in filing briefs in the Supreme Court. So we just count, going back to 1970, how many times have you done this. So it's a pretty nuanced measure, but those briefs look really specific. And these are just some of our findings. We actually have, you know, like, 37 findings in that chapter. But those are the, kind of, interesting ones. Experience does some of the things we would expect it to do.

We also look at the coordination between parties in amici, because it turns out all these briefs aren't written on islands; these people talk to each other, right? So, if you're submitting on the side of the respondent or the petitioner, you know, if you want the court to do a conservative thing or you want the court to do a liberal thing, you're talking with all the other folks on your side who are asking for the same thing. So, we find that when the attorneys on both sides are more experienced, there's more overlap between their briefs. Which kind of fits. Later on, we find repetition is a good thing, so it makes sense if you're more experienced, you understand that repetition is a good thing. If you’re Carter Phillips, who writes a zillion briefs to the Supreme Court and argues a zillion times, you don't need our empirical study to tell you the repetition is good. Everybody else now has their study, so there we go, we're all on the same page now. But they figure that out. So with more experienced folks there's more overlap. They're saying the same things as each other. And we argue that's the strategic choice. They understand what works and it signals policy coalescence to the court. Multiple people want the same thing, and in a democracy that has a certain kind of weight to it.

The second half of the book, we look at consequences. What is the outcome, or, how do these briefs and the people writing them affect what happens in the court, how justice's vote, who wins and loses, and how that majority opinion is written. So, there's kind of two cool findings here. The information in the briefs does matter, we measure it a bunch of different ways. It especially matters for the content of the majority opinion, which makes sense. The text of your written arguments affects the text of the written law the Supreme Court establishes. Now, in aggregate, turns out it also affects who wins and loses. The side that presents more information to the court - just pure bulk - when you add it all together, stack it up in a pile, and compare the stack for the two sides, the side with the higher stack is more likely to win.

Edgar: Wow.

Rachael: So, the other kind of cool thing that comes out of our look at consequences is, we've had this research for a while that says, if you have resources you can hire the best attorneys, you litigate in court a lot, you're more likely to win. It's the haves versus the have-nots. The haves do better. We actually find that continues to be the case after we control for information. If you've been in the court a lot, you can hire attorneys who've been in court a lot, they can present higher quality briefs. Those higher quality briefs have an impact. Even after you control for that impact, just the fact that you have a more experienced attorney, that in and of itself is going to help you win your case. It's going to help you impact the majority opinion more, right?

Which suggests that attorney experience isn't just about knowing the court, it's about the court knowing you. The court knows you appear in front of them a lot, and if you appear in front of them a lot, you can't lie to them. Because if you lie to them today, they're gonna be mad at you tomorrow. And if you expect to be there tomorrow, you care about that. The information matters, but the people who present it also matter. And we find lots of cool stuff, and we find, like, one brief impacting the majority opinion, and all the briefs on a side impacting the majority opinion. We kind of cut the data a bunch of different ways and a lot of these - these are the broad themes that appear throughout. A bunch of other little cool findings too, but those are some of the highlights.

Edgar: Were there any, kind of, specific cases that you want to get into in detail with that?

Rachael: For the most part, we look at these, kind of, broad patterns. We do use the case of Florida versus Harris, it's a dog sniff case, that we use throughout the book as a motivating example. It's kind of a shockingly good example in almost every chapter, so, we use that to kind of put some context into our measures, right? So like, if this measure is seven, what does that mean? Well, in this dog sniff case, this measure, for each of these different brief writers, was this. We talk about who submits the briefs. It's an interesting case. There's kind of five briefs on one side and four on the other. It was unanimously decided in favor of upholding the search as right, the dog sniff as a legal search.

So, it's a way we kind of motivate the fact that even if all the justices rule the same, you can still get interesting nuance about how the briefs in that specific case influence the majority opinion differently. So we can see, is that brief filed by 38 states and a ton of attorneys with massive experience? Well yeah, that has a bigger impact on the majority opinion than a litigant brief filed by the two lawyers for the poor person in jail who, his lawyers have never been in the Supreme Court before, he's never been in the Supreme Court before. Yeah, their brief wasn't reflected so much in that majority opinion. So, we use that case as, kind of, a motivating theme throughout the book, just to stay a little bit grounded in what a specific case looks like instead of just the massive abstract, like, you know, this process from 30,000 feet, which is what the empirical data gives you.

Edgar: Wow! So, um, you said the book's coming out in October, right?

Rachael: That is the plan, yes. Yes. If everything stays on schedule.

Edgar: Where can our listeners get a copy?

Rachael: It is coming from the University Press of Kansas, and I would imagine electronic versions will be available through their website, that it would be available on Amazon. We haven't talked all of those details with the marketing team yet, but the title of the book is, Persuading the Supreme Court: The Significance of Briefs in Judicial Decision Making.

Edgar: Wonderful. All right, thank you very much for coming on today.

Rachael: Thank you. Thank you for having me, it was a pleasure.

Edgar: It was a pleasure having you, and, please come back.

Rachael: Well I have two more books, I can come back to discuss them. [both laugh]

Edgar: Good, very good.

Very good. I'm Edgar Girtain, and that was Dr. Rachael K. Hinkle, Associate Professor in the Department of Political Science here at the University at Buffalo. This has been the Baldy Center for Law and Social Policy podcast. You can learn more about the Center on our website, buffalo.edu/BaldyCenter. We would also like to hear your thoughts about this podcast. Please tweet us, @BaldyCenter, or send us an email at baldycenter@buffalo.edu. The theme music for this season was composed by Matias Homar, a PhD student in the Department of Music here at the University at Buffalo. I'm Edgar Girtain, your host. Thank you for listening, and be well.