A Taxonomy to Assess the Usefulness of Amicus Briefs

Measuring whether the arguments are unexpected, and whether the signatories are unexpected.


In recent years, the number of amicus briefs submitted to the Supreme Court, as well as to the lower courts, has increased. Generally, at the beginning of every brief is a "Statement of Interest," which explains why the brief was filed. The statement of interest tends to convey two broad ideas. First, the brief purports to introduce some argument that is different than the arguments presented by the parties. An amicus brief that merely repeats what the parties argued is not very helpful. Second, the brief is signed by people who bring novel perspectives to the case. For example, they have relevant experience or expertise.

We can graph these two related concepts on an x-y axis.

First, the y-axis plots the novelty of the argument, ranging from expected arguments to unexpected arguments. Expected arguments in amicus briefs are already well-developed by the parties, and do not add much value to deliberations. By contrast, unexpected arguments in amicus briefs are not developed by the parties, and offer the Court valuable new ways to think about the case.

Second, the x-axis plots the identity of the signatories. There are many ways to measure this characteristic. To mirror the y-axis, I consider whether those who signed the brief are expected, or unexpectedExpected signatories are those people who are expected to support a specific result. At the other end of the spectrum are unexpected signatories. These are people you would not expect to support a particular side.

This plot separates the types of amicus briefs into four quadrants. Type I briefs present unexpected arguments from unexpected signatories. Type II briefs present unexpected arguments from expected signatories. Type III briefs present expected arguments from expected signatories. Type IV briefs present expected arguments from unexpected signatories. I also drew two lines that measure usefulness, which I'll explain later.







Expected and Unexpected Arguments

One way to assess the usefulness of an amicus brief is to measure whether the arguments advanced are expected or unexpected. Briefs at one end of the spectrum simply parrot the sorts of arguments advanced by the parties. Perhaps amici provide a new spin on these arguments, but there is not much new. These briefs are not particularly helpful. Briefs on the other end of the spectrum make novel contributions. For example, scholars who write in a niche area of the law introduce some new argument that was not developed in the litigation. These briefs may be filed in support of neither part. Or, the brief offers an innovative way to read or reconcile older precedents. Most lawyers would boast that their arguments are unexpected. I'm doubtful. To be candid, most amicus briefs do not tread new ground–present company included. Usually, by the time a case gets to the Supreme Court, the issues have been fully and thoroughly vetted. These sorts of cases are won and lost based on well-honed briefing by the parties. From time to time, these unexpected briefs are written. But they are rare.

Expected and Unexpected Signatories

More often than not, the signatories on a brief are entirely expected. Scan through the docket on any Supreme Court case. If you look at the party who submitted the brief, you can guess–with a high degree of accuracy–what the brief says. For example, in a case where the plaintiffs challenged a gun control law, there is no surprise if a gun-rights group supports the plaintiffs. Or, in a case where the plaintiffs challenge an abortion law, there is no surprise if an abortion-rights group supports the plaintiffs. These ideologically-sympathetic briefs are predictable. Briefs with expected signatories can still advance unexpected arguments. That metric is plotted on the y-axis. But any such novelty must be assessed against–and even discounted by–the philosophical simpatico between the group and the supported litigant. Expected signatories are not neutral arbiters. They have a vested interest in the outcome of the case.

Briefs with unexpected signatories are far more rare. For such briefs, it is surprising that certain people supported a specific legal position. I can think of four categories of unexpected signatories.

First, there is the strange-bedfellows brief. Here, people on opposite sides of the ideological spectrum, who tend to see things differently, reach a common consensus on a specific legal issue. These briefs signal that a given case need not be resolved on a strict ideological divide. Often, the Cato Institute and the Constitutional Accountability Center join forces for these sorts of briefs. In McDonald v. Chicago, a cohort of right-of-center and left-of-center scholars filed a joint brief to reinvigorate the Privileges or Immunities Clause. These briefs signal that there is broad support for a position.

Second, there is the against-interest brief. Here, people who previously took some position on an issue, now take a position that is against their prior interest. To use a familiar example, my colleagues Jonathan Adler and Ilya Somin filed amicus briefs in the ACA litigation. Both scholars were involved in other Obamacare challenges and did not support the law as a policy matter. But in California v. Texas, they argued that the courts should not rule against the ACA based on severability doctrine. I think these briefs were especially effective because these signatories were unexpected: their current positions stood in principled tension with their prior positions.

Third, there is the former-government-official brief. Here, the signatories previously served in government, and they argue that current administration is doing something different than they did. These types of submissions are especially effective where government officials who served in both Republican and Democratic administrations agree. These briefs were very prominent during the Trump Administration. For example, in Department of Commerce v. New York, former directors of the Census Bureau filed a brief. They argued that inquiring about citizenship status would yield a lower response rate. Chief Justice Robert favorably cited this brief to show that the Trump Administration failed to justify its policy.

Fourth, there is the what-did-you-expect brief. Here, the signatories are expected to take one position based on their reputations, but in fact take the opposite position. In other words, you might think they favor X as a policy matter, but they support ~X as a legal matter. For example, conservatives may be thought to favor gun rights as a policy matter, but they file a brief that urges the Court to uphold gun control laws. Or liberals may be thought to favor abortion rights as a policy matter, but they file a brief that urges the Court to uphold abortion restrictions. These briefs suggest that law and policy can be separated. But these sorts of briefs only work where the signatories have never opined on the issue before. Some conservatives may not favor gun rights. And some liberals may not favor abortion rights. But once the cat is out of the bag, these unexpected briefs become expected briefs. I recently criticized an amicus brief filed in the New York Second Amendment litigation. My post was regrettably misread. My criticism was narrow. When one of the lead signatories stated that he supported gun control laws well before Heller, without regard to the Second Amendment, the utility of the brief was diminished; an unexpected brief became an expected brief. The usefulness of the brief was diminished along the x-axis, even if the novelty along the y-axis was unchanged.

Measuring the Usefulness of Amicus Briefs

In an ideal world, the most useful briefs would be Type I and Type II. These submissions provide the Court with unexpected arguments that do not otherwise appear in the briefing. And, in an ideal world, the identity of the signatories should be irrelevant. But we do not live in an ideal world.

Type I briefs are the most useful briefs: unexpected arguments are advanced by unexpected signatories. These positions are treated as more reliable because the signatories bucked conventional wisdom. Perhaps they aligned with cross-ideological groups. This sort of brief would minimize differences and find common ground. Or former government officials repudiated their successors–always a fraught task. Or, in some cases, a brief is filed in support of neither party.  In any event, the Type I brief is the gold standard. Very few briefs fit within this taxonomy.

Type II briefs are the second-most useful briefs. They advance novel arguments that may facilitate deliberations. But there may be reason to doubt the objectivity of the brief because of the identity of the signatories. To use my prior examples, briefs from abortion rights groups who file a brief in an abortion case are not neutral. Ditto for gun rights groups who file a brief in a gun rights case. These groups have a vested stake in a particular outcome. They may be less friends of the Court and more friends of the cause. For Type II briefs to be useful, their novelty must exceed the novelty of a Type I brief. On the plot, I have drawn a "Non-cynical Usefulness" line in red. It illustrates how briefs can be cited in a non-cynical fashion. This line favors unexpected arguments, and gives a preference to those unexpected arguments advanced by unexpected signatories.

I also plotted a "Cynical Usefulness" line in blue, which stretches between Type III briefs to Type I briefs. Why is this line dubbed cynical? Here, the focus is less on the novelty of the arguments, and more on the identity of the signatories. For example, imagine that an unexpected group of signatories puts forward a mediocre brief. The only valuable aspect of the brief is that a cross-ideological group of people agreed on an issue; or that former government officials were willing to criticize the incumbent administration; or people who were thought to favor some policy filed a brief against that interest. With these submissions everything between the statement of interest and the signature block is window dressing. This brief is quite literally judged by its cover. Still, these briefs can be cited in a cynical fashion–not to develop reasoning, but to make an ideological point. I think these missives are especially useful in dissents: see, even people who usually agree with the majority reject the majority's argument! Tu quoque!  Of course, the cynical line stretches to Type I briefs. Where useful amici make useful arguments, there is a plus. But that benefit is not necessary.

Finally, Type III briefs are the least useful: the usual suspects round up the usual arguments. These sorts of briefs are not designed to change hearts or minds. At best, they may be useful tools for fundraising or virtue signaling.


This taxonomy is not a substitute for studies on citation patterns. In many cases, briefs from expected signatories are far more useful than briefs from unexpected signatories. The ACLU, Cato, CAC, and other groups are frequent players, but routinely submit high quality briefs with novel arguments that aid the deliberations. And I suspect that ideologically-sympathetic Justices favor briefs from certain groups over others. Rather, this taxonomy is designed to asses how the arguments presented in statements of interest affect–in an abstract sense–the utility of a brief.

NEXT: Today in Supreme Court History: September 19, 1907

Editor's Note: We invite comments and request that they be civil and on-topic. We do not moderate or assume any responsibility for comments, which are owned by the readers who post them. Comments do not represent the views of Reason.com or Reason Foundation. We reserve the right to delete any comment for any reason at any time. Report abuses.

  1. I was hoping for a taxonomy based on effect on judicial decisions. What are the most influential briefs?

    1. Perhaps AI can be applied to data analytics to determine the content of the most influential amicus briefs. Then, perhaps, the same app can write the briefs and submit them.

      Experts have influence on decisions. For example, Roe v Wade was copied from an adopted AMA Resolution.

  2. Unexpected arguments is a good measure of usefulness. I fail to see how unexpected signatories adds any usefulness to the issue, however. If an argument is legally true and relevant, it should remain so regardless of the person making the argument. To say otherwise is to commit either an ad hominem fallacy or the inverse (whose name I can’t remember) where you grant inappropriate deference because of the virtue of the speaker rather than the validity of their arguments.

    The NRA filing an against interest brief is no more (or less) compelling than the same brief filed by the most virulent anti-gun group.

    1. We should try to introduce the Ivy indoctrinated, big government biased, stinking, failed lawyer profession to critical thinking of the past 700 years. One Fallacy is the Appeal to Authority, or as it is called in the law, precedent. Most authorities are idiotic, biased, self interested selfish pricks. You are promoting those interests, and nothing else. The sole validation of 99.9% of precedent is a guy with a gun. Almost none has the empirical validation that national policy deserves and requires under the Fifth Amendment Procedural Right to Due Process. The process that is due is empirical validation. The Fallacy of Appeal to Authority applies to any AI algorithm written for the tyrannical dumbass scumbags.

      One simple solution is to make a one year course in Critical Thinking and in the Scientific Method and Its Problems, a mandatory pre-law course requirement. The other course that would help would be Economics 101. It must include a major segment about the Rent Seeking Theory, since this profession is solely about rent seeking and money grubbing. A course on Ethics would be too much to ask.

    2. Sort of, but I wonder whether the anti-gun control brief from the NRA, for example, gets much attention.

      “Hey, they’re against this law. Big deal. Next!”

      And this is not unreasonable. Josh can draw axes all he wants, but I suspect these two characteristics are far from independent of each other. Expected signatories are going to give you expected arguments much of the time.

      I also wonder why unexpected arguments are necessarily useful. They can be lousy arguments too. In fact, on well-worn issues won’t lots of unexpected arguments be bad? If we’ve spent decades fighting over something how likely is it that someone is going to come up with a brilliant new insight?

      Anyway, isn’t the test of whether a brief is useful whether it changes anything, either the outcome or important parts of the decision?

      In short, this exercise seems not very useful.

      1. re: unexpected arguments – I think it’s a labeling problem. He is not saying that they are necessarily useful – he is saying that they are potentially useful.

        An unexpected argument may be lousy but it is at least new. It’s something I haven’t already heard before. There is no potential utility to yet another person telling me the same things I’ve already heard.

        I would agree, by the way, that the ultimate test of a brief is whether it changes anything – but you won’t know that until long after your brief is submitted. Expected/unexpected argument, on the other hand, is a filter that you can apply before your brief is even submitted.

        If all you have is a rehash of the same old arguments, don’t waste everyone else’s time asking them to read your retread.

    3. Argument from authority. I don’t recall if there is a commonly used Latin phrasing of that.

      1. The Latin name is argumentum ab auctoritate, but Google says the English version has been more common for at least the last two centuries.

      2. Any legal utterance in Latin is void because it promotes the Catholic Church. That is not allowed in our secular nation. There is something deeply wrong with the lawyer profession, 10 times more toxic than organized crime.

        1. David,
          That is just a silly statement. It is hard to believe that you are serious

          1. None but the Church uses Latin today. To use it, promotes the Church, and should void the utterance in our secular nation. Utterances plagiarizing the catechism, word for word, not inspired by it, like the criminal code describing first degree murder should be void.

            1. I read the Sharia. 10% is objectionable. 90% is really good, and would make a better basis of the common law than the catechism. Is that OK with you, copying the Koran, the Sharia, using ancient Koranic Arab phrases in legal utterances? I would prefer it, “Charging interest on a loan is haram, a sin.” But, that sentence would be illegal in our secular nation. It promotes Islam.

      3. Argument from authority is only a subset of the broader fallacy I was thinking of. Argument from authority is when you grant greater deference because the speaker is an “expert” (even if not an expert in this area). But it’s still a fallacy (with a different name) when you grant greater deference because the speaker is a “good person” even if they are not expert.

    4. “I fail to see how unexpected signatories adds any usefulness to the issue, however. If an argument is legally true and relevant, it should remain so regardless of the person making the argument”

      The real question there is “what is the unexpected argument”?

      Sometimes it will be something unexpected, but basically useless. For example, Ilya Somin’s repeated arguments in favor of open borders. Ilya’s always going to argue in favor of open borders, expected or unexpected. Many of the unexpected arguments will be useless.

      However, when the unexpected argument picks up an unexpected signatory, then you pay attention. For example, if one of Ilya’s arguments in favor of open borders suddenly picked up the signatory support of the GOP of Texas, THAT may be useful.

  3. This could have been explained more simply, without a graph.

    1. In fact the graph was confusing with the red and blue arrows.

  4. JB’s taxonomy is not very clever and not very useful

    1. A lot of words to say a pretty boring thought: judges will likely listen carefully to people they think are smart when they offer a unique perspective.

      1. Exactly.
        You summed up the post in 17 words. Maybe even fewer would be possible.

  5. Josh badly needs some sort of mechanism to filter his thoughts before he posts.

    Quite frankly, he would be wise to just take a day after he writes something, and before posting, to both edit it down and think about whether it’s worth posting at all.

    1. And yet, you felt compelled to read it and comment on it, regardless.

      1. Which reply serves no purpose

  6. Blackman is, as ever, utterly fascinated with the remarkable architecture of his own thoughts.

  7. This taxonomy is not a substitute for studies on citation patterns.

    This says everything you pretty much need to know about this worthless piece. You have no evidence to suggest that this “taxonomy” has any basis in observed reality. Indeed, the only thing you have to say about the empirical support for your “taxonomy” is that the simplest way to potentially verify it is suspect for other reasons.

    So, no, this is almost as useless as the ancient philosophers speculating about the humors and elements. “I can imagine various ways that amicus briefs could be rhetorically more or less persuasive” is a great way to start a discussion where you find reasons to discredit evidence that counts against your preferred conclusion and amplify evidence that supports it.

    Josh, if you want this bullshit to go anywhere at all, you’ll spend less time drawing meaningless graphs and scribbling notes for the VC and more time thinking about ways of measuring the usefulness of amicus briefs in a way that controls for their varying quality and can rely on fairly objective metrics, such as citation practice or ultimate outcomes. From there, you can see if any of the data supports a higher value of amicus briefs that carry unexpected arguments or alliances.

    I rather suspect that the evidence is unlikely to clearly support any claim that unexpected alliances sway the justices, which is likely to be a problem for your putative “taxonomy.” So you’re going to have to see if you can make a persuasive case for some kind of metric that can pick up the effect of those kinds of amicus briefs. I suspect, further, that will be quite beyond your abilities.

  8. I’d love to see this put to the test, sorting the NYSRPA vs Bruen briefs into quadrants. ????

Please to post comments