Introduction
Justice Neil Gorsuch is a public-meaning originalist.
He believes that the Constitution’s meaning was fixed upon adoption, the Constitution’s meaning reflects how the public at the time of adoption would have read the Constitution’s text, and such meaning remains binding until the Constitution’s text is authoritatively changed.
Gorsuch thus cites originalist sources,
advances originalist arguments,
and criticizes nonoriginalist approaches.
And he is not alone. Justices Clarence Thomas, Brett Kavanaugh, and Amy Coney Barrett describe themselves the same way,
and Justice Samuel Alito calls himself a “practical originalist.”
Even Justices who are not card-carrying originalists, moreover, often appeal to the historical record as at least relevant.
Accordingly, advocates have increasingly framed their arguments using public-meaning originalism.
In Ramos v. Louisiana, one of his most significant opinions, Justice Gorsuch made a point to rely on public-meaning originalism.
Writing at times for the Court and other times for a plurality, he concluded that the Sixth Amendment requires unanimous juries and that the Fourteenth Amendment incorporated that right against the states.
To reach his first conclusion, Gorsuch surveyed historical sources “to determine what the term ‘trial by an impartial jury’ meant at the time of the Sixth Amendment’s adoption,”
including Blackstone, Founding-era practice, and opinions and commentary from soon after ratification.
After doing so, he concluded that the Sixth Amendment’s requirement for jury unanimity is incorporated against the states.
In reaching that second conclusion, Gorsuch avoided the Fourteenth Amendment’s Due Process Clause but rather explained that incorporation occurred through “the Fourteenth Amendment” generally
—apparently because he believes, like many originalists, that the correct clause is the Privileges or Immunities Clause.
Whatever one thinks of Ramos, it is hard to describe Justice Gorsuch’s opinion as anything but originalist.
He is a principled jurist who works hard, and it is apparent from Ramos that he wants to get history right. Despite such care, however, the likelihood that Ramos accurately reflects the Constitution’s original public meaning could well be about the same as a coin flip coming up “heads.” This is not because Justice Gorsuch overlooked historical evidence.
Nor is it because either of his steps is plainly wrong. No, Ramos’s problem is not its history but its math.
It is a basic rule of probability that, all else being equal, it is less likely for two necessary things to be true than just one. For example, there’s a 50% probability that an unweighted coin flip will come up heads, but the odds that it will come up heads twice in a row are only 25%.
Scholars have attempted to apply this probability concept—often called “aggregation” or “the conjunction paradox”
—to the law. For example, Professors Ariel Porat and Eric Posner have explained that when broad multistep propositions like “this criminal defendant should be convicted of (i) conspiracy to (ii) defraud” require running the table for multiple truth claims, it can be bad math to ignore aggregate uncertainty even if each claim is more likely than not correct.
Returning to Ramos, even if historical evidence suggests that (i) the original public meaning of “trial[] by an impartial jury”
more likely than not requires unanimity and (ii) the original public meaning of the Fourteenth Amendment more likely than not requires “that incorporated provisions of the Bill of Rights bear the same content when asserted against States as they do when asserted against the federal government,”
it still need not follow that the Court should overturn a state conviction for lack of a unanimous jury. Ramos required the Court to answer at least three distinct questions, including whether: (i) the Sixth Amendment requires unanimous juries, (ii) states are bound by that requirement, and (iii) the U.S. Supreme Court has the power to resolve such questions to begin with. None of these questions has an answer that is 100% certain. Aggregate uncertainty thus creates a puzzle, at least with respect to assessing the validity of the result—who wins or loses—in a case involving multistep originalist analysis.
To see the puzzle, assume it is 70% likely that the Sixth Amendment’s original public meaning requires jury unanimity and that it is also 70% likely that the Fourteenth Amendment’s original public meaning (through whatever clause) incorporates the Sixth Amendment.
Scholars will disagree about those assumptions, but given that the Constitution does not plainly answer either,
70% seems reasonable enough for a thought experiment. Next, assume (generously) that it is 99% likely that Article III’s original public meaning allows the U.S. Supreme Court to determine on direct review whether a state criminal judgment violates the federal Constitution.
And finally, assume that these questions are all independent
—that is, the answer for one tells us nothing about the answer for another.
Under these assumptions, it is more likely than not that the judgment in Ramos does not reflect the Constitution’s original public meaning. To find the probability of three independent events, the product rule requires multiplying the probability of each event occurring: 70% × 70% × 99% = 49%.
Of course, these assumed probabilities might be too high or too low. But precise percentages do not matter yet; this is just a thought experiment. The point is that even if the answer to each step in a multistep problem is more likely right than wrong, the ultimate conclusion can still probably be wrong. Evidence theorists have recognized the challenge that aggregation’s probability rules pose for the law,
but no one has systematically assessed originalism through the lens of aggregation, even though originalism also involves the search for historical facts through the use of evidence.
The purpose of this thought experiment is not to pick on Justice Gorsuch or anyone else. On the contrary, similar probabilistic questioning could be done for almost any opinion (originalist or otherwise) written by any judge or justice. Almost no one approaches law probabilistically.
What’s more, the likelihood of correctness for each question in Ramos could be greater than 70%, so the judgment there may be more likely than not correct anyway. Instead, the point of this thought experiment is to show how courts address multistep analysis. What often happens is the Court considers one question, decides who has the better of the argument with respect to that question even though the answer is not 100% certain, and then proceeds to the next question as if the answer to the first question was 100% certain.
Setting aside the risk of aggregated error in this way may not be a problem for some methodologies, or at least not as much of a problem. Some pragmatists, for example, are not necessarily trying to figure out the answer to multiple independent questions; they generally want to know the pragmatically best ultimate answer.
Or for someone who does not believe that meaning is a fact at all,
or at least rarely a knowable one,
perhaps aggregated uncertainty also doesn’t matter. Some other methodologies have also mitigated aggregation’s effect through tools such as stare decisis and deference to fact finders (what happens in the jury room often is essentially a black box
). But one of the premises of originalism is that courts, for reasons sounding in legitimacy, should issue decisions that reflect the fixed meaning of the text being interpreted.
So, if it is more likely than not that a judicial judgment is not actually supported by the constitutional provisions at issue, it should be harder for a committed originalist to proceed.
In other words, although aggregation could be relevant to many methodologies, including nonoriginalist ones, it certainly should be relevant (at least presumptively) to originalism.
By assuming the answer to each question in multistep analysis to be 100% certain rather than assigning each step a probability of correctness, courts applying originalist analysis may side with parties whose ultimate claim more likely than not is wrong as a matter of historical fact. Indeed, even modest degrees of uncertainty about historical meaning can undermine ultimate judgments. Yet courts regularly decide cases with multistep analysis.
As with Ramos, for example, any case in which a party argues to the U.S. Supreme Court that a state official has violated a right in the Bill of Rights requires answering (at least) whether the right was violated and whether the right was incorporated.
Many of the Court’s most significant cases fall within this category.
Likewise, every case in which a criminal defendant asks a court to exclude evidence because of an alleged constitutional violation requires evaluating at least whether: (i) the claimed right exists; (ii) the claimed right was violated; and, if so, whether (iii) exclusion is required.
And courts often decide (i) whether a right or power exists and then (ii) whether an “edge” case falls within that right or power.
If originalists wish to issue judgments reflecting what is most likely the Constitution’s original public meaning, why would probability-sensitive rules designed to minimize aggregate error not play at least some role? Ignoring the possibility of being wrong “could leave us less rather than more faithful to the original plan.”
There are significant obstacles, however, to incorporating aggregation into originalism. For one, doing so raises questions about legal interpretation itself and how likely propositions of law must be before courts should enforce them. Should judges only accept propositions about meaning that have a greater than 50% likelihood of being accurate?
Originalists presumably should not accept Interpretation A when historical evidence suggests that competing Interpretation B is more likely, but is a “more likely than not correct” standard appropriate or is a greater showing required?
This threshold question of how one “proves” the law is even more significant when multistep analysis—and with it, aggregation—is considered.
Aggregation, moreover, can be difficult to do. It is easy to posit as a thought experiment that incorporation is 70% likely; it is more challenging in the real world to accurately assign such a probability.
A strong form of aggregation also may change the outcome of many cases. For example, putting aside stare decisis (which lower court judges, of course, can’t do), originalists may be hard pressed to ever exclude evidence on Fourth Amendment grounds in state criminal trials—even if the violation is egregious and even assuming (probably a stretch) that the Constitution’s original public meaning more likely than not requires the exclusionary rule.
After all, assuming for argument’s sake that there is 60% originalist support for the exclusionary rule, uncertainty about whether the Fourth Amendment is incorporated by itself could prevent exclusion even when it is 100% certain that the right at issue was violated.
Aggregation, moreover, may destroy “congruent” constitutionalism, which treats federal and state governments the same when enforcing the U.S. Constitution, because of uncertainty about the incorporation of the Bill of Rights.
Yet many of the Court’s highest-profile cases in recent years involve state officials.
Aggregation’s potentially disruptive implications thus elevate the importance of other already significant debates, including those about the strength of stare decisis,
the manner and scope of construction,
the legitimacy of appeals to liquidation and tradition,
the permissibility of a “compensating adjustment that approximates” what originalism might require,
and the value of corpus linguistics.
If, for example, a court could treat the likelihood of incorporation as 100% certain because of stare decisis, one step would disappear. The opposite, of course, is also true; if such doctrines should play little role, many cases may have to be revisited. Even “fearless originalists,”
however, may stagger before discounting every constitutional decision by the possibility (even if slight) that Marbury is wrong.
Aggregation, however, need not be only disruptive—it also can lend at least some additional stability to constitutional law by offering conceptual support for features that originalists struggle to defend.
For example, is there a place for hybrid-rights analysis?
Maybe, because the Court might have misapplied more than one provision, making it more likely than not overall that the alleged right exists somewhere. Should courts embrace a presumption of constitutionality?
Perhaps, not because any discrete provision creates the presumption, but because judicial review itself is subject to some historical doubt (even if slight) that it may be wrong. So why wouldn’t close questions favor judicial restraint? Similarly, should there be clear-statement rules?
Possibly, at least sometimes, because when it is not certain the federal government has power to act in the first place, statutory clarity is more important. And what about qualified immunity, another doctrine often attacked by originalists?
Even setting aside historical support for such immunity,
this defense may also be supported by everyday math. If the doctrine is applied correctly, the cases in which qualified immunity bars recovery should be those in which it is not obvious that a constitutional violation occurred.
Finally, although aggregation matters for originalism because debates about stare decisis are particularly sharp among originalists, aggregation may also help shore up public-meaning originalism itself because certain competing methodologies may be less able to account for aggregation. For example, purposivists who believe they can “cull legislative ‘intent’ or ‘purpose’” from sources other than the text
must also account for the possibility that they have performed that difficult—perhaps impossible
—task wrongly, thus increasing uncertainty; such risk becomes more important given multistep analysis.
These are not the only potential consequences, but they illustrate why aggregation creates a puzzle for constitutional theorists, including and perhaps especially originalists. But that puzzle comes with an opportunity. By confronting aggregation, originalists can better appreciate the stakes of other significant debates, while also finding a conceptual path to defend some familiar aspect of constitutional law. This may be particularly true, moreover, if judges adopt what this Article dubs “soft” aggregation that focuses on aggregation’s intuition against stacking inference upon inference rather than taking the concept so far as to exceed judicial competence. Just because it may not be realistically possible for courts to do the types of calculations that the purest form of aggregation requires, it does not follow that courts should disregard aggregation’s intuitive value—everyday logic teaches that, all else equal, it is more likely for one proposition to be true than for a party to run the table for multiple necessary propositions, or for two easy questions to be both right rather than two difficult ones. By itself, following that intuition should lead to greater accuracy for judgments requiring multiple propositions to all be resolved in favor of the same party.
* * *
This Article proceeds as follows. It first explores the basics of originalism. It then explains aggregation and how evidence scholars have addressed the product rule and conditional probability. Next, it shows how the premises of public-meaning originalism support aggregation and offers five exemplars of what such analysis may include. The Article then sketches aggregation’s disruptive and stabilizing effects. Finally, it concludes with a potential path forward that relies on a softer, more administrable form of aggregation dubbed soft aggregation. Even if the most demanding form of aggregation is beyond judicial competence, there still may be a place for soft aggregation.