How Many States Make Up a Historical Tradition? Emerging-Consensus Doctrine and Lessons from International Law

  • Date:
  • September 16th, 2022

By: Andrew Willinger

Bruen itself is not clear on how many laws, or laws from what time period specifically, are needed to establish a historical tradition of regulation.  The opinion tells us it is “doubtful” that three colonial-era regulations “could suffice to show a tradition of public-carry regulation,” and suggests that laws from the Founding era are accorded greater weight in the analysis—so it could be that fewer laws from the Founding era are needed to constitute a historical tradition of regulation, than laws from the 19th century.  But Bruen does not provide lower courts with clear numerical guideposts on the quantum of laws needed.  Namely, the decision leaves open crucial questions about the correct denominator to use: should historical laws be measured as a percentage of states at that time, as a percentage of historical U.S. population, or against some other denominator?

Two recent federal court decisions have suggested one way to apply Bruen: only when a majority of states at the relevant time had analogous laws in place is there a historical tradition of regulating guns in that manner.  In Antonyuk v. Bruen, Chief Judge Suddaby of the Northern District of New York wrote in dictum that New York’s sensitive-places list was unconstitutional because a “vast majority of [] states . . . did not have statutes restricting firearms at those very locations,” in 1791 or 1868 (emphasis added).  In Firearms Policy Coalition v. McGraw, Judge Pittman of the Northern District of Texas found that age-based restrictions on gun use and possession enacted “between 1856 and 1899 [in] approximately twenty jurisdictions (of the then 45 states)” were insufficient to establish a historical tradition of restricting the public carry of weapons by 18-20-year-olds. 

Of course, there’s nothing in Bruen that requires a majority-of-states approach.  As I observed in discussing Antonyuk, it’s somewhat strange to read Bruen in this way because the Court spends a great deal of effort analyzing and substantively distinguishing historical laws that don’t appear to have been in place in anywhere near a majority of states at the relevant time.  But looking to whether laws were enacted in a majority of states is one way to implement Bruen and, at the very least, it would make the historical test more manageable (and mechanical) for lower courts.  Of course, this approach would also make it very difficult for courts to uphold even laws that Heller (and Bruen) say are presumptively constitutional, such as prohibitions on gun possession by felons and the mentally ill, and laws barring those convicted of domestic violence offenses from possessing guns.  Such laws were not in place in a majority of states at any time close to the Founding—indeed, many of these laws were not enacted in any state until the 20th century.

The denominator problem is a recurring one in U.S. constitutional law, and there is no trans-substantive rule for which states to measure against when determining, for example, whether a particular punishment is “cruel and unusual,” whether a particular practice is “deeply rooted,” or whether material is “obscene” and thus not protected by the First Amendment under “community standards.”  As Ernest Young has observed, the issue of whether to include foreign jurisdictions in the denominator in Eighth Amendment cases is often outcome-determinative because a “substantial minority position on the domestic plane becomes an aberrational practice . . . when judged against the backdrop of world opinion.”

This same issue, it turns out, also arises in rights litigation in other contexts, especially in judicial tribunals enforcing international conventions such as the European Court of Human Rights (or ECHR).  As Shai Dothan has explained, the ECHR determines minimum human-rights standards for European countries under the “emerging consensus” test:

The emerging consensus doctrine is used by the ECHR to discover the minimal human rights standards that are respected by at least a majority of the countries in Europe. This minimal standard is then required from all European countries. States that do not grant these minimal standards will be found in violation of the [European Convention on Human Rights] by the ECHR.

Emerging consensus can be justified in light of a simple mathematical model called the Condorcet Jury Theorem. This model stipulates that a majority vote in a group of similar decision-makers, who decide independently, is more likely to be correct than the choice of each individual decision-maker. The problem with using the Jury Theorem to justify emerging consensus is that not all countries in Europe are similar in every respect, and some countries may be motivated not to decide independently.

For example, in a 2002 case dealing with the right of transgender individuals to change their gender on identification documents, the ECHR emphasized that a slight majority (54%) of EU contracting states recognized the right of “post-operative transsexuals to marry a person of sex opposite to their acquired gender.”  However, in light of the concern (observed by Professor Dothan) that “not all countries in Europe are similar in every respect,” the ECHR does recognize departures from even a majority-supported approach when a country is deemed sufficiently unique that it should enjoy a “margin of appreciation” in domestic legislation.  According national governments a “margin of appreciation” is “based on the notion that each society is entitled to certain latitude [from even consensus standards adopted by the majority of other European countries] in resolving the inherent conflicts between individual rights and national interests or among different moral convictions.”  For example, in a 2005 challenge to Turkey’s headscarf ban, the applicant argued that “no European State prohibited students from wearing the Islamic headscarf at university.”  However, the court found that the Turkish situation was unique because most of the population is Muslim and such laws might be thought more necessary in Turkey to encourage secularism and protect democracy—according to the court, this context supported allowing a margin of appreciation from the majority approach.

It’s important not to overlook the word emerging in “emerging consensus” doctrine.  The ECHR often appears to use the framework when evaluating standards that might be on the very cusp of majority acceptance, and to nudge states slow to adopt such norms.  The ECHR uses the test to evaluate the present-day situation, not to draw lessons from the historical record.  In some instances, the analysis may even be forward-looking, as ECHR decisions occasionally look to emerging international trends outside of the EU.[1]  But, in many ways, emerging consensus doctrine is similar to the way that Antonyuk and McGraw seem to approach Bruen:  laws are only highly-persuasive evidence when adopted by a majority (even a slight majority) of jurisdictions within a system at a given time.

I believe there are a couple of potential lessons to draw from this comparative example.  First, as the Turkey headscarf case described above illustrates, even a tribunal that places great emphasis on majority consensus among jurisdictions recognizes certain substantive differences in setting which can sanction departures from the majority rule.  This works both ways: it could be that laws enacted in a vast majority of countries or states are not appropriate in a certain unique setting, or it could be that a unique jurisdiction may enact laws that would not be appropriate in other jurisdictions.  The Supreme Court has largely rejected the argument that cities are fundamentally distinct from rural areas in ways that justify different types of gun regulation—including in its discussion of sensitive places in Bruen.  Moreover, Joseph Blocher has described how self-defense considerations and state preemption laws pose real barriers to firearms localism.  But the urban-rural divide may not be the only relevant factor at play here.  How about jurisdictions that have recent experience with a tragic mass shooting?  Are those jurisdictions potentially substantively distinct, such that they should have some margin of appreciation in passing gun legislation?  For example, could Highland Park prohibit guns at its next July 4 parade, even if Dallas or Chicago could not?

Second, as Professor Dothan notes, there is concern that judges might “not really attempt[] to discover the laws of the majority of states on which to base their decisions, but rather mak[e] their decision for their own reasons and afterwards us[e] states with similar laws to support that decision.”  Professor Young’s analysis similarly suggests that consensus-based tests are especially susceptible to being coopted as legal argumentation: “Opponents of the death penalty who have striven in vain to persuade their fellow Americans to abandon the measure will find more support by extending their sphere of argument to take in foreign opinions and practices.” 

This is a major issue with the Bruen test—Justice Breyer in dissent is particularly concerned that the test will “permit judges to reach the outcomes they prefer and then cloak those outcomes in the language of history.”  Judges can also distort the denominator of states to reach their preferred outcome.  The United States grew from 13 to 37 in number from 1790 to 1868.  If a judge is dealing with a type of regulation that existed in some form during that period, but wishes to find no supportive historical tradition, he or she can simply pick a year in which the number of states with such a law in place was slightly below 50%.  Similarly, if a judge wishes to find historical support for a law, the judge can pick a different year or look, instead, at state practice over a range of several decades.  These concerns are, primarily, a product of the lack of guidance Bruen provides to lower courts.

In the ECHR context, Professor Dothan suggests that “judges can be prevented from cherry-picking the laws of states that suit their own wishes even if they do not have to survey all states, but only a certain predetermined group of states.”  This concern could, in theory, be addressed in two ways:  require courts to always evaluate every country or state, or limit judges to a subset of countries or states similar in terms of demography, population, GDP, or other measures to the specific jurisdiction at issue.   While the former is easy to do when looking for a modern-day consensus, it’s much trickier when looking at the historical record, in part because judges may be tempted to dictate the outcome by selecting the relevant denominator, as described above.  But some version of the latter approach might be possible in the Second Amendment context.  Instead of using as the denominator all states at a given point in history, such as 1791, a court might be limited to a subset of states that shared certain characteristics with the state whose modern-day law is at issue (whether in terms of demography, population, or availability and use of firearms).  The court would then determine whether a majority of that selected group of states had analogous laws in place at the relevant time.

[1] This post does not argue that international laws regarding firearms should play any role in courts’ analysis of Second Amendment challenges, but merely recognizes that this kind of empirical determination based on practice among different jurisdictions is not a problem unique to American constitutional law.