The Senate’s bipartisan working group on election law reforms, led by Sen. Susan Collins (R‑ME) and Joe Manchin (D‑WV), today unveiled their long-awaited deal on a variety of issues including the Electoral Count Act, the law at the center of the casting and counting of Electoral College votes every four years. 


In addition to Collins and Manchin, the working group also includes Rob Portman (R‑OH), Kyrsten Sinema (D‑AZ), Mitt Romney (R‑UT), Jeanne Shaheen (D‑NH), Lisa Murkowski (R‑AK), Mark Warner (D‑VA), Thom Tillis (R‑NC), Chris Murphy (D‑CT), Shelley Moore Capito (R‑WV), Ben Cardin (D‑MD), Todd Young (R‑IN), Chris Coons (D‑DE), Ben Sasse (R‑NE), and Lindsey Graham (R‑SC). 


At Cato, my colleagues and I have been working on the Electoral Count Act, including analysis of particular provisions, meetings with policymakers, and a recent policy analysis outlining our overall conclusions and providing a model template with a section by section set of recommendations.


From the description released, the proposed “Electoral Count Reform Act of 2022″ covers the most important points, and reflects broad consensus among the scholars and organizations across the political spectrum. These changes include a process for ensuring only a single conclusive set of votes is sent by each state, expedited judicial review by a three-judge panel to handle possible rogue state officials, a higher threshold to raise objections in Congress, and clarifying the “failed elections” provision to ensure that it only covers extreme natural disasters. Between concerns that this bill would be too narrow and only make cosmetic changes at the behest of Republicans, versus prior Democratic plans that went too far and were overcomplicated, this announcement strikes a happy medium: a bill that is broad in scope but simple and conservative in substance. 


The status quo Electoral Count Act, as we saw in 2020, is a ticking time bomb and an invitation to a constitutional crisis. This announcement is a major step forward in fixing that problem. It should be a top priority for every member of Congress who takes seriously their oath to defend the Constitution. 

The Constitution gives federal judges the right to serve “during good behaviour.” In practice, that means federal judges serve for life, choosing whether and when they retire. Studies have shown that judges are more likely to voluntarily retire under presidents of the same party that appointed them, most likely out of a preference to be replaced by judges of a similar ideology.

The tendency of judges to strategically time their retirements is a natural one. Justice Antonin Scalia once noted that he “would not like to be replaced by someone who immediately sets about undoing” what Scalia himself had tried to do during his judicial career. So long as judges have the power to choose when they are replaced, the practice of strategic retirement is likely to continue.

But recently, some judges have seemingly taken strategic retirements a step further. Last year Judge Robert King of the Fourth Circuit revoked his planned transition to senior status (a kind of semi-retirement that triggers a judicial vacancy), reportedly out of dissatisfaction with the particular person the Biden administration planned to choose as his replacement.

A similar story played out in 2018, when Judge Michael Kanne of the Seventh Circuit reversed course on a plan to take senior status. As Judge Kanne candidly admitted, his original decision to take senior status had been “on the consideration that [a former clerk] would be named” as his replacement. When that former clerk told Judge Kanne that he would be passed over for the nomination, Judge Kanne replied “If you’re not going to be named, then I’m not going to take senior status.”

In April of this year, Judge Johnnie Rawlinson of the Ninth Circuit publicly urged that a particular former clerk of hers be nominated as her successor, even though she had not yet decided whether to retire. With little subtlety, Judge Rawlinson told Reuters she could be “persuaded” to take senior status, in an interview in which she further boosted her preferred nominee.

And just this past month, strategic retirements were put back in the spotlight with the news that Judge Karen Caldwell of the Eastern District of Kentucky had made her intention to take senior status known to a select few political actors (including President Joe Biden and Senator Mitch McConnell) and that a deal had been struck to name a particular replacement before her decision became public. At least one report alleges that Judge Caldwell too conditioned her retirement on that particular choice of replacement. (It should be noted that the deal for her intended replacement has since fallen through, but there is no indication that Judge Caldwell intends to revoke her transition to senior status.)

Whatever the full story may be in any of these particular cases, the prospect of judges influencing the selection process for their successors raises serious concerns. When judges use their power of discretionary retirement to influence the political branches into picking a particular judicial nominee, they arguably cross the line into impermissible political activity. At the very least, it is not ideal for judges who must often sit in judgment of the other two branches to at the very same time be involved in negotiations with the political actors in those branches. As Professor Richard M. Re observes in a forthcoming law review article, “interbranch bargaining is ill-suited to the assertedly independent federal judiciary.”

Judicial influence over appointments also arguably warps the Constitution’s intended separation of powers, under which it is only the president and the Senate who have a role in nominating and confirming judges. As Laurie Lin and David Lat put it in a Wall Street Journal op-ed, “judicial seats aren’t property to be bequeathed.” Acceding to a judge’s demand for a particular replacement in the face of a threat to un-retire would, in Professor Josh Blackman’s words, “give judges a veto power over presidential nominations.” And Professor Re calls the idea of judges exerting such control “inimical to the role of federal courts and at odds with the constitutionally delineated process for selecting officials[.]”

If this emerging practice is indeed problematic, what can be done to stop it? Professor Re proposes making judicial retirement letters formally binding and irrevocable (either by court rule or by statute), so that judges cannot renege on a commitment to retire out of opposition to the choice of their successor. Lin and Lat propose that the Code of Conduct for U.S. Judges should be revised to make explicit that “judges putting conditions on their retirement, or entertaining offers from the White House regarding their successors, is unacceptable.” (This approach is also supported by Gabe Roth of Fix the Court.) And Professor Blackman urges the president, when faced with a judge’s threat to un-retire, to stick with his chosen nominee and “hold the line to avoid setting a precedent,” even if it means losing the opportunity to fill a particular seat.

These are all promising reform proposals, but there is an even more sweeping option that would prevent judges from engaging in any retirement-related gamesmanship in the future: entirely decoupling judicial appointments from judicial retirements.

Federal law currently defines the jurisdiction of the various district and circuit courts and assigns each of those courts a fixed number of seats. Thus, the timing for when a president may make an appointment to any particular court is dependent on when particular incumbent judges leave. Because the number of seats on each court is fixed, an appointment can only be made when a particular seat falls vacant.

But nothing in the Constitution requires that the size of the courts be defined by a fixed number of seats. Instead, federal law could define the size of each court by setting new appointments to occur at some regular, fixed interval. The shorter the interval for a particular court, the more frequent would be appointments to that court, and the more judges would sit on that court at any given time (on average). Under such a system, the exact total number of judges on each court at any given time would fluctuate a bit, because these appointments at regular intervals would be entirely unaffected by when any sitting judges choose to leave the court.

The advantage of such a system is that it would completely eliminate the leverage federal judges currently have over the political branches, by removing judges’ power to grant (or revoke) the boon of an appointment opportunity. And unlike judicial term limits (another proposal that would lead to appointments at regularly intervals), such a system would retain judicial life tenure and thus could unquestionably be attained merely by statute, without the need for a constitutional amendment.

For multiple reasons, decoupling appointments from retirements would be easiest to accomplish at the district court level. At the Supreme Court level, any fluctuation in the number of sitting justices at a given time could significantly alter the balance of power (and when the total number of justices is an even number, there is a much greater likelihood of tie votes). At the circuit court level, fluctuations in the total number of active judges could likewise shift balances of power in en banc proceedings and could lead to temporary circumstances where an en banc court becomes so large as to be unwieldy.

But none of these concerns are present at the district court level, which makes that level the natural place to experiment with such a reform. District court judges preside alone, and there is no such thing as an en banc proceeding at the district court level. This means that fluctuating total numbers of active judges in any particular district would in no way affect the proceedings in any particular case. True, shifting numbers of judges can affect the odds of being assigned a judge of any particular ideology in a given district. But shifts in the total number of judges already regularly occur whenever one district judge takes senior status and another judge joins the court, or whenever a senior status judge fully retires. Finally, district judges can be temporarily reassigned to other districts with relative ease, making temporary fluctuations in particular districts less problematic from a workload perspective.

In addition, it would be relatively straightforward to “translate” the current statutory size of each district court into an appointment interval that would sustain that same approximate size (on average). The only variable Congress would need to know is the average total tenure of active federal district judges, a statistic that would be simple to calculate. If we suppose for the sake of easy math that the average such tenure is currently 20 years (likely a reasonable estimate), then Congress could simply define the appointment interval for each district court as one appointment every 20/N years, where N is the number of seats assigned to that district under current law. So a 5‑judge court under current law would switch to a court with an appointment every four years, a 10-judge court would see an appointment every two years, and a 20-judge court would see an appointment every one year.

Such a change would certainly be more drastic than an amendment to the judicial ethics code, and other practical difficulties could make it a less attractive option. Keeping each court at a reasonable number of judges for its caseload might require temporary judicial reassignments that last for years (or short-term statutory adjustments). When the Senate and presidency are controlled by different parties, this might result in appointments not being made at their scheduled times. Judges may remain in active service for longer, on average, if unable to influence their successors, and this may lead to unintended consequences. I do not claim to have fully worked out every potential issue, nor do I speak for any of my colleagues in suggesting that this option should be considered.

Rather, my point is to emphasize that Congress shouldn’t keep any options off the table at this stage, no matter how out-of-the-box. If explicit influence over the judicial selection process continues to emerge as a norm accepted among some judges, then it may be necessary to significantly rethink the inherent political power that the current system gives to sitting judges. The only way to ensure that judges stay out of the appointments process entirely is to remove their power to trigger that process in the first place.

By a 6–3 vote, the Supreme Court in West Virginia v. Environmental Protection Agency took an important step toward restoring constitutional balance to federal policymaking. Though Washington Post columnist George Will probably overstates the case in calling it the “term’s most momentous decision,” he is right that this is a big deal. As I discuss below, the Cato Institute perhaps played a role in this welcome result.

So, what happened? Setting aside the case’s backstory (explained here), the immediate result is that the EPA doesn’t have the power to impose a nationwide cap-and-trade climate policy based on an “ancillary” part of the law that no one had heard of before the Obama administration. The Biden administration is working on a significant climate rule based on the very statutory provision at issue in West Virginia v. EPA, so the Court’s holding provides guidance as to what the EPA cannot do.

But it’s how the Court reached this result that will have lasting consequences. In ruling against the government, Chief Justice Roberts’s majority opinion “announces the arrival of the ‘major questions doctrine,’ as put in a dissent by Justice Kagan.

And what is the “major questions doctrine”? It is, the Chief Justice explains, no more than “common sense” regarding how Congress works. Basically, it’s the Court’s belief that Congress will be clear when it assigns major policymaking authority to regulatory agencies. In practice, this means that “major” domestic policy must emanate from the votes of elected lawmakers rather than from expansive legal interpretations devised by unelected bureaucrats. Again, this is common-sense stuff. Under our Constitution, lawmakers are supposed to pass laws to make major policy.

Prior to yesterday, in a handful of decisions over the last 25 years, the Supreme Court had relied on reasoning that resembled what scholars came to call the “major questions doctrine.” But it was all circumspect and indirect. Indeed, no majority opinion even used the term “major questions doctrine.” As a result, this interpretive principle was inchoate. Regulated parties often invoked the concept in challenging agency rules, but lower courts had no idea how to identify a “major question.”

All that changed yesterday. Chief Justice Roberts didn’t just officially recognize and rely on the major questions doctrine to check the EPA’s (ludicrous) statutory interpretation. He also provided lower courts with guidance how to identify major questions. Of course, an agency’s rule must be economically and politically significant to trigger the doctrine. Also, the rule would have to be based on an expansive interpretation of ambiguous statutory text. Other red flags include whether the agency is doing something unprecedented, or if the agency is attempting to do something that Congress failed to do, or if the “nature” of the law doesn’t comport with the agency’s claims to power. Still another red flag is when the agency is operating outside its expertise. All these boxes were checked with the EPA’s climate rule at issue in West Virginia v. EPA.

Many commentators are up in arms about the decision; they claim it will take a wrecking ball to the administrative state. They are wrong. Just because the EPA can’t impose a nationwide cap-and-trade for climate change, it doesn’t mean the agency is “gutted.” As the Chief Justice noted, the EPA retains broad authority to regulate greenhouse gases. More generally, as noted by Chief Justice Roberts, the major questions doctrine will come into play only for those “extraordinary” regulations that evince the circumstances identified in the prior paragraph. In the past, agencies issued these sorts of “major” policies only a handful of times per presidential administration. Anyone who claims this decision would undermine the administrative state simply doesn’t know what they’re talking about.

That’s not to discount the decision’s effects! Even though there haven’t been many regulations that would run afoul of the major questions doctrine, it doesn’t mean those instances weren’t highly deleterious. After all, Congress only passes, at most, a handful of major laws during any given presidential administration. The practical problem with the executive branch interpreting vague old laws to make “major” policy is that there’s no permanency. What any one presidential administration can do, the next can undo. We saw this with the EPA’s climate rule at issue in West Virginia v. EPA. Obama ordered the EPA to issue the rule; Trump ordered the EPA to replace Obama’s rule; and then Biden ordered EPA to replace Trump’s rule. The electricity sector—a very important industry!—was caught in the spin cycle. Chief Justice Roberts just put a stop to the chaos engendered by the worst excesses of executive lawmaking.

Unlike regulations, laws endure. If Congress directs an agency to take on a major question, then the agency will perform that role, regardless of the president’s political orientation. From now on, when it comes to “major” policymaking, the ball is in Congress’s court, as the Constitution intended.

A couple final notes:

• Chief Justice Roberts’s majority’s opinion was masterful. In fact, the Chief Justice has a history of making important (and welcome) administrative law doctrine, as I explain at this Notice & Comment post.
• At the risk of sounding self-important, I believe that the Cato Institute might have influenced the Court’s thinking. While other parties argued that the EPA’s climate rule clearly implicated the major questions doctrine, our amicus briefs (joined by the Mountain States Legal Foundation) were the only ones to ask the Court to flesh out the doctrine to resolve confusion in the lower courts. To that end, we had pitched a framework for identifying major questions. And the majority opinion tracks our framework. The upshot is that we either contributed to the court’s thinking, or we predicted exactly what the Court would do. Regardless of whether our input was coincidental or influential, yesterday was a great day for liberty.

Today the U.S. Supreme Court unanimously overturned lower court decisions in the consolidated cases of Ruan vs. United States and Kahn vs. United States. The two physicians were convicted of prescribing opioid pain medicine “outside the usual course of [medical] treatment” and were sentenced to prison.

The jury in Ruan was not instructed to consider Dr. Xiulu Ruan’s “good-faith defense,” i.e., that he was indeed prescribing the drug “legitimately” to treat pain based upon his good-faith assessments of his patients’ medical contexts and requirements. The Eleventh Circuit Appeals Court denied Ruan’s request to vacate the lower court decision on the grounds that he was denied a “good faith” defense.

The Supreme Court held “the Government must prove beyond a reasonable doubt that the defendant knowingly or intentionally acted in an unauthorized manner.” The Cato Institute filed an amicus brief in Ruan, as did I, separately, as a private practicing physician along with the Association of American Physicians and Surgeons (AAPS).

While the decision was unanimous, Justice Alito was joined by Justices Thomas and Barrett in filing a separate concurring opinion. I will defer judicial analysis of the two separate opinions to my colleague Trevor Burrus. Here I offer the perspective of medical doctor in clinical practice.

There are always robust debates among clinicians regarding the proper and rational treatment of a host of conditions, from high blood pressure to diabetes to acute and chronic pain—and that patients and their clinical contexts vary—that there is no ONE RIGHT WAY to treat a wide range of medical conditions, treatment must be individualized, and clinicians must remain open to modifications and adjustments along the way. In Ruan vs. United States, the government failed to consider this.

Prescribing medications in an unusual manner, or a manner that falls outside the mainstream, might be a “standard of care” or malpractice issue, but should not be automatically considered a criminal issue. Approximately 20 percent of medications approved by the Food and Drug Administration (FDA) are legally prescribed “off-label,” i.e., for different purposes than those for which the FDA approved them. The originators of off-label uses fall outside the mainstream of prescribers, but they are not treated as criminals. And many off-label uses are later approved by the FDA. This is one of the ways clinical medical science advances.

The lower court convictions of Drs. Ruan and Kahn treated what, at worst, could have been medical malpractice or standard of care violations as criminal matters. This amounted to, in effect, cops practicing medicine. Cases and convictions like these have sent chills up the spines of other health care practitioners who are trying to help their patients in pain, causing many pain patients to be under-treated or, worse, abandoned.

Today’s Supreme Court decision was a victory for the “good faith” defense. But it was also a victory for physician autonomy, the patient-doctor relationship, and patients in pain.

Surprising no one in the wake of the leaked draft opinion, the Supreme Court today overturned Roe v. Wade and Planned Parenthood v. Casey and held that there is no constitutional right to elective abortion access. Some will celebrate that decision for returning one of the most contentious policy issues in American life to the legislative branch, while others will denounce it as a repudiation of personal autonomy and a further example of the judiciary’s historic failure to afford coequal legal status to women.

Cato did not file an amicus brief in Dobbs v. Jackson Women’s Health Organization because we believed—and continue to believe—that reasonable libertarians can disagree about whether the Constitution, properly understood, has anything to say about abortion and if so, what. In this post, we will explain why libertarian ideology does not supply a clear answer to the moral question of abortion, why the constitutional question is not as straightforward as absolutists on either side contend, and what lessons we can draw from the majority opinion in Dobbs.

I. Libertarianism tolerates a wide range of views on the policy question of abortion access

Libertarianism, in the most general sense, is the political philosophy that holds individual liberty as the central political value. It is a broad concept without rigid boundaries, but those who embrace its tenets hold that people should be presumptively free to make choices about their own life and that what people do with their bodies and their property is up to them, provided only they do not cause harm to non-consenting parties. Thus, any claim of authority to infringe upon this domain of individual liberty—whether made by a government or a private party—necessarily requires strong justification.

From this perspective, it is hardly surprising that state regulation of abortion access is a grave concern for many libertarians. If bodily autonomy entails the right to decide, for example, what food to eat, whether to use drugs, and what medical treatments to accept or refuse, then it surely entails the right to reproductive liberty as well. There are few decisions more intimately personal than deciding whether to have a child, and few developments with greater consequence for one’s life and body than pregnancy and childbirth.

Moreover, there is ample reason to doubt the efficacy of state attempts to control an activity that a huge number of people are determined to engage in regardless. Alcohol and drug prohibition appear to do little to discourage most people from consuming drugs and alcohol; but they certainly do set the stage for black markets that empower criminal organizations, reduce safety for consumers, and entangle otherwise law-abiding citizens in the many pathologies of the criminal-justice system. Abortion prohibition, whatever its merits in principle, will certainly entail some or all of these problems and will, like all prohibitory policies, inevitably raise the question, “How much violence are we prepared to employ against people who disagree with this policy in order to enforce it?”

For these and other reasons, a clear majority of self-professed libertarians describe themselves as “pro-choice.” But of course, abortion access is, at least debatably, not solely a question of personal bodily autonomy. The heart of the “pro-life” position is that unborn children—at some point during pregnancy, and perhaps as early as conception—become distinct rights-bearing entities entitled to moral consideration for their own sake. To those who start with such premises, “my body, my choice” is no more persuasive an argument than “my property, my choice” would have been to an abolitionist. Both slogans beg the relevant question, because whether it is just the pregnant woman’s body (or just the slaveowner’s property) is the precise issue under debate.

To be sure, there are many sensible arguments for holding that fetuses are not entitled to the same quasi-deontological moral consideration that other members of a polity are entitled to—for example, that they lack the sort of reflective self-awareness that gives someone an independent sense of their own self and their future. Or that, at early stages of pregnancy, they lack the capacity to feel pain. Or that even granting that a fetus has a right to life, that doesn’t give it a right over the woman’s body.

The point of this particular discussion isn’t to resolve these debates. Rather, it’s to illustrate that these are questions of moral philosophy. And libertarianism is, fundamentally, a political philosophy that can arise out of many different ethical traditions (consequentialism, natural rights, common-sense moral intuitions, etc.). Thus, even if libertarians were in complete accord on the state’s practical obligation to rights-bearing entities, they might still disagree on what makes something a rights-bearing entity in the first place. What of living humans with brain damage that prevents them from achieving conscious awareness? Or the higher animals outside our political community, which nevertheless demonstrate signs of true self-awareness? Or animals that, even if lacking a higher sort of consciousness, nevertheless have the capacity to suffer? Or, more speculatively, computer-emulated human brains?

There is a not an obvious “libertarian position” on any of these questions because libertarianism does not purport to be a complete theory of personhood and moral obligations. Thus, it is hardly surprising that libertarians disagree on abortion, not because of fundamental differences in (or compromises with) their general political philosophy, but because of different views on what sort of moral consideration is due to fetal life.

Of course, the fact that libertarians can reasonably disagree on this issue doesn’t mean individual libertarians should refuse to take a position, and the fact that there are reasonable arguments on both sides doesn’t mean that the issue is incapable of resolution. But we do not think Cato, as an institution, should attempt to offer a definitive answer to a moral question dividing libertarians generally, as well as Cato scholars and staff specifically.

II. Libertarian legal theory and the constitutional status of abortion

Granting that libertarians might have different views about abortion rights as a policy matter, is there nevertheless a clearly correct position for libertarians to take regarding the constitutional status of abortion? To be sure, there is nothing inconsistent with a libertarian supporting abortion rights as a policy matter but opposing the Court’s abortion jurisprudence as a legal matter. Indeed, many libertarians in fact hold that precise view. But we believe the constitutional question is also one about which reasonable libertarians can disagree and that aspects of it may well be as difficult as the intractable disagreements among leading lawyers, judges, and scholars suggest that it is—especially given that courts had recognized a constitutional right to abortion access for half a century.

Most libertarians start with a shared set of beliefs about determining what rights the Constitution does and does not protect. First, human beings possess certain “unalienable” natural rights that are neither conferred by nor dependent upon any legal charter, including the Constitution. Second, the U.S. Constitution neither lists nor purports to list all of the distinct legally enforceable rights that people possess. To the contrary, libertarians recognize and give force to the many provisions of the Constitution that provide explicit textual support for protecting that broader category of liberty often labelled, perhaps misleadingly, “unenumerated rights”—most obviously, the Ninth and Tenth Amendments and the Fourteenth Amendment’s Privileges or Immunities Clause.

Uncontroversial examples of so-called “unenumerated rights” include the freedom to travel around the country (nowhere explicitly mentioned in the text of the Constitution but never seriously questioned), the ability to direct the upbringing of one’s children, and the right to defend oneself against violent and unjustified attack. The Supreme Court has also held, in a decision that even most self-professed textualists and originalists appear to accept, that the Constitution protects a right of access to contraception.

Finally, many libertarians believe in what Professor Randy Barnett calls “the presumption of liberty,” which entails that the government bears the burden of explaining and justifying its use of force to constrain people’s freedom—and not merely with respect to the exercise of rights deemed “fundamental” by the Supreme Court, such as freedom of speech or religion.

In short, most libertarians reject the idea that the Constitution only protects a discrete number of specifically identified rights, and instead understand the Constitution to protect a fairly broad—though admittedly finite and difficult to formulate with precision—set of rights, particularly including those with momentous implications for a person’s life, such as where to live, which career to pursue, whether and whom to marry, whether to have children, and how to raise them if so.

What about the historical argument that an originalist interpretation of the Fourteenth Amendment cannot support a right to abortion because, as the Dobbs opinion asserts, “when the Fourteenth Amendment was adopted, three quarters of the States made abortion a crime at all stages of pregnancy”? This is certainly a more sophisticated argument then the ipse dixit that the Constitution doesn’t use the word “abortion,” and it may indeed be persuasive evidence.

But that history is still not conclusive on the constitutional question—at least not to us—because Cato scholars generally embrace the version of originalism based on original public meaning, not original intent. In other words, originalism is about interpreting words and phrases as they would have been understood at the time of adoption, not about the subjective intent of those who adopted them. And especially with respect to constitutional provisions written at a high level of generality, those approaches may well conflict. After all, the same Congress that adopted the Fourteenth Amendment also segregated public schools, and yet even Robert Bork acknowledged the correctness of Brown v. Board of Education on originalist grounds. As he wrote in The Tempting of America:

Since equality and segregation were mutually inconsistent, though the ratifiers did not understand that, both could not be honored. When that is seen, it is obvious the Court must choose equality and prohibit state-imposed segregation. The purpose that brought the fourteenth amendment into being was equality before the law, and equality, not separation, was written into the law.

It is thus unsurprising that the Fourteenth Amendment, properly understood, would lead to many conclusions unanticipated by its Framers, yet consistent with the original meaning of the broader principles that actually comprise the text they adopted. Cato, for example, filed multiple amicus briefs arguing that the text and original meaning of the Equal Protection Clause protect all people from arbitrary and invidious discrimination, and thus preclude states from limiting marriage to opposite-sex couples.

The point of this discussion isn’t to suggest that the answer to such constitutional questions is straightforward or obvious; rather, it’s to illustrate that sophisticated originalism, at least with respect to those constitutional provisions written at a high level of generality, requires not just interpretation, but construction. In other words, just knowing the semantic meaning of say, the Privileges or Immunities Clause or the Equal Protection Clause isn’t enough to decide individual cases. You also need doctrinal rules that flesh out how to apply these broad principles in specific circumstances.

Of course, even accepting this sophisticated understanding of originalism, there are still many valid grounds to criticize how the Court has constitutionalized abortion rights in particular. Unlike nearly all of the other “unenumerated rights” cases—such as Meyer v. Nebraska (raising and educating children), Griswold v. Connecticut (access to contraception), or Lawrence v. Texas (homosexual sex)—the right to obtain elective abortions involves not just the liberty of the person subject to state regulation, but also, at least arguably, the life and liberty of another person. And this is why the Dobbs opinion described the abortion right as “critically different from any other right that this Court has held to fall within the Fourteenth Amendment’s protection of ‘liberty.’” Indeed, one of the most powerful objections to the reasoning in Roe itself is that, while the opinion asserted that the Court “need not resolve the difficult question of when life begins,” the Court was necessarily deciding that the state could not constitutionally protect the fetus for its own sake until the third trimester.

Yet even conceding that the reasoning of Roe itself is quite muddled, the question today is complicated by the fact that a constitutional right to abortion access had been acknowledged and preserved for half a century. And while the Court has regularly reversed itself in the direction of expanding constitutional rights, it has rarely reversed itself in the opposite direction of revoking constitutional rights once protected. Indeed, the only other explicit example of that in the Court’s history is West Coast Hotel Co. v. Parrish, which abandoned constitutional protection for the liberty of contract and labor previously acknowledged in Lochner v. New York and Adkins v. Children’s Hospital.

That is not to say that rescinding protection for constitutional rights should be unthinkable, but rather that there should be an extraordinarily high bar for such a momentous act. The legal deficiencies of Roe and Casey may well be “obvious” to some, but there are some scholars and judges who see errors just as “obvious” in cases like Citizens United v. FEC (campaign finance), District of Columbia v. Heller (gun possession), or Janus v. AFSCME (compelled support of public employee unions). Proponents of reversing Roe and Casey should at least consider what a reversal might portend in future years, when the ideological makeup of the Court might be quite different.

III. The Court’s process for identifying unenumerated rights that are entitled to meaningful judicial protection is un-originalist, ends-oriented, and deeply flawed

Even assuming the Dobbs majority is correct in its ultimate conclusion—that the Constitution does not protect a general right to elective abortion—the opinion as written falls well short of answering that question with sufficient originalist rigor, because its analysis relies almost entirely on the “Glucksberg test” for assessing so-called “unenumerated rights.” Washington v. Glucksberg was a 1997 Supreme Court decision holding that there is no constitutional right to assisted suicide, in which the Court held that “substantive due process” protects only those fundamental liberties that are “objectively, deeply rooted in this Nation’s history and tradition.” But the Dobbs majority offers little justification for this test itself, and there is ample reason to question it.

First, Glucksberg concerns the doctrine of substantive due process, yet Dobbs all but ignores that the Privileges or Immunities Clause may be the more natural textual basis for protecting “unenumerated rights” than the Due Process Clause. In the Slaughter-House Cases in 1873, the Supreme Court effectively read the Privileges or Immunities Clause out of the Constitution, holding that this provision protects only those rights that “owe their existence to the Federal government,” such as the right to access federally navigable waterways and protection against piracy on the high seas. Thus, it excluded from protection anything that might be termed natural liberty, such as the rights to acquire and possess property, make contracts, pursue safety and happiness, along with all the liberties mentioned in the Bill of Rights. However, despite overwhelming consensus that this narrow interpretation is wrong, Justice Thomas and Justice Gorsuch are the only members of the current Court that have expressed interest in revisiting this grievous mistake.

To be sure, the Dobbs majority dropped a footnote asserting that its conclusion would be the same under the Privileges or Immunities Clause, which would require the same “deeply rooted” analysis as Glucksberg. But that ipse dixit ignores the rich scholarship explaining how this long-ignored provision is best understood as requiring the state to demonstrate legitimate rationales when it restricts liberty. In other words, the Court entirely ignores the argument that the Privileges or Immunities Clause requires a “presumption of liberty.”

Second, the Glucksberg test requires that a right be “carefully defined” before assessing whether it is “deeply rooted,” but there are different levels of generality at which a right can be defined. For example, the dispute in Bowers v. Hardwick—which upheld a state prohibition on “sodomy”—turned in part on whether the relevant right at issue was “a fundamental right [of] homosexuals to engage in sodomy” or the right of adults “to decide for themselves whether to engage in particular forms of private, consensual sexual activity.” One of these rights is much easier to characterize as “deeply rooted,” yet Glucksberg offers no real guidance on the “correct” way to frame such questions. This problem of generality-shifting is thus closely related to another major criticism of Glucksberg, which is that a narrow emphasis on rights that are “deeply rooted” has the inevitable effect of disfavoring rights that would specifically benefit historically marginalized groups.

Similarly, consider the D.C. Circuit’s 2007 en banc decision Abigail Alliance v. Eschenbach, which ruled against terminally ill patients who wanted to use experimental drugs not yet approved by the FDA. The majority framed the relevant question as “whether terminally ill patients have a fundamental right to experimental drugs that have passed Phase I clinical testing.” Of course, given the relatively recent history of FDA regulation in this country, a right defined so narrowly would, nearly by definition, not be “deeply rooted.” Yet as the dissent explained, the “right to act to save one’s own life” absolutely was deeply rooted in our nation’s history. Assuming this was the appropriate level of generality, then the majority erred by “conflating the right with the deprivation.”

Third, the Glucksberg test fails even to protect those fundamental rights that clearly are “deeply rooted” by any reasonable standard. The most obvious example here would be the right to pursue a lawful occupation, and thus to be free of occupational-licensing restrictions enacted solely for arbitrary or protectionist reasons. In the Slaughter-House Cases themselves, Justice Field explained in his dissent how “privileges or immunities” included “the right to pursue a lawful employment in a lawful manner, without other restraint than such as equally affects all persons.” Moreover, protecting this particular aspect of liberty—which had been so flagrantly denied to both slaves and freedmen—was a core motivation for passage of the Fourteenth Amendment itself. John Bingham, principal framer of the Fourteenth Amendment, wrote that its enactment would protect “the liberty … to work in an honest calling and contribute by your toil in some sort to the support of yourself, to the support of your fellowmen, and to be secure in the enjoyment of the fruits of your toil.”

Nevertheless, at least since the infamous Williamson v. Lee Optical decision of 1955, the Supreme Court has given no meaningful protection to this fundamental right, declaring that occupational-licensing barriers are subject only to the “rational basis test”—which explicitly requires judges to ignore the actual reasons for restrictions on occupational freedom, and instead imagine possible rationales a legislature might have had. The utter failure of Glucksberg to check infringements on vocational liberty is the clearest indication that the test was never intended to be a principled doctrinal mechanism for identifying and protecting fundamental rights; rather, it is a results-oriented, endlessly manipulable artifice that allows judges to sidestep protecting fundamental rights entirely.

Again, there are reasonable arguments that—even taking into account a more robust understanding of the Privileges or Immunities Clause and fundamental rights—the Fourteenth Amendment still doesn’t protect a general right to elective abortion. But as Professor Evan Bernick recently stated, it makes no sense to hail Dobbs as a “triumph of originalism” when the opinion “barely perceives an obligation to be originalist at all.”

Conclusion

In sum, reasonable minds—including reasonable libertarian minds—can disagree about the moral nature of abortion itself, the constitutional status of abortion rights, and the proper legal framework with which to evaluate asserted claims of a right to abortion access in particular circumstances. Likewise, critics of the Court’s abortion cases can reasonably disagree on the most principled way to address precedents whose reasoning arguably has not withstood the test of time.

In the aftermath of Dobbs, some states may well attempt to regulate abortion in ways that do pose grave concerns for libertarians generally and Cato specifically, not because of abortion itself but because of more general constitutional principles. For example, many states are pursuing plans not just to prohibit abortion within their own borders, but also to punish residents who cross state lines to receive an abortion in a state where its legal. Similarly, if Congress attempted to either prohibit abortion or enshrine abortions rights nationwide, that would inevitably raise questions about whether such acts are within the scope of Congress’s enumerated powers at all.

If and when such questions arise, Cato will not hesitate to advance the principles of federalism and limited government that we always have. But as to the issues in Dobbs itself, we don’t believe these questions have an answer so obvious or uncontroversial that principle demands adopting an institutional position that will undoubtedly be at odds with the legal judgment of one group or another of our respected colleagues.

Every lawyer remembers the unpleasant rite of passage that is taking the LSAT, the standardized test for law school admissions. The most distinctive component of the LSAT is known colloquially as the “logic games” section (though its formal name is “analytical reasoning”). This section, full of hypothetical dinner parties with picky guests and distinctive seating requirements, is meant to test the ability of future lawyers to apply the types of overlapping and interacting rules that we often must untangle to apply the law.

A much-simplified version of the type of question one might see in the logic games section is as follows: A restaurant calls its happy hour menu the “5:00 PM and after” menu, but the menu is actually available up to an hour before 5:00 pm as well as up to an hour after 5:00 PM. Yesterday, John ordered off the “5:00 PM and after” menu. Today, John ordered at the same restaurant an hour earlier than he did yesterday. Was the happy hour menu necessarily available when John ordered today?

If you answered “no,” you just successfully avoided a logical fallacy that multiple federal courts of appeals had fallen victim to, before finally being corrected by the Supreme Court yesterday. The history of these decisions is a case study in how flimsy logic and basic errors can be perpetuated by too much deference to precedent, as well as (perhaps) some motivated reasoning.

Yesterday’s decision in United States v. Taylor concerned whether attempted robbery under federal law qualifies as a “crime of violence.” If it does, then using a gun during that attempted robbery can result in a second conviction and a longer sentence.

The relevant law defines a “crime of violence” as a crime that “has as an element the use, attempted use, or threatened use of physical force.” Because the law’s text requires that at least one of these three options must be “an element” of the crime, the Supreme Court has previously held that a “crime of violence” must be a crime that involves at least one of these options in every single case. This approach, known as the “categorical approach,” asks whether conviction of a crime necessarily requires that the defendant must have performed at least one of these options. In other words, if it is logically possible to be convicted of a crime without using, attempting to use, or threatening to use physical force, then that crime is not a “crime of violence.”

The defendant in yesterday’s case had been convicted of attempting to commit robbery under a federal law called the Hobbs Act (references to “robbery” in the remainder of this piece will mean robbery as defined in the Hobbs Act). That law defines completed robbery as unlawfully taking personal property “by means of actual or threatened force.” There is thus little doubt that completed robbery qualifies as a crime of violence: to be convicted of robbery, a person must either use or threaten force. Thus, it is logically certain that every person convicted of completed robbery must have been found to have committed at least one of the three options for a “crime of violence.”

But what about a person only convicted of attempted robbery? Such a person, under the relevant law, has intended to commit robbery and taken a substantial step toward committing robbery, but failed to complete the robbery. And since robbery itself can be accomplished by either “actual” or “threatened” force, attempted robbery can be accomplished by attempting to commit either “actual” or “threatened” force.

In those cases where the attempted robbery consists of an attempt to commit actual force, the crime of course involves the attempted use of physical force, which is one of the three options for a crime of violence. But what about the other possibility, those cases where the attempted robbery consists of an attempt to commit threatened force? As the Supreme Court explained yesterday, that possibility does not entail any of the three options for a “crime of violence,” which means attempted robbery itself cannot qualify as a crime of violence.

Justice Neil Gorsuch, writing the opinion of the Court and joined by six other justices, laid out a simple hypothetical to show why:

Suppose Adam tells a friend that he is planning to rob a particular store on a particular date. He then sets about researching the business’s security measures, layout, and the time of day when its cash registers are at their fullest. He buys a ski mask, plots his escape route, and recruits his brother to drive the getaway car. Finally, he drafts a note—“Your money or your life”—that he plans to pass to the cashier. The note is a bluff, but Adam hopes its implication that he is armed and dangerous will elicit a compliant response. When the day finally comes and Adam crosses the threshold into the store, the police immediately arrest him. It turns out Adam’s friend tipped them off.

In such a case, Adam is guilty of attempted robbery because he attempted to threaten the cashier for money. But “Adam did not ‘use’ physical force. He did not ‘attempt’ to use such force—his note was a bluff and never delivered. And he never even got to the point of threatening the use of force against anyone or anything. He may have intended and attempted to do just that, but he failed.”

Put simply, a failed attempt to threaten force is itself neither an attempt to use force nor a threat to use force. And so attempted robbery cannot be a crime of violence under the categorical approach. Indeed, even the two dissenting justices, Clarence Thomas and Samuel Alito, did not dispute this point. Rather, they each contested the premise as to whether the categorical approach (as currently defined) is in fact the best approach to applying the statute. None of the nine justices disputed that under the categorical approach as it has been previously defined, attempted robbery cannot qualify as a crime of violence.

Yet remarkably, three federal appellate courts in a row had held just the opposite, until the Fourth Circuit finally became the first federal appellate court to reach the same conclusion that the Supreme Court would later affirm. Each of those three appellate courts, the Eleventh, Seventh, and Ninth Circuits, had held that because completed robbery itself is a crime of violence, attempted robbery must entail the attempted use of force. The Eleventh Circuit, the first to reach this conclusion, reasoned that “a completed Hobbs Act robbery itself qualifies as a crime of violence … and, therefore, attempt to commit Hobbs Act robbery requires that [the defendant] intended to commit every element of Hobbs Act robbery, including the taking of property in a forcible manner.” The Seventh and Ninth Circuits later followed this reasoning to reach the same result.

This is an error as basic as failing the logic puzzle at the start of this piece. Even though the happy hour menu is called the “5:00 PM and after” menu, it is available as early as 4:00. Its mere name does not mean John actually ordered at 5:00 PM or later yesterday. And since John might have ordered as early as 4:00 yesterday, he might have ordered as early as 3:00 today, before the happy hour menu was available.

Similarly, despite the name, not every crime of violence actually involves the use of violence. Indeed, courts know that crimes of violence don’t necessarily entail the use of violence when they perform the categorical approach and apply its three-pronged definition. Yet immediately after performing this test, the same court can seemingly forget this definition and act as if, based on the name, all crimes of violence must entail the actual use of violence (and thus that all attempted crimes of violence must entail the attempted use of violence). The possibility of an attempt to threaten is completely overlooked, just like overlooking the possibility that yesterday’s happy hour order was itself in the earliest possible hour.

The fact that three consecutive panels of intelligent jurists confidently made such a basic error is a reminder to be wary of reflexive deference to precedent. Perhaps it seemed to these judges that attempted robbery must be a crime of violence, and this intuition led them to excuse less-than-rigorous logic. And perhaps these judges believed that ruling the other way would have been viewed as unorthodox, counterintuitive, and more likely to be reversed. When all these factors point toward the same “safe” answer, those forces can be difficult to overcome.

Both the Fourth Circuit and the lawyers for the defendant Taylor should thus be praised for sticking with their conviction that their logic was simply correct and would win out in the end. As the Fourth Circuit succinctly put it, the premise that an attempted crime of violence must itself be a crime of violence “simply is not so.” Had the Fourth Circuit not bucked the trend of decisions holding otherwise, it may have taken much longer for the question to reach the Supreme Court and for the error in those other circuits to be corrected.

Justice Scalia used to award an “E pur si muove” prize to district judges whose opinions were finally vindicated at the Supreme Court despite being erroneously reversed by an appellate court. A new award may be in order for appellate panels willing to reject unanimous sister-circuit precedents who later find themselves similarly vindicated.

As Colleen Hroncich and Solomon Chen have observed nearby, the Supreme Court yesterday ruled that the state of Maine violated the Constitution when it excluded schools that engage in religious instruction from an otherwise generally available program of tuition assistance payments to parents. That is the outcome the Cato Institute urged in our Carson v. Makin amicus brief last fall, written by Ilya Shapiro and Trevor Burrus. For more on the ruling and on the distinctive features of the Maine program, which assists parents who live in remote areas without a public secondary school, see also Neal McCluskey’s new Cato Podcast with Caleb Brown and Ilya Somin’s opinion piece for NBC.

It’s been widely observed that religious-liberty litigators have been on a winning streak lately, but it is worth noting specifically how many of those cases have been won by framing each dispute in terms of anti-discrimination principles. Masterpiece Cakeshop and Fulton v. Philadelphia come to mind, but also the Court’s string of rulings in favor of churches against local COVID restrictions on grounds that they were not treated as well as one or another secular institution.

To some extent this must reflect the simple truth that anti-discrimination principles have become a ruling theme both in law and in public life generally, exceedingly hard to argue against. If you can successfully invoke them you’re well on the way to winning an argument.

Examples from the progressive side are legion. But conservatives too are now apt to couch gripes about, say, social media moderation in terms of discrimination. Companies with a no-guns-on-premises rule are “discriminating” against gun-owning employees, we’ve heard.

And so on. Why, it was asked, should cruise ship lines or nurse staffing agencies be permitted to engage in “discrimination” toward the unvaccinated?

Now, most of us would rather win arguments than lose them, and it is only natural to gravitate toward whichever framing is most likely to do that. Compare “freedom to marry” (for the record, the phrase I favor) to the more widely employed “marriage equality.” I’d argue, though, that something is often lost in the pressure to re-couch claims as equality claims. Sometimes what’s really on our mind is something else — individual liberty, rights of self-defense or bodily integrity, fair process, institutions that work.

In the religious freedom context, there is a strong logic to the idea that the Free Exercise Clause implies a robust principle that churches and their institutions may not be singled out for worse treatment as against otherwise comparable entities. But that’s just one component of free exercise; there have to be others. In Cato’s brief, for example, Ilya and Trevor point out that parents’ right to direct the religious upbringing of their children is a basic component of free exercise encroached upon by the Maine law. To take a different example, the Court has recognized in the Hosanna-Tabor line of cases that free exercise requires the state to respect churches’ internal autonomy, quite aside from equality concerns about whether it treats secular institutions the same way. 

Maybe churchgoers want freedom of worship for its own sake, not because they want to ensure churches are treated at least as well as drugstores in COVID regulations. Maybe fairness and benefit to kids are even better arguments for school choice in Maine than equality! But — in today’s climate of jurisprudence — we may not get a chance to find out any time soon.