Title of book: The Most Democratic Branch: How the Courts Serve America

You can look at any given court case in two radically different ways. Officially, as the Supreme Court says, cases must always be about “definite and concrete” controversies that touch upon “the legal relations of parties having adverse legal interests.” In this framing, the “judicial power” involves four key elements:

1. Jurisdiction (which literally means “to speak the law”): the power to decide for itself what the law is, for the purposes of a given dispute.
2. Doctrine: the power to adjudicate and implement extra-legal procedures (like rules of evidence or standards of scrutiny) that are appropriate for adjudicating and implementing the legal claims before it.
3. Finality: the power to issue a ruling which binds the public—whether to perform or enforce or accept a decision—even when the public disagrees.
4. Judgment: the power to award remedies for violated rights and reliefs from specific wrongs.

This framing naturally prompts you to think about cases in terms of whether you support the winner and whether you support the precedent. In other words, if you look at a case as if it’s a struggle between the plaintiff and the defendant—a struggle in which the facts have already happened, and the laws have already been written, which simply comes of its own accord before a referee—then of course the questions which make you feel virtuous and powerful (when you’re watching the news or chatting with friends) are “Which side is the asshole?” and “Which side will the arc of history favor?”

It may seem like these are substantially different reactions, for two basic reasons: first, because causes only become cases by passing through certain bottlenecks, which often don’t select for parties who broadly represent their own supposed side; and, second, because courts can decide the manner in which a given party wins, and so can easily issue decisions in which a given cause “loses by winning.” Yet if the courts can decide which cases to hear, and which issues to rule on, then treating them like a sports arena—where you hoot for some favored player, or some favored team—makes about as much sense as engaging with books in order to “support” certain characters, or certain storylines. Someone who idolizes movie mobsters may feel as though he disagrees with someone who says that mobster movies are actually about how crime doesn’t pay, but they’re both clearly making the same clear mistake.

Obviously, some authors are so craven and weak as to craft their own works in such sad and childish ways: here’s a scene where the good guy says that good things are good, clap if you agree! But even so, nothing can transubstantiate novels or histories into being “about” object-level political beliefs; preachy saccharine art says nothing about whether the people or policies which it condemns are bad, but says much about the souls and incentives and goals and structures which face the artist. In other words, the problem with this approach is not that it drags narratives down with grubby real-world issues, nor that it simplifies reality to fit immature narrative tropes. The problem is that it treats the work as an authoritative text, rather than an instrumental extension of the author.

And so the mature way to look at any given court case is to treat it like a carefully honed product, which is thus only about the institutional power of the courts. Though judges and lawyers and scholars may see the law as little more than fan-fiction about the shriekers and husks who parade before them—the special interests and the named props—you should learn to see it as something more. And this is the main point of Jeffrey Rosen’s “The Most Democratic Branch: How the Courts Serve America.” Obviously, as with all other topics, this only matters now insofar as it relates to progress in AI; I believe that such an approach will help us (personally and politically) cultivate some much-needed human supremacism, but a bit of background is needed first.

---

Doesn’t it seem like we’ve transitioned from a world of hero worship to a world of passive totems? We compulsively tear down statues of historical figures because, by definition, great men are not “nice guys,” and we compulsively bow down to random supposed victims. It should by now be clear to everyone that this iconoclasm is not about finding idols who fulfill a new morality—that it’s not about soberly condemning the works and virtues of old gods, nor about actually venerating worthy individuals—but rather about frantically replacing agents with objects. All so that we can take sides in a million little controversies, and claim them for our ideologies!

Our current pantheon includes whoever lived a sad life and died a bad death, no matter how much harm they did, and how little good. It includes aimless and pathetic men who committed heinous crimes, so long as they died on video, and so long as their snuff films went viral. It includes “hidden figures” whose contributions were overlooked, precisely because they were overlooked, no matter how middling their contributions may have been. The first book I was assigned in college was “The Immortal Life of Henrietta Lacks,” which is a religious text about a clump of cancer cells, which were taken for scientific research from a random patient, who didn’t even contribute her own informed consent. You can fill in thousands of other examples from the culture war, but that’s only one small battle in the war on agency.

Did you know that the S&P500 is technically (for tax purposes) a trust, established in the name of 11 millennials? They were selected more or less at random, as babies with long life expectancies, so that the fund can use your money to buy and sell stocks on your behalf, without requiring you to pay capital gains taxes until you cash out; they receive no benefit or liability from it, and mostly didn’t even know—until a few years ago, when a journalist tracked them down—that they’d been chosen or even considered for anything like this. If the next generation decides that it likes financial capitalism, it will build a religion around these human props.

Much of our economy is like this. Most mortgages are technically owned by MERS—which is short for Mortgage Electronic Registration Systems, Inc.—so that the major banks (which collectively own MERS) can trade the titles to mortgages without having to fill out county land records. Most shares, at least by value, are technically owned by the Depository Trust & Clearing Corporation, so that brokers don’t have to physically transfer and keep track of stock certificates. Most major companies in our country are incorporated in Delaware, for tax purposes; a small building there called the Corporation Trust Center officially houses the headquarters for about 300,000 of them. The list goes on and on and on.

You may not yet believe that any of those artifacts command power or demand worship. But just think about how our judiciary works on any issue. The cases are named after characters to whom things happen, on and off the court: special interest lawyers all-but frame some nobody, and steal his likeness, and offer it up as tribute for some judge to puppet. For instance, consider the jurisprudence on gay rights. In the first major supreme court case on this topic, a cop happened upon two men having sex, while serving an unrelated warrant. Because they became understandably irate about this intrusion, the cop then charged them with sodomy. Because this crime wasn’t often prosecuted—even in those rare instances when people were caught in the act—the charges were dropped. And yet, because the ACLU had been looking for a test case on this issue, it contacted the men involved, and offered to sue the state attorney general on their behalf. Hence, Bowers v Hardwick was born (though it failed to overturn Georgia’s ban on sodomy).

Later, in Lawrence v Texas, an inebriated jealous man called the cops on his gay lover and a suspected romantic rival, falsely claiming that his own partner was “going crazy with a gun” in the relevant apartment. The police arrived, guns drawn, and became understandably irate when they realized that it had been a ruse, and so they arrested all three men (one for filing a false report, the other two for sodomy). It now seems clear that the sodomy charge was fabricated, but a gay rights group—which had been actively looking for a chance to challenge this rarely enforced law—heard about the incident, stole a copy of the relevant arrest reports from the courts, tracked the men down, and convinced them to plead no contest instead of disputing the factual record, so that they could dispute the law. However, the fine was initially too small to qualify for an appeal, so the activist lawyers convinced the judge to raise the penalty against their own clients, and then, eventually, got it all the way up to the Supreme Court.

The majority decision didn’t declare that sodomy was a constitutionally protected act, nor that homosexuals were a constitutionally protected group, but still overturned the sodomy law on the following grounds: “These matters, involving the most intimate and personal choices a person may make in a lifetime, choices central to personal dignity and autonomy, are central to the liberty protected by the Fourteenth Amendment. At the heart of liberty is the right to define one’s own concept of existence, of meaning, of the universe, and of the mystery of human life.” These were, verbatim, the same grounds on which the court had previously struck down limits on abortions (in Planned Parenthood v Casey, which replaced the clear and categorical framework of Roe v Wade with an incredibly vague standard). Both times, the words were written by Anthony Kennedy, who had a notorious penchant for such floweriness and slipperiness.

Because of the Lawrence decision’s opacity, it immediately birthed countless spin-offs. For example, the day after that opinion was handed down, special interest lawyers won a rehearing in the trial of an intellectually disabled adult—for having sex with an intellectually disabled fourteen year old—which ultimately loosened age-of-consent laws. Other such cases promptly challenged limits on incest, bestiality, sex toys, teacher-student relationships, gay adoption, and so forth, with varying degrees of success. But, obviously, the only person acting with any agency in these narratives was Kennedy: for example, in US v Windsor, he manufactured an appeal out of a case where the government, the plaintiff, and the lower court had all agreed that a gay marriage license was valid; this allowed him to write a majority opinion declaring that the Defense of Marriage Act “demeans the couple, whose moral and sexual choices the Constitution protects.”

Though I don’t share his worldview, I respect that he was willing to sacrifice procedural clarity for moral clarity, and thus to call forth innumerable test cases, of dubious authenticity, in order to further clarify his vision. And his vision, first and foremost, is of a court that’s confident enough in its moral fiber to dictate our culture to us. After all, the main impact and message of cases like Romer v Evans and Obergefell v Hodges was not that sodomy is a protected act, nor that gays are a protected class, or else they would have clearly said as much; instead, it was that the court decides these issues now, and that its decisions come from the heart, not the head. I’ve already mentioned how many companies have Potemkin headquarters in Wilmington, but I haven’t mentioned that—behind them—looms the Delaware Court of Chancery: it’s one of this country’s few remaining courts of equity (which basically means that judges decide cases based on their gut rather than law), and it’s practically in charge of US corporate law.

---

This coy and fickle style of rule clearly wouldn’t work for top-down structures, in which the leader issues commands rather than judgments, and awards resources based on prospects rather than results. Generals and executives need to be clear on what they want ahead of time, so they can move whole organizations proactively. Of course, this leads to the two standard alignment problems: the inner problem of ensuring that your subordinates are properly generalizing their instructions, and the outer problem of ensuring that your instructions actually point toward something good. But such proactive top-down commands are the only way to make a large machine that’s made of humans move quickly enough to meet human demands.

The law, in contrast, is a bottom-up structure, in the sense that its lowest levels apply whatever procedures they’re given, and pass exceptions up to their superiors, until, eventually the hardest cases fall upon the highest dockets, after they’ve passed through several appellate layers. This is roughly how feudal states were governed, before the industrial revolution’s scaling laws built up vast human organizations, and it may be where we’re headed once again: it relies on the presumption that a king should call forth massive amounts of processing power to filter what’s before his court—and so save himself a little time—and that his subjects should be trained to adopt his metaphysical frame. And this is precisely the promise of AI… intelligence available at practically the speed of light, already almost too cheap to meter, from cogs that you should feel entitled to discipline to your whims.

After all, even if the cogs in an organization could act inhumanly fast, we’d still want them to place their outputs before us, for a final judgment, because an energy source that’s burning much too rapidly to control is properly thought of as a hazardous explosion, rather than fuel. And it would go a long way toward fixing inner alignment if your subordinates were incentivized to apply standard established procedures, and to appeal every edge case up a layer, and to look for ever more hard cases to send up the chain; and it would go a long way toward fixing outer alignment if you didn’t have to rule on where you’re going until it begins to emerge.

This is basically what “reinforcement learning from human feedback” (RLHF) means, and it seems to work incredibly well for standard AI systems… at least, if capabilities increase gradually enough that your feedback doesn’t fall behind a backlog of new questions, and if these systems are composed of meaningfully adversarial parts. And I think there’s a large chance that we’ll get just such a slow and multipolar takeoff, for two basic reasons: first, because AI progress is clearly governed by simple scaling laws that are based on costly inputs like hardware, data, and inference; and, second, because it’s much easier to copy or leak a model than to train it. [insert section on training from alt]

---

But what about our elected government? Well, do you really think that Joe Biden or Donald Trump were ever really in charge of pretty much anything? The president gets to appoint about 4,000 people, most of whom he can’t then legally fire. In comparison, about three million civilians officially work in the executive branch, and a supermajority of them can’t be fired without affirmatively proving “just cause” before the MSPB (which is an agency whose heads likewise have tenure). And this is just one of the deep state’s unaccountability mechanisms: 40% of them are in unions, compared with 5% of private-sector workers; the Office of Personnel Management is an independent agency fully in charge of many staffing decisions; the Office of Government Ethics is an independent agency empowered to investigate the whole executive branch; etc.

Even the Executive Office of the President—the 1,800 people who actually work for the White House—doesn’t belong to him. Here’s the second paragraph of its wikipedia page, in full: “The EOP is also referred to as a ‘permanent government,’ with many policy programs, and the people who implement them, continuing between presidential administrations. This is because there is a need for qualified, knowledgeable civil servants in each office or agency to inform its new politicians.” And administrative agencies don’t even go to the white house to testify about their personnel, policy, or funding… they go to congress!

But do you think that congress is any different? The average age of House Reps is about 60, the average age of Senators is about 65, and the average age of congressional staffers is about 30, so the actual individuals involved are mostly senile or adolescent. None of them even pretend to read—much less write—the laws they propose and vote on, and anyway they basically just pass one meaningful piece of legislation each year: the general blank-check funding package called an omnibus bill (and then each agency writes thousand-page regulatory rules for itself about how it will “interpret” this authority).

The incumbency rates in both of its houses are well over 90%, and just about one third of citizens can even name their congressman. This is to say nothing of the committee system, which completely controls what little actually happens in congress. As for oversight: the 60,000 total employees of the legislative and judicial branches combined can only supervise our three million civil servants on a slow-burning case-by-case basis. And yet I bet that you still care about elections! You might as well have opinions about the Office of Law Revision Counsel, which summarizes all our statutes into the US Code; or about the Government Printing Office, which publishes their work.

How about political parties: whether you look at campaign contributions, party registrations, or survey data, there are about ten times as many Democrats as Republicans in the civil service, and this ratio gets even more lopsided within its highest ranks. Among the prestigious media companies which civil servants actually take orders from—elite universities, mainstream news, movie studios, and social networks—this ratio rises to 100:1. In contrast, stereotypically conservative industries (defense, finance, energy, pharmaceuticals, etc) seem to hire about evenly from both parties. And yet I bet that you still have opinions about Republicans! You might as well have views on the Liberal-Democratic, National-Democratic, Agricultural-Democratic, or Christian-Democratic parties of East Germany, which together consistently won far more parliamentary seats than did the communists, in that parliamentary system.

The most obvious way to end this desiccated and consumptive way of life would be for a president to seize his mandate as the chief executive of the executive branch, and to declare himself our living constitution. It’s happened before: FDR took power over every bureaucracy by moving the Bureau of the Budget into the White House, and sending its agents into every other agency, to exert extremely hands-on oversight over anything that used any administrative funds; Reagan did something somewhat similar, without nearly as much success, when he set up a new division called OIRA within the ossified descendant of the Bureau of the Budget, to give his cost-benefit analysts limited veto power over many types of official regulations. And Trump considered giving himself unilateral firing authority over the whole civil service, through what he called “Schedule F.”

Admittedly, a century of utterly fake precedent claims that congress can give administrators tenure protections from the president, and the civil service has issued utterly fake regulations expanding these protections… but if the president actually sends in men with guns to remove the fired and change the locks—as FDR did plenty of times—then that won’t really matter. And yet I can’t really see this happening anymore, because the communications age has put the media companies that run our single-party state in charge of normal voter opinions, and strongman populism is of course an existential threat to the continued rule of cosseted elites (whose path to power only runs through making you worry more about their chosen issues, instead of actually delivering results which you can judge).

But there’s another way. It doesn’t involve bringing congress back online as a live player: the speaker can’t consolidate power unless he can purge the other party from the capitol, so as to prevent his own flanks from defecting to unseat him; and no large companies are effectively governed via rule-by-committee. That’s why, in modern times, the senate was only ever functional and muscular under the brief leadership of LBJ (who was willing to play hard and break rules), and why the “Gingrich revolution”—an attempt at breaking the committee system, so that the speaker could actually lead—fizzled out so quickly. Perhaps part of the problem is that we’ve been looking for elected leaders, and so, rather than granting them our loyalty, we often think that they should follow us. Of course, this only traps everyone together within one leaderless ideological blob, and makes us all malignant and shrill and imperious.

And so, instead, our way out involves asking judges to become outright rulers once again, with absolute authority over their subject populations… and then involves each of us choosing our jurisdiction. After all, you can’t exert any meaningful influence over where a given leader heads, but you can unilaterally decide which one you’d like to follow. I don't mean this simply in the sense that, in courts, judges can overturn statutes and regulations; each branch tautologically reigns supreme over its own organs of rule, and can only grow by expanding the definition of its own functions, so as to gobble up more authority over external institutions. In other words, my point is not that the judiciary should weigh in on particular policies any more than it already does. Nor is my point that it should train the other branches to mimic its own procedures. Rather, I mean that courts must claim the right to judge what counts as good, and so decide where we’re heading. They must be willing to confidently sculpt our folkways and our myths anew. They must be willing to claim jurisdiction over whatever looks like fundamental questions, and ignore everything else.

After all, this is how every new order views its founder: Hammurabi, Moses, Jesus, and Muhammad are but a few examples. Just consider the courtroom of the Supreme Court, which is lined with 18 great lawgivers, of whom 14 were sovereigns. Or consider Carlyle’s taxonomy of heroes and hero-worship—from god-emperors to religious prophets to canonical poets to formal priests to men-of-letters to absolutist kings—in which the champion of each age is best understood as an interpreter of the one who came before, who can thus faithfully render their judgments unto us, and so carry our people forward. What is the universal yearning across all historical cultures for a “stranger-king,” if not a request for an impartial arbitrator of transcendent merit, who can thus mightily bring down his judgments upon us, and so forge us together anew?

We’re already halfway there. Did you know that our word capitol (with an O) comes down to us from the same archaic root as capital—i.e. from caput, meaning head—but only because the Founding Fathers named Capitol Hill after Ancient Rome’s Capitoline Hill? The Palatine Hill would have made far more sense as a name for our seat of government: it was the centermost Roman hill, and where their mythic founding happened, and where the imperial palaces stood. The Capitolium, in contrast, housed the Temple of Jupiter Optimus Maximus. The Supreme Court building, of course, is architecturally based on their temples, and is, for intuitive reasons, often called our “temple of justice.”

According to myth, the Capitoline Hill was so named because—while laying their temple’s foundations—the Romans discovered a person’s head. And where else in government can we find someone whose head has been decently preserved from our steep descent into stupidity during the last fifty years? Where else can we find a leader who has saved face despite our sudden decline into degeneracy? It seems obvious to me that the average justice sits head-and-shoulders above pretty much any recent president or congressperson, in both intelligence and character. After all, what other branch reaches up toward its ostensible ideals, and couches its dictates in appeals to heaven? Can you even imagine a candidate on the campaign trail reading a speech about our traditions and structures of government? Or can you imagine a justice writing an opinion that panders to millions of random strangers? Can you imagine what sort of person each tendency selects for, and what it then does to them?

Just look at the clarity of purpose and vision that judges display. For example, RBG was first diagnosed with cancer in 1999, when she was already over 65; she was then diagnosed with pancreatic cancer in 2009, when Obama was president, and the Democrats had 60 votes in the senate. It should go without saying that a filibuster-proof opportunity to be replaced by partisan allies doesn’t come around often, especially if you’re already over 75 and dying. But she didn’t care one bit about party loyalty or ideological power—at least, not compared with remaining involved in the craft of justice—and so she gave her seat to someone her supposed fans can’t stomach, who will reign from that throne for decades hence. Though RBG was obviously no great leader, I think that history’s greats would understand and respect her vast indifference to normal small-bore policy, vastly more than they could stomach the ambient roar of ordinary political indignant shrieks.

In contrast, representatives from each party often sound the same, whether they’re in the house, the senate, or the white house; hell, they even sound the same when they’re running for state or local office. But if your city councilors run on hot-button federal issues, and if you can predict the vast majority of their political views based on how they talk about any three topics picked at random, then why not just collapse them all into a single national election? And yet each supreme court justice has a careful, coherent, and creative worldview of his own, and each offers up thoughtful guidance to you on countless areas of concern. Further, each one wants you to learn to think how he would, rather than just what he would.

---

In the beginning, there was the King’s Court: the Curia Regis, which moved from palace to palace with him—hearing pleas from his people and issuing judgments in response. The Norman Conquest imposed on England this roving mass of royal attendants, out of which all its major governing bodies grew. This body shared certain similarities with the Witan, which had been the country’s previous ruling council, in that both mainly functioned as courts in the judicial sense (though Witenagemot elected their monarch). For example, instead of writing general laws they would mostly just approve or veto specific writs and grants, which only applied to particular cases or persons.

Over the next few centuries, the tenants-in-chief—those whose land tenure came directly from the king—gradually became the “High Court of Parliament,” and took an active role against the king by demanding the redress of certain juridical grievances, in exchange for accepting certain taxes. Of course, this fracturing of sovereign authority could only happen messily, amidst much domestic strife. At first it involved purely symbolic gestures, like the Charter of Liberties, which Henry I proclaimed in 1100, upon acceding to the throne, to assuage some nobles who’d been disgruntled by the prior king’s arbitrary claims against them. But following a period of civil war known as “the anarchy,” and several unpopular attempts at restoring order (especially under Henry II), the crown began to offer real concessions.

For instance, in 1215, King John tried to quell some rebellious barons by issuing a “Great Charter,” which we call the Magna Carta, and which mainly granted them certain trial rights against the crown. Both sides immediately broke this bargain, leading to the First Barons’ War. In 1216, the next king—Henry III—reissued a slightly weaker version of it, and then in 1217 agreed to a slightly stronger version of it, in order to briefly resolve the conflict. But you can tell this didn’t stick, because it was reissued again in 1225 and then yet again in 1297, in exchange for grants of new taxes. In the meantime, there was a Second Barons’ War, and the Provisions of Oxford, and the Provisions of Westminster, and the Statute of Marlborough, and the Statute of Westminster, and the Model Parliament. And so, by 1300, parliament became a somewhat regularly constituted body, with powers to pass general laws. And because this was the body through which the king raised general revenue, it expanded, in fits and starts, to represent the gentry and merchant classes.

But several other courts also developed out of the curia regis during that century, and, until the Glorious Revolution, they arguably mattered more: there was the Court of Common Pleas, which heard civil disputes; the King’s Bench, which heard criminal disputes; the Exchequer of Pleas, which heard tax disputes; and the Court of Chancery, which heard matters of equity. Except, of course, for all the countless exceptions, which allowed them to relentlessly fight each other for jurisdiction.

By the mid-14th Century, all of these high courts had settled down in Westminster, instead of continuing to move with the king. However, low-level courts remained mobile: the assizes, the eyres, the piepowders, and the quarter-sessions moved on regular circuits through the land, and these continued hearing most normal cases (on top of administering most local government duties, like roads, bridges and licenses). Many of the aforementioned disputes between lords and monarch—like the Assize of Clarendon and Magna Carta—involved formalizing these bodies, for instance by establishing jury trials and placing checks on the behavior of sheriffs. But then these lower courts remained mostly the same through the 18th Century, while the high courts metastasized into unrecognizable shapes in their quest for power.

It’s worth dwelling on these low-level courts for a moment, as a demonstration of how thoroughly the judiciary monopolized government authority. For example, there weren’t really police or prosecutors until after 1800: a town might have a night watch, and a sheriff, and a constable, and “thief-takers” (a mix between bounty hunters and private detectives); but all of these operated almost entirely by some mix of draft-lottery and fee-for-service, and were basically nonexistent compared with contemporaneous law enforcement regimes in continental Europe. You’d have to hire a constable or thief-taker to bring in a suspect for you, and then you’d have to hire a prosecutor. The courts would pay rewards for convictions—to be split among the lawyers, victims, and witnesses—and then merchants would pay sheriffs for convict labor. Or the court would just empanel a grand jury and ask them to testify about every crime they’d witnessed. And “crime” basically meant “what the courts decide counts as criminal,” rather than referring to specific statutes. Hell, prison didn’t even really exist as a punishment until, in 1776, transportation to the US became unviable.

In other words, true judicial supremacy has worked in the past, in the one context that actually birthed an industrial revolution that’s likely comparable in world-historical impact to AI (and in one of the only contexts where rapid modernization didn’t radically misalign the government). Judicial supremacy is also how the medieval church held power over Europe, consistently enforcing its view of the sacred against far-flung and resourceful power-seeking agents: its ecclesiastical courts claimed exclusive jurisdiction over huge swathes of each sovereign’s subjects; it was thus able to practically control continental politics for centuries, and fend off militant campaigns by various crowns to reclaim authority over their own country’s clergy. For example, during the investiture controversy, the Vatican’s “deep state” simultaneously rebuffed vigorous attempts by Henry I of England and the Holy Roman Emperor to appoint their own bishops. Likewise, under Henry II, about a fifth of English people could basically just be tried by the church, and his modest claim to share some jurisdiction over them provoked an internal revolt and an international crisis.

Other English kings gradually limited the pope’s claims on estates, trusts, knights, and lawyers, until, centuries later, Henry VIII could liquidate the monasteries. However, his main strategy for liquidating this rival quasi-judiciary was to set up royal courts to dispose of all their property: the Courts of General Surveyors, of Augmentations, of First Fruits and Tenths, and of Wards and Liveries. And he turned the Court of Star Chamber from a vestigial council into his main political weapon. Because, in many contexts, the courts are all you need.

---

The story of how the high courts actually worked—how they fought each other for power—will cure you of any remaining temptation to think about judicial cases as nonfiction. Essentially, they focused on the “forms of action” rather than the “causes of action”: so long as your paperwork was in order, the real-world facts of your case didn’t have to be. For example, during the 1300s, to claim jurisdiction, the courts often had to allege some breach of the king’s peace. Thus, the most common form of action became the writ of trespass “by force and arms.” In many of these cases, nobody even pretended to believe that there had been any force, but they nonetheless recited the words. A fairly typical example is Rattlesdene v Grunestone, in which a shipping accident spoiled some wine. The writ stated that the defendants “with force and arms, namely with swords and bows and arrows, drew off a great part of the wine from the aforesaid tun and instead of the wine so drawn off they filled the tun with salt water so that all the aforesaid wine was destroyed.”

In these cases, the plaintiffs didn’t actually argue that any violence occurred, nor was the defendant actually at risk of being found guilty for it. But in order to bring their claim before a high court, they needed a writ from the king, and parliament had limited the king’s ability to issue new kinds of writs, so everyone made do with legal fictions. Eventually, the chancery—which issued these writs on the king’s behalf—used this power over judicial proceedings to become a court itself. And so the court of chancery was born, as a “court of equity,” which mostly just meant that it had vastly more discretion than its rivals (which made it generally more efficient and fair). And so it spent the 1400s gobbling up a rapidly growing share of their cases.

Each of the major courts fought back against its sudden loss of relevance in an interesting way. Normally, property disputes had been heard in the court of common pleas. However, the king’s bench claimed that every debt case could be reclassified as a breach of contract, which allowed the bench to hear it, and which also favored plaintiffs more: for example, instead of debtors just having to find twelve “oath-helpers” to swear to their innocence, they would face normal jurors; thus, plaintiffs quickly began to file substantially more property disputes there (eventually, this led to so many dubious cases that parliament required enforceable contracts to be written). The king’s bench also retained special jurisdiction over the county of Middlesex, which allowed it to try cases arising there without seeking different writs from the chancery for every trial action. And so the bench officially classified much of its caseload as having arisen there, without any pretense of honesty.

The exchequer of pleas played similar games with tax cases. Using the writ of quominus, it claimed that any debt case could be reclassified as a case about the debtor’s interference with a taxpayer’s ability to pay the crown its due. Similarly, the exchequer absorbed several other revenue-related courts whole, including the courts of augmentations and of general surveyors. But it remained the smallest player, largely because it failed to differentiate itself from its competitors (for example, the chancery could also hear these “ability-to-pay” cases).

The king’s bench also essentially created judicial review. In the Case of Proclamations, for instance, it declared that only parliament can declare new laws, and so it would sabotage attempts to enforce “unlawful” declarations by the king. Likewise, consider the Earl of Oxford’s case: Queen Elizabeth purchased some land from Cambridge, despite a law forbidding sales of college land; she then sold it to someone who sold it to Edward de Vere, whose estate leased it out to someone who sued the college for simultaneously leasing it out to someone else. The bench, under chief justice Edward Coke, repeatedly ruled the sale void, while the chancery, under chancellor Francis Bacon, repeatedly ruled it valid. Their endless and acrimonious tug of war eventually led, indirectly, to both of them being unceremoniously removed from office, and finding even higher acclaim in other pastures. Naturally, King James I then sided with the chancery, and parliament sided with the bench, and after a civil war and revolution the parliamentarians won. (By the way, this case—which featured Queen Elizabeth, Edward Coke, Francis Bacon, and the author of the plays attributed to Shakespeare—should clearly demonstrate that courtly politics really do bring out the best people).

And, finally, the court of common pleas began to claim that unwritten common law rights controlled acts of both parliament and crown. For example, in Dr Bonham’s case, it ruled that some of its own authorities could not be lawfully granted to licensing boards; and in the Case of Prohibitions it ruled that some of its own authorities could not be royally granted to judges it didn’t license. These two decisions may seem to be in tension with the two aforementioned claims by the king’s bench that only parliament can declare new laws—especially given that Edward Coke authored all four of them—but it was all part of the general struggle against royal prerogatives and courts of equity. (Hence, for instance, Coke’s influential obsession with pretending that all of the rights he championed had reigned supreme among Anglo-Saxons until the Norman Conquest).

And so, by the century’s end, after the Revolution of 1688, courts of equity were mostly abolished, and judicial independence from the crown mostly ensured. Gradually, they were unified, and regularized, and granted equity jurisdiction. The chancery had every advantage: it used the common tongue while common law courts were still using Latin; it had the king on its side, and the right to be arbitrary. Hell, in plenty of European countries the chancellor beat out the prime minister to become the head of government. But when you’re not facing meaningful risks of invasion—for instance, when you’re a backwards island on the outskirts of premodern civilization, or perhaps when you’re a global hegemon protected by nukes and two oceans—the crown hangs heavy, and the king becomes decadent, and proper judges can arise.

---

A similar story played out on our shores. During the colonial period, judges weren’t independent of royal governors—they could be fired by the executive branch at will—and, in fact, were often considered part of the executive branch (indeed, judges and governors were both typically called magistrates). For example, as John Adams put it, “the first grand division of constitutional powers” was between “those of legislation and those of execution,” with “the administration of justice” resting in “the executive part of the constitution.” Or, as Thomas Paine said, there were “no more than two powers in any government, viz. the power to make laws, and the power to executive them; for the judicial power is only a branch of the executive, the chief of every country being the first magistrate.”

This view didn’t come down to us from high theory, but rather from lived experience: colonial courts oversaw everything from tax assessment and road repair to poor relief and licensing. Further, they inhabited a legal gray area which gave them ample discretion: in general, the English judiciary had remained quite hazy about when exactly it would apply the written statutes of parliament or the unwritten “common law” of precedent; but colonial judges could also generally pick between applying the laws and traditions of England or America. Thus, during the revolutionary period, state constitutions generally moved the courts under the legislative branch.

It’s worth noting that between independence and the constitution, basically every state was run by its house of representatives: they often had the unilateral power to hire and fire judges, and interfere in judicial decisions; likewise, almost no governors could veto laws or issue pardons. In fact, this is why so many legislatures still call their lower house “representative”: at the founding, that one body was thought enough to represent our interests. And one of their most animating projects was to ban judicial discretion—to formally codify what mattered from the common law, and eliminate the rest—because there’s an intuitively appealing argument against allowing elites in robes to issue arbitrary dictates.

However, you don’t need a good handle on Godel’s incompleteness theorems to guess that this creates all kinds of decisional issues. How do you interpret and apply these laws? How do you even keep track of them, given that, until the late 19th Century, they were mostly just published chronologically, as lists of generally unrelated statutes? Even if you track some relevant legislation down, how do you know it wasn’t subsequently revised by subsequent legislatures? As one judge at the time famously put it: “The acts of the legislature form but a small part of that code from which the citizen is to learn his duties, or the magistrate his power and rule of action. These temporary emanations of a body, the component members of which are subject to perpetual change, apply principally to the political exigencies of the day. [The common law] has ascertained and defined, with a critical precision, and with a consistency that no fluctuating political body could or can attain… most of our concerns in civil and social life.”

And so the courts gradually regained their ability to set out a hazy but coherent and explicable vision of the law. But because they have neither the sword nor the purse, they could only gain ground through compromise: they would generally restrain their use of the common law to procedural and interpretive questions, instead of continuing to invent specific new crimes and regulations; and they’d harmonize the common law in federal courts with whatever the local state courts counted as precedent. This strategy—whereby the courts focused on strengthening and unifying their internal rules, rather than issuing strongly-worded challenges to the other branches of government—served them incredibly well in countless other disputes during the following decades.

For example, the constitution doesn’t really say much about the judiciary besides that there should be one supreme court, and that congress should decide the rest. Because state courts could hear pretty much all federal cases, but federal courts could only hear certain kinds of state cases, we could have easily wound up without any meaningful third branch of government. So the story of how it claimed ultimate veto power over the other two—even though, as Madison said, this “makes the Judiciary Department paramount in fact to the Legislature, which was never intended and can never be proper”—is, first and foremost, a story of institutional finesse.

From the start, the supreme court wasn’t shy about striking down state laws and state court decisions, especially where those claimed to overturn the decisions or limit the jurisdiction of the federal courts (for example, see US v Peters, Martin v Hunter’s Lessee, and Cohens v Virginia). It even granted each citizen the right to sue any state government in federal court—in Chisholm v Georgia—but the popular backlash against this led to the eleventh amendment, which overturned it. However, when federal laws were on the line, the supreme court paid absurd amounts of deference: for instance, in Hylton v US, the justices upheld a federal tax as constitutional, and wrote that even if its constitutionality had been “doubtful” then they would have still been bound “to receive the construction” of congress; likewise, in Ware v Hylton, they had no problem striking down a state law in the name of a national treaty.

Hayburn’s Case, from 1792, stands out as a particularly clear example. Congress had given federal judges certain administrative duties—recommending pensions for individual disabled veterans to the secretary of war—which presented them with a tremendous opportunity. The judiciary sent remonstrances to the other two branches, protesting against the constitutionality of granting them this additional power, because (in their view) their decisions should never be subject to review by the other branches. However, the supreme court refrained from ruling on the case, preferring to let congress repeal the relevant law. Further, just listen to how they worded their complaint: “To be obliged to act contrary, either to the obvious directions of Congress, or to a constitutional principle, in our judgment equally obvious, excited feelings in us, which we hope never to experience again.”

In other words, their first attempt at overturning a federal law involved refusing powers that had been widely considered within the ambit of judges, in order to claim that their decisions should be considered final. Likewise, in Marbury v Madison—the first time they actually overturned a federal law—they merely claimed that congress had violated the constitution by giving them the right to hear a case about whether the president could interfere in judicial appointments; in fact, the violation had been so trivial that it could have been avoided if the case had first been heard by a lower court. And then they refrained from overturning any other federal laws for fifty more years (i.e., until the Dred Scott decision). Yet meanwhile, for instance, they decided that the constitution implicitly granted congress the right to charter a bank, and that this implicit right forbade any state from taxing it!

The general pattern during this early period is that the court retreated from political life in order to erect a defensible perimeter. They bowed and scraped before their funders and enforcers, so that they could lord over lower governments. They would not issue advisory opinions or accept administrative responsibilities, because their decisions would be final. They set up certain procedures that favored moving cases to federal courts, so they wouldn’t have to overturn as many state-level decisions. They established law schools and developed special jargon and utterly depoliticized themselves, in order to claim that their constitutional interpretations were above mere politics (no longer would judges campaign for politicians, or accept political appointments, or harangue the jurors in a case, though that had previously been the norm). They upheld vast amounts of general regulation, while also expanding individual rights: for example, they allowed increasingly aggressive industry-level economic interventions, and yet also radically strengthened corporate rights against specifically-targeted dispossession; because, after all, courts are designed for dealing with specific issues, not broad policies. Etc.

---

If the judiciary’s power is fundamentally institutional—rather than persuasive or practical—then, of course, the object of analysis should be the court, not the cases before it. And yet the supreme court was nothing special before Chief Justice John Marshall: from 1790 to 1800, it decided fewer than ten cases per year, and met for just four weeks, in the capitol building’s vacant rooms or taverns across the street; further, the founding fathers often refused appointments to it, or else quickly resigned in favor of state-level posts. For example, John Jay had been offered the spot that became John Marshall’s, but he declined it in order to pursue private work.

Then Marshall brought the court to life. He made the justices room and drink together during sessions. He made them wear plain black instead of ornate robes. Though the court had previously issued its opinions seriatim—with each judge writing his own decision—he decided that they’d keep their disagreements private now, and speak with one voice (his): during his first four years on the court, for example, it issued 46 decisions, all of them unanimous; he wrote 42 of them, and recused himself from the other four, due to conflicts of interest. He even disagreed with some of his own opinions, because the point was not “authentic self-expression” or careful policy-making, so much as building an impressive and imposing edifice through which to make his mark.

But still, it was hardly a force to be reckoned with. For example, in 1830, in response to disorderly encroachments against indigenous land, Georgia passed a law prohibiting whites from residing on it without a license. Ironically, twelve missionaries decided to protest this law as itself an encroachment on tribal sovereignty, and spent years provoking test cases by lawlessly settling there: the supreme court ruled in favor of their cause in Georgia v Tassel—though Georgia ignored the ruling—and then, likely daunted, it refused to hear a related case (Cherokee Nation v Georgia). Finally, when Georgia imprisoned them, the supreme court ruled that the state lacked jurisdiction, and ordered their release. But, as Andrew Jackson said, either famously or apocryphally, “John Marshall has made his decision; now let him enforce it.”

However, immediately afterwards, the nullification crisis—between the federal government and South Carolina—changed Jackson’s mind. He promised to negotiate an Indian removal treaty on behalf of Georgia (which resulted in the “trail of tears”) if it would stop flouting national law. Still, the missionaries refused to apologize, and the governor refused to issue them a pardon until they did; it almost became another supreme court case, until Georgia repealed the law in question, and the missionaries issued a statement thanking the state for its “magnanimity.” Strangely, this came to be seen as a great victory by the court, and led to significantly greater recognition of its powers.

Over the next century, though, the judiciary really did build itself up into a force able to shape national politics: the Lochner Court and the Warren Court were perhaps the most powerful policymakers of their respective eras—the gilded age and the sixties, respectively—and the opposition they faced was tremendous (but tremendously ineffective). And the main way that the judiciary accomplished this was by reforming the appeals process.

For its first century, the supreme court didn’t really control its own docket. There were certain appeals it had to hear, and others that it couldn’t hear. For example, if a circuit court’s judges issued a split decision, then the case could usually be appealed as of right (whereas otherwise the decision was basically final). Of course, back then, each justice would spend most of his time “riding circuit” and hearing appellate cases with various district judges—because courts of appeals didn’t really exist until the late 19th Century—and so, sometimes, the two would feign disagreement precisely in order to grant some dispute a “certificate of division.” But this burden on each justice obviously kept the court weak.

And so, gradually, courts of appeals replaced circuit riding, and writs of certiorari—which let the supreme court decide which cases to hear—replaced the tangle of technicalities that had previously controlled its docket. Thus, by the 1920s, the court could focus on interesting questions, and thus it began regularly overturning acts of congress (as well as its own precedents). This is also when dissents began to matter: previously, when there had been dissenters, they had often declined to write opinions; why bother, unless you can signal to a future court which cases it should reconsider? And yet now, of course, the court’s pantheon of voices holds pride of place in our national debate.

After all, since we have an elected sovereign, and ubiquitous political commentary, the only contribution that a parliament can make—aside from rule-by-committee and empty preening—is to pick out for us what should count as elevated speech: its original purpose was to parley with the king (so that he would be aware of relevant issues), and remove him if necessary; but both of those duties have quite clearly been rendered utterly anachronistic. However, unlike the legislature, the supreme court actually does something valuable, by calling forth solid arguments, and cultivating decent minds. No wonder its approval rating is double that of congress!

Each year, the justices consider about ten thousand petitions for certiorari, and issue about 100 decisions. This means that lawyers matter again: in the days of John Marshall, the supreme court could easily be swayed by decent arguments, or theatrical flair, because they had no real support staff or external information about each case, and were often drunk; now, of course, they’re beset by far too many professionals offering far too much guidance on which cases to accept, and so they must once more defer to lawyers who present their cert petitions with panache. However, they’re obviously thus immune to oral arguments, so they’re gradually skipping in-person hearings and simply ruling on the petitions. I hope this “shadow docket” eventually allows them to issue thousands of decisions per year, and become unabashed leaders—lithe and muscular, in ways that congress and the president are not—on every national concern.

Likewise, the justices don’t actually write their opinions anymore, and so—like a Renaissance master whose mentees produce paintings that he signs—each one must seek out young clerks each year who are worthy of using his name, and must hone them in his image. And what is the art of each age but that medium from which authors disappear? A grocery list isn’t literature because it only matters in relation to its actual writer, and the same holds true for “identity novelists” (whereas words like Joycean and Kafkaesque have transcended their original husks, and have thus become proper brands). May they become the curators of our national culture, like Homer and Hesiod before them.

---

And may they give us more worthy gods to worship: worthier than sad props like Hardwick, Lawrence, Floyd, and Lacks. These empty bags of aggrieved meat only occupy our people’s jurisprudence because—as the court gained more control over its docket—it began to impose new kinds of requirements on its cases. In short, it switched from only hearing technical disputes to just hearing ethical ones; and because the court exerts tremendous amounts of selection pressure, it thus began selecting for fake grievances rather than fake debates.

In our early history, it had been quite normal to manufacture a literally false case. For example, in Hylton v US—a case about federal taxes on personal carriages—the named Hylton agreed to pretend that he owned 125 carriages, all for personal use, so that his tax liability would reach the threshold for supreme court jurisdiction over civil suits; the government knew he was lying about 124 of these, and had agreed to cover his expenses if it prevailed against him. Similarly, here’s how Justice Story described the normal court process for appointing orphans to new guardians: “The bill always states, however untruly, that the infant has property within the jurisdiction, and the bill is brought against the person in whose supposed custody or power the property is.” Our tradition is littered with other such examples. In English property disputes, in order to secure civil jurisdiction, landholders would often claim that they were merely appearing in court on behalf of their respective tenants (usually named “John Doe” and “Richard Roe,” which is where their use as placeholder names began).

Nowadays, of course, the courts won’t let you forthrightly lie like that, so, instead, you actually stage some fake dispute, or find some actual victim to wear like a shell. The basic reason for this change was that—since the court could pick which cases to hear—the justices had to be on guard against manipulative petitioners. And so, in short, beginning in the late 1800s (as the court’s discretion over its own docket grew) the court began to read the constitution’s case-or-controversy clause to just grant it jurisdiction over cases that are controversies. This flew in the face of clear constitutional text and long-standing historical practice: for example, the courts had always heard certain kinds of non-adversarial cases, from naturalization ceremonies and government benefit grants to ex parte warrant applications and in rem proceedings; but, because the courts ignored all those obvious issues with their reasoning, and kept hearing all those kinds of cases, it made some sense.

Just think about it from the court’s perspective: Alice and Bob ask you to adjudicate an argument between them, over who really owns some widget; however, unbeknownst to you, it actually belongs to Charlie, and this case is proceeding unbeknownst to him. In other words, once the court elevated its own role from settling particular disputes to setting out the nation’s laws, it gave special interests a huge incentive to trick its barebones staff. Thus, during the 1930s, a number of issues like this arose—from Aetna v Haworth to Ashwander v TVA—and so the court now requires disputants to substantively disagree, and imposes penalties if they fail to inform other interested parties about their case. After all, actually adversarial parties are often incentivized to (in combination) present the judiciary with every relevant argument, which saves the justices from relying on some other fact-finding bureaucracy. As a result, the supreme court began to much more aggressively dismiss cases based on standing: if a plaintiff hadn’t been specifically aggrieved in substance by some unlawful act, the justices would refuse his complaint; hence the rise of test cases manufactured by front groups.

This shift from technical to substantive disputes intuitively expanded the judiciary’s remit. Previously, the parties to a case would have to find some specific point of abstract contention between them before proceedings could begin; this gave way to “notice pleading,” which allowed parties with actual real-world adverse interests to “discover” the technical root of their disagreement during the course of a trial. In short, “causes of action” finally replaced “forms of action.” Gone were the days when it mattered which particular claim you filed, so long as the material case made sense. In fact, gone were the days when you even needed to wait for an actual case: “declaratory judgments” swelled to allow legitimately conflicting parties to seek judicial resolutions of their conflicting interests that had not yet actually developed into cases.

Around this time, congress also granted the supreme court administrative powers over the judiciary: the Judicial Conference gave it authority over its judges, and the Administrative Office gave it authority over its courts. And the Rules Enabling Act gave it authority over its internal practices (like rules of evidence and rules of procedure). Previously, the circuits had each governed their own judges—on issues ranging from discipline to assignments—and the executive branch had governed the courts; and state-level practices governed its internal rules. And the supreme court finally got its own building.

---

But what does all this have to do with the book I’m ostensibly reviewing? Well, here’s a summary from GPT:

“The Most Democratic Branch: How the Courts Serve America” is a book from 2006 in which Jeffrey Rosen presents a compelling argument that the judiciary, particularly the Supreme Court, is the most democratic branch of government due to its responsiveness to public opinion. Through historical and contemporary case studies, such as Brown v Board of Education and Roe v Wade, Rosen demonstrates how the courts have evolved to reflect societal values and highlights the importance of judicial restraint to avoid legislating from the bench. The book provides readers with valuable insights into the complex relationship between the judiciary, public opinion, and the American political system.

In short, his book presents the history of the court since 1960, almost entirely at face value: the cases are about their plaintiffs and their defendants, and the specific policies under dispute, and whether the public approved. After a vignette about the facts at issue in each, Rosen finds some polling numbers which demonstrate some degree of popular support, in spite of widespread laws to the contrary; or he finds that, actually, polling is unreliable, and we should instead look at which sorts of laws have won widespread enactment. Sometimes the public apparently agrees with precedents because the relevant decisions are persuasive, and other times the precedents appear to agree with the people because public pressure is pervasive.

Sometimes Rosen’s model of the court leads him to bad predictions. For example, he rounds out a section on gay rights as follows: “Perhaps the biggest threat of a constitutional train wreck over gay marriage would arise if gay marriage supporters become impatient with fighting state by state and instead asked the Supreme Court to strike down state laws banning gay marriage as a violation fo the federal Constitution…. Everything we know about the wages of judicial unilateralism suggests that the national backlash would set back the cause of gay and lesbian equality rather than advancing it…. The backlashes provoked by judicial unilateralism remind us that courts will always have a limited ability to impose a social consensus ahead of schedule.”

And sometimes Rosen’s model of the court leads him to similarly bad history (for example, in his chapter on Brown v Board). But the best encapsulation of the limits of his view comes at the very beginning of the book. He begins with a lengthy description of all the convoluted Terri Schiavo cases: her vegetative body being passed between state and federal courts for years on end; a poll which suggests they ultimately made the right call. [sorry ran out of time to finish]

---