By James Grimmelmann

Why Access?

“But the plans were on display…”

“On display? I eventually had to go down to the cellar to find them.”

“That’s the display department.”

“With a flashlight.”

“Ah, well, the lights had probably gone.”

“So had the stairs.”

“But look, you found the notice, didn’t you?”

“Yes,” said Arthur, “yes I did. It was on display on the bottom of a locked filing cabinet stuck in a disused lavatory with a sign on the door saying ‘Beware of the Leopard.’”

—Douglas Adams, The Hitchhiker’s Guide to the Galaxy

Why do we care so much about access to the law? I think we can agree that it’s clearly wrong to make the documents that constitute the law accessible only by putting them on display on the bottom of a locked file cabinet stuck in a misused lavoratory with a sign on the door saying “Beware of the Leopard.” But why? What’s wrong with this approach to government? At least four things:

• Democracy: Rule by secret law is perhaps indistinguishable from rule by no law whatsoever. It’s fundamental to democracy and the rule of law that rules be announced and applied all prospectively. A government that only uses secret laws can rule with impunity; it can change the laws on the on the fly; it can simply act and make up the laws in hindsight so that whatever it does turns out to be allowed. Thus, requiring laws to be published acts as a restraint on the exercise of arbitrary government power. It’s the first measure of transparency and accountability, the one on which every subsequent measure depends.
• Fairness: Even in a functioning democracy, deficient access to law is dangerous. The rule of law only works if people actually know the laws. There are legal maxims that express this idea, such as “Everyone is presumed to know the law” and “Ignorance of the law is no defense.” These maxims are only fair—they are only reasonable principles rather than cruel jokes—if the public in fact has an opportunity to learn the content of the law. It’s ridiculous, both morally and pragmatically, to expect someone to comply with a set of rules that he has no reasonable chance of learning. It’s literally Kafkaesque, and incidentally violates the Due Process Clause of the constitution.
• Consistency: Access to the law is also vital to the idea of law as a system of rules. You can’t act consistently if you have no memory of what happened last time. While either of two rules might be fair if applied consistently, oscillating back and forth between then is pure chaos. There’s a rule in administrative law that agencies aren’t free to reverse their own prior decisions without providing some reason for the switch. This is the rule of stare decisis in the courts and it obviously depends on having some accurate information about what happened in previous cases.
• Equality: Unequal access to the law creates substantive inequality. If law is only available to those who have the resources to go and seek it out—a well paid lawyer, a better library, access to expensive services like Westlaw and Lexis—then people who can afford better access can afford better outcomes. This means that the rich can take advantage of law in ways that the poor can’t. Worse, it means that they can set traps to ensnare their less legally educated opponents. Law law becomes the servant of those with better access to it; it takes their side, exacerbating inequality.

A Page of History

None of these things are new problems.

Babylon

Consider the the Code of Hammurabi etched on to a rock. It’s a bit under four millennia old and even there, you can see some very modern things. While this is a highly visible artifact—8 feet tall and carved out of solid black basalt—literacy in Babylonian society was low. Only members of the priestly caste and a few others were able to read this and even a skilled reader would have a difficult time reading the 282 laws inscribed on the surface. That doesn’t, however, stop the code from declaring in its epilogue that “Hammurabi did teach the land these laws.” Even at this early stage, the Babylonian legal system was taking the problem of notice seriously.

Note also a few other features of this piece of legal access technology:

• Quality There’s a technological choice here. The laws are inscribed on an 8-foot stele, rather than on clay tablets. Hammurabi wanted to make sure that his laws were permanent and that their authenticity was indisputable. He succeeded. Today, we still know the contents of almost all of his laws (except for the one or two that have been chipped away over the years).
• Organization: Technology isn’t just what you write on; it’s also how you arrange and index the things you write. The Code uses some clever organizational technologies. In an early example of a citation system, the Code’s laws are consecutively numbered from 1 to 282. The Code also contains what we’d call today headnotes or legislative history. True, they’re mostly accounts of Hammurabi’s deeds and dedications to the gods, but they still provide us extra information about the history of the Code and its purposes.
• Cost: Something this large and substantial doesn’t come cheap. It took a substantial investment to physically produce this stele; it also took labor to compile the laws themselves. Hammurabi could have used those resources to engage in additional smiting, but instead he chose to spend them on distributing his laws.
• Copyright (!): Lest anyone else think to tamper with his Code, Hammurabi calls down the curses of the gods on any later ruler who should “corrupt my words, change my monument, efface my name, write his name there, or on account of the curses, commission another to do so.” It’s not exactly a copyright, but it comes from the same impulse to exercise control over words.

England

Fast-forward a few millennia, and legal publication is still grappling with these same issues. Here are a few snapshots from England’s long and halting progress towards widespread access to the law. If there’s a recurring theme, it’s that good legal publishing is hard.

Within 150 years after the Norman Conquest, England’s royal courts started keeping parchment plea rolls to track the results of cases. They reported the nature of the lawsuit and who won, but in formulaic terms that provided no factual details or legal explanation. They were thus completely inadequate at meeting the needs of actual lawyers and their clients—who need to know which arguments will win in court and which will lose.

Instead, by 1300 or so, lawyers were consulting anonymous and often sketchy outlines of the arguments made in court, compiled in volumes called the year books. While they became standard sources for lawyers and judges to consult in looking up the law, over the years their disorganization became a serious problem. It’s a painful task to have to search through centuries of handwritten volumes without an index, especially when you’re looking for rough analogies rather than exact matches.

Eventually, the anonymous yearbooks were displaced by nominative (or “named”) reports, i.e. volumes of cases produced by specific editors who named their reports after themselves. This is a familiar shift in copyright history; going from script to print means both that the author becomes more important and that charging for copies becomes the dominant business model. By modern standards, the early nominative reports were woefully incomplete. In 1571, Edmund Plowden produced two carefully-edited volumes, but they covered a grand total of 62 cases. The second attempt was an 1582 collection of notes by Sir James Dyer (who had died in 1579), which mixed reports of actual cases with reports of dinner conversations. Sir Edward Coke, another judge, reported many cases starting in 1606, but sometimes falsified precedents, when doing so would support his position in a case. After Coke, no one tried again until the 1640s. Meanwhile, the yearbooks had petered out in the 1530s, which leaves quite a gap.

Over the years, there were more and more of these nominative reports. Some were excellent; many were terrible. An 1849 report on the system noted the inconsistencies among various competing reporters, the resulting inaccuracies and gaps, and the enormous cost to lawyers who needed to buy copies of multiple reporters’ sets in order to have everything relevant. It concluded:

To sum up in a few words the evils and inconveniences of the existing system of Law Reporting, there is no guarantee afforded to the public that the judicial exposition of the law will be reported at all, or reported correctly—or in time to prevent mistakes—or in such a manner, with respect to conciseness, form, and price, as to be accessible to those whom it so vastly affects.

The Law Reports produced by the independent, unpaid committee ultimately set up in response to the report became the English standard. And yet the profession hated them for many decades: they were slow, badly indexed, badly arranged, and—astonishingly enough—printed some cases repeatedly and other, important ones not at all. Nonprofit, quasi-public reporting turned out to have most of the same defects as private control over case law had.

Another instructive example involves not case law, but legislation. Consider the Statutes of the Realm, the first official attempt to provide a comprehensive collection of the legislation passed by Parliament in all of English history. It appeared in nine volumes and took twelve years to compile. The result of this extensive effort, completed in 1822, was to produce a relatively complete set of laws all the way up through … 1714. The Statutes of the Realm were over a century out of date on the day they first came off the printing presses! Perhaps unsurprisingly, practicing English lawyers turned instead to a privately-produced set, Chitty’s Statutes of Practical Utility. This is an important point, and it bears repeating: From the perspective of the people who most regularly needed to consult these legal materials, a private editor did a better job than the men with the official governmental commission to make the laws widely available.

Henry Wheaton

Legal publication in the U.S. didn’t get off to a good start. In its early years, the U.S. Supreme Court didn’t produce printed opinions. Even today, we don’t have copies of all Supreme Court cases. The first published reports of Supreme Court cases come from Alexander Dallas, who was actually reporting Pennsylvania cases and decided to report the U.S. Supreme Court because it was sitting at the time in Philadelphia. His first volume, published in 1798, contained a grand total of five cases from the Supreme Court—decided between 1791 and 1793. As for his accuracy, the best that we can say is that we have nothing to compare it against, since we don’t have the originals. This sad state of affairs continued through the term of Dallas and his successor, William Cranch, until Henry Wheaton entered the picture in 1816. If this story has a hero, it’s Henry Wheaton. If it has a villain, it’s also Henry Wheaton. His 11-year tenure as Supreme Court reporter was marked by a number of innovations, some of which may sound interestingly familiar.

• Quality: He achieved nearly unheard-of standards of timeliness and accuracy. His reports of a given year’s Supreme Court term would typically be produced before the start of the next term.
• Organization: His enthusiasm also extended to the organization of his reports. He produced marginal notes and scholarly essays following the cases that put the opinions in a broader context.
• Cost: These high standards of scholarly editorial quality came at a quite literal cost: his volumes were expensive. A single volume of Wheaton’s reports cost $7.50, which works out to over $100 today, an even more princely sum considering the distribution of income at the time. Only a small group of lawyers could afford his reports.
• Copyright: Perhaps most interestingly for our story, Wheaton used copyright to fight lower-cost competitors.

In 1829, Wheaton’s successor, Richard Peters, brought out inexpensive, smaller versions of cases first reported by his predecessors. He omitted the marginal notes and deleted the scholarly essays, but he took the text of Wheaton’s cases word for word. Wheaton sued. The resulting case, Wheaton v. Peters, decided by the Supreme Court in 1834, is famous in American law as the first Supreme Court case on copyright.

(Among other things, it’s the source of the statement that United States copyright is exclusively statutory. Copyright covers only those things that Congress wishes it to cover.) Right at the end, almost as an afterthought, the Court says:

It may be proper to remark that the Court is unanimously of opinion that no reporter has or can have any copyright in the written opinions delivered by this Court, and that the judges thereof cannot confer on any reporter any such right.

So this is the initial statement of rule that the law is not a proper subject for copyright. Wheaton tried to get a copyright by intermingling his additions with the public domain texts underneath. Peters got around Wheaton’s trick by extracting the underlying material, reprinting only the stuff that came from the court. So Wheaton v. Peters, in effect, says that Wheaton’s intermingling doesn’t take the underlying kosher public domain materials and render them treyf. Peters was free to publish his cheaper editions.

But think about the incentives that this decision gives to potential reporters. Until Henry Wheaton’s innovations, Supreme Court reporting had been a money-losing business. No one had been able to make it work. Dallas’s and Cranch’s reports were too slow and too inaccurate to meet lawyers’ needs. Wheaton did the job right. But doing the job right proved to be nearly impossible without copyright’s incentives. While Peters nominally won, the decision left him and his successors with no clear way to assert copyright in their work. For the next several decades, the general consensus was that the Supreme Court reports were no good.

John West

Let’s jump ahead to the story of how later reporters in the 19th century figured out how to do, more or less, what Wheaton hadn’t. Starting in 1876, John B. West produced a series of reporters tied to geographic regions. Each reporter covered several states’ worth of courts (including the federal courts in those states). By 1887, he had expanded to national coverage, very much on Wheaton’s model; careful editing, lots of additional material, and strong copyright claims. His production process was a good example of late 19th-century information workflow. Very carefully structured, it took the handwritten decisions as they came from the judges and passed them through a group of clerks who not only typeset the opinions but also proofread them and arranged them. The process took advantage of substantial economies of scale in printing and in managing the clerical work.

To this, West added a system of key numbers, essentially a hierarchical taxonomy of legal topics, the idea being that any legal question whatsoever could be classified using some number. For example, 99 is “Copyrights and Intellectual Property,” 99.I is “Copyrights,” 99.I(A) is “Nature and Subject Matter,” and 99.I(A)k14 (which can be abbreviated to 99k14) is “Statutes and Law Reports,” the key number that covers essentially this entire essay. If you look under that number in one of West’s printed indexes (or on Westlaw, its online service), you’ll find Wheaton v. Peters and other cases dealing with these issues of copyright in the law.

This is a fairly typical page from a modern reporter in the West system. Let’s pull it apart into its constituent elements, to see how the copyright claim fits together.

1. We have, for example, the text of the opinions themselves. Thanks to Wheaton v. Peters, these are public domain. Banks v. Manchester, later in the century, extended the rule to state courts, such as the Kansas courts whose opinions are reported here.
2. We also have a number of things that copyright law would classify as “facts”: the names of the parties in the case, the court deciding it, the court’s own number for the case, and the date. All of these things are treated as uncopyrightable; they’re not considered original to the reporter writing them down.
3. On the other hand, there are some things in here that are indubitably copyrightable. The headnotes written by West and his employees consist of summaries of the points of law discussed in the opinion, rephrasing the court’s reasoning and making it shorter and more concise. They make it easier for a lawyer to skim through quickly, looking for relevant cases. (In the system of West Digests, which extract the headnotes and link them with the key numbers, you’d simply read through a list of headnotes until you’d found one reasonably close to what you were looking for, and then go consult the entire case.)
4. But we’re missing one other very important thing that West added to this system: the page numbers. And this is the key to the American system of legal citation.

The tradition hereabouts is that one cites a case by listing the name of the reporter in which the version cited appears, the volume number within that series, and the number of the page on which the case starts. Thus State v. Commemorative Services Corporation, the case we’re looking at here on the right-hand-page, would be as “823 P.2d 831.” That’s volume 823 of the second series of the Pacific Reporter (abbreviated to P.2d) at page 831.

While technically, a lawyer could cite to any reporter in which the case appears, there are enormous network effects at work. If you produce a reporter that’s more commonly used than other reporters, as West’s reporters were by the early twentieth century, then courts come to expect citations to those reporters. Lawyers stock the dominant reporter in their libraries; the value of providing citations to any other reporter drops. Ultimately, the citation system becomes the hook that makes West’s reporters the ones every lawyer has to buy. Today, there are about 20 states where the West reporter is the only available source for published court decisions.

This standardization on West page numbers gives West ammunition to argue that it has a copyright not merely in its editorial additions, but also its arrangement of cases within the volumes. More precisely, the page numbers become artifacts of the order of cases in a volume, and the editorial decisions West makes in choosing how to order them (for example, here West is grouping two cases from Kansas) become the basis of a claim to copyright.

The State of Play

West is hardly the only one trying something like this. Plenty of other entities are claiming that their annotations of the law or citations to the law are copyrighted:

• West has gone computerized, and so has its arch-rival, LexisNexis. Both of them have extensive online databases of primary and secondary legal materials. These databases—which one used to access via direct dial-up are now on the Web, wrapped in dire copyright warnings and extensive terms of service.
• The American Medical Association produces a standardized taxonomy of medical billing codes. The taxonomy is privately drafted, but the federal government requires that Medicare and Medicaid reimbursement paperwork use its codes. The AMA claims that its taxonomy is copyrightable. A federal appeals court agreed. as did one considering a similar code produced by the American Dental Association.
• Similarly, cities require that all new buildings be constructed to standards specified in building codes. The private construction industry trade associations that draft the building codes assert copyright over them. In a closely contested 8-6 decision, a federal appeals court threaded the needle: It’s legal to make copies of a town’s law, but not of a building code as such. Thus, what would otherwise be an infringing copy of a building code becomes legal if it’s made as a copy of “the law.”
• How about tax maps? New York City makes maps of property lines to determine the assessment boundaries for property taxes. It’s gone to court to assert its copyright in those maps—and won. (That case isn’t publicly accessible online, oh the irony, but here’s an earlier one on the same issues.)
• Out in Wisconsin, a private company has even attempted to use copyright to limit the use of tax assessment data. Not data that it created, not data that it gathered, but just data that city employees used its software to gather. The court that heard this case rejected this argument, saying, “It would be appalling if such an attempt could succeed.”

The lay reader may be forgiven for wondering whether lawyers have lost their minds, that such claims are not immediately laughed out of court. So let’s talk briefly about why these things are not obviously forbidden under copyright law.

Copyright Law

The first line of defense is 17 U.S.C. § 105:

Copyright protection under this title is not available for any work of the United States Government, but the United States Government is not precluded from receiving and holding copyrights transferred to it by assignment, bequest, or otherwise.

If the United States government writes it, it’s not copyrightable. This is in one sense a natural consequence of the rule from Wheaton v. Peters. But the rule is both broader and narrower than we’d like if we’re thinking about access to the law. It’s broader in that there are lots of other federal works that are not copyrightable. For example, NASA’s space photographs aren’t copyrighted and they’re freely usable by anyone. But the rule is limited to the federal government. Congress hasn’t prohibited states or cities from taking copyright in the works they produce.

Also, the federal government can be the assignee of a copyright. In such a case, the work may be copyrightable even though it seems to be a government work prima facie.

Another common approach is to emphasize 17 U.S.C. § 103, which deals with “compilation” copyrights. If you gather together a bunch of uncopyrightable things, the things themselves don’t suddenly become copyrightable merely because you collected them. If you take a group of public domain poems, such as Shakespeare’s sonnets, you can’t recopyright them. This is an important principle, but it’s also limited. The rule that compilation copyrights don’t cover underlying material is part of a larger provision that says that copyright is available for the compilation as a whole. If I took Shakespeare’s sonnets and arranged them into an original order that I thought displayed the logical progression of the author’s emotional relationship with the Dark Lady that might well be copyrightable as an original arrangement. The same would be true if I wrote an introductory essay, or short summaries of each sonnet. Add additional material, and you get a copyright in that additional material.

The result is that these modern Wheatons engage in a process of intermingling. They take public-domain legal material, add some copyrightable bells and whistles, and make their combined version the easiest, most accessible form of the law. If it should so happen that their copyrightable contributions are now hard to separate from the underlying public-domain matter, well, that’s just too bad for the rest of us.

(This essay is licensed. It is canonically available and has been written by James Grimmelmann with inputs from others.)