Tomorrow, the Supreme Court will hear arguments in Oil States vs. Greene’s Energy Group, et al. on whether the Patent and Trademark Office, acting through the Patent and Trial Appeal Board (PTAB), can decide the validity of existing patents. The question, in particular, is whether the PTAB is unconstitutionally extinguishing private property rights in a non-Article III forum without a jury. At stake, therefore, is a range of vital issues, including patents, property, and the right to be heard in a real court, with a jury.

In the debate over this case, what has not been sufficiently understood is the significance of the government’s granting property in form of a patent. This was a form of grant that ideally could be invalided only by a court of record.

As a preliminary matter, I should mention why I am turning my attention to the PTAB case. Just before the weekend, I learned that, at a Washington DC conference on Oil States, John Thorne held up my book Is Administrative Law Unlawful? and made a strange claim. He said that “the worry about the administrative state is not about canceling patents, as I read Philip Hamburger, it is about patents that interfere with other people’s existing rights.” (See the video at 48:34.) Well, not so fast.

It is true that my book discusses the executive’s issuance of patents, including the danger of issuing patents that interfere with existing rights. But rather than discuss contemporary patent law, it merely examines late eighteenth-century patent practices in order to illustrate that my constitutional objections to administrative power leave plenty of room for lawful executive power. In the case of patents, the point was simply that my objections do not stand in the way of executive decisions that mimic judicial hearings for purposes of allocating these grants of property rights.

In contrast, the question in Oil States is whether, after the executive has issued a patent, it can hold an adjudication to cancel the patent by declaring it invalid. This looks like an exercise of administrative power to take property rights, and it is difficult to understand why anyone would claim that my book suggests otherwise.

In the circumstances, it seems necessary not merely to reject the strange interpretation of my book but also to explore why the PTAB is unconstitutional. The word “patent” is revealing.

Patents were a form of grant, which could be used to grant a wide range of things. The Crown granted patents for land, for annuities, for offices, and for exclusive rights to inventions. In contrast, the Crown tended to grant other things in other forms—for example, it ordinarily granted personal property without a patent. And when it wanted to create a revokeable privilege rather than a property right, it could simply issue a license.

What was distinctive about the form of grant known as “letters patent”? The name alludes these letters being open, rather than sealed shut. More to the point here, they were issued by the Chancellor under the Great Seal and were understood to be acts of the Court of Chancery, and indeed were enrolled in its records. Echoing this practice, Congress in 1790 authorized the issuance of patents for inventions under the Seal of the United States. Why did these formalities matter? Being acts of record, patents were understood to put third parties on notice of what was granted. Moreover, in the absence of misrecitals, false suggestions, etc., they were considered dispositive evidence of the granted rights, and as a practical matter, their status as records meant they ideally could not be held void or otherwise invalidated except by a court of record.

In the seventeenth-century, the Privy Council had invalidated patents for inventions and thus had evaded the regular courts. And it is not unsurprising that, amid royal assertions of absolute power, the Privy Council thus challenged what were understood as property rights. But this soon became a constitutional anomaly, and in the eighteenth-century the Privy Council gradually relinquished its power to recall patents. Thus, notwithstanding the Privy Council, patents remained, in the ideals of the common law, grants made of record, which were vulnerable only in courts of record.

It is therefore no small matter that, since the adoption of the Constitution, Congress has authorized the grant of “patents.” The Constitution authorizes Congress to “promote the Progress of Science and useful Arts, by securing for limited Times to Authors and Inventors the exclusive Right to their respective Writings and Discoveries.” The use of the word “patent” echoes the Constitution’s words about “securing” an “exclusive right.” The Constitution evidently was speaking about a right to property—a sort of right protected by the courts. And Congress’s use of the phrase “letters patent” in 1790 and later is powerful evidence as to how the Constitution’s authorization was understood at the founding and for a long time thereafter.

Patents were grants of property rights—rights that could be invalidated only in the courts.

Indeed, throughout American history (until very recently) patents have been property rights, vulnerable in the courts, not in administrative adjudications. The Symsbury Case is a good illustration. Connecticut had issued conflicting patents for a substantial parcel of land—initially to Symsbury and then to Windsor. Later, in 1727, the colonial legislature authorized a survey to determine which town had a right to the disputed property, and after the surveyor reached his decision in favor of Windsor, the legislature enacted his adjudication in a statute. Nonetheless, more than a half century later, a plaintiff holding property under Symsbury’s patent challenged the surveyor’s determination and prevailed—in 1785 in the state’s Superior Court and the next year on appeal in the state’s Supreme Court. (Incidentally, one of the judges on the Superior Court was Oliver Ellsworth—a future framer and Chief Justice of the U.S. Supreme Court.)

The Superior Court judges predictably resolved the case on the narrow and familiar ground that an act of the Assembly “could not legally operate to curtail” the prior grant. But the constitutional question was not entirely below the surface. Indeed, the lawyers for the Symsbury plaintiff squarely presented it. They pleaded that because of the legislatively authorized survey, Symsbury’s “Original Grant By the General Assembly” and “their Deed & pattent are Greatly Altered Infringed Reduced & Curtailed[,] Which was Not in the power of the General Assembly Constitutionally to Do.”

The lesson remains potent. Whether granted for tangible or intangible things, patents can be altered, reduced, and curtailed only by courts—not by legislatures or the administrative tribunals they authorize.

Of course, intellectual property is different from land, but just because it is different from an ancient sort of property does not mean it is not property. And as already noted, the key point is not whether intellectual property could be viewed as something less than property, but what it means to protect it in the form of a patent and, more fundamentally, what it means to “secur[e]” it as an “exclusive right.” Evidently, in the United States, once a patent is granted, it is a property right.

Although it has long been conventional to talk about patents as a type of “intellectual property,” the arguments in favor of the PTAB suggest that this is a misnomer. It is to be believed that everyone who has hitherto spoken about “intellectual property” has been under a profound misimpression? Or that property can become non-property at the drop of a hat? If the PTAB is upheld, it will be curious to see if the name of the field changes as rapidly.

In fact, one of the defenses of PTAB is that patents for inventions are nothing more than grants of continuing benefits. From this point of view, one might conclude that the field should be called “intellectual benefits.” Of course, the legislature can authorize the executive to distribute benefits of all sorts, and can authorize the executive to work through judicial like procedures to allocate or withdraw the benefits. But mere benefits are not what the Constitution “secur[es]” as an “exclusive right” or what Congress has always granted in the form of “patents.” And the courts traditionally recognized this.

Finally, it must be mentioned that some justices may fear that if they hold the PTAB unconstitutional, then the entirety of the administrative state will come tumbling down with it. For better or worse, this is improbable. Oil States concerns a very specific sort of property right—one secured as a property right by the Constitution and granted in the form of a patent. It also concerns a very specific type of administrative tribunal—one created for the sole purpose of hearing petitions from anyone in the world seeking executive interference in what the Constitution specifically secures property rights. If understood narrowly in these ways, the case has no necessary consequences for other sorts of property rights, let alone for executive discretion over government benefits.

Of course, patents come with a host of problems that cannot be discussed here. But however these difficulties are to be resolved, it must be recognized that patents are property, which cannot be invalidated outside the courts.

Philip Hamburger is the Maurice & Hilda Friedman Professor of Law at Columbia Law School.