The phrase “invention” can mean many different things. In the technical context (as opposed to economic, humanistic, or cultural), the common usage of “invention” can be different from its legal meaning, which also can vary by country/jurisdiction. Regrettably, under US patent law and practice, the meaning of “invention” is inconsistent, logically flawed, and confusing. Let me explain. Stick with me on this.
For any given patent, the courts have proclaimed that, under US law, “the claims define the invention”. So if a patent contains only a single claim, must that one claim define the invention? There seems to be no other logical choice but “yes”.
But what if three claims, which could be presented together in a single patent, are each instead presented in their own patent, such that each patent presents only one claim? Must each of those patent’s “inventions” be defined by all three claims, two of which are found in other patents? There also seems to be no other logical choice but “yes” if it is the “claims” (plural) that define “the invention” (singular). That is, whether the claims are presented in one patent or several, it must remain true that the “claims define the invention”.
Yet US courts do not (always) take that stance. Instead, in practice the courts generally constrains the exclamation to essentially “the claims of this patent define the invention”. Yet even that constraint is insufficient to adequately explain the law, as sometimes courts rely on the premise that only the asserted claims, or even only the appealed claims, of a patent define “the invention”. And if only a single claim (out of many in a patent) is asserted (or appealed), then apparently only that single claim defines “the invention”.
But the inconsistency goes further, as courts will sometimes “read into” one or more of the claims certain “limitations” found in the “specification” (another inconsistent term that usually means that part of the patent other than its claims). In other words, courts sometimes use what’s said about “the invention” in the descriptive portion of a patent to limit the scope of one or more of its claims, and thus ostensibly the scope of “the invention”. Courts also sometimes constrain the scope of one or more claims based on what was stated about “the invention” during the “prosecution” of a patent application (i.e., the back-and-forth negotiations between the patent applicant and the USPTO’s patent examiner that finally result in the application being granted, appealed, or abandoned).
On occasion, courts will even constrain the scope of one or more allegedly infringed claims based on what was stated about “the invention” during the prosecution of a “related” patent or patent application, i.e., an application that relies for its filing date on the same common ancestor application as the patent containing the allegedly infringed claims of the lawsuit.
Looking at another example, the US patent statute, which both the courts and the USPTO are obligated to apply, states that: “If two or more independent and distinct inventions are claimed in one application, the Director [of the USPTO] may require the application to be restricted to one of the inventions.” Since restricting a given patent to claiming only a single invention is at the discretion of the Director of the USPTO, then it logically must be possible for the Director to not choose to do so. As a result, it must then be possible for the claims of a single patent to define more than one “independent and distinct” invention. In other words, as logically inferred from the patent statute, a given patent can define more than one “invention”. Imposing that inference on the initial exclamation, the entire sets of claims of a patent do not necessarily define “the invention”, but rather can define “the inventions” (plural).
Shifting our focus from patent enforcement to patent procurement, the USPTO’s patent examiners are required to comply with the USPTO’s Manual of Patent Examining Procedure (the “MPEP”). On this topic of restricting patent claims, the MPEP commands, in all caps, that “IT STILL REMAINS IMPORTANT FROM THE STANDPOINT OF THE PUBLIC INTEREST THAT NO REQUIREMENT BE MADE WHICH MIGHT RESULT IN THE ISSUANCE OF TWO PATENTS FOR THE SAME INVENTION.” That is, the claims of two patents are not permitted to be directed to the “same invention”, even if those claims could have both been presented together in a single application. Thus, different patents must mean different inventions. And if each patent is permitted to contain only one claim (which it is), then different claims must mean different inventions.
That logic points us to the inescapable conclusion that each claim must define its own invention, at least from the USPTO’s perspective. That understanding of “invention” is at least consistent with the reality that the USPTO examines each individual claim for patentability, such that the USPTO might find one claim of a patent application (or patent) to be patentable while another is not, i.e., one claim might define an “invention“, yet another does not. For that to be the case, two claims (whether of the same patent or different patents) can’t define “the invention”, but instead must each define its own distinct invention.
So from the USPTO’s perspective, regardless of what the courts say, each claim defines its own unique invention. Unless what’s claimed isn’t inventive, in which case that claim defines a non-invention. Or is it simply an unpatentable invention? But if it’s unpatentable, can it be an invention at all?
To make this puzzle even harder to resolve, one can’t truly know whether a patent’s claim is patentable (and thus whether it defines any invention whatsoever) until the patentability of that claim is challenged in court, the court’s decision is appealed, the appeals court’s decision is again appealed to the Supreme Court, and the Supremes announce, either directly or by refusing to hear the appeal, the final verdict on whether the claim defines any invention at all. Given that the patentability of so few claims is ever pursued that far, the best we can typically say is that a given claim might define an “invention”.
If your head isn’t spinning yet, you haven’t been paying attention. If you have, not only do you deserve an award, but I suspect you agree that under US patent law and practice, the meaning of “invention“ is at best inconsistent, logically flawed, and confusing. At worst, its evolution and use reflect gross negligence (or worse). So, if you want to avoid being caught up in the mess, simply steer clear altogether from the word “invention”.