It’s been a while since I wrote about how the Israel PTO deals with the issue of “overlap” between the claims of applications, but that doesn’t mean the issue has gone away. It’s still there. In fact, rumor has it that the higher-ups at the ILPTO will be reconsidering this policy this year. In light of that, this (admittedly long) post will address the matter again, presenting the issues involved, why I think Israel should implement a terminal disclaimer policy similar to that in the USA, and why I think the legal machinery to implement such a policy is already in place.
One point of clarification at the outset: this discussion does not pertain to situations in which there is overlap between the claims of a later application and the claims of an earlier application (or patent) which was published before the earliest priority date of the later application. In that situation, under sections 4 and 5 of the statute, the earlier application is available as prior art against the later application, and the normal assessments of novelty and non-obviousness are made.
Rather, the issue of “overlap” arises in situations in which there are two different patent applications with claims of partly overlapping scope (e.g. “A table comprising a top and at least three legs” and “A table having a top and four legs”), and neither application was published before the earliest priority date to which the claimed subject matter is entitled. The ILPTO long ago adopted the view that in such cases, the overlapping material must be excised from the later-filed application, even though the earlier application cannot be cited against the novelty or inventiveness of the later application. (The “why” of this policy will be explained shortly.) Within this fact pattern there are two possible situations: the two applications are co-owned (same applicant), or they’re not (different applicants). Although the ILPTO ultimately treats both cases similarly, it uses different legal mechanisms to get there.
Overlapping Claims in Applications Filed by Different Applicants
In the case of applications filed by two different applicants, the ILPTO’s practice of requiring the later applicant to excise overlapping material from its claims is supposedly based on section 9 of the statute:
9. First Comer Wins
If a patent was applied for on one invention by more than one applicant, the patent shall be granted on the invention to the one who first applied for the patent in accordance with the law.
In general, the ILPTO interprets “invention” in this context to mean anything included within the scope of the claims (but see below re: selection inventions). Thus, by requiring the later applicant to amend its claims to eliminate overlap with the claims of the earlier-filed application, the ILPTO believes it is upholding section 9. From a strict legal construction standpoint, there is a problem with this interpretation. Section 13(a) of the statute says that “The specification shall conclude with a claim or claims that define the invention, provided that each such claim shall arise in a reasonable manner from the description in the specification”. This would seem to mean that claims of differing scope define different inventions, even if those claims overlap, and therefore there is no basis to require limitation of the scope of the claims of the later filed application. Be that as it may, the rationale for this policy, evidently, is to avoid the situation in which potential infringers will have to contend with more than one patent covering the same aspect of an invention.
Overlapping Claims in Applicant’s Own Applications
Section 9 is limited to situations in which “patent was applied for on one invention by more than one applicant”. Thus in the case of two applications belonging to the same applicant, section 9 is irrelevant. Nevertheless, in such situations the ILPTO will require the applicant to eliminate the overlap between the claims of the two applications. Again, the rationale is that if two patents are granted and there is overlap between the claims, then one of the patents could be assigned to a different entity, leaving potential infringers with two parties to negotiate with, thus raising the transaction costs of the negotiations (in the words of Ronald Coase), which is economically inefficient. The legal justification for this practice is an alleged violation of section 2 of the statute:
2. Right to apply for a patent
The owner of a patentable invention is entitled to apply, in accordance with the dictates of this statute, that a patent be granted on that invention.
Section 2 says a patent (singular), not patents (plural), and therefore an applicant may not receive two patents for the same invention. As with its interpretation of section 9, in taking a broad view of section 2, the ILPTO ignores section 13(a)’s statement that it is the claims that define the invention. Hence, rather than conclude that claims of different scope are not claims to the same invention, the ILPTO interprets “that invention” in section 2 to mean anything falling within the scope of the claims, so that overlap between the claims of an applicant’s different applications is precluded. It will be appreciated that such rejections are sometimes raised with respect to applications having a parent/divisional relationship, i.e. in which they have the same disclosure, filing date and priority date.
As we have just seen, the rationale behind the ILPTO’s “no overlap” policy is to reduce transaction costs for parties interested in licensing patents (or considering mounting challenges to those patents). It will be appreciated that this policy consideration is waived in the case of so-called “selection inventions”, in which the later claims are dominated, at least in part, by the earlier claims, but (a) there is no specific disclosure in the earlier application of what is claimed in the later application (let alone claims that specifically claim what is claimed in the later application), (b) the degree of overlap is small relative to the earlier disclosure, and (c) an inventive step can be demonstrated for what is claimed in the later application that overlaps with the earlier claims. Apparently here the thinking is that since the later-claimed grouping is inventive over the earlier generic disclosure, the two claimed inventions are not the same, so section 9 (or section 2, in the case of the same applicant) doesn’t apply; but why the “sameness” of the invention should be contingent upon inventive step considerations is unclear. While the selection invention doctrine is normally applied in the context of published prior art in the obviousness context, applicants may invoke this doctrine to overcome “overlap” rejections, whether between their own applications or their application and that of a third party.
The Present “Overlap” Practice is Inefficient
The concern about an applicant effectively multiplying its patent pool so that multiple patents cover the same aspect of the same invention, and the potential for mischief that such multiplication engenders, is not necessary an illegitimate concern. Indeed, as will be discussed shortly, US courts long ago recognized this as a potential problem. But in Israel, having adopted the view that, in the name of promoting efficiency in hypothetical negotiations that may or may not take place, “overlap” between co-pending applications must be eliminated at all costs, the ILPTO has adopted policies that themselves promote inefficiency of a different kind.
Specifically, examiners spend time looking for and identifying overlap in applicants’ co-pending applications. Applicants and their patent counsel must then spend time – in most cases, significant amounts of time – assessing the veracity of the examiners’ claims, and, where necessary, amending claims to eliminate the overlap. The examiner who receives such an amended set of claims must then carefully check to ensure that the “overlap” has been removed – another time-consuming task, one that inter alia prevents the examiner from working on other cases. (The problem is compounded in the not uncommon situation in which an applicant has filed not just two but multiple applications on closely-related subject matter.)
Alternatively, as noted above, in some cases, applicants may invoke the selection invention doctrine, but this too necessitates adducing (in the lab) proof of unexpected superiority, and presenting the results to the ILPTO.
In addition, the amendment of claims to eliminate “overlap” may in turn result in a finding of disunity, necessitating the division of an application as a consequence of the claim amendments. This imposes an additional burden and cost on the applicant. (The burden is further increased under the ILPTO’s incorrect no-divisional-of-a-divisional policy, which has been discussed in earlier posts and hopefully will be revisited in the future.)
US Terminal Disclaimer Policy
As noted above, US courts long ago recognized the problem of claims of overlapping scope in an applicant’s different patent applications, and how the grant of several patents having overlapping claims could lead to “vexatious litigation” if different patents were to be assigned to different entities. In fact, in the US, the problem was more acute than it is in Israel: under the pre-1995 US statute, in which patents had a term of 17 years from the date on which they were granted, but enjoyed the benefit of the filing date of the earlier application in the continuation chain, a savvy applicant could theoretically obtain a series of patents spaced years apart but having overlapping claims. Since the claims weren’t identical, they were deemed to be directed to different inventions. Therefore no section 101 rejection (akin to the Israel section 2 rejection described above – applicant is entitled to a single patent, not multiple patents, on its invention) would arise; and with the continuation priority in place, the earlier patents in the chain could not be cited as prior art. Hence an applicant could in principle extend his patent rights beyond the initial term by obtaining patents on obvious variations of the same invention, by using continuation practice to ensure that the patents issued at different times.
What resulted from the courts’ sensitivity to this potential problem was the judicially created doctrine of “obviousness-type” double patenting. If the claims of a pending application are obvious in view of the claims (not the specification or disclosure, but the claims) of an applicant’s earlier-issued patent (or earlier-filed-but-still-pending application), then the pending application will not be allowed, on the grounds that the later claims are an obvious variation of the earlier claims and therefore in principle could have been included in the earlier patent.
However, this rejection will be waived if the applicant promises two things: first, that the two patents will always be owned by the same entity; second, that the term of the later patent will not extend beyond the term of the earlier patent. In this way, the possibility of “vexatious litigation” is averted, as is the possibility of receiving an effective time-wise extension of exclusivity.
This dual promise by the applicant is made in the form of a document called a “terminal disclaimer”, and its use is so common that the USPTO even has standard terminal disclaimer forms (here and here) that applicants can use if they wish. The terminal disclaimer includes the following wording:
“The owner [of the application in which the terminal disclaimer is being filed] hereby agrees that any patent so granted on the instant application shall be enforceable only for and during such period that it and any patent granted on the [earlier] reference application [or prior patent, as the case may be] are commonly owned. This agreement runs with any patent granted on the instant application and is binding upon the grantee, its successors or assigns.”
The terminal disclaimer is part of the patent’s file history, and a notice that a terminal disclaimer was filed is published on the front of the issued patent. And by its very wording, the terminal disclaimer is binding upon all subsequent owners of the patent.
Thus the terminal disclaimer addresses the same concerns that underlie the current ILPTO practice for dealing with overlap between the claims of an applicant’s own applications. The terminal disclaimer model, however, is a more efficient way to address those concerns: it spares the applicant the need to (i) assess the degree of overlap between the claims of co-pending applications, (ii) devise claims that eliminate the overlap, or (iii) possibly file divisional applications as a result of the amended claims. And it frees examiners from assessing whether or not overlap has been eliminated. Moreover, since both patents will expire simultaneously, it also allows potential infringers to more easily identify the date on which they may commence activities that, if performed before that date, would constitute infringement.
Of course, under US practice, an applicant reserves the right to challenge the propriety of the obviousness-type double patenting rejection, or to amend its claims to overcome that rejection. But it’s the applicant’s choice – the applicant isn’t forced to excise the overlap.
The Legal Mechanism Behind the “Unenforceable if” Provision of Terminal Disclaimers
The terminal disclaimer effectively allows an applicant to say, “The claims of these two applications/patents really should be kept together in a single patent, because the two inventions claimed are not patentably distinct from one another, and therefore I promise that if the two patents cease to be commonly owned, the later-issuing patent won’t be enforceable.” But the legal mechanism by which the terminal disclaimer operates may be explained in a number of different ways.
For example, the terminal disclaimer may be conceived of as a partially conditional dedication of the later patent to the public: if at any time prior to a certain date (viz. the date on which the first patent will expire), the ownership of the two patents is split, then the later patent is dedicated to the public (and is therefore unenforceable). Even if the ownership of the two patents remains in the same hands, the later patent is dedicated to the public as of the date on which the earlier patent expires.
Alternatively, the legal mechanism of the terminal disclaimer maybe thought of as the partially conditional granting of a paid-up, royalty-free license to anyone who is interested exploiting the patent: if prior to a certain date, the ownership of the two patents is split, then a license has been granted to anyone who wishes to practice the later patent. After that date, such a license is granted even both patents remain owned by the same party.
Other ways to conceive of the mechanism of making the later patent unenforceable include the partially conditioned grant to the public of a covenant not to sue; the patentee’s partially conditional forfeiture of the rights conveyed by the later patent; and the patentee’s acceptance of less than all the rights to which it would otherwise be entitled (i.e. the patentee explicitly accepts a patent term in the later patent of less than 20 years from filing if certain events occur subsequent to the grant of the patent, namely the splitting of ownership). The latter may be explained in terms of contract law: a patent is often likened to a contract between the state and the patent applicant, in which the state agrees to give the patentee exclusivity for a claimed invention in exchange for adequate disclosure of a way to make and use the invention. Although the state cannot grant the patentee more rights (e.g. a longer period of exclusivity) than the law allows, the patentee can agree to accept less than maximum that the state is allowed to give. The terminal disclaimer is an agreement by which the patentee does that.
The terminal disclaimer also states that it is binding on anyone who steps into the shoes of the applicant, either before or after grant of the patent. This too can be explained in a number of ways. Through the lens of contract law, the patentee can be thought of as agreeing to condition the grant of the patent on the patentee’s other self-imposed conditions being binding on subsequent owners of the patent. It can also be viewed as analogous to an easement or other encumbrance that runs with the land.
If this section has seemed pedantic, it’s been for a good cause: to show that the terminal disclaimer operates using legal mechanisms and frameworks that are familiar to American lawyers – and to their Israeli counterparts. It therefore shouldn’t be too difficult for Israeli lawyers to get their heads around the terminal disclaimer idea – and in the next section, I discuss how Israeli law already contains the legal mechanisms for implementing terminal disclaimer practice.
There is a Mechanism in Israel Law for Introducing Terminal Disclaimer Practice
Terminal disclaimer practice in the USA utilizes familiar legal mechanisms, and seeing how the American model works, it doesn’t take a lot of effort to understand how Israel law likewise has mechanisms that can be put to similar use. For example, sections 82-88 of the Israel patent statute deal with patent assignments and licenses. As long as such assignments or licenses are in writing and recorded with the ILPTO, they are enforceable against third parties. If, as suggested above, the terminal disclaimer is conceived of as the granting of a conditional, paid-up, royalty-free license, then these sections of the statute provide a mechanism for the implementation of terminal disclaimer practice in Israel: the terminal disclaimer would be a non-exclusive license (i) granted to everyone (sections 84 and 86), (ii) in writing (section 84), (iii) having effect toward everyone by virtue of its being recorded (section 87), (iv) with the license’s coming into force being conditioned (section 86) on the splitting of the ownership of the particular patent and one or more additional patents listed in the terminal disclaimer. If the ownership of the patents is split, the license comes into force and effectively dedicates the later patent to the public. If the ownership remains in the same hands, the license has no effect against anyone until the earlier patent expires.
Alternatively, Israel patent law provides for the cancellation of a patent by its owner, either actively by informing the Commissioner that he wishes to cancel the patent, or passively by failing to timely pay a maintenance fee. A terminal disclaimer could thus be viewed as a conditional cancellation of the patent, triggered by the splitting of the ownership or the expiration of the earlier patent. In a comparable fashion, it could be deemed to be a conditional gift to the public, in accordance with the Gifts statute.
Similarly, Israel law provides for imposing conditions on that are binding upon subsequent owners of property. The Patents statute in particular specifically provides for patents serving as pledges against obligations, so that ownership can change hands if the obligation is not fulfilled. The imposition of the conditions of enforceability upon subsequent owners of the patent can thus be viewed, for example, as being the emplacement of a limitation on subsequent owners of the patent, or the pledging of the patent to the benefit of the public, should the ownership thereof and that of the earlier patent ever split.
Thus the legal tools for implementation of terminal disclaimer practice in Israel already exist. That being the case, rather than continuing requiring examiners to look for “overlap” between applications (instead of examining substantive patentability issues), and rather than requiring applicants to spend time and money figuring out ways to eliminate such overlap, the Commissioner should offer applicants an easy out: file a terminal disclaimer and make the “overlap” problem between the applicant’s different applications disappear. This being the Hebrew month of Elul, with Rosh Hashana just around the corner, now would be a good time for the ILPTO to admit the error of its ways and repent by making terminal disclaimer practice available to patent applicants in Israel.