Proving a Negative Claim Limitation: A Tale of Three Nonprecedential Cases

In the past 10 months, the issue of proving negative claim limitations has cropped up on appeals to the Federal Circuit. At least three panels of the Federal Circuit have addressed the issue, to one degree or another. Each panel has contributed, to some extent, to an understanding of the law. In dealing with the issue, each panel cited no precedent for doing so.

Curiously, all three decisions were designated as nonprecedential.

“An opinion or order which is designated as nonprecedential is one determined by the panel issuing it as not adding significantly to the body of law.” So says Federal Circuit Rule of Practice 32.1(b). Rule 32.1(d) adds: “The court may refer to a nonprecedential disposition in an opinion or order and may look to a nonprecedential disposition for guidance or persuasive reasoning but will not give one of its own nonprecedential dispositions the effect of binding precedent.”

The negative claim limitations issue is one that is almost certain to recur. Perhaps the next time the issue comes up, the decision ought to be precedential, and therefore, binding.

What Is a Negative Limitation?

Negative limitations in claims, generally speaking, curtail the breadth of a claim by expressly reciting an element that is absent or excluded. Furthermore, the negative limitation makes clear that absence or exclusion of the element from the claim is deliberate, not that the element is overlooked or optional or of little importance.

In the past, the Federal Circuit has addressed negative claim limitations in contexts such as definiteness and written description support. But the Federal Circuit has, until recently, given scant attention to a particular legal question:

When a claim is challenged as obvious or anticipated, what (if anything) must be proven to show that a negative claim limitation is disclosed by the prior art? (Proving a negative limitation in the context of an infringement action involves some different considerations, so this scenario is reserved for a later day.)

This Is an Easy Question, Isn’t It?

At first glance, since a negative limitation is something that the claim lacks, the answer would seem to be that no proof is needed, because there is nothing to prove. Yet saying that there is nothing to prove runs counter to the general rule that a challenger to patent claims has the burden to prove all elements when asserting anticipation or obviousness. Claim elements are never assumed to exist, and it is ordinarily not enough to show that a reference might include an element. So, if the challenger must prove something, what is it the challenger must prove?

At second glance, it would seem that if the cited reference is silent as to the element in question, then that ought to be proof enough. Yet, that does not quite satisfy the burden of proof either, since a negative limitation involves an element that is deliberately absent or excluded. Silence about an element is not excluding it, nor is silence the same thing as saying that the element is absent.

At third glance, perhaps one would need to have the prior art reference expressly or impliedly or inherently exclude the element. Such a level of proof, though likely sufficient to prove the negative limitation, is probably more proof than ought to be required. Just as a matter of experience and practicality, patents and printed publications do not routinely exclude elements.

At fourth glance, maybe it ought to be sufficient for a reference to say that the element is optional. This probably would be sufficient, as an optional element is, by definition, one that can be omitted without adversely affecting the basic operations being described. In the case where an element is not expressly identified as optional, a rule of reason must apply to determine whether an element is optional or not. It may be possible to drive an automobile without tires, for example, but doing so would have a significant adverse effect upon functionality, and it would be unreasonable (wouldn’t it?) to say that tires on a conventional car are optional. But once again, as a matter of experience and practicality, patents and printed publications do not routinely list an array of elements that might or might not be present.

Now up to our fifth glance, then, we come to three nonprecedential cases decided within the past 10 months.

The Google Case: November 2018

The first nonprecedential case is Google LLC v. Conversant Wireless Licensing S.A.R.L., No. 2017-2456 (Fed. Cir. Nov. 20, 2018) (before O’Malley, Chen & Stoll, Circuit Judges; opinion by Judge Chen). The negative limitation in the case was “wherein the method is performed without pre-registering the mobile station for the location finding service.”

The Google case turned more upon procedure than upon substance, but there is an interesting discussion there regarding proof of a negative limitation. The challenger in an inter partes review (IPR) argued that the element was shown because the “Staack [reference] nowhere discloses that pre-registration is required to access the location-based services.” In support of this argument, the challenger relied upon previous IPR rulings, one of which said that “[n]egative limitations may be satisfied by silence in the prior art”; the other ruling said that “a negative limitation requiring the absence of an element may be adequately described by a cited prior art reference if that reference does not otherwise require the presence of the element recited in the negative limitation.”

The Federal Circuit panel contributed to an understanding of the issue by acknowledging that these two standards are distinct. In the former, mere silence is all that need be shown. In the latter, it must be affirmatively shown that the element is not required. If silence alone were sufficient, that would remove any need to address whether the prior art indicated that the element was optional.

The patent owner responded that the cited reference did show the element in question, which was not excluded; and that what the reference indicated to be optional was “not germane.”

The Patent Trial and Appeal Board (PTAB) erred, said the Federal Circuit panel, by conflating the two standards: “Without explanation, the Board’s final written decision … conflated its analysis for Petitioners’ negative-limitation argument into its analysis for the non-instituted optional-feature argument.” Since the PTAB failed to address all key arguments, the panel decided that the appropriate course of action was to vacate and remand.

So, the Google case identified at least two distinct proposals by which a challenger may prove a negative limitation but did not say which (if either) was correct or permissible.

The IBM Case: April 2019

Less than six months later, the Federal Circuit decided another nonprecedential case, International Business Mach. Corp. v. Ianacu, No. 2018-1065 et al. (Fed. Cir. Apr. 1, 2019) (before Moore, Taranto and Chen, Circuit Judges; opinion by Judge Taranto). The negative limitation in the case was an authentication operation called a “single sign-on operation.” The use of the word “single” excluded two or more such operations. The PTAB found in an IPR that the patent owner’s claims were anticipated, and this appeal followed.

Without getting too bogged down in the details, the cited art dealt with “associating a meCard with an accessCard,” and the patent owner argued that this was not disclosing a “single sign-on.” The PTAB determined that the cited art “is silent as to what information is included in the accessCard,” so it was not shown that there really was another sign-on.

Unlike the Google case, this time the Federal Circuit panel did have a view as to whether mere silence was enough. “The overall finding that this portion of [the reference] teaches a process involving only one user authentication action is not supported by substantial evidence,” said the Federal Circuit. “To begin with, even if the Board were correct that [the reference] is ‘silent’ about the content of the accessCard, that characterization would not alone support a finding that there was no user authentication action in this scenario if, as appears, the Board meant that it simply could not tell one way or the other whether the accessCard contains credentials. Silence in that sense would not by itself suffice for the Petitioner to meet its burden to prove, by a preponderance of the evidence, that there was no user authentication action in this scenario” (emphasis in original).

Cited as indirect support for this proposition was 35 U.S.C. § 316(e): the statutory requirement that the challenger has the burden of proof. And for good measure, the panel mentioned that the defender had no burden to prove the opposite.

In the IBM case, “silence” seemed to be all that the PTAB had to go on. Other evidence in the case, including the prior art’s own disclosure and IBM’s expert testimony, indicated that that a “single sign-on operation” was not disclosed by the prior art.

So, the IBM case puts forth at least three important principles: One is that the challenger has a burden to prove something in order to establish an excluding limitation. A second principle is that proof of mere silence about the limitation in the prior art is not enough. And a third is that a defender can refer to the prior art disclosure and expert testimony to rebut the challenger’s challenge.

The WAG Acquisition Case: August 2019

Less than six months after the IBM case, the Federal Circuit issued a nonprecedential opinion in WAG Acquisition, LLC v. WebPower, Inc., No. 2018-1617 (Fed. Cir. Aug. 26, 2019) (before Newman, Chen, and Stoll, Circuit Judges; opinion by Judge Stoll). The limitation in question was “said server does not maintain a pointer into a buffer established within said server, for each said user.”

In an IPR, the patent owner argued that a pointer is inherent in the cited art. The PTAB disagreed and ruled that the claims were anticipated.

As a general rule, an element is inherent if it is necessarily present, even if it is not mentioned. E.g., Endo Pharmaceuticals Solutions v. Custopharm, Inc., 894 F.3d 1374, 1381 (Fed. Cir. 2018). In the WAG Acquisition case, the Federal Circuit focused upon whether the use of a pointer in the prior art was necessary. An unspoken assumption was that, if a pointer were indeed inherent in the prior art, then the pointer would not be an optional element, and the negative limitation would have been unproven.

Applying a substantial evidence test, the panel determined that there was sufficient basis in the record for finding that the pointer was an optional element in the prior art, and thus the negative limitation was indeed proven. The panel did not say that mere silence was enough to prove a negative limitation (and it does not appear from the opinion that any party so argued). Nor did the panel say that the patent owner had any sort of burden to prove inherency.

“A reasonable fact finder could find,” said the panel, “that [the] Carmel [reference] does not require use of a pointer for the reasons stated by the Board: Carmel emphasizes client control, lacks specialized server software, and uses pointerless protocols. While Carmel does not specify that a pointer is not used, nothing in the record suggests that a pointer must be used. The Board’s findings are therefore supported by substantial evidence” (emphasis in original).

This brief analysis says quite a lot. The challenger proved that the particular reference is not only consistent with a missing pointer, the reference also describes circumstances in which a pointer is expressly or impliedly unnecessary. The prior art need not expressly exclude the element, but there must be some sort of factual basis for finding that the element could be excluded; and that factual basis ought to be premised upon what the prior art reference does say, rather than upon what it does not say.

The rule of reason, mentioned above, is implied in the analysis of WAG Acquisition: if the element were excluded from a scenario described by the prior art, the prior art affirmatively indicates that that scenario would still reasonably work. It makes little sense to say that an element in the prior art is optional if removing that element would make the prior art scheme non-functional or otherwise problematic.

Now Would Be a Good Time for Some Binding Authority

We can leave aside the question of whether any of these three cases ought to have been precedential in the first place. If a question arises again as to what standard of proof ought to be required when a negative limitation is in question, the Federal Circuit may have no binding precedent to which it may turn. The three nonprecedential cases may actually offer some “guidance or persuasive reasoning,” however, and it may be time for a ruling to be precedential.

It is noteworthy that the three nonprecedential cases are mostly (though not entirely) consistent with one another, and the Federal Circuit may be able to look to the particular facts and reasoning of the cases to make the following principles binding:

1. To establish anticipation or obviousness of a claim that has a negative limitation, a challenger has a burden to prove the limitation in some fashion. The negative limitation cannot be summarily ignored nor deemed self-proving.
2. Mere silence in the prior art is not sufficient to prove a negative limitation. There must be other evidence showing that the element is or could be excluded.
3. A prior art reference that expressly or impliedly or inherently excludes the element would be sufficient to meet the challenger’s burden; but a lesser showing by the challenger may also be sufficient.
4. When the prior art expressly or impliedly discloses that the limitation is optional, then that would be sufficient to disclose the negative limitation; but there may be other ways to satisfy the burden of proof as well.
5. If the defender can show that the element is inherent or otherwise necessary, that will defeat the challenge; but the defender does not have the burden to prove such a thing, and the defender may be able to defeat the challenge in other ways as well.
6. Whether an element is necessary or optional in a prior art reference depends upon what the prior art reference does say, rather than upon its silence. This may involve some detailed analysis of the prior art, and whether it includes any disclosure or suggestion that the element in question could be omitted without substantially compromising the operation of what is disclosed in the prior art.