Patent prosecutors eventually encounter a familiar phenomenon during examination: an examiner adopts a claim interpretation that appears to disregard not merely the specification or the prior art, but basic English grammar itself. The issue often arises under the banner of the “broadest reasonable interpretation” standard, where a claim term is stretched beyond ordinary usage in order to read on a cited reference. At times, the interpretation is aggressive but defensible. At other times, the interpretation crosses into linguistic revisionism, where nouns become verbs, structures become functions, and claim language is effectively rewritten in order to sustain a rejection.
A recent PTAB decision in Ex parte Wu, Appeal 2026-000809, provides an unusually direct example. The issue on appeal centered on a claim limitation reciting a seal arranged on a precision-machined surface between a cover and the precision-machined surface. The examiner relied on a prior art reference that disclosed mating surfaces which created a sealing effect, but which did not disclose a separate physical seal component. The applicant argued that the examiner improperly transformed the noun “seal” into the verb concept of “sealing.” The examiner acknowledge using the verb form, but maintained it was a reasonable construction. The Board agreed with the applicant and reversed the rejection.
Interestingly, the Board addressed the issue in plain grammatical terms. The majority explained that the plain meaning of the claim requires a seal on the surface not just that a seal that is formed by mating surfaces. The Board further emphasized that the claim requires the noun form ‘a seal,’ which includes the article ‘a’ before the object, ‘seal.’ The Board concluded that both the plain meaning and the broadest reasonable interpretation of the claim consistent with the Specification require ‘a seal’ in the noun form. It is hard not to notice that the decision essentially came down to the Board explaining elementary grammatical distinctions in order to cabin an overextended claim interpretation.
The examiner’s position, as reflected in the Answer, was that the claim does not preclude the verb form interpretation of ‘a seal.’ The examiner further argued that requiring the seal to be only in the noun form is inconsistent with how the original disclosure describes the seal. The dissent adopted a similar view, reasoning that the precision-machined surfaces themselves could constitute the claimed “seal” because the surfaces created a sealing relationship. The dissent therefore concluded that the cited prior art satisfied the limitation through metal-to-metal sealing contact alone.
The majority, however, focused on the actual structure of the claim language and the specification. The specification repeatedly described a physical seal arranged between components. The Board quoted the specification’s disclosure that a seal, e.g., a flat seal, which is arranged on the surface, is arranged between the cover and the surface. The Board also noted that the specification separately discussed “sealing surfaces,” thereby reinforcing the distinction between a surface capable of sealing and a separate physical seal element. The majority further invoked claim differentiation, observing that the examiner’s interpretation would render later claim limitations superfluous.
The case is a useful reminder that claim construction disputes are frequently won or lost on small linguistic details. Patent prosecutors often focus heavily on technological distinctions while underestimating the importance of grammatical precision. Examiners, meanwhile, sometimes exploit ambiguity by shifting between different parts of speech or by conflating structure with function. A claim term drafted as a concrete noun can gradually morph during prosecution into an abstract capability, a functional result, or a process. Once that shift occurs, prior art mappings that initially appeared impossible suddenly become available.
These disputes are not limited to words like “seal.” Similar issues arise when examiners reinterpret “controller” as merely “controlling,” “fastener” as “fastening,” “filter” as “filtering,” or “connector” as simply “connecting.” In each instance, the examiner effectively dissolves a claimed structure into a functional concept broad enough to encompass the prior art. The problem is particularly acute under the broadest reasonable interpretation standard because some examiners treat “broadest” as if it were untethered from grammar, syntax, or ordinary usage. Of course, the governing standard is not the broadest conceivable interpretation. It is the broadest reasonable interpretation consistent with the specification.
The PTAB’s decision illustrates a practical reality of prosecution. Once an examiner and supervisory examiner become entrenched in a claim interpretation, interviews and repeated written responses may yield little progress. Examiners understandably become invested in positions they have already defended through multiple office actions. At that point, an appeal can materially change the dynamic because the issue is presented to a panel that approaches the claim language with fresh eyes and without institutional investment in the rejection. Many prosecution attorneys have experienced situations where a strained interpretation that persisted for months during examination quickly collapsed once presented to the Board in a clean appellate posture.
The appeal record in this case also demonstrates the value of directly framing the issue in ordinary language. The applicant did not merely argue that the reference lacked the limitation. Instead, the applicant identified the precise interpretive maneuver being used by the examiner: converting a noun into a verb. That framing appears to have resonated with the Board because the opinion repeatedly returned to the grammatical distinction itself. Sometimes the most effective appellate argument is not technologically sophisticated at all. It is simply showing that the examiner’s reading does not comport with ordinary English.
The case also offers useful drafting lessons. Patent applicants should carefully distinguish structural components from functional results throughout the specification. If a claim recites a physical element, the specification may repeatedly describe it as a concrete structure and provide examples, positional relationships, and interactions with surrounding components. Prosecutors should also consider using contrasting terminology where appropriate, as occurred here with “seal” versus “sealing surface.” Those distinctions can become highly valuable during prosecution when an examiner attempts to blur structural boundaries.
Similarly, prosecutors should pay close attention to articles, prepositions, and grammatical framing during claim drafting. Small words often carry significant interpretive weight. The Board in this case specifically relied on the article “a” preceding “seal.” That seemingly minor grammatical detail became central to the reversal. Careful drafting can therefore create linguistic anchors that later limit unreasonable interpretive expansion.
During prosecution, practitioners should also avoid assuming that an examiner’s strained interpretation is harmless or merely semantic. If left unchallenged early, those interpretations can become embedded in the prosecution record and later complicate appeal or enforcement positions. When an examiner begins shifting claim terms between noun and verb forms, or between structures and functions, it is often worth confronting the issue directly and expressly. Responses can explain not only why the prior art fails to disclose the limitation, but why the examiner’s interpretation itself is linguistically unreasonable in view of the claim language and specification.
Finally, Ex parte Wu serves as a reminder that precision in patent law is inseparable from precision in language. Patent claims are not merely technical disclosures; they are legal sentences. Grammar matters. Syntax matters. Parts of speech matter. And although the broadest reasonable interpretation standard permits flexibility, it does not authorize rewriting claims into something the applicant did not actually say.

Leave a comment