Construing Claims “Without Reference to the Accused Device” Is Put to the Test

AIPLATalk174Wilson Sporting Goods v. Hillerich & Bradsby (Fed. Cir. 2006, 05–1103).

By Jeff Steck

Following Federal Circuit precedent that “claims may not be construed with reference to the accused device,” it is common for district courts to construe patent claims before they know anything about the products at issue in the case. Through an uncommon procedural twist leading up to Wilson Sporting Goods Co. v. Hillerich & Bradsby Co. (05-1103), the Federal Circuit found itself in the same position as those district courts. 


Wilson owns a patent on a non-traditional softball bat that is best described as being “corked” with, essentially, a beer can. The defendant, H&B, was accused of infringing this patent. Although the Federal Circuit did not have “any information about the accused products,” it is safe to conclude that H&B’s most celebrated product, the solid ash Louisville Slugger, was not at issue.


Before the district court, Wilson essentially lost the case when the court issued a narrow construction of the shape and rigidity of the beer-can insert. Instead of pressing forward with an unfavorable construction, Wilson stipulated that there was no infringement under that construction, and the district court entered a final judgment.


Wilson appealed the claim construction to the Federal Circuit. With no record on the accused products, the Federal Circuit found itself in the position of so many district courts: both sides were fighting over claim construction, and the court had no idea why. With no information about the accused products, the court complained that “this sparse record lacks the complete context for accurate claim construction.” As a result, the court refused to provide its own claim construction, choosing instead to identify just enough errors in the district court’s construction to send the case back on remand.   < ?xml:namespace prefix ="" o />


Although the only issue in the case was claim construction—ostensibly an issue of pure law—the Federal Circuit wanted to understand the accused device. Faced, however, with its own rule that “claims may not be construed with reference to the accused device,” the court had difficulty explaining the contradiction between what it permits in claim construction and what it wants in claim construction. Cf. CVI Beta Ventures Inc. v. Tura LP, 43 U.S.P.Q. 2d 1860, 1861 (Fed. Cir. 1997) (Mayer, J. dissenting) (“[T]he court once again shows that we are solecists…. [W]e are bound to interpret law according to specific rules. By violating these interpretive rules, or by exposing them to exceptions so that our court can amass decisionmaking authority … we inappropriately encumber the process with ambiguity….”).


Hoping, perhaps, that contradiction would prove more enlightening than troublesome, the court restated the relevant law on claim construction in a form more characteristic of a koan than a rule of law 

while a claim is not to be construed in light of the accused device,
in an infringement case,
it must inevitably be construed in context of the accused device.

 (citing SRI Int’l v. Matsushita Elec. Corp. of Am., 775 F.2d 1107, 1118 (Fed. Cir. 1985)) (scansion added, odd grammar in original). 


The panel tried unsuccessfully to show that it was not the first to require evidence of the accused product before conducting claim construction. Citing Bayer AG v. Biovail Corp., 279 F.3d 1340, 1340 (Fed. Cir. 2002), the court noted that it had previously declined to construe claims based on an incomplete record. In Bayer, however, the record was “incomplete” because it was missing “evidence of the meaning of the terms to one of skill in the art at the time of invention”—i.e. the usual evidence in claim construction—not because it was missing evidence on the accused product.


The court then tried to restate the rule against construing claims “with reference to the accused device,” concluding that the rule is simply a prohibition against “biasing the claim construction process to exclude or include specific features of the accused product.” As such, the rule “does not forbid any glimpse of the accused product” before claim construction. The issue, however, is not whether a court is permitted to see the accused product, but whether the court is required to see the accused product.


By remanding this case without a conducting a complete claim construction, the Federal Circuit has sent the message that a district court’s claim construction must include some analysis of the accused products. The district court had given hope to the promise of Markman: an infringement allegation was resolved with no need for a costly trial, or even summary judgment proceedings; the Federal Circuit’s remand showed that promise to be illusory.

Jeffrey Steck is a partner at MBHB and an experienced litigator. He was an editor of the California Law Review at the University of California at Berkeley, where he received his JD.

6 thoughts on “Construing Claims “Without Reference to the Accused Device” Is Put to the Test

  1. 6

    When we (Australians) interview a witness for the purpose of establishing claim construction, we would not necessarily be concerned whether or not that witness had knowledge of the “accused device”. However, that witness would forever be tainted for the purpose of establishing obviousness. Here, it is important not to blur the line between establishing a meaning of a claim and testing for obviousness. A court, not being a person skilled in the art, can certainly decide, with optional witness evidence, what a claim means. As for obviousness, the court is simply not permitted to decide that without witness evidence.

  2. 5

    I think Rader’s parenthetical summary of SRI v. Matsushita is not an esoteric koan, but something rather more mundane–a typo. Here is the sentence from SRI that Rader appears to be paraphrasing: “It is only after the claims have been construed without reference to the accused device that the claims, as so construed, are applied to the accused device to determine infringement.” Given that Rader underlined the first occurence of “construed” in his parenthetical, I think he meant to distinguish it from “applied” which is what the second occurrence of “construed” was meant to be.

  3. 4

    Dan, I think that you would agree that considering accused devices to determine what claim terms should be construed is quite different from needing to see the accused device to determine the meaning of those terms.

  4. 3

    Looks like some CAFC judges talk to each other, but your contributors don’t: the decision in Wilson (Lourie, Rader & Bryson) was foreshadowed in footnote 10 in the previous day’s Exigent Technology case (Gajarsa, Dyk & Prost): “it is appropriate for the court to consider the accused device when determining what aspect of the of the claim should be construed”, among other quotes.

  5. 2

    In response, Robert Groover of Groover & Holmes ( provided the following “Haiku’d Koan Headnote” reflecting on the Wilson Sporting Goods decision:

    Court can’t construe claim
    in light of besmirched device:
    context darkly guides.

  6. 1

    Other than “rigid,” it looks as though the CAFC construed every term that the parties asked for. I’m not sure that that undercuts “the promise of Markman.”

    The wishy-washy language and the koan are really strange and disappointing, but the ultimate holding doesn’t look so bad, unless I’m missing something.

Comments are closed.