Post Mortem of the Reverse Doctrine of Equivalents

The Federal Circuit in Tate Access Floors, Inc. v. Interface Architectural, 279 F.3d 1357, 1368 (2002), announced the death of the Reverse Doctrine of Equivalents (RDOE). The Supreme Court created the RDOE as an equitable release valve for accused devices that literally infringe claims. The RDOE applies “where a device is so far changed in principle from a patented article that it performs the same or a similar function in a substantially different way, but nevertheless falls within the literal words of the claim.” Graver Tank & Manufacturing Co. v. Linde Air Products Co., 339 U.S. 605, 608-609 (1950); see also Boyden Power-Brake Co. v. Westinghouse, 170 U.S. 537 (1898). In such a case, the RDOE “may be used to restrict the claim and defeat the patentee's action for infringement.” Graver Tank, 399 U.S. at 609.

In Tate Access, the court reasons that the RDOE is superfluous because its function was replaced, after Graver Tank, with '112 of the Patent Act. According to the court, the requirements of '112, such as written description and enablement, are “co-extensive with the broadest possible reach of the [RDOE]” because '112 prevents excessively broad claims from being valid and, especially in the case of means-plus-function claims under '112 paragraph 6, acts to reduce the scope of claims. Tate Access, 279 F.3d at 1368. As discussed below, not only are the court's reasons for the RDOE's supposed death inaccurate, they reveal how the ghost of the RDOE lives on in claim construction and how it may be stronger in death than in life.

The RDOE Is Not Co-extensive with '112

The Tate Access court's pronouncement of parallel coverage between '112 and the RDOE is surprising because, in theory, they are separate and distinct. Since Markman v. Westview Instruments, Inc., 517 U.S. 370, 384 (1996), all patent infringement cases occur in two steps: 1) claim construction, which includes a '112 analysis, and 2) an infringement determination, which includes the RDOE. According to Markman and its progeny, these two steps involve separate questions that should not have cross-influence. The first is a question of law determined by a judge and the second is a question of fact determined by a jury. As a result, construction of the claims is to occur without referencing the accused device while infringement and the RDOE involves a direct comparison of the claims and accused device. See NeoMagic Corp. v. Trident Microsystems, Inc., 287 F.3d 1062, 1074 (Fed. Cir. 2002).

In essence, the Tate Access court believes that a rigorous '112 analysis in claim construction is sufficient to limit broad claim language, thereby satisfying the equitable tension addressed by the RDOE. Despite this sharp line drawn by Markman and Tate Access, the courts are still drawn to the equity considerations that fueled the creation of the RDOE.

While Tate Access may be right when it says the RDOE is often mentioned but rarely applied (never by the Federal Circuit), the RDOE does allow a valuable comparison not found in '112. Because a patent only has to meet the '112 requirements for technology that existed on its filing date, see Vas-Cath Inc. v. Mahurkar, 935 F.2d 1555, 1563-64 (Fed. Cir. 1991), defendants using technology developed after the filing date have no '112 recourse to limit broad claim language. There needs to be a mechanism to limit claim language by referencing later-developed technology before '112 is truly “co-extensive” with the RDOE. In order to fill this void and achieve equity, the court has turned to incorporating RDOE principles into claim construction.

The RDOE Lives On in Claim Construction

An example of this includes Biogen, Inc. v. Berlex Labs, Inc., 318 F.3d 1132, 1140 (Fed. Cir. 2003), where the court conflated a RDOE-type analysis into its claim construction. The Biogen court used a narrow disclosure to limit otherwise facially broad claims rather than invalidating them. In doing this, the court cited Texas Instruments, Inc. v. Int'l Trade Comm'n, 846 F.2d 1369, 1371-72 (Fed. Cir. 1988) as properly using “the principle of the RDOE” to preserve the validity of facially broad claims, and Netword, LLC v. Centraal Corp., 242 F.3d 1347, 1352 (Fed. Cir. 2001) as using embodiments disclosed in the Specification to limit broad claims without invalidating them.

But the court's motivation in limiting broad claim language is not always as transparent as it was in Biogen. Currently there is a set of inconsistent opinions that use a variety of ways to restrict claims having broad ordinary meanings. Examples include limiting the claims when the inventor used boilerplate language such as “the invention is” or “this is important to the invention.” See, e.g., Toro Co. v. White Consol. Indus., 199 F.3d 1295, 1301, 1302 (Fed. Cir. 1999) (the court limited the otherwise broad ordinary meaning of the claim term because the patentee had referred to that embodiment as “important to the invention”). The court has even gone as far as saying the claim was “inherently limited” based on statements within the intrinsic evidence stressing a particular functional limitation. Netword, LLC v. Centraal Corp., 242 F.3d 1347, 1352 (Fed. Cir. 2001).

But there are many other cases that go the other way, refusing to limit broad claim language, making the current doctrine confusing and irreconcilable. See, e.g., Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313, 1326-28 (Fed. Cir. 2002) (in construing the claim term “clip” the court stated that a mere reference to a single embodiment is not enough to limit the term's plain meaning even when it is the only embodiment disclosed); Advanced Cardiovascular Sys. Inc. v. Scimed Life Sys. Inc., 261 F.3d 1329, 1339 (Fed. Cir. 2001) (stating that a consistently narrow depiction of a claim element throughout the drawings is not enough to limit an otherwise broad claim term); compare Johnson Worldwide Assocs. v. Zebco Corp., 175 F.3d 985, 990-92 (Fed. Cir. 1999) (in rejecting the argument that a patentee had redefined a broad claim term, the court stated that a varied use of a disputed claim term in a Specification supports a broad ordinary meaning) with Datapoint Corp. v. Standard Microsystems Corp., 2002 U.S. App. LEXIS 3617; 31 Fed. Appx. 685, 689 (Fed. Cir. 2002) (unpublished opinion) (a lone reference to a broader genus does not overcome a narrow disclosure and a narrow claim construction).

The starkest contrast to the narrowing cases may be Gart v. Logitech, Inc. In Gart the court did not limit a broad claim term even though all of the drawings depicted the term narrowly and the Specification distinguished the prior art in part based on this narrow embodiment. 254 F.3d 1334, 1342-43 (Fed. Cir. 2001). The court stated that while the drawings only depict a single preferred embodiment, that alone was not enough to limit the unambiguously broad language of the claim. Id. Furthermore, the Specification distinguished only the “preferred embodiment” (not the invention as a whole) from the prior art using several alternative distinguishing features; the patentee therefore had not limited all of his claims to any one of those specific limitations. Id.

In principle these cases are accomplishing a goal of the RDOE albeit within claim construction: “The purpose of restricting the scope of such claims is not only to avoid a holding of infringement when a court deems it appropriate, but often is to preserve the validity of claims with respect to their original intended scope.” See Texas Instruments, 846 F.2d at 1372. If the court were up front about this motivation, then greater predictability and rationality could be injected into the case law surrounding claim construction. The only obstacle seems to be Markman and the canon of not construing the claims in reference to the accused device. But as discussed below, many other canons surrounding extrinsic evidence are fading, which may someday lead to a softening of Markman.

Claim Construction Is Already Moving Toward a More Inclusive Test

Despite strong early case law to the contrary, the Federal Circuit has been expanding the type of permissible evidence considered during claim construction. One example is the entrance of extrinsic evidence into the forefront. This is a very surprising (although rational) move from the Federal Circuit case largely considered the early bible of claim construction, Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576 (Fed. Cir. 1996). In Vitronics, the court indicated that extrinsic evidence is not to be considered in claim construction if the intrinsic evidence alone may resolve any ambiguities. Id. at 1583. “Such instances will rarely, if ever, occur” when expert testimony will be needed to enable the court to construe disputed claim terms. Id. at 1585.

Fast forward to present day, “failure to take into account the testimony of persons of ordinary skill in the art may constitute reversible error.” AFG Indus., Inc. v. Cardinal IG Co., Inc., 239 F.3d 1239, 1249 (Fed. Cir. 2001). “[U]nder Vitronics it is entirely appropriate, perhaps preferable, for a court to consult trustworthy extrinsic evidence. Pitney Bowes, Inc. v. Hewlett-Packard Co., 182 F.3d 1298, 1309 (Fed. Cir. 1999). Thus, even if the meaning of the claim term is clear from the intrinsic evidence, the court must at least consider expert testimony; but cannot rely upon it if it contradicts the reasonably ascertainable meaning of the claim term. Id. at 1308. Such a dramatic swing in so little time is fascinating: from rarely, if ever appropriate to perhaps preferable and reversible error not to consider.

The importance of extrinsic evidence does not end with expert testimony; dictionaries and treatises, although extrinsic, are consulted in many Markman hearings, even before the Specification or file history. Texas Digital, Inc. v. Telegenix, Inc., 308 F.3d 1193, 1202 (Fed. Cir. 2002); Dow Chem. Co. v. Sumitomo Chem. Co., 257 F.3d 1364, 1372 (Fed. Cir. 2001) (“in determining the ordinary meaning of a technical term, courts are free to consult scientific dictionaries and technical treatises at any time); Vitronics, 90 F.3d at 1584 n.6. The court finds dictionaries and the like particularly useful in ascertaining the ordinary and customary meaning of claim terms. Thus, since claim construction begins with the claims, a dictionary is the next resource consulted.

Another dramatic advance in allowable extrinsic evidence is the emergence of the “court appointed technical advisor.” The district courts are now free to appoint an ex parte expert to advise them on any and all aspects of the technology including scientific and factual issues; apparently this includes advisement on both the patented and accused device. TechSearch L.L.C. v. Intel Corp., 286 F.3d 1360, 1376 (Fed. Cir. 2002). And this power is not strictly regulated by the Rules of Procedure; it is largely up to the district court's wide discretion whether or not to use these advisors. Id.

Finally, undeniable proof the canon excluding accused devices in claim construction is eroding comes from the universal practice of almost every claim construction brief having some section comparing the accused device to the patent claims, regardless if it is after developed technology. For example, an accused product may enter claim construction to help the court frame the issues of infringement. This is especially true when the patentee appears to be basing its broad definition specifically to entrap the accused product. In Multiform Desiccants, Inc. v. Medzam Ltd., the court summarily resolved the issue of literal infringement right after construing the claims precisely narrow enough to exclude the accused product. 133 F.3d 1473, 1476 (Fed. Cir. 1998). This type of decision is neither unusual nor surprising. The remarkable part of the opinion, however, is where the court expressly recognized that the claim construction was done “in light of the mode of action of the accused device.” Id. at 1478. Usually, the court is not so open about relying on the accused device during claim construction, given the historical rhetoric against it. The court even cited the authority that states this position. Nonetheless, the court agreed with the claim construction seemingly approving of the methods used by the trial court.

The reasoning of the court is very pragmatic, realizing that a judge should not have to sit in a vacuum when interpreting claims. The defendant argued that the patentee's claim construction was based entirely upon their understanding of how the accused product worked. Id. at 1477. They further alleged that the Specification warranted a narrower construction eliminating the possibility of infringement. Id. The only real basis for the court's decision seems to be found within the proposition that infringement and claim construction may sometimes be resolved at the same time: “On occasion the issue of literal infringement may be resolved with the step of claim construction, for upon correct claim construction it may be apparent whether the accused device is within the claims.” Id. at 1476 (citations omitted).

Thus, despite the canon denying reference to the accused device, this is an available option for a party to get the accused device before the judge during claim construction. And since this strategy does not have a time bar, as does '112, it is a good way to get after developed technology before the court.

But the confusion and inconsistency under the Federal Circuit claim construction law still remains. In order to untangle claim construction law, the RDOE-type analysis must either be fully included (or fully excised) so it can be dealt with in a systematic manner. Following the demise of canons against extrinsic evidence, it may follow that a full RDOE-type analysis may someday formally enter claim construction, alleviating the tension in equity created when claims cover more than they disclose.

Conclusion

The RDOE is not as dead as the Federal Circuit claims in Tate Access. Its policies are found in the current menagerie of opinions in claim construction using written description and enablement arguments to limit broad claims. Given the eroding canons against extrinsic evidence, a full RDOE-type analysis may someday formally enter claim construction. Not only would this help reconcile conflicting case law, it would help bring some certainty and predictability in construing claims. Once the real motivations are revealed and able to be formally confronted in litigation, a comprehensible set of rules should follow. Otherwise, confusion will continue to reign when trying to decide if broad claim language will be given full effect by the courts.

The Federal Circuit in Tate Access Floors, Inc. v. Interface Architectural, 279 F.3d 1357, 1368 (2002), announced the death of the Reverse Doctrine of Equivalents (RDOE). The Supreme Court created the RDOE as an equitable release valve for accused devices that literally infringe claims. The RDOE applies “where a device is so far changed in principle from a patented article that it performs the same or a similar function in a substantially different way, but nevertheless falls within the literal words of the claim.” Graver Tank & Manufacturing Co. v. Linde Air Products Co., 339 U.S. 605, 608-609 (1950); see also Boyden Power-Brake Co. v. Westinghouse, 170 U.S. 537 (1898). In such a case, the RDOE “may be used to restrict the claim and defeat the patentee's action for infringement.” Graver Tank, 399 U.S. at 609.

In Tate Access, the court reasons that the RDOE is superfluous because its function was replaced, after Graver Tank, with '112 of the Patent Act. According to the court, the requirements of '112, such as written description and enablement, are “co-extensive with the broadest possible reach of the [RDOE]” because '112 prevents excessively broad claims from being valid and, especially in the case of means-plus-function claims under '112 paragraph 6, acts to reduce the scope of claims. Tate Access, 279 F.3d at 1368. As discussed below, not only are the court's reasons for the RDOE's supposed death inaccurate, they reveal how the ghost of the RDOE lives on in claim construction and how it may be stronger in death than in life.

The RDOE Is Not Co-extensive with '112

The Tate Access court's pronouncement of parallel coverage between '112 and the RDOE is surprising because, in theory, they are separate and distinct. Since Markman v. Westview Instruments, Inc., 517 U.S. 370, 384 (1996), all patent infringement cases occur in two steps: 1) claim construction, which includes a '112 analysis, and 2) an infringement determination, which includes the RDOE. According to Markman and its progeny, these two steps involve separate questions that should not have cross-influence. The first is a question of law determined by a judge and the second is a question of fact determined by a jury. As a result, construction of the claims is to occur without referencing the accused device while infringement and the RDOE involves a direct comparison of the claims and accused device. See NeoMagic Corp. v. Trident Microsystems, Inc., 287 F.3d 1062, 1074 (Fed. Cir. 2002).

In essence, the Tate Access court believes that a rigorous '112 analysis in claim construction is sufficient to limit broad claim language, thereby satisfying the equitable tension addressed by the RDOE. Despite this sharp line drawn by Markman and Tate Access, the courts are still drawn to the equity considerations that fueled the creation of the RDOE.

While Tate Access may be right when it says the RDOE is often mentioned but rarely applied (never by the Federal Circuit), the RDOE does allow a valuable comparison not found in '112. Because a patent only has to meet the '112 requirements for technology that existed on its filing date, see Vas-Cath Inc. v. Mahurkar, 935 F.2d 1555, 1563-64 (Fed. Cir. 1991), defendants using technology developed after the filing date have no '112 recourse to limit broad claim language. There needs to be a mechanism to limit claim language by referencing later-developed technology before '112 is truly “co-extensive” with the RDOE. In order to fill this void and achieve equity, the court has turned to incorporating RDOE principles into claim construction.

The RDOE Lives On in Claim Construction

An example of this includes Biogen, Inc. v. Berlex Labs, Inc., 318 F.3d 1132, 1140 (Fed. Cir. 2003), where the court conflated a RDOE-type analysis into its claim construction. The Biogen court used a narrow disclosure to limit otherwise facially broad claims rather than invalidating them. In doing this, the court cited Texas Instruments, Inc. v. Int'l Trade Comm'n, 846 F.2d 1369, 1371-72 (Fed. Cir. 1988) as properly using “the principle of the RDOE” to preserve the validity of facially broad claims, and Netword, LLC v. Centraal Corp., 242 F.3d 1347, 1352 (Fed. Cir. 2001) as using embodiments disclosed in the Specification to limit broad claims without invalidating them.

But the court's motivation in limiting broad claim language is not always as transparent as it was in Biogen. Currently there is a set of inconsistent opinions that use a variety of ways to restrict claims having broad ordinary meanings. Examples include limiting the claims when the inventor used boilerplate language such as “the invention is” or “this is important to the invention.” See, e.g., Toro Co. v. White Consol. Indus., 199 F.3d 1295, 1301, 1302 (Fed. Cir. 1999) (the court limited the otherwise broad ordinary meaning of the claim term because the patentee had referred to that embodiment as “important to the invention”). The court has even gone as far as saying the claim was “inherently limited” based on statements within the intrinsic evidence stressing a particular functional limitation. Netword, LLC v. Centraal Corp., 242 F.3d 1347, 1352 (Fed. Cir. 2001).

But there are many other cases that go the other way, refusing to limit broad claim language, making the current doctrine confusing and irreconcilable. See, e.g., Teleflex, Inc. v. Ficosa N. Am. Corp., 299 F.3d 1313, 1326-28 (Fed. Cir. 2002) (in construing the claim term “clip” the court stated that a mere reference to a single embodiment is not enough to limit the term's plain meaning even when it is the only embodiment disclosed); Advanced Cardiovascular Sys. Inc. v. Scimed Life Sys. Inc., 261 F.3d 1329, 1339 (Fed. Cir. 2001) (stating that a consistently narrow depiction of a claim element throughout the drawings is not enough to limit an otherwise broad claim term); compare Johnson Worldwide Assocs. v. Zebco Corp., 175 F.3d 985, 990-92 (Fed. Cir. 1999) (in rejecting the argument that a patentee had redefined a broad claim term, the court stated that a varied use of a disputed claim term in a Specification supports a broad ordinary meaning) with Datapoint Corp. v. Standard Microsystems Corp., 2002 U.S. App. LEXIS 3617; 31 Fed. Appx. 685, 689 (Fed. Cir. 2002) (unpublished opinion) (a lone reference to a broader genus does not overcome a narrow disclosure and a narrow claim construction).

The starkest contrast to the narrowing cases may be Gart v. Logitech, Inc. In Gart the court did not limit a broad claim term even though all of the drawings depicted the term narrowly and the Specification distinguished the prior art in part based on this narrow embodiment. 254 F.3d 1334, 1342-43 (Fed. Cir. 2001). The court stated that while the drawings only depict a single preferred embodiment, that alone was not enough to limit the unambiguously broad language of the claim. Id. Furthermore, the Specification distinguished only the “preferred embodiment” (not the invention as a whole) from the prior art using several alternative distinguishing features; the patentee therefore had not limited all of his claims to any one of those specific limitations. Id.

In principle these cases are accomplishing a goal of the RDOE albeit within claim construction: “The purpose of restricting the scope of such claims is not only to avoid a holding of infringement when a court deems it appropriate, but often is to preserve the validity of claims with respect to their original intended scope.” See Texas Instruments, 846 F.2d at 1372. If the court were up front about this motivation, then greater predictability and rationality could be injected into the case law surrounding claim construction. The only obstacle seems to be Markman and the canon of not construing the claims in reference to the accused device. But as discussed below, many other canons surrounding extrinsic evidence are fading, which may someday lead to a softening of Markman.

Claim Construction Is Already Moving Toward a More Inclusive Test

Despite strong early case law to the contrary, the Federal Circuit has been expanding the type of permissible evidence considered during claim construction. One example is the entrance of extrinsic evidence into the forefront. This is a very surprising (although rational) move from the Federal Circuit case largely considered the early bible of claim construction, Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576 (Fed. Cir. 1996). In Vitronics, the court indicated that extrinsic evidence is not to be considered in claim construction if the intrinsic evidence alone may resolve any ambiguities. Id. at 1583. “Such instances will rarely, if ever, occur” when expert testimony will be needed to enable the court to construe disputed claim terms. Id. at 1585.

Fast forward to present day, “failure to take into account the testimony of persons of ordinary skill in the art may constitute reversible error.” AFG Indus., Inc. v. Cardinal IG Co., Inc., 239 F.3d 1239, 1249 (Fed. Cir. 2001). “[U]nder Vitronics it is entirely appropriate, perhaps preferable, for a court to consult trustworthy extrinsic evidence. Pitney Bowes, Inc. v. Hewlett-Packard Co., 182 F.3d 1298, 1309 (Fed. Cir. 1999). Thus, even if the meaning of the claim term is clear from the intrinsic evidence, the court must at least consider expert testimony; but cannot rely upon it if it contradicts the reasonably ascertainable meaning of the claim term. Id. at 1308. Such a dramatic swing in so little time is fascinating: from rarely, if ever appropriate to perhaps preferable and reversible error not to consider.

The importance of extrinsic evidence does not end with expert testimony; dictionaries and treatises, although extrinsic, are consulted in many Markman hearings, even before the Specification or file history. Texas Digital, Inc. v. Telegenix, Inc., 308 F.3d 1193, 1202 (Fed. Cir. 2002); Dow Chem. Co. v. Sumitomo Chem. Co., 257 F.3d 1364, 1372 (Fed. Cir. 2001) (“in determining the ordinary meaning of a technical term, courts are free to consult scientific dictionaries and technical treatises at any time); Vitronics, 90 F.3d at 1584 n.6. The court finds dictionaries and the like particularly useful in ascertaining the ordinary and customary meaning of claim terms. Thus, since claim construction begins with the claims, a dictionary is the next resource consulted.

Another dramatic advance in allowable extrinsic evidence is the emergence of the “court appointed technical advisor.” The district courts are now free to appoint an ex parte expert to advise them on any and all aspects of the technology including scientific and factual issues; apparently this includes advisement on both the patented and accused device. TechSearch L.L.C. v. Intel Corp., 286 F.3d 1360, 1376 (Fed. Cir. 2002). And this power is not strictly regulated by the Rules of Procedure; it is largely up to the district court's wide discretion whether or not to use these advisors. Id.

Finally, undeniable proof the canon excluding accused devices in claim construction is eroding comes from the universal practice of almost every claim construction brief having some section comparing the accused device to the patent claims, regardless if it is after developed technology. For example, an accused product may enter claim construction to help the court frame the issues of infringement. This is especially true when the patentee appears to be basing its broad definition specifically to entrap the accused product. In Multiform Desiccants, Inc. v. Medzam Ltd., the court summarily resolved the issue of literal infringement right after construing the claims precisely narrow enough to exclude the accused product. 133 F.3d 1473, 1476 (Fed. Cir. 1998). This type of decision is neither unusual nor surprising. The remarkable part of the opinion, however, is where the court expressly recognized that the claim construction was done “in light of the mode of action of the accused device.” Id. at 1478. Usually, the court is not so open about relying on the accused device during claim construction, given the historical rhetoric against it. The court even cited the authority that states this position. Nonetheless, the court agreed with the claim construction seemingly approving of the methods used by the trial court.

The reasoning of the court is very pragmatic, realizing that a judge should not have to sit in a vacuum when interpreting claims. The defendant argued that the patentee's claim construction was based entirely upon their understanding of how the accused product worked. Id. at 1477. They further alleged that the Specification warranted a narrower construction eliminating the possibility of infringement. Id. The only real basis for the court's decision seems to be found within the proposition that infringement and claim construction may sometimes be resolved at the same time: “On occasion the issue of literal infringement may be resolved with the step of claim construction, for upon correct claim construction it may be apparent whether the accused device is within the claims.” Id. at 1476 (citations omitted).

Thus, despite the canon denying reference to the accused device, this is an available option for a party to get the accused device before the judge during claim construction. And since this strategy does not have a time bar, as does '112, it is a good way to get after developed technology before the court.

But the confusion and inconsistency under the Federal Circuit claim construction law still remains. In order to untangle claim construction law, the RDOE-type analysis must either be fully included (or fully excised) so it can be dealt with in a systematic manner. Following the demise of canons against extrinsic evidence, it may follow that a full RDOE-type analysis may someday formally enter claim construction, alleviating the tension in equity created when claims cover more than they disclose.

Conclusion

The RDOE is not as dead as the Federal Circuit claims in Tate Access. Its policies are found in the current menagerie of opinions in claim construction using written description and enablement arguments to limit broad claims. Given the eroding canons against extrinsic evidence, a full RDOE-type analysis may someday formally enter claim construction. Not only would this help reconcile conflicting case law, it would help bring some certainty and predictability in construing claims. Once the real motivations are revealed and able to be formally confronted in litigation, a comprehensible set of rules should follow. Otherwise, confusion will continue to reign when trying to decide if broad claim language will be given full effect by the courts.