Opinion
Civil Action No. 97-CV-427 (DGT)
MAY 22, 2001
Kenneth King, Esq., SCULLY, SCOTT, MURPHY PRESSER, Garden City, NY, for Plaintiffs.
Kenneth P. George, Esq., AMSTER, ROTHSTEIN EBENSTEIN, New York, NY, for Defendant.
MEMORANDUM AND ORDER
Spotless Enterprises, Inc. and Spotless Plastics Pty. Ltd. (together, "Spotless"), manufactures of clothing hangers, brought this action against Carlisle Plastics, Inc., now known as AE Products Group, L.P. ("AE"), for infringing certain claims of a patent held by Spotless. Both parties now move for summary judgment as to the infringement claims. The parties agree that the dispositive issue in this motion is the proper interpretation of the term "parallel edges" in the Spotless patent.
Background
The Patent and Trademark Office ("PTO") issued patent number 5,509,587 (the "'587 Patent") to Spotless on April 23, 1996. The '587 Patent was issued for a clothing hanger invented by Spotless that was designed to maximize display area for retail clothing stores (the "Spotless Hanger"). The basic concept embodied in the '587 Patent is a hanger with two upswept arms, which raise garments held by the hanger. Employed in retail stores, this hanger offers a distinct advantage over a hanger with straight or downward arms by allowing more garments to be displayed in the same amount of vertical space.
At the time the '587 Patent was issued, the concept of a hanger with upswept arms was not new to the industry. Hangers of this general type are called "low profile" hangers. In fact, when Spotless presented an earlier version of the Spotless Hanger to the Patent and Trademark Office in May of 1994, it was rejected because the examiner deemed it to be obvious from a previous patent, a hanger designed by DeVito. Zuckerman Decl., Ex. B, pp. s000319-21. The DeVito hanger, like the Spotless Hanger, shows upswept arms extending from the hook. Silfin Decl. Ex. F. Consequently, the examiner rejected the earlier patent application, stating that "DeVito discloses a structure for a garment hanger including . . . first and second upswept arms." Zuckerman Decl., Ex. B, p. s000319.
In order to overcome this rejection, Spotless amended the claims in its patent application, emphasizing the fact that the arms of the Spotless Hanger were divided into two parts. The first part of each arm, called the "inner arm," extended horizontally from the hook, while the second part of each arm, called the "upswept outer arm," extended upward from the end of each inner arm. In this revised application, Spotless argued that the horizontal inner arms distinguished the hanger from prior art, specifically DeVito, in that it allowed promotional materials to be displayed on the hanger. Id. at s000332. A retailer could display a promotional card on the hanger by folding the card in half, passing the hook of the hanger through a hole in the fold and resting the folded part of the card on the flat inner arms of the hanger. Id. The arms on the DeVito hanger, on the other hand, immediately curve upward from the hook; thus, they do not provide any flat surface to support a promotional card. Id. at s000332-33. The amendments in this revised patent application, submitted to the PTO in June of 1995, id. at s000326, were, nonetheless, insufficient to define over the prior art without one more amendment. Id. at s000346.
Spotless and the examiner agreed to the substance of this final amendment during a conversation on September 28, 1995. Id. at s000346. The amendment altered Claims 1, 9, and 16 to include the requirement that the horizontal inner arms and upswept outer arms have "first and second parallel edges." Id. at s000348. Originally, the examiner had proposed that the claims require "first and second parallel webs," but the word "edges" was substituted, although it is not evident from the prosecution history why this change was made. Wilczynski Decl., Exs. 8, 9. The webs are molded plastic pieces that run along the body of the hanger for structural support, appearing as the top and bottom portions of the body in a cross-sectional drawing. See infra, Figs. 1-3. For some unexplained reason, the faxed proposal from the examiner containing the words "parallel webs" and the subsequent proposal containing the words "parallel edges" were not included in the prosecution history of the '587 Patent, although the claims were amended to include the "parallel edges" language. After this amendment, the PTO allowed the claims, and the hanger received patent number 5,509,587.
Although the parties agree as to which part of the hanger body constitutes a "web," it should be noted that the parties' definition seems to be in conflict with the dictionary definition. While the parties use the term "web" to describe the plastic pieces running along the top and bottom of the hanger body, the relevant dictionary definition seems to define the term as the piece of the structure connecting the top piece to the bottom piece. See Webster's Third New International Dictionary 2591 (1993) (definition 6(b), "the vertical plate or portion connecting the upper and lower flanges or parts of a girder or rail"). Nonetheless, the parties' agreed-upon definition will be used for this opinion.
Included in the '587 Patent were three claims for slightly different versions of the Spotless Hanger, Claims 1, 9, and 13. In their final form, these claims describe, in relevant part, the inner and outer arms as follows:
Plaintiffs address Claims 1 and 13 in the present motion, while defendants address Claims 1, 9, and 13. Pls' Mem. in Support of Motion for Part. Summ. J. ("Pl. Mem."), p. 1; Def. Mem. in Opp. to Pls' Mot. for Part. Summ. J. ("Def. Mem"), p. 12, n. 5. Because the claim language in dispute is identical in all three, this opinion addresses an infringement claim based on any of the three claims.
(b) a body member comprising first and second horizontally opposing extending inner arm portions with first and second parallel edges. . .
(c) first and second upswept arm portions with first and second parallel edges. . .
Zuckerman Decl., Ex. A ('587 Patent).
The '587 Patent contains seventeen drawings, including front view, rear view, cross-sections of the arms of the hanger, and right and left end views. Id. None of the seventeen drawings, however, provides a view from the top of the hanger looking down or from the bottom of the hanger looking up.
Different Dimensions, Inc. ("DDI") once offered a low profile hanger known as the DDI 726 which was very similar to the Spotless Hanger. Zuckerman Decl., ¶ 6. When Spotless received a patent for the its '587 hanger, AE, which had purchased DDI, was concerned that the DDI 726 may infringe the '587 Patent. Id. As a result, Alex Zuckerman, who was president of AE at the time, designed two new low profile hangers, known as the DDI 344 and the DDI 345. Both the DDI 344 and 345 are known by the model number 735 (the "735 Model"). Id. In an attempt to distinguish the 735 Model from the Spotless Hanger, the arms of the 735 Model have a straight top web and a curved bottom web; thus, the top edge of the arm and the bottom edge of the arm are not parallel. Id. The 735 Model is the hanger design which Spotless alleges infringes the '587 Patent.
On April 22, 1998, after this litigation began, Spotless filed an application to reissue the '587 Patent. The differences between the original patent and the reissued patent all related to the definition of the word "edges." Specifically, Spotless sought: 1) to alter two of the cross-sectional drawings by adding labels to the front and back edges of the webs; 2) to add a further description of the hanger in the specification clarifying that these newly labeled edges must be parallel; and 3) to include the omitted examiner proposals concerning the words "parallel webs" and "parallel edges" in the prosecution history. The examiner rejected the first application for reissue, finding, among other things, that it was not necessary because the arms would "clearly inherently" have parallel edges. Wilczynski Decl., Ex. 14, p. 3. Spotless's second application for the reissue application was granted, although there is no indication in the '873 Reissue prosecution history as to how the examiner's concerns were alleviated. The patent was reissued as Reissue Patent 36,873 (the "'873 Reissue"), Pl. Reply Mem., Ex. A. The claims themselves were not altered in any relevant way by the '873 Reissue. Id.
Discussion
The determination whether a product infringes a patent involves a two-step process. Gentry Gallery, Inc. v. Berkline Corp., 134 F.3d 1473, 1476 (Fed Cir. 1998). The district court must first interpret the language of the claim, and, second, the properly interpreted claim must be compared to the challenged product to determine if there is infringement. Id. (citing Carroll Touch, Inc. v. Electro Mechanical Sys., Inc., 15 F.3d 1573, 1576 (Fed. Cir. 1993)). The first question, interpretation, is exclusively for the court to resolve. See Markman v. Westview Inst., Inc., 517 U.S. 370, 116 S.Ct. 1384 (1996).
The parties agree that the only term that requires interpretation in the present motion is the term "parallel edges." The parties further agree that, with respect to the literal infringement claim, the resolution of the first issue, i.e., the proper interpretation of the term "parallel edges," will be dispositive as to the second.
(1) Intrinsic Evidence
When interpreting a term in a patent, a court must first look to the intrinsic evidence. Bell Howell Doc. Mgmt. Prods Co. v. Altek Sys., 132 F.3d 701, 705 (Fed. Cir. 1997). This evidence includes the claims themselves, the written description in the specification, the drawings in the patent, and the prosecution history of the patent. Id. This evidence is usually sufficient to interpret a term, and, if it is sufficient, it is improper for a court to resort to extrinsic evidence. Id. at 705-06.
a. '873 Reissue
The first issue that must be resolved, therefore, is what consideration to give the information in the '873 Reissue that was not included in the '587 Patent. Spotless argues that, because the '587 Patent was surrendered by Spotless when the PTO allowed the '873 Reissue, the latter is the only patent before the court, and all additional information contained in the '873 Reissue is intrinsic evidence.
To be sure, § 252 of Title 35 requires that "every reissued patent shall have the same effect and operation in law, on the trial of actions for causes thereafter arising as if the same had been originally granted in such amended form . . ." See 35 U.S.C. § 252 (Supp. 2000). That statute continues, however:
but in so far as the claims of the original and reissued patents are identical, such surrender shall not affect any action then pending nor abate any cause of action then existing, and the reissued patent, to the extent that its claims are identical with the original patent, shall constitute a continuation thereof and have effect continuously from the date of the original patent.Id.
The term "identical" has been interpreted by the Federal Circuit to mean "without substantive change." Kaufman Co., Inc. v. Lantech, Inc., 807 F.2d 970, 976 (Fed. Cir. 1986); Seattle Box Co., Inc. v. Industrial Crating Packing, Inc., 731 F.2d 818, 828 (Fed. Cir. 1984). Although some minor changes were made to the claims in the '873 Reissue, the parties agree that, for the purposes of this motion, no substantive changes were made to the claims in dispute. Accordingly, it would seem that the additional specification language, drawing labels, and prosecution history in the '873 Reissue should not be considered at all. After all, the statute clearly states that the surrender of the original patent "shall not affect" any pending litigation, and the present litigation was pending at the time the PTO allowed the '873 Reissue. However, the language quoted above was added to protect the patent holder by keeping alive any infringement claims based on the original patent in the event that the patent was reissued during the litigation. Before the enactment of § 252, such claims were dismissed for failure to state a claim due to the fact that the original patent had been surrendered. See Seattle Box, 731 F.2d at 827 (explaining history and purpose of § 252). Thus, the question becomes whether the language of § 252 also protects the alleged infringer by preventing a patent holder from changing the specification, drawings, or prosecution history after litigation has commenced in order to bolster the patent holder's pending infringement claim.
The answer to this question is apparently "no." The Federal Circuit has held that the prosecution history of a reissued patent should be considered in interpreting the language of the claims. See Howes v. Medical Components, Inc., 814 F.2d 638 (Fed. Cir. 1987). In Howes, the Federal Circuit reversed the district court for failing to consider the prosecution history of a reissued patent, despite the fact that the patent was reissued during the pendency of the lawsuit. Id. at 645-46; see also E.I. DuPont de Nemours Co. v. Phillips Petroleum Co., 849 F.2d 1430, 1439 (Fed. Cir. 1988) (finding that district court erred where it "seemed to ignore arguments made during the reissue/reexamination proceeding" when interpreting claims).
It would be troublesome, however, to find that AE's 735 Model infringes the '873 Reissue based on information that was not available to AE when it was designing the 735 Model. The purpose of requiring courts to try to resolve patent infringement claims on the basis of intrinsic evidence and to avoid, if possible, considering extrinsic evidence is that competitors should be entitled to rely on the contents of the file wrapper when attempting to design a product which does not infringe on the patent. Vitronics Corp. v. Conceptronic, Inc., 90 F.3d 1576, 1583 (Fed. Cir. 1996). This is precisely what AE did, based on the original '587 Patent. It would appear not only unjust, but also inconsistent with the underlying policies of the patent law, now to base a determination that AE infringed the patent on information that was not part of the public record at the time AE designed the 735 Model, regardless of whether that information substantively altered the claims of the original patent. To do so would expose any subsequent inventor who seeks to improve the art to the risk of paying substantial damages by the simple expedient of allowing a patent holder to obtain a reissued patent and then use the changes in that patent in a pending infringement suit in order to prevail with its interpretation of a disputed term. See Stairmaster Sports/Medical Prod. Inc. v. Groupe Procycle, Inc., No. Civ. 97-396 (MMS), 1998 WL 290296 (D. Del. May 20, 1998) (recognizing risk of allowing patent holder to "self-servingly make a reissue declaration that broadens the scope of the claims beyond what was intended by the inventor when the initial patent was first filed").
It is notable that, where the claims are substantively altered in a reissued patent, the alleged infringer would be protected by the grant of intervening rights, see 35 U.S.C. § 252, but where, as here, there is no substantive change, the alleged infringer would be subject to damages even for the time prior to the reissued patent. See Kaufman Co., 807 F.2d at 977-79 (affirming award of damages where claims were amended during litigation only to "clarify the completely sufficient language used in the original claims"). While such a distinction makes sense, it also serves to underscore the importance that new information in the reissued patent not alter in any substantive way the claims of the original patent, either by changing the claims themselves or the specification.
This potential risk is balanced to a certain extent by the ability of the court, in interpreting the claim, to take into consideration the motivation for adding or altering the specification in a reissue applied for during the litigation. Nevertheless, it is far from clear that new information added to a patent by way of a reissue during litigation should be given any substantial weight where that information was added for the admitted purpose of "clarifying" the very term which is at the center of the dispute between the parties.
Notwithstanding this concern, the Federal Circuit has been clear on the point, and, accordingly, the '873 Reissue will be considered in full in interpreting the term "parallel edges." In doing so, however, it must be remembered that the first consideration in interpreting a claim is the language of a claim itself. Neither the specification, DuPont, 849 F.2d at 1433, nor the prosecution history, Markman v. Westview Instr., Inc., 52 F.3d 967, 980 (Fed. Cir. 1995), aff'd 517 U.S. 370, 116 S.Ct. 1384 (1996), can be used to "enlarge, diminish or vary the limitation in the claims." Id. This applies with at least equal force to specification language that is altered or added in a reissued patent.
b. Interpretation
Spotless argues that the term "edges" refers to the front and back edges of each web, not to the top and bottom edges of the arm segment itself. Spotless largely predicates its interpretation on two points: first, in the '873 Reissue, the added language in the specification and the amended cross-sectional drawings make clear that the front and back edges of the top web must be parallel and the front and back edges of the bottom web must be parallel; and, second, Spotless's refusal to accept the examiner's proposed amendment altering the claim language to require first and second "parallel webs," shows that the edges described in the claim cannot merely refer to the top and bottom edges because that requirement is synonymous with the rejected "parallel webs" language.
1. The Patent Claims, Specification and Drawings
Several pieces of intrinsic evidence, however, weigh heavily against Spotless's interpretation. First, and most important, is the relevant language of the disputed claims, which, as previously explained, is the prime consideration in their interpretation. Indeed, the language of the claims in the present case is virtually dispositive. In order to understand why this is so, a brief discussion of the construction of the hangers in question is necessary.
To facilitate this discussion, below are three figures depicting cross-sections of various embodiments of the arms of the Spotless Hanger. Figure 1 below depicts a cross-section of the hanger arm with an I-beam construction. Figures 2 and 3 represent cross-sections of the hanger arms in a U-beam and E-beam construction, respectively.
These figures correspond to Figures 6-8 in the '873 Reissue, but have been recreated and re-labeled as an aid to this discussion. See Zuckerman Decl., Ex. A, Figs. 6-8.
The hanger body is constructed such that there are "webs" at the top and the bottom of the arms. When the hanger body is of an I-beam construction, as illustrated in Figure 1, these webs would appear as the top and bottom serifs of the "I" shape in the cross-sectional drawing. Each of these webs has a front and back planar side, "a" and "b" respectively in Figure 1. If the hanger is constructed as a U-beam or E-beam, as illustrated in Figures 2 and 3, each web has a front planar side "a" and the two webs share a back planar side "b."
Spotless's argument is that the specification and the cross-sectional drawings in the '873 Reissue make clear that the front and back top edges, "1a" and "1b" in the Figures above, should be parallel, and the front and back bottom edges, "2a" and "2b" in the Figures above, should be parallel. Spotless argues that because the specification makes clear that these edges should be parallel, the "parallel edges" described in the claims are these same edges. See Vitronics, 90 F.3d at 1583 (in interpreting a term in a patent claim, the specification can often "act as a dictionary" when it implicitly or explicitly define terms).
The problem with Spotless's final leap — the conclusion that these are the edges referred to in the claims — is that the claims describe only two edges of the arms. If Spotless were correct that the claims require that the front and back top edges (1a and 1b) must be parallel and the front and back bottom edges (2a and 2b) must be parallel, then it is four edges in total that are described. But the claims describe only two edges, specifically "first and second" edges of the "arm portions," not four. Put simply, the claims do not describe enough edges for Spotless's interpretation to be plausible. If Spotless were correct, the claims would have to refer to "first and second parallel edges of each web" or, more directly, "front and back parallel edges of each web," suggesting that there are more than two edges described. Because the claims describe only two edges, Spotless's interpretation is inconsistent with the claim language itself.
The language in the specification and labels on the cross-sectional drawings do make it clear that the front and back edges of the webs should be parallel, but that is not what is in dispute in the present case. The relevant question is which edges are referred to in theclaims. As explained in the preceding paragraph, a comparison of the drawing labels and language in the specification with the language in the claims reveals that the "edges" referred to in the specification are not the same edges referred to in the claims.
Moreover, the specification is not as helpful to Spotless's argument as it asserts. A principal of claim construction is that "[w]here a specification does not require a limitation, that limitation should not be read from the specification into the claims." DuPont, 849 F.2d at 1433 (quoting Specialty Composites v. Cabot Corp., 845 F.2d 981, 987 (Fed. Cir. 1988)) (emphasis in original). Here, the additional specification language merely states that it is "preferred" that these edges be parallel, not that this is a requirement. See Pl. Reply Mem., Ex. A, Col. 6, ln. 49-64. Furthermore, labels were only added to two of the six cross-sectional drawings in the '873 Reissue. If the parallel edges described in the specification were required in the claims, one would think that the labels representing these edges would be included in all of the embodiments described in the patent. Thus, the specification and drawings do not necessarily require that the edges referred to be parallel, and, accordingly, that language cannot be read into the claims. Consequently, the specification and labels in the drawings upon which Spotless relies do not resolve the question of which edges are referred to in the claims; they merely suggest that the front and back edges of each web should be parallel — a suggestion that is not included as a requirement in the claims.
In addition, although not as supportive of AE's position as the last point, in another context, the term "edge" is unquestionably used in the specification to refer to the top and bottom edges of the webs. Specifically, toward the end of the upswept arms on the Spotless Hanger are protrusions on top of the upper web and on the bottom of the lower web. In describing the location of these protrusions, the specification states that they are each located "on an outside edge . . . of the upswept arms." Zuckerman Decl., Ex. A, Col. 10, 1. 47-53. Thus, in this specification, the top and bottom of each arm are clearly referred to as "edge[s]," albeit "outside edge[s]." Id. It could be argued, however, that the use of "outside edge[s]," as opposed to just "edges" or "parallel edges," could have been a means of distinguishing the description in that part of the specification from the claims in dispute. Furthermore, the fact that the top and bottom of the webs are described as "edge[s]" does not preclude the same word being used to describe the front and back of the webs. Still, there is no question that the word "edge[s]" in this specific context is being used to refer to the top and bottom edges of the arm. While not conclusive, this specification language does show, at a minimum, that the top and bottom of the webs are referred to as "edges" in the specification, just as the front and back edges are.
This portion of the specification, however, does conclusively refute one of Spotless's other arguments. Spotless points out that the term "edge" in geometry normally refers to the line where two faces of a solid object join. See Webster's Third New International Dictionary 722 (1993). Using this definition at oral argument, Spotless argued that the claims cannot refer to the top and bottom of the arms because those areas are not edges at all — they are flat planar surfaces. However, dictionary definitions of ordinary words used in a patent are "rarely dispositive," because "[a] word describing patented technology takes its definition from the context in which it was used by the inventor."Anderson v. International Eng'g Mfg., Inc., 160 F.3d 1345, 1348-49 (Fed. Cir. 1998). Here, the above specification language clearly refers to the top and bottom surfaces as "edges," notwithstanding the fact that they may, in fact, be planar surfaces. Thus, the patent itself clearly uses the term "edges" in a broader sense than Spotless urges. Spotless's definitional point is, therefore, without merit.
It should also be noted that, in contrast to the flat surfaces on the top and bottom of most of the cross-sectional drawings in the '873 Reissue, the actual hangers brought to the oral argument do not have flat, planar surfaces on the top and bottom of the arms. The surfaces each have a peak, creating more of a true "edge" on the top and bottom. In addition, in one embodiment shown in cross-section in the '873 Reissue, the top and bottom webs are curved, thereby creating something closer to an "edge," than a flat surface. See Pl. Reply Mem., Ex. A, Fig. 10 ('873 Reissue).
In sum, the claim language, specification and drawings ultimately fail to support, and, in large part, contradict Spotless's contention. The language added to the specification and the labels added to two of the cross-sectional drawings in the '873 Reissue do not dictate what is meant by the term "parallel edges" in the claims. Spotless's argument to the contrary is inconsistent with the claim language itself in that Spotless's interpretation would require that the claims describe four edges, but the claims describe only two. Moreover, the suggestion added to the specification is not a requirement, and, consequently, cannot be read into the claims. Therefore, although the issue remains as to which edges are referred to by the term "parallel edges" in the claims, Spotless's interpretation of that term lacks support in the added specification language and drawing labels of the '873 Reissue and is actually refuted by the claims themselves.
2. The Prosecution History
The second piece of evidence relied upon by Spotless is the prosecution history related to the addition of the "parallel edges" language. Spotless points to its rejection of the examiner's initial suggestion that the phrase "parallel webs" be used, and its agreement to use "parallel edges" instead. Because a requirement that the arm portions have parallel top and bottom edges would have the same effect as the initial suggestion that the arm portions have parallel webs, Spotless argues that "parallel edges" cannot simply mean the top and bottom edges. To conclude otherwise, Spotless maintains, would ignore the fact that Spotless rejected the "parallel webs" language in the first instance.
(i) Origin of the "parallel edges" term
The first problem with Spotless's argument is the fact that the examiner himself suggested both the "parallel webs" language and the "parallel edges" language. His choice of the phrase "parallel webs" shows that initially he was concerned that the top and bottom edges must be parallel — if it were the front and back edges that he wanted to be parallel, he would never have suggested the phrase "parallel webs." Therefore, when he suggested the phrase "parallel edges" as a replacement, he must have understood that term to refer to the top and bottom edges because that was his concern, and there is absolutely no evidence that the examiner's concern had changed. If the "parallel edges" language defined a different trait of the Spotless Hanger than the "parallel webs" language, as Spotless suggests, how is the former language consistent with the examiner's desire for the latter? It would seem that Spotless would have been required to include both terms if they referred to different attributes of the hanger. To accept Spotless's position, it would have to be assumed that the examiner wanted additional descriptive language in the claim, but he did not care what part of the hanger was more carefully described. This is an untenable position. Clearly, then, he thought the two phrases were essentially synonymous — i.e. that the word "edges" referred to the top and bottom edges of the webs.
(ii) Reissue application
Spotless's argument is further undermined by the prosecution history relating to the first reissue application filed by Spotless. The initial application for the reissue patent was rejected by the same examiner who had allowed the original '587 Patent. In doing so, the examiner explained that the additional specification language and the labels on the cross-sectional drawings showing parallel front and back edges on the webs were not necessary because the webs "clearly and inherently would be understood to have `parallel edges.'" Wilczynski Decl., Ex. 14, P. 3.
The examiner's choice of language is revealing as to which "edges" are referred to in the claims. Spotless argues that the examiner, by adding the "parallel edges" language to the claims, was requiring that the front and back edges of the webs be parallel. However, in describing the front and back edges of webs in the context of the reissue, the examiner describes them as being "inherently" parallel. If these edges are the same edges that are expressly required to be parallel in the claims, as Spotless argues, the examiner would not describe this quality as inherent; he would have said that the claims already require these edges to be parallel. Conversely, there would have been no reason for his insistence that the claims include the "parallel edges" language in the claims if that language merely described what he felt was an inherent quality. Thus, the examiner's choice to describe the front and back edges of the webs as "inherently" parallel indicates that, contrary to Spotless's position, these are not the edges described in the claims.
Moreover, in rejecting the initial application for reissue, the examiner also stated that "[a]n amendment to the specification to describe `parallel edges' for these cross-sections is not considered required since this has been clearly disclosed in the drawings . . . and terms have been used in the specification . . . which clearly inherently describe `parallel edges.'" Wilczynski Decl., Ex. 14, P. 3. Nowhere does the examiner refer to any requirement in the claims. Why would he state that the drawings and specification disclose that the front and back edges must be parallel if that was an express requirement in the claims? The examiner's statements in this context further indicate that he was not referring to the front and back edges of the webs in adding the "parallel edges" language to the claims.
(iii) Distinguishing DeVito
If the intrinsic evidence discussed thus far leaves any question about to which edges the claims refer, it is answered by reviewing the prosecution history related to distinguishing the DeVito hanger. As mentioned above, the Spotless Hanger was not the first "low profile" hanger on the market, the DeVito hanger also showed upswept arms. It was in the context of distinguishing the DeVito hanger that the "parallel edges" language was added to the '587 Patent, and there can be little doubt that the inclusion of this term was what finally defined over the DeVito hanger. Indeed, the examiner described the conversation that resulted in the agreement to add the "parallel edges" requirement by stating that "an agreement was reached which would [allow the claims] . . . [t]he agreed upon amendment defines over the prior art of record." Zuckerman Aff., Ex. B, p. s000346 (emphasis added). Because the only change made pursuant to the "agreement" to which the examiner refers was the addition of the "parallel edges" language, and because the DeVito hanger was the prior art of concern to the examiner, the only reasonable conclusion from this statement is that the "parallel edges" language is what finally, in the examiner's determination, defined over that prior art. The issue then becomes, how did including the term "parallel edges" distinguish the '587 Patent from the DeVito patent? The answer to that question reveals what was meant by the disputed term.
Relying on a declaration of one of its attorneys, Spotless attempts to refute this point by recounting a conversation the attorney had with the examiner. Because this can only be considered extrinsic evidence, it will be addressed in that section.
Depending on how the examiner understood the word "parallel," two possible answers emerge as to how he might have been attempting to distinguish DeVito. Specifically, the reason behind the examiner's request to include the "parallel edges" language depends on whether he understood the word "parallel" to encompass curved lines, or whether he understood it implicitly to mean "parallel and straight." There is nothing in the intrinsic evidence to indicate clearly which of these two possible interpretations the examiner was following, and, accordingly, each is examined below.
If the examiner allowed for the possibility that parallel lines can be curved, as both parties seem to agree was the case, the examiner s use of the "parallel edges" requirement could have distinguished DeVito in that the top and bottom edges of the arms of the DeVito hanger are not entirely parallel. Silfin Decl., Ex. F. Although the center section of each arm on the DeVito hanger is curved, the edges appear to be parallel in that section insofar as they are equidistantly spaced throughout this portion of their curvature. Id. However, in the portion of the arms around the clips and in the portion near the hook of the DeVito hanger, the top and bottom webs have different curvatures and are not parallel. Id. Because of this, entirely parallel top and bottom edges on the Spotless Hanger would distinguish it from the DeVito hanger.
On the other hand, under Spotless's interpretation, i.e., that the front and back edges of each web must be parallel, and again assuming that parallel edges can be curved, the term would have done nothing to distinguish the DeVito hanger. The DeVito patent includes top and bottom views, and these drawings show parallel front and back edges. See Silfin Decl., Ex. F; see also Vitronics, 90 F.3d at 1583 (in examining the prosecution history to interpret a claim, it is proper for the court to examine prior art cited therein). Accordingly, the examiner could not possibly have understood the term "edges" to refer to the front and back edges of the webs. The only reasonable explanation, therefore, is that the "parallel edges" referred to in the claims were the top and bottom edges, thus helping to distinguish the Spotless Hanger from DeVito, which does not have parallel top and bottom edges around the clips or near the hook.
That being said, what is problematic is how these seemingly insignificant differences would have represented any kind of improvement over the prior art; they appear to serve no practical purpose whatsoever. This question does not find a ready answer in the intrinsic evidence if one follows the assumption above that the examiner understood the definition of parallel to include curved lines. The answer to the question is considerably clearer, however, if the examiner understood the word "parallel" implicitly to mean "parallel and straight."
If the examiner understood the term in this way, his employment of the term "parallel edges" does help to distinguish meaningfully DeVito. Such a requirement would relate directly to the improvement Spotless claimed over DeVito. Specifically, Spotless had argued to the examiner that the Spotless Hanger improved the art described in DeVito because the inclusion of flat and straight inner arm portions on the hanger allowed a promotional card to rest on the hanger, whereas the DeVito hanger provided no such flat surface. Zuckerman Decl., Ex. B, p. s000332-33. This being the case, what the examiner was probably attempting to accomplish by including the "parallel webs" or "parallel edges" language in the claims was to describe arm segments that were straight, thereby creating this flat surface. Construed in this manner, the amendment adding the "parallel edges" language would relate directly to the improvement that Spotless claimed over the art described in DeVito.
One could argue that it is rather odd that the examiner would choose to use the word "parallel," as opposed to simply using "straight" or "level," to achieve this goal. No explanation seems to explain fully this word choice, but it makes some sense if the examiner was attempting to describe the continuity of the two inner arms, emphasizing that one top edge and one bottom edge run across both inner arm portions. Thinking of the top and bottom edges of the inner arm portions as parallel carries with it an image of a straight and continuous arm, well-suited for supporting a folded promotional card. This is likely what the examiner was trying to communicate in choosing the word "parallel" instead of "straight." Simply describing the edges as "straight" may have resulted a description that included inner arm portions that are each straight but angle up from the hook portion of the hanger. The term parallel, at least on some level, is more indicative of continuous top and bottom edges. In any event, regardless of the examiner's reasons for choosing the word "parallel" as opposed to straight, he was clearly referring to the top and bottom edges for the reasons explained in the text.
If the examiner was following this line of thought, i.e., that he understood parallel implicitly to mean "straight and parallel," AE's interpretation, albeit not quite as strong as under the earlier assumption, is, nonetheless, still far more likely than Spotless's interpretation. To be sure, if we assume that curved lines cannot be parallel, either party's interpretation of the term "edges" would result in straight, flat arms because a requirement that any of the edges of the arm segments be straight would necessarily result in straight arms. Even given this fact, however, it is unlikely that the examiner would have chosen to use the front and back edges to describe the arms. Doing so would have been an extremely roundabout method of accomplishing the goal of describing straight arms in light of the fact that the only edge that was of any moment in distinguishing DeVito was the top edge. Because the promotional card rests on the top edge of the arm portions, it was that edge which the examiner wanted to be straight. Accordingly, it is only natural that the examiner would have been referring directly to that edge in adding the "parallel edges" language. Therefore, the only reasonable conclusion is that the "parallel edges" referred to in the claims are the top and bottom edges of the arm segments.
It is worth noting that, if one assumes that the examiner interpreted parallel to mean straight and parallel, Spotless's infringement claim would ultimately fail under its own interpretation of the word "edges." As explained above, Spotless contends that the claims require that the front and back bottom edges be parallel. But the bottom web on AE's 735 Model is curved, and, as a result, its front and back edges cannot possibly be parallel. Consequently, the 735 Model would not infringe under Spotless's own interpretation of the claims because it would not have "parallel edges" on the bottom web.
Unfortunately for Spotless, the examiner chose, and Spotless agreed to, language that poorly described what the examiner clearly believed was the improvement over DeVito. The language chosen resulted in an unnecessary limitation in the claims; instead of merely stating that the top edge must be straight, the examiner related this requirement to the bottom edge by requiring that the top and bottom edges be parallel. In reality, of course, the shape of the bottom edge is completely unrelated to the placement of the promotional card. Nevertheless, the language chosen requires that the bottom edge be parallel with the straight top edge.
In sum, regardless of whether the examiner was considering the possibility that curved lines can be parallel, he was clearly referring to the top and bottom edges when he suggested the language "parallel edges." His initial suggestion that the phrase "parallel webs" be used indicated that his concern related to the top and bottom edges of the arms, and there is no indication that this concern changed. Moreover, parallel front and back edges would have done little or nothing to distinguish DeVito.
(iv) Spotless's rejection of "parallel webs"
The discussion thus far explains why the examiner intended the "parallel edges" language in the claims to refer to the top and bottom edges. It does leave open, however, the question of why Spotless would agree to "edges" but reject "webs" if the two words were being used to achieve the same result. Because the intrinsic evidence discussed to this point discloses what was meant by the term "parallel edges," it is sufficient to resolve this motion, and, consequently, we need not finally resolve what Spotless's motivation was in rejecting the initial suggestion of "parallel webs." Nevertheless, Spotless focused intently on this point at oral argument, so it will be addressed to the extent possible here.
There is little in the prosecution history to answer the question of why Spotless rejected "parallel webs" but agreed to "parallel edges." The only pieces of information available are the two faxed proposals, neither of which explains why the language was proposed or changed. The evidence discussed thus far answers why "parallel webs" was proposed, but the issue of why it was changed to "parallel edges" remains. Although Spotless argued that the parallel edges requirement was preferable because it would result in a flat hanger, allowing the hanger to be used in conjunction with certain automated machinery, Pl. Reply Mem., p. 7., it points to no intrinsic evidence to support this theory beyond a brief reference to the patent for the machinery in the '873 Reissue. The '873 Reissue does not include any reference, however, to why there was a need for parallel front and back edges in conjunction with this machinery. Pl. Reply Mem., Ex. A, Col. 10, ln. 42-48. To the contrary, the machinery is mentioned only in the context of describing how a plastic cap indicating the size of the merchandise displayed on the hanger can be attached to the hook. Id. at ln. 18-48. Thus, there is no intrinsic evidence from which one could discover any connection between the automatic machinery and the need for parallel edges.
In any case, as discussed infra at (2), even after considering Spotless's extrinsic evidence on this point, their position remains untenable.
The patent itself, on the other hand, does provide a possible reason for Spotless's decision. Two of the embodiments suggested in the cross-sectional drawings of the '873 Reissue have irregularly shaped webs, and one does not have distinct webs at all. See id. Figs. 7, 9, 10. In such situations, a requirement that the top and bottom webs be parallel would be somewhat difficult to interpret and apply. On the other hand, using "parallel edges" provided a flexibility that allows for these embodiments. This is particularly true for the embodiment which appears to be one solid piece with no distinguishable webs. Id. at Fig. 10. A parallel edges requirement clearly protects such constructions, and this would have been preferable to Spotless.
In sum, the intrinsic evidence supports the interpretation that the word "edges" in the claims refers to the top and bottom edges of the arms. The fact that the claims describe only two edges clearly contradicts Spotless's position that the four edges referred to in the specification and drawings of the '873 Reissue are the same edges as those described in the claims. Moreover, the use of the term "edge" in another portion of the specification to refer to the top and bottom of the arms also supports the interpretation that these surfaces can be referred to as "edges" even though they may be planar surfaces. Finally, the fact that the examiner initially suggested the language "parallel webs" and his own statements in the '873 Reissue show that he was referring to the top and bottom edges, especially in light of the process of distinguishing the DeVito hanger. Accordingly, the intrinsic evidence leads to the conclusion that the claims require that the top and bottom edges of the arms be parallel. As a result, AE is entitled to summary judgment with respect to the interpretation of the term "parallel edges."
(2) Extrinsic Evidence
Because the interpretation of the disputed term can be resolved based only on the intrinsic evidence, there is no need to, and it would in fact be improper to, turn to extrinsic evidence to achieve a different result. Bell Howell, 132 F.3d at 705-06. Nonetheless, even if Spotless could create some ambiguity as to which "edges" are referred to in the claims, AE would still prevail based on the extrinsic evidence.
The only extrinsic evidence offered by Spotless is a declaration from one of its attorneys recounting the circumstances surrounding the addition of the "parallel edges" language. The declaration is offered for two purposes. First, Spotless argues these events demonstrate that the "parallel edges" requirement referred to the front and back edges of the webs. Second, using this declaration, Spotless attempts to counter AE's argument that the "parallel edges" language was inserted as a means to distinguish the DeVito patent. In other words, Spotless argues that, because the requirement of "parallel edges" had nothing to do with defining over the DeVito patent, AE's arguments based on the premise that the "parallel edges" language was related to defining over that patent are without merit.
In her declaration, the attorney recounts two conversations she had with the examiner concerning the addition of the "parallel edges" language. The attorney states that she spoke with the examiner just prior to the addition of the "parallel edges" language, and the examiner informed her that he had determined that the segmentation of the arms into horizontal inner arms and upswept outer arms had resolved "all prior art issues" but that he wanted "additional structure to the claims before the case would be allowed." Wilczynski Decl., ¶ 11. At the attorney's request, the examiner faxed her a proposal that included this additional "structure." This fax contained the proposed addition of the "parallel webs" language. Id., ¶ 11.
After receiving the fax, Spotless's attorney called the examiner to reject his proposal. She told him that the "parallel webs" language was "unacceptable because the amendment would unduly limit the claims." Id., ¶ 13. Consequently, after further consideration, the examiner faxed a second proposed amendment, this time using the "parallel edges" language. Id., ¶ 16. Spotless agreed to that amendment.
The declaration is lacking in any information to support Spotless's first point — that the "edges" referred to are the front and back of edges of each web. In fact, the amount of information that one would expect to find in the declaration, but is missing, is remarkable. Nowhere does the declaration state that the examiner ever explained what "edges" his proposal was referring to; nor did anyone representing Spotless ever make this inquiry. There is absolutely nothing in the attorney's declaration to suggest that either party understood the language to refer to the front and back edges, as Spotless now claims. Further, and perhaps most significantly, the declaration contains no explanation whatsoever, or even reference to, the reason for the inclusion of the language in the first place. The declaration merely credits the addition of the language to the need for more "structure," without any further explanation as to what was meant by this term, or why or what "structure" was needed. Finally, the attorney does not explain what, exactly, was too "limiting" about the word "webs." Such information would greatly help to explain the reasons for the change in the language from "webs" to "edges."
Spotless's current position would be far more credible if any of this information were included in the declaration. It is not included, and its absence is striking in light of the fact that, if the discussion on these issues were favorable to Spotless, it seems highly unlikely that it would have been omitted. By failing to explain any of the reasoning for the alteration in language, Spotless asks that I infer that the examiner had completely changed his requirement from a hanger with parallel webs to a hanger that had parallel front and back edges, and that Spotless understood that this is what was meant by the term "parallel edges," without either party ever discussing the reasons for, or the nature of, this altered requirement. The only evidence Spotless supplies in support of its theory are the two faxed proposals, neither of which contain any discussion as to the reason for the change in language, and neither of which specifies which edges are to be "parallel." Such an inference would be completely inconsistent with the intrinsic evidence, and I decline to draw it. Accordingly, the declaration does not shed any light on the question of which edges are referred to in the patent claims.
Nor does the declaration support Spotless's second contention, i.e., that the addition of the "parallel edges" language was not related to defining over DeVito. Spotless argues that the "parallel edges" language was not necessary to define over DeVito because the examiner stated that "all prior art issues had been resolved," Wilczynski Decl., ¶ 11, prior to the addition of that term. Specifically, the attorney states:
On September 21, 1995 I received a telephone call from Examiner Mohanty who indicated that all prior art issues had been resolved but that additional structure had to be inserted into the broad claims before the case would be allowed. He indicated that he had a proposal that would place the claims in allowable condition. I asked him to fax me his proposal. . . .Id. It was the faxed proposal referred to in this paragraph that first suggested the "parallel webs" language which later became "parallel edges." No further description of this conversation is included in the declaration.
In reading the declaration carefully, it is clear that the conversations could just as easily be understood to mean that the examiner found that all prior art issues had been resolved so long as it was clear from the claim language that the top and bottom webs must be parallel and straight. In other words, the hanger, as revealed by the drawings, already showed parallel top and bottom edges on the hanger arms, and perhaps the examiner simply wanted to include this detail in the claims to give them more "structure." As described above, such additional structure would be directly related to allowing the promotional aid to be placed on the hanger, which was the precise improvement which Spotless claimed in defining over DeVito. This interpretation is, at a minimum, equally consistent with the intrinsic and extrinsic evidence discussed thus far as the interpretation Spotless now suggests.
In fact, Spotless does not, and cannot, explain why the examiner would want to add the "parallel edges" language if not to define over prior art. This shortcoming is especially hard for Spotless to explain in light of the fact that much of the prior art in question — and the DeVito hanger in particular — had parallel front and back edges. See Zuckerman Decl., Ex. C, Figs. 2, 4 (DeVito Patent, top and bottom views). What could the examiner possibly have been hoping to accomplish by adding the "additional structure" of the "parallel edges" language if he was referring to the front and back edges and this "structure" did nothing at all to distinguish prior art?
The only explanation offered by Spotless for the inclusion of the additional "parallel edges" language is that parallel front and back edges are necessary for the hangers to be used with "automatic machinery that automatically attaches the top sizer to the hanger." Pl. Reply Mem., p. 7. As further explained at oral argument, because the machinery requires that the hanger be flat as it lays on its front or back, parallel front and back edges are necessary. Spotless provides no evidentiary support for this proposition either in the specification of the hanger or by way of affidavit or declaration. Notably, the attorney's declaration makes no claim that this was ever a subject of her conversation with the examiner. The argument is merely an unsupported claim by an attorney in a brief, and, thus, carries no weight.
In any event, the explanation suffers from three faults. First, it does not answer the key question above — how was this requirement responsive to the examiner's request for parallel webs? There is no reason to believe that the examiner was aware of the "automatic machinery" issue, and Spotless provides no documentation or other evidence that the examiner was so informed beyond pointing out that there is a reference to the patent for the automated machinery in the '873 Reissue. The need for parallel edges in conjunction with this machinery, however, is mentioned nowhere in the original '587 Patent or the '873 Reissue. As noted, the attorney who spoke with the examiner makes absolutely no mention of this issue in her declaration. Moreover, it was the examiner — not Spotless — who brought up the necessity of including the "parallel" language; if this was necessary only for industry reasons known to Spotless, why would the examiner have been the one to suggest it?
Second, and perhaps most importantly, the intrinsic evidence contradicts Spotless's position that the parallel edges were not necessary to overcome prior art. The examiner himself describes the conversations he had with Spotless's attorney regarding the "parallel edges" amendment by stating that "an agreement was reached which would [allow the claims] . . . [t]he agreed upon amendment defines over the prior art of record." Zuckerman Aff., Ex. B, p. s000346 (emphasis added). This statement was made after the examiner had already reviewed the application including the segmentation of the arms, which is the feature which Spotless claims finally defined over DeVito. Spotless does not explain how this statement, which referred to the agreement reached to add the "parallel edges" term, can be reconciled with its position that the parallel edges language was not necessary to distinguish DeVito.
Third, the explanation would not even result in a flat hanger. The description relates only to the arm portions, there is no requirement that any portion of the hook or clips be flat. If the entire hanger must be flat, as suggested by Spotless, one would expect the parallel edges language to be included in the descriptions of the hook and clips. Thus, the most that could possibly be accomplished by the requirement is that the body portion of the hanger is flat, and, in truth, it doesn't even accomplish that; even if the front and back top edges are parallel and the front and back bottom edges are parallel, it does not follow that the hanger body will be flat. Indeed, the top web could be narrower than the bottom web, as illustrated in Figure 4, or could be askew, as illustrated in Figure 5, and still have parallel edges.
In such situations, the hanger body would not be flat, but would still have parallel edges as described by Spotless. Accordingly, Spotless's explanation for the additional "parallel edges" language would not accomplish what they claim it was meant to accomplish.
To be sure, the specification language does say that in a certain "preferred" embodiment, the top and bottom webs should extend equally from the wall, see '873 Reissue, Col. 6 ln. 53-55. In this "preferred" embodiment parallel front and back edges would lead to a flat hanger body, but the claims of the patent are not limited to that one embodiment, and nothing in the claims indicates such a requirement.
Thus, Spotless has not resolved the ambiguity, assuming such ambiguity exists, as to which edges are referred to in the claims. The attorney's declaration does not resolve the ultimate issue of which "edges" are referred to in the claims; nor does it support Spotless's conclusion that the parallel edges were not necessary to define over DeVito.
In contrast, the remaining extrinsic evidence offered by AE definitively resolves the ultimate issue of which edges are referred to in the claims. AE offers the depositions of two former Spotless employees: Roland Harmer, who was one of the inventors of the Spotless Hanger, and John Havrilla. At his deposition, Harmer testified that the word "edges" in the claims referred to the top and bottom edges of the arms. Silfin Decl., Ex. I. Similarly, Havrilla testified that the term "edges" referred to the top and bottom edges, and further went on to clarify that he has never heard of the front and back of the webs being referred to as the "edges." Id., Ex. H. While this testimony may be subject to some attack, coming, as it does, from former Spotless employees who may or may not have ill will toward their former employer, it is, nonetheless, unrebutted and unattacked. In light of the fact that Spotless has presented no convincing evidence to the contrary, this testimony is sufficient to demonstrate that the term "edges" should not be read to refer to the front and back edges of the webs.
To conclude, the extrinsic evidence in the case does not support Spotless's position. Accordingly, even assuming there is some ambiguity as to the definition of the term "edges," requiring the consideration of extrinsic evidence, AE's interpretation of the claim as referring to the top and bottom edges of the arms is correct.
(3) Infringement
a. Literal Infringement
Where the accused product contains every element set forth in a claim of a patent, the accused product is said to "literally infringe" the patent. See Hi-life Prods, Inc. v. American Nat'l Water-Mattress Corp., 842 F.2d 323 (Fed. Cir. 1988). The parties have agreed that if AE's interpretation is correct, which it is, there is no literal infringement. Consequently, for the reasons stated above, there is no literal infringement of the '873 Reissue because the 735 Model does not have parallel top and bottom edges, as required in the disputed claims.
b. Doctrine of Equivalents
Although plaintiffs represented at oral argument that they are not claiming that there is infringement based on the doctrine of equivalents, the issue was addressed both in AE's motion for summary judgment and in Spotless's reply, and so will be discussed here.
Even where there is no literal infringement, an accused product may be found to infringe under the "doctrine of equivalents." This doctrine stands for the proposition that an accused product infringes a patent "if it performs substantially the same function in substantially the same way to obtain the same result." Hughes Aircraft Co. v. United States, 717 F.2d 1351, 1361 (Fed. Cir. 1983) (quoting Graver Tank Mfg. Co. v. Linde Air Prods. Co., 339 U.S. 605, 608, 70 S.Ct. 854, 856 (1950)). The relevant inquiry focuses on each element of the claim, not on the invention as a whole; if the accused product has elements that perform the same function in the same way as each element of the disputed claim, infringement may be found. Warner-Jenkinson Co., Inc. v. Hilton Davis Chemical Co., 520 U.S. 17, 29, 117 S.Ct. 1040, 1049 (1997).
This doctrine is limited, however, by what is known as "prosecution history estoppel" or "file wrapper estoppel." The doctrine of file wrapper estoppel works to prevent a party from claiming equivalency with respect to an element that was amended to allow the claim. Jonsson v. Stanley Works, 903 F.2d 812, 817 (Fed. Cir. 1990). If the party claiming infringement made an amendment to the element in question in order to obtain claim allowance, that party has the burden of proving that file wrapper estoppel should not apply. Warner-Jenkinson, 520 U.S. at 31, 117 S.Ct. at 1051.
The parties make the same arguments on this issue as they do in arguing for their interpretation of the term "parallel edges." Specifically, Spotless claims that file wrapper estoppel should not apply because it was the addition of segmented arms, not the addition of parallel edges, that allowed it to obtain the '587 Patent over DeVito's prior art. Conversely, AE argues that, because it was the addition of "parallel edges" that allowed the '587 Patent to be issued, file wrapper estoppel dictates that a hanger without parallel edges cannot be found to infringe on the '587 Patent based on the doctrine of equivalents.
As explained above, Spotless's position on this point is unconvincing. Spotless does not explain, and provides no evidence showing, why the examiner would have wanted parallel edges if not for the purpose of defining over the DeVito Patent. The examiner himself described the final agreement to amend the claims to include the "parallel edges" language by stating that the amendment "defines over the prior art of record." Zuckerman Aff., Ex. B, p. s000346. This statement, standing alone, is dispositive on the issue. Moreover, Spotless's only proffered explanation for the addition of this language — that the parallel edges were necessary for use with the automated machinery — would not result in the flat hanger Spotless claims it was intended to create and is irreconcilable with the fact that it was the examiner, not Spotless, who suggested the language. Consequently, there is nothing to support Spotless's conclusion that the addition of the "parallel edges" language was unrelated to defining over prior art. Accordingly, file wrapper estoppel prevents the application of the doctrine of equivalents in the present case.
In any case, even if it is assumed that the "parallel edges" language was not necessary to define over prior art, Spotless's argument would, nonetheless, fail. File wrapper estoppel is not limited to an amendment necessary to overcome prior art. To the contrary, the doctrine applies to all amendments that are "related to the statutory requirements for a patent." Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., Ltd., 234 F.3d 558, 566 (Fed. Cir. 2000). It is undisputed that the examiner refused to allow the '587 Patent without the "parallel edges" language, and it follows that this additional language was related to the statutory requirements for the patent. Consequently, file wrapper estoppel would apply to that element regardless of whether it was added to define over the DeVito patent.
Moreover, if we accept Spotless's claim that the examiner did not require the parallel edges language in order to define over prior art, we are left with an amendment for which the prosecution history provides no explanation. A court cannot consider a declaration submitted by one of the parties in determining the reason for an amendment; "only . . . the patent's prosecution history can be a basis for such a reason." See Pioneer Magnetics, Inc. v. Micro Linear Corp., 238 F.3d 1341, 1345 (Fed. Cir. 2001) (internal quotation omitted). Where no reason is given for an amendment, there is a presumption that file wrapper estoppel will apply.Id. For the reasons stated above, Spotless cannot overcome this presumption because it has failed to prove that the amendment concerning the parallel edges language was not related to the statutory requirements of the patent.
On the whole, the evidence firmly supports the proposition that the addition of the "parallel edges" language was necessary in order to obtain claim allowance of the '587 Patent — whether it was to define over DeVito or for some other reason. It may be true that less restrictive language could have been used to create a flat, horizontal top edge — thereby designing over the DeVito patent — which did not also require a parallel, straight bottom edge, but less restrictive language was not used. Spotless agreed to include the suggested "parallel edges" language, and the 735 Model does not literally infringe because it does not have parallel edges. Moreover, because file wrapper estoppel operates to prevent the application of the doctrine of equivalents in the present case, the 735 Model cannot be found to infringe under that doctrine either. As a result, there is no infringement in the present case.
Conclusion
This is a classic case where, if a valid patent is to be obtained on an improvement to an everyday product, the limited improvement must be narrowly and carefully defined. The inevitable consequence is that the patent is often easy to design around, especially where, as here, the language chosen may not have precisely defined the improvement. The pressure to satisfy the Patent Office is almost inevitably followed by an attempt on the part of the patent holder to stretch the claim to cover what it views — perhaps rightly — as an unfair taking advantage of its improvement. But, nonetheless, the improvement is defined by the claim, and a post hoc stretch cannot support an infringement claim.
Because the 735 Model does not infringe the '873 Reissue, AE's motion for summary judgment is granted, Spotless's motion is denied, and the infringement claim based on AE's 735 model is dismissed.
SO ORDERED: