Phillips v. AWH Corp.: dictionaries in litigation

The en banc review in Phillips v. AWH Corp., 376 F.3d 1382 (CAFC 2004), promises to give guidance on the mechanism for interpreting terms within a patent claim. One commentator stated that many patent law people expect that the CAFC will select dictionaries as the primary source for resolving claim construction disputes. The intrinsic evidence, which is the patent document and the associated file history, would control when no relevant dictionary definition
exists and when the intrinsic evidence clearly defines the disputed language. A different commentator stated: “...it is not clear that the Federal Circuit, given the recent trend in its decisions to rely heavily on dictionaries, will use its en banc reconsideration of Phillips to diminish their use significantly. Indeed, none of the judges appears to be a vociferous opponent of dictionaries playing a major role. The judicial differences focus more on the relationship
between dictionary definitions and intrinsic evidence as sources for claim construction.”

The commentators are probably correct in predicting the survival of general purpose, conventional dictionaries in the interpretation of terms in patent claims. One does wonder if this is the right result or merely an artifice to streamline litigation. Is the court correct in determining that dictionaries give the ordinary meaning that would be attributed to claim terms by persons
skilled in the relevant art, or are dictionaries perhaps one of the last sources that a skilled artisan
would use to determine the ordinary meaning of claim terms? Perhaps, the court has improperly
translated the use of dictionaries as extrinsic evidence that may be used to aid "the court's
understanding of the patent" (or the understanding of non-technically trained litigators or legal
academics) to an imputed acceptance by persons of ordinary skill of the technical accuracy of
descriptions in conventional dictionaries. The language of patent documents is directed to one of
ordinary skill in the art, and that is the starting point for any inquiry. It is unlikely that definitions
in conventional dictionaries are explicitly contemplated in most patent applications or
prosecutions. Although courts, litigators, and academics might use dictionaries to get up to
speed for discussion, one should not substitute that exercise for the understanding of the term by
a person already trained in the art.

Contemplate some of the cases.

TEXAS DIGITAL

The Texas Digital case, 308 F.3d 1193 (CAFC 2002), elevated the status of dictionaries to
construe claim terms in suggesting that consulting intrinsic evidence prior to consulting a
dictionary "invites a violation of our precedent counseling against importing limitations into the
claims." In the Phillips case, the patentee argued unsuccessfully for a dictionary definition of the
word --baffle-- to give broader scope than that in the illustrated embodiments. Texas Digital
limited situations in which the intrinsic record would control, as, for example, if the intrinsic
record contains clear lexicography or disavowal of claim scope. Texas Digital, 308 F.3d at 1204;
additionally, the intrinsic record could be used to select among diverse dictionary definitions
(“the intrinsic record must always be consulted to identify which of the different possible
dictionary meanings of the claim terms in issue is most consistent with the use of the words by
the inventor,” Id. at 1203)

NOVARTIS

In Novartis v. Eon, 363 F.3d 1306 (CAFC 2004), the term –hydrosol– was not defined in the
specification. The patentee argued unsuccessfully for a dictionary definition of hydrosol: solid
particles dispersed in an aqueous medium. The patentee also invoked Liebel-Flarscheim v.
MedRad, 358 F.3d 898 (CAFC 2004) (holding that claim terms are given the full breadth of their
ordinary meaning unless a clear disavowal of scope is stated in the specification.) Nevertheless,
the majority limited the term --hydrosol-- to a medicinal preparation consisting of a dispersion of
solid particles in a liquid colloidal solution prepared outside the body. Part of the reasoning
underlying this definition involved importing a definition of the term –solution– into the term
hydrosol, even though a true aqueous solution (wherein the solute is dissolved in the solvent
water) is basically the antithesis of a colloidal dispersion (wherein the added material is not
dissolved in the surrounding water). The dissent noted that the majority’s interpretation required
the use of one particular dictionary, and did not flow from other dictionaries.

ISSUES

A definition in a conventional dictionary, which is more directed to brevity than to detail, is apt
to give broad scope to a claim term, and thus generally will be advocated by the patentee, as was
the case in Novartis and Phillips. The accused infringer will seek to narrow claim scope, by
limiting a term to the embodiments disclosed in the patent specification. In terms of the present
formalism, where there is no clear-cut surrender of claim scope (the case in Novartis), the court
will select a limiting definition from several available to conform to (perceived) limitations in the
specification. The arbitrariness of this is seen in Novartis.

The underlying tension arises from the disparity in the scope of embodiments in comparison to
the scope of the claims. A battleground can be the claim terms, as with Phillips, or it can be
written description or enablement, as with the Rochester COX-2 case.

The use of conventional dictionaries for claim terms is an expedient. In this approach, rather
than getting bogged down with testimony of conflicting experts or analyses of technical
documents beyond the scope of the training of the judges or of the litigators, the court relies on
the conventional dictionary to set an initial baseline against which the specification can be
measured. After that, litigation skill may matter. In the real technical world, one of ordinary skill
would probably define a term contextually in terms of a patent document, refer to technical
textbooks, treatises, and publications or ask another technical person long before the skilled
artisan would look to, or rely upon, Webster’s.

To reduce difficulties with claim terms, the patent applicant should present definitions and
examples in a non-limiting way. Rather than rely on dictionary roulette, the applicant should
make sure that a suitable baseline definition of key claim terms is within the specification. The
breadth of embodiments should be stressed, up to the point of intrusion on prior art or 112 issues.
In conjunction with Festo and Johnson & Johnston, if in prosecution an examiner presents a
more limiting definition than contemplated, appropriate steps should be taken.

[Submitted to Int. Prop. Today, Dec. 7, 2004]

****
The discussion about dictionaries, and separately the discussion about patent reform (by the FTC and the NAS), reminds one more of an academic debate than one about policies affecting business. I don't think that practitioners are much enamored of the use of general purpose dictionaries, anymore than the PTO is. The NAS report flitted from law review article to another without getting much input from anyone who actually does prosecution.

A column by Ellen Goodman in December 2004 started --I like the old maxim that academic politics are vicious because the stakes are so small.-- Oddly, the stakes aren't necessarily small for patents, but it does not look like we're getting into the real issues in the debate so far.