By Kevin E. Noonan —
In Anna Karenina, Leo Tolstoy
wrote, "All happy families resemble one another, each unhappy family is
unhappy in its own way." This
sentiment could also be applied to the briefs submitted to the Federal Circuit
for its en banc consideration of the
written description requirement. A
large proportion of those briefs support the current construction of 35 U.S.C. § 112, first paragraph, to include a separate written description requirement,
but like unhappy families each amicus takes its position in a slightly
different way. The Federal Circuit
Bar Association's brief is one example: their position is that there is
a separate written description requirement, but that there is much less of a
difference between those who agree and those who believe the only requirement
is enablement, including even Ariad who (understandably) has made the strongest
argument that the statute should not be construed to contain a separate written
description.
The brief begins with the argument that there is "substantial common
ground" among all the parties involved, including Ariad, Lilly, and the
various amici. The basis for this assertion is that
all parties agree (according to this amicus) that the "invention"
must be disclosed in the specification. "Ultimately," they write, "the real issue on written
description is not whether the law
requires a description of the invention in a patent application, but whether
that condition is its own statutory requirement or merely an 'aspect' of
enablement without independent force."
"Law and logic" converge in the conclusion that written
description is a separate requirement, they argue. In reaching this conclusion, they don't rely merely on the
language of 35 U.S.C. § 112, first paragraph, but cite the provisions
of the second paragraph, that the claims be directed to subject matter that an
inventor "regards as his invention." That statutory requirement requires that patents be granted
only on what an inventor has invented, but there are instances where enablement
is "insufficient to protect against overclaiming," according to this
brief.
The "central truth" providing "fundamental common ground"
for the parties is the "simple common sense" that "a patent must identify what the inventor believes
to be his or her invention," say the amicus (emphasis in original). This requirement is based on the quid
pro quo of the patent system. The brief cites Ariad's principal brief for "the undisputed
proposition that the claims must be
directed to an invention that is identified in the specification"
(emphasis in original). Indeed, amicus make a strong argument
based on this section of the brief that the dispute is one more semantical than
real. Amicus' brief further cites Ariad's
principal brief that "[t]he specification must state what the invention is, for otherwise it fails to inform a person of
ordinary skill in the art what to make and use" (emphasis in original). (This "admission" is so significant that amicus feels
compelled to state that it is "not a slip of the pen or poor proofreading.") Further (and from the first page of
Ariad's brief), "[p]roperly interpreted, the statute requires the
specification to describe (i) what the
invention is, and (ii) how to make and use it" (emphasis in original).
Not content to rely on the "common sense" of its position, the
brief cites precedent from the CCPA (In
re Moore, In re Sus, Jepson v. Coleman) for the requirement
that the specification disclose what the inventor regarded as her
invention. Specifically discussing
In re Ruschig, the amicus states that
"all parties agree that the specification must disclose what the inventor
actually invented."
Turning to the specific questions asked by the Court, the amicus asserts
that the "common ground" discussed above provides the "starting
point" for resolving the debate. However, they argue that there is a "much narrower" question
before the Court, specifically whether it is "possible to satisfy
the enablement requirement without also satisfying" the written
description requirement. In most
cases, the brief argues, these requirements are "so closely related"
that satisfaction of one provides satisfaction of the other, citing LizardTech Inc. v. Earth Resource Mapping,
Inc. and University of Rochester v.
G.D. Searle & Co. Inc. (tellingly, cases involving electromechanical
and biotech/pharma claims, respectively). The problem with Ariad's argument, according to this amicus, is that "it
fails to provide any statutory safeguard to keep an inventor to what he or she
has actually invented in the limited class of cases where enablement is
insufficient to prevent abuse." This is because the two requirements — written description and enablement — serve independent
purposes. These purposes are
related to the fact that "the heart of innovation" is not always its implementation, i.e., there are instances where the inventor's "true
contribution" is in implementing an invention, and other cases where an
inventor recognizes that an "already-enabled system or method can be
applied to solve a problem or otherwise advance a field of art." In the latter instances, it is the
identification of the solution that is the invention, and in these instances,
this amicus argues, the enablement requirement "can be inadequate to
ensure that the inventor's patent monopoly is limited to his or her
contribution." (This brief,
as have other briefs, uses cases like Rochester
and Amgen v. Hoescht Marion Roussel
to support this point, some of the very cases Ariad and those challenging the
separate written description requirement contend is a misapplication of the
law.)
The brief sets forth a hypothetical to illustrate the point: an invention
involving an "easy-to-make compound" to be used in a specific medical
or biological application. The
inventor did not contribute by identifying how to make the compound; rather,
the inventor's contribution is "to identify for the first time that the
compound will be effective for the desired application." Often, perhaps frequently, the inventor
will disclose this new application in the patent specification. But, the brief questions, "what
happens when the inventor's patent application . . . evolves over the course of
amendments to his or her claims?" It is possible, amicus argues, that the claims "may shift to identify
for the first time . . . compounds that are useful in solving a problem
underlying the invention, but that the inventor had not identified in his or
her original application." The brief also raises the specter of such a change in the "invention" — i.e., what is claimed — as not even involving the inventor but being accomplished
by a patent attorney (which while true, incongruously directs the source of the
problem of misapplication of the written description requirement towards a
substantial portion of the FCBA's membership). Under these circumstances, the brief argues, enablement "can
be wholly inadequate as a vehicle to prevent abuse . . . the newly-claimed
compounds are undeniably enabled, even if they were never identified as part of
the invention in the specification." The brief also cites for this principle situations of belated claiming
of a species falling within the scope of a disclosed genus but not expressly
described in the specification as filed. Other instances cited in the brief include those where "the
specification describes a feature of the invention that is indispensible to the
inventor's contribution but where the inventor — or his or her patent attorney — failed to include it in the claims as properly construed." (This was the
situation in the LizardTech
case.) In LizardTech, according to the amicus, without the written
description requirement "neither principles of enablement nor claim
construction would always keep the scope of the patent monopoly faithful to the
inventor's invention — the claims could be completely untethered to his or her
technical contribution." (As
an aside, the brief casts the written description requirement as being needed
to constrain the "patent monopoly" much more frequently than might
have been expected from a group that should understand the inaccuracy inherent
in the phrase and the extent to which the Supreme Court has used the concept of
"monopoly" to unduly restrict patent rights in the past.) This portion of the brief closes by
describing the written description in a patent specification to be "a
snapshot in time" of the invention that is "indispensible" to
identify the invention, and that serves to limit the scope of granted claims to
what the inventor invented and to which she is "fairly entitled."
Secondly, the brief argues that Ariad's position that the "requirement"
that the inventor disclose what she thinks is her invention devolves to the
enablement requirement is "not faithful to the text of Section 112." Calling the first paragraph of §
112, "far from a model of clarity," the brief argues that
interpretation of the statute to require a separate written description
requirement "flows not only from the statutory text and history" but
also "from the overall context of the patent statute . . . itself." This is the error in focusing (as Ariad
and some amici have) on how § 112, first paragraph, should be parsed as a
matter of English language sentence construction. Rather, they argue that the statute should be construed with
regard to context, specifically the interplay of the first paragraph of §
112 with the second paragraph. The
amicus construes the second paragraph as having two requirements: that the
claims "set forth what the applicant regards as his invention" and "that
is to do so with sufficient particularity and distinctiveness." While the second requirement is the
usual focus of analysis regarding fulfillment of the second paragraph of § 112, the amicus argues that the first requirement "provides
powerful proof that an inventor's claim scope is tied directly to his or her
invention as defined in his or her patent specification," a reading the
brief argues is supported by Allen Engineering
Inc. v. Bartell Industries Inc. (Fed. Cir. 2002). By this interpretive legerdemain, this amicus moves the statutory
situs of the written description requirement from the first paragraph (where it
is inconveniently linked to the enablement requirement) to the second paragraph
where it is free of the encumbrances of enablement. The brief also argues that the Allen Engineering precedent,
and the statutory context, confirm the "common sense principle" that
a patentee is entitled to claims that extend as far and no farther than what
the inventor sets forth as her invention in the specification.
The brief then delves briefly into the waters of English language
sentence analysis, on what it terms Ariad's "grammatical argument." The amicus notes that this
argument "largely revolves around the presence of a comma between the
phrases 'of the manner and process of making and using it' and 'in such . . .
terms'." The objection to finding a separate
written description requirement based on the position of this comma advanced by
Ariad and amici in agreement with Ariad was also voiced by Chief Judge Markey
in dissent to the Court's opinion in In
re Barker (CCPA 1977) and by Judge Linn dissenting from the Rochester opinion. But this amicus refutes the argument
with a simple "not so." For amicus, "[a]n at least equally-plausible reading of the [and
manner and process of making and using it] phrase is that the comma between the
phrases . . . represents a natural pause between the two clauses." The amicus tests this analysis by
assessing the meaning of the first paragraph of § 112 if its drafters had
removed the phrase "written
description of the invention" from the statute. The comma still has a function, as a pause according to the
amicus, and thus "the comma between the phrases . . . — relied upon so
heavily by opponents of a separate written description requirement — does not
undermine a separate written description requirement because proper English
would include it in the statute even if
the statute did not include language requiring a written description of the
invention." In short,
according to this amicus, Ariad's principal grammatical argument is a
nullity. Moreover, the brief
argues, if Congress wished to limit the scope of § 112, first paragraph, to require enablement alone, Congress could have achieved this end by "a
couple of minor revisions." Of course, Congress has not availed itself of this opportunity. And the brief further notes that the
phrase "of the invention" is set off from the phrase "the manner
and process of using is" by its own comma, and thus Congress could have
more clearly indicated an intention to equate a description of the invention
with enablement by deleting this
comma.
The brief ends its survey of what it calls the "comma conundrum"
by performing its own "grammatical deconstruction" of § 112,
first paragraph, saying "it can hardly be a coincidence that the
enablement requirement directly parallels the requirement contain the manner
and process of making and using the invention but not the written description of
the invention," saying that "[t]his parallelism of language reflects
a parallelism of thought," that "an enabling disclosure must allow
those of ordinary skill in the art to make and use the manner and process of
making and using the invention."
Finally, the brief surveys and rejects Ariad's contention that the
written description requirement is but a relic of a time before the statute
mandated peripheral claiming, saying "[t]his theory reads far too much
into the statutory history of Section 112."
The brief also addresses the second question posed by the Federal Circuit
for en banc consideration. According to amicus, the written
description should be applied on a case-by-case basis, the brief arguing
that there is "no one-size-fits-all approach" that can be properly
used. However, they do recommend
to the Court two principles important for guiding development of written
description jurisprudence. First,
claims should only be found valid to the extent that they are directed to "what
the inventor 'regards has his invention'." This is "[t]he key to a fair and balanced written
description requirement," amicus argue.
Second, fulfillment of the written description requirement should be
assessed as a matter of law based on underlying facts, rather than as a factual
matter. Their basis for this
contention is that "ultimately" patent validity is a question of law
(citing Graham v. John Deere) and the written description requirement is
something "fundamentally focused on the text of the patent's disclosure
itself," in analogy to claim construction under Markman v. Westview Instruments (seemingly disregarding the
difficulties and criticisms the Court's application of Markman decision to claim construction has engendered, by members
of the FCBA as well as several current Federal Circuit judges). Because satisfaction of the written
description requirement goes "hand-in-hand" with satisfaction of the
enablement requirement (except, as amicus argues in another portion of the
brief, when they don't), both should be assessed as a matter of law, based on
the "inventor's disclosure within the four-corners of the patent
specification." The argument
focuses on the "black-and-white" nature of these analyses, meaning
(it seems) that both are objective rather than subjective tests (as opposed,
for example, to the best mode requirement). However, the argument ignores the other distinction, made by
other amici, that enablement permits consideration of what was known by one
having ordinary skill in the art, while written description involves only what
the inventor has described as being her invention. The brief characterizes en
banc consideration as being a "golden opportunity" for the
Federal Circuit to address this "inconsistency."
The FCBA's brief thus invites the Federal Circuit to once again take an issue of
patent validity away from the jury, and potentially to add yet another aspect
of patent infringement litigation to its de
novo review. As the history of
Markman claim construction has amply showed, de novo review results in the opposite of what Congress intended
when instituting the Federal Circuit — less consistency and less predictability
in outcomes. Unless the Court
intends to revise its treatment of such de
novo review situations — such as reviewing the factual underpinnings of
legal conclusions for substantial evidence or clear error and basing its de novo review of the legal conclusions
on trial court-determined facts — following the FCBA's recommendations seem
curious coming from an amicus that would be expected to best understand the
burden de novo review has been to the
Federal Circuit, and the negative effects it has had on U.S. patent law and
litigation.
For additional information regarding this topic, please see:
• "Amicus Briefs for Ariad v. Lilly: Intellectual Property Owners Association," November 25, 2009
• "Amicus Briefs in Ariad v. Lilly: GlaxoSmithKline," November 24, 2009
• "Amicus Briefs in Ariad v. Lilly: United States," November 23, 2009
• "Amicus Briefs in Ariad v. Lilly: Google, Verzion Communications Inc. and Cisco Systems, Inc.," November 22, 2009
• "Amicus Briefs in Ariad v. Lilly: Professor Christopher Holman," November 19, 2009
• "Lilly Files Principal Brief for Ariad v. Lilly Rehearing En Banc," November 16, 2009
• "Next Up: Ariad v. Lilly Rehearing En Banc," November 10, 2009
• "Federal Circuit Grants En Banc Review in Ariad v. Lilly," August 21, 2009
• "Ariad Files Petition for Rhearing in Ariad v. Lilly," June 3, 2009
• "Ariad Decision Voids Attempt to Use Broad Claiming to Avoid the Written Description Requirement," April 14, 2009
• "Ariad Pharmaceuticals, Inc. v. Eli Lilly and Co. (Fed. Cir. 2009)," April 6, 2009
Patent Docs 