More on Phillips v. AWH and dictionaries in claim construction

from an endnote in my article in Intellectual Property Today for August 2005:

In the en banc decision in Phillips on July 12, 2005, Phillips got a remand on the infringement issue, and a dictionary-centric interpretation of Texas Digital was questioned. Recall that the dissent in the first Phillips case argued that the Webster’s definition of baffle applied (Phillips, 363 F.3d at 1218-1219). Thus, in the end, one obtained the result of the dissent (district court decision overturned; Mr. Phillips gets another chance) but for (some) different reasons (emphasis on intrinsic evidence preferred over extrinsic evidence such as dictionaries). However, one recalls that the majority in the earlier decision had based its conclusion on intrinsic evidence: that the specification disclosed baffles only at angles other than 90 degrees, and thus that this was a "written description" issue. Phillips, 363 F.3d at 1214. This wasn’t a "dictionary case;" it was an "interpretation of the specification" case. Separately, of the "Ulysses S. Grant" issue, the en banc decision of July 12 acknowledged that it did not reach the 7th question of 376 F.3d 1382: is it appropriate for this court to accord deference to any aspect of trial court claim construction rulings?

from Bloomberg News:

Dictionaries Are Set Aside in Patent Suits

Inventors, not dictionaries, get to define the terms in patents
involved in lawsuits, a federal appeals court said in a ruling that sets guidelines for judges on how to interpret patents.

Dictionaries, encyclopedias and other reference books can be used only sparingly in patent trials, the United States Court of Appeals for the Federal Circuit in Washington said yesterday.
A judge must first look at the patent and the history of the
application when seeking to clarify its claims, the court ruled.
"To construe patent terms, you have to look at the document," said
Edward D. Manzo, a Chicago patent lawyer who submitted a brief in the case. "I'm happy we don't have to have a battle of dictionaries anymore."

The circuit court, which handles almost all patent appeals, has issued contradictory rulings in the last decade, including a 2002 decision that told judges to look to dictionaries first.

Judges' patent rulings are overturned at least a third of the time in cases where their rulings hinge on interpretations of so-called "claim terms."

Circuit Judge William C. Bryson, in an opinion for 9 of the 12 judges on the court, wrote that "The main problem with elevating the dictionary to such prominence is that it focuses the inquiry on the abstract meaning of words rather than on the meaning of claim terms within the context of the patent."

The ruling touches on every patent lawsuit, said David W. Long, a
patent lawyer who submitted a brief on behalf of the wireless industry.

The dispute before the court was over the word "baffle," used to
describe part of a wall for prisons that can deflect sound, heat and bullets. The patent owner, Edward H. Phillips, had accused the AWH Corporation of infringement and said a judge improperly dismissed the case after refusing to use the dictionary definition. [!!]

The majority sent the case back to the trial judge for reconsideration. Circuit judges Alan D. Lourie and Pauline Newman dissented on grounds that the trial judge had relied upon the correct standard.

Another judge, Haldane R. Mayer, veered from the majority opinion,
saying it did not resolve the broader issue of how much deference to give trial judges.

From January 2003 to February 2004, the circuit court modified judges' interpretations in 63 percent of the precedent-setting cases that relied on them, according to a study by the law firm Jenkens & Gilchrist. Not all of the cases were overturned.