Tuesday, October 9, 2012

Does the BPAI require a definition from the time of filing?


If you read this blog regularly, you'll know that claim construction is often the dispositive issue on appeal. Claim construction arguments are, in the end, arguments about the meaning of claim terms. I've posted several times about using dictionary definitions to make persuasive arguments about the meaning of a claim term. (See this list of posts tagged with "definition"). But what source should you use for a definition? And since claim terms are to be interpreted according to meaning at the time of filing, do you need to show a date for the definition?

Today's post will address the second question. I plan to cover the first question in a later post. (In the mean time, check out the post "What does the BPAI think of Wikipedia definitions?" at the Florida Patent Lawyer blog.)

The Board has on several occasions refused to consider a definition that is undated. For example:

[The Applicant's] citation lacks a date to ensure that it reflects knowledge as of the time of their invention.
Ex parte Hyland, BPAI 2011.

However, both the Examiner and Appellants failed to provide corresponding dates indicating that such definitions come from dictionaries that precede the effective filing data of the present patent application. We therefore decline to adopt the respective dictionary definitions offered by the Examiner and Appellants.
Ex parte Raith, BPAI 2012. See also Ex parte Hicks, BPAI 2011 and Ex parte Moran, BPAI 2011.

Though I don't know of a case in which the Applicant provided a dated definition that postdated the Applicant's filing date, I suspect the Board would refuse to consider such a definition for the same reason.

The Board often introduces a definition of its own when the parties do not. When doing so, the BPAI is inconsistent about whether its definitions have dates before Applicant's filing date. See Ex parte Beer (dictionary dated 1971); Ex parte Raith (dictionary dated 2000); Ex parte Schmitz (dictionary dated 1996); Ex parte Reiter (citation with copyright 2012); and Ex parte Gerards (citation to online dictionary "visited Apr. 21, 2012").

Does the meaning of claim terms change so rapidly that we really need to introduce dated definitions? Would it make more sense to presume that a definition reflects the meaning at the relevant date, and let either side rebut the presumption?

The Board recently took this approach in the reexamination appeal Nike v Adidas (PTAB 2012). In the Nike reexam, the parties argued over the meaning of "inflection." In deciding the issue, the Board introduced a dictionary definition of this word. However, instead of noting the date of the definition, the Board instead applied a presumption about the date:
The definitions of general terms cited herein are believed to accurately reflect the meanings of those terms as of the effective filing date(s) of the rejected claims.

I think this approach makes sense during prosecution. After all, most applications and even issued patents are appealed within 10 years of the time of filing, and I doubt that many dictionary definitions change much in that period. So this seems like a reasonable presumption.

I wonder if this presumption is available to Applicants? Maybe. But don't rely on this. Instead, you should provide the date of any definition you rely on. And better yet, provide some evidence of this date rather than a mere assertion.

6 comments:

  1. Karen, the PTAB didn't exist until 2012. Your link to the Nike opinion has an incorrect date.

    ReplyDelete
  2. BPAI -- "do as I say, not as I do." They are a bunch of hypocrites. They'll admit dictionary definitions when it suits their purpose (to affirm) and they'll refuse to enter them when it doesn't.

    That being said, I recommend that a smart patent attorneys opens up their Bluebook and find the correct citation form for a dictionary and use it. Also, I suggest using McGraw Hill's Dictionary of Scientific and Technical Terms. The Federal Circuit uses that dictionary all the time, so the BPAI cannot disparage its use.

    "I doubt that many dictionary definitions change much in that period. So this seems like a reasonable presumption."
    It is a reasonable presumption – so long as it suits the BPAI's (now PTAB's) purpose.

    ReplyDelete
  3. Even if the BPAI does not accept that definition, you can bet that during claim construction in a subsequent lititgation, in view of the Philips case, such definition may be considered intrinsic evidence that will be given greater deference than a definition proferred during the litigation itself. Due to prosecution history estoppel, this could work for or against the patent owner.

    ReplyDelete
    Replies
    1. >such a definition [provided by
      >Applicant] may be considered
      >intrinsic evidence

      Good point. And I do appreciate that an Applicant may not want to pin himself down to a definition during prosecution.

      But I feel like some Applicants are too hesitant to make substantive arguments during prosecution out of a generalized fear of file wrapper estoppel.

      The way I see it, sometimes you gotta put something down on the record in order to even get a patent to litigate.

      Delete
  4. I don't see much inconsistency in the cases that are cited. If the claim term is a word such as "smooth" or "above," the date doesn't matter a lot because the definitions have not changed much over the centuries, nor are they expected to. If the claim term is a term of art and a technical dictionary needs to be consulted, dates matter. A young word in, say, database systems, IC manufacture, or Internet commerce can change quite a bit over time (e.g., ten years, or less).

    ReplyDelete
  5. Two comments:
    1. Re: word meanings don't change. In general, word meanings don't change. But, I'd think in patent law, where we are drafting about new technologies, many words may not have established meanings or are being used in new ways more than in newspapers and such. So be careful with that one.
    2. Re: "you gotta put something down on the record in order to even get a patent to litigate." Perhaps, but then you might as well put it in the claim.

    ReplyDelete