The Requirement that the Written Description be Concise

ProlixBy Dennis Crouch

The Patent Act requires that the written description of an invention be “concise.”  35 U.S.C. 112p1. The meaning of that requirement is unclear, but but the language of Section 112 at least suggests that a written description needs to be concise in order to satisfy the enablement requirement. Of course, there is a tension between providing a disclosure that is “concise” and at the same time “full.” In Markman v. Westview, the Supreme Court recognized this tension — noting that the description must be “complete yet concise.”

In a 1967 district court case, the court focused briefly on the requirement — noting that the “descriptions contained in the patent in suit are sufficiently concise and clear to enable one skilled in the art to construct that which is taught by the patent.” San Marino Electronic Corp. v. George J. Meyer Mfg. Co., 155 U.S.P.Q. (BNA) 617 (C.D. Cal. 1967).

In my brief search, I found no cases invalidating a patent because its description was not concise. The most on-point case appears to be a pre-1952 Eighth Circuit case where the court came close to invalidating a family of patents based on their wordiness:

We think that the patents are unnecessarily prolix in verbiage and there should have been fewer claims more simply stated. They border on failure to comply with the statutory requirements concerning succinct description and particular pointing out and claiming. But consideration of the file wrappers and the trial court record has persuaded that those skilled in the art have understood them and the courts have had them explained.

National Aluminate Corp. v. Permutit Co., 145 F.2d 175 (8th Cir. 1944). It is interesting that part of the court’s problem with the patents was that they had too many claims. The USPTO does reject individual claims that it deems overly wordy or “prolix.” The idea behind those rejections is that sometimes overly-wordy claims may have the “net result of which is to confuse rather than to clarify.” However, the USPTO prolix rejections are grounded in Paragraph 2 of Section 112 — for failing the requirement of “particularly pointing out and distinctly claiming the subject matter.”

The probably more accepted use of the “concise” requirement is explained in the decision captioned Application of Knowlton, 481 F.2d 1357 (C.C.P.A. 1973). In that case the precurser to the Federal Circuit relied on the requirement to limit the power of the enablement requirement.

[I]t must be borne in mind that the disclosure need not only be full, clear and exact to satisfy the statute, it must also be concise, and that the disclosure is directed to those skilled in the art. The amount of precision necessary in any given case is always a matter of degree. Absent special circumstances it is not required that every nut, bolt and rivet actually used in mechanical inventions be described, or, in chemical cases, that the electron orbital patterns for a claimed compound be set forth.

In the right case, a defendant may well convince a court to give the concise requirement its full weight and invalidate a patent whose specification is written in confusing and overly-wordy language.

Notes:

  • I searched the BPAI database of decisions (1997–2009) and found only one case reviewing a rejection where the Examiner found claims invalid as prolix “since they contain long recitations or unimportant details which hide or obscure the invention.” Ex parte Nagano, App. No. 1996–4094 (Bd. Pat. App. Inter. 1999). In Nagano, the Board reversed the rejection – finding no support for the Examiner’s generalized assertions that the claims were confusing because of their length or contained unimportant details.

49 thoughts on “The Requirement that the Written Description be Concise

  1. 49

    Why are U.S. applications as long as they are?

    I think that most often it is defensive. I routinely have to prosecute applications where the specification is to sparse. You know there is an invention there, esp. after talking to the inventors. But you can’t claim it, because it isn’t in the specification. And, if you do, you get new matter rejections. So, you put in a lot of extra detail JIC you find yourself in the position I have found myself in so much recently, where you need more detail to overcome obviousness rejections.

    One type that I am facing a lot right now, esp. in view of Bilski, are applications that require computers to operate, but since that was so obvious, the drafter skipped including a stock computer drawing. So, you find yourself responding to the, now fairly standard, 101 and 112 P2 rejections by trying to argue the inherency of the computer that you now have to link to all the claims. Luckily for me, I have been habitually and automatically including computer drawings in the applications that I have written, so haven’t run into this with the apps that I have written.

  2. 48

    By my count, Apple’s i-Phone pinch patent, 7,479,949 has twenty six inventors (including Steve) 293 sheets of drawings and 126 columns of text. It still can be ordered from the PTO for three bucks. I think this is prolix designed to overwhelm in an adversarial situation

    . Yeh, well, just wait until all 26 inventors are deposed for what they contributed to which claims. And, then if there are really 25 inventors, or 27 inventors, fraud on the patent office.

    When I was in-house patent counsel, and we had mega-applications, before they would issue, we would do inventor reviews to identify who should really be named as inventors, esp. since this may change during prosecution.

  3. 47

    “Better keep that gov service job. Chop chop 6, come serve me.”

    You’re the kind of girl whom I make sure has an issue with an unnamed and largely unknowable LIE that I simply cannot fix myself.

  4. 46

    NWPA,

    …only those that 6 himself writes for as long (or as short) as people allow him to write applications (if and when he leaves the gov and can find employment as a practicioner).

    My prediction: such a career would be VERY short for the one known as 6. Better keep that gov service job. Chop chop 6, come serve me.

  5. 44

    I think there is a relationship between the court cases that have invalidated patents or narrowly interpreted claim language and the length of the patent application. So cases necessitate adding matter and some necessitate removing matter from the patent application.

    It is a complicated relationship, but patent application guidelines in large corporations are driven by the court cases.

  6. 43

    >>BTW Bentwater, you’re North American, not European, right?<< Just a solid North American hockey fan. Not a US patent law fan. Broadest unreasonable interpretation.... good grief, where to they get folks to examine claims using that charged anti-matter strainer for patentability ??

  7. 41

    Prolixity used to be a ground of rejection when I began practicing IP law, until the early 1970s. It applies to the claims (excessive wordiness and/or excessive number of claims). It was dropped (along with a number of other typical rejections) as an acceptable ground of rejection following CCPA decisions that required rejections to have a stautory basis. Currently prlixity is not an acceptable basis for rejection of claims unless the examiner can show the claim is vague or indefinite.

  8. 40

    “One can have the best of both worlds by filing a prolix provisional application and a more concise utility application based on it. That way the provisional application can function like a vast reservoir from which you can retrieve additional support on an as-needed basis.”

    And once your more concise utility grants, you have per se dedicated to the public anything not claimed in your utility, but which was in your prolix provisional?

  9. 39

    Just for entertainment: A recent client of mine initially insisted that I include all ~70 dwg sheets of his self-written priority applic – for a relatively simple invention). While sorting through the mess, I discovered about 15 dwg sheets that were identical copies ! And, many others that were side views showing much less than his excellent perspective views. I think I ended up filing 25 dwg seets in the end. I cut his text from 25 pages to 12. He would have paid over $1,000 more in PCT fees.

  10. 38

    By my count, Apple’s i-Phone pinch patent, 7,479,949 has twenty six inventors (including Steve) 293 sheets of drawings and 126 columns of text. It still can be ordered from the PTO for three bucks. I think this is prolix designed to overwhelm in an adversarial situation. My opinion, people.

  11. 37

    I agree that a provisional must satisfy 112 and therefore concisely describe “the invention.” But since a provisional needs no claims, there is no “the invention” against which to determine whether or not 112 is satisfied.

    And of course the inventorship issue also goes away. If “the invention” is nowhere defined, inventorship is meaningless.

    This is a bit of a logical quandary in the statute.

  12. 36

    “Maxdrei, I’ve been busy. Ironically, I have been trying to invalidate a software patent.”

    HAHAHAHAHAHAHAHHAHAHAHAHAHHAHAHAHAHAHAHAHAHHAAH

    If you need any help you give me a mail.

  13. 35

    They should change the incorporation by reference rules to allow incorporation of applications that have not yet been published, but then automatically make such applications available to the public online. I include a lot of content for enablement and/or best mode that I’d rather incorporate by reference, especially when I file a set of related cases.

  14. 34

    I’ve received a Prolix rejection once in 8 years (and never made one in the three years examining before that). That rejection was replaced on the next Office Action by a 112-2 rejection based solely on the number of claims (35) — not to-maa-to, but to-mah-to. The next Office Action was an allowance. The Examiner in question just didn’t want to search the case.

  15. 32

    “Concise” – free from all elaboration and superfluous detail.

    There is nothing like superfluous detail to obscure the real details of an invention. Its like submitting 100 documents in an IDS.

    I don’t see any tension between full disclosure and conciseness… I do see tension between conciseness and flexibility and depth.

    “Ganja-tek” No doubt an obscure drug related reference. Next thing ya know, Mooney will be talking about his crack habit again.

  16. 31

    The above suggestion to use a single provisonal application as a large data dump for whatever formal application claims you maight want to file later may create enforcement issues. The CAFC has made it clear that in order to obtain the benefit of a provisional filing date for a claim, the provisional has to meet all the same full 112 requirements as a formal application for whatever is being later claimed. That includes a required description of “the invention”, not just a whole lot of possible enablements. Also it might create inventorship issues in some cases.

  17. 30

    Night, you might try T_1143/06. It is very recent, and canters through earlier cases including the landmark Duns licensing case T_0154/04 on the unpatentability of “presentations of information”. But, as you know, there are loads of T Decisions, to the extent that it is hard to keep up. To get a feeling of what will “fly” (or what you can bring down), you have to be somehow “in the flow”, which I am not.

  18. 28

    Maxdrei, you put up some recent cases from the EPO. I’m just looking for those, so I can read them.

  19. 27

    NWPA, not sure whether or not to be helpful here. Are you seriously asking where the EPC refers to 35USC101 and, if so, is that supposed to be sarcastic?

    I guess not. Are you asking me to write “Article 52(2)(c) EPC”?

  20. 26

    Fisher is correct.

    Actually Max, Assuming each of claims 3 – 6 fully uses multiple dependency, doesn’t it work out to be 16 US claims for 6 ROW claims?

    Claim 3 would be equivalent to 2 US claims, claim 4 would be equivalent to 3 US claims, claim 5 – 4 US claims, and claim 6 – 5 US claims. Claims 1 and 2 of course are not able to be multiple dependent. Therefore it’s one plus a summation to N-1 US claims for N ROW claims.

  21. 25

    OOPS. Memory failing me. Thanks Fisher. Not x but +. What I meant was 5+4+3+2+1 = 15. That evidently has another name. Can you put you to the further inconvenience of reminding me?

    But even that’s wrong, now I think harder, before posting. What I meant was (if I understand correctly how to compute USPTO claims fees) 1+1+2+3+4+5 = 16

  22. 24

    Maxdrei, I’ve been busy. Ironically, I have been trying to invalidate a software patent. Where were those references to 101 in the EPO. I wanted to read them, but can’t seem to locate them.

  23. 22

    Too many notes. Great quote Roger. I remember that bit in the film too. Got a great laugh, didn’t it?

    What are you driving at though? Should filings at the USPTO have fewer claims? Or are they, like Mozart’s writings, already as perfect as can be?

  24. 21

    MaxDrei

    I believe that subject matter in a provisional application is available to support claims in a patent that claims the benefit of the priority date of the provisional, whether before or after issuance. So I would think that suggestion would work.

  25. 20

    Great posting. It is hard to explain to Americans that, when an app has no more than claims 1 through 6, the EPO says there are “6” claims, no more than that. Conversely, for the USPTO

    6 = factorial 6 (that is, 15).

    Question for readers: Where else in the world is it forbidden to have a multiple dependent claim itself dependent on a multiple dependent claim?

    Bargains at the EPO. 15 claims for the price of 6. Roll up, roll up.

  26. 19

    Reminds me of the movie “Amadeus”, where Emperor Joseph II complains to Mozart that his operas have “too many notes”.

  27. 18

    Contained the concise 367 page, 354,702 word page application, the September 20, 2001 U.S. Patent Publication 20020093648 had 6,632 claims and 2,269,086 characters. Filed nine days after the September 11 attacks via express mail with first named inventor “Mehrdad Nikoonahad”, the practitioner filing the application counted the claims as 6 independent and 88 total claims and figured the total amount owed as $4,430. They were slightly off.

    The October 29, 2001 missing parts requested $119,016 for 6,612 claims over 20 and $13,680 for 171 independent claims over 3. They also wanted their $130 surcharge for late fee for a total owed of $133,536.

  28. 17

    Is that a serious suggestion? Is such retrieval possible, even after issue? I had thought that the prolixity was there because it’s needed in court.

    Now a more serious suggestion might be to file simultaneously a non-pro (in the name of the inventor) and (for everywhere in the world except the USA) a PCT in the name of the Owner. Then, for exactly the same invention, one could have the optimal drafting for the US courts, and the optimal drafting for the courts everywhere else.

    Taking a long term view, that would be cheaper and better than trying to get the same document to do duty both inside the USA and outside.

    Some corporations are doing it that way already, no?

  29. 16

    One can have the best of both worlds by filing a prolix provisional application and a more concise utility application based on it. That way the provisional application can function like a vast reservoir from which you can retrieve additional support on an as-needed basis.

  30. 15

    The current issue of the AIPLA Quarterly includes a trenchant piece from Bey and Cotropia that savages the BRI standard. Bent, those who draft for the USA, they need double the length don’t they, in order to address two completely different ways of interpreting the plain words of a patent claim, namely BRI in the PTO, and “anything else but BRI” in the courts.

    B + C do a good job of exposing how inane this idea from the CCPA judges really was, when they came up with it, all those years ago.

    BTW Bentwater, you’re North American, not European, right?

  31. 14

    TO add to Max Drei’s list from a US Perspective…..

    The major reason that US cases are lengthy is because US patent jurisprudence is totally screwed up.

    1. You have to provide a lot of explanation and background and examples, cover as many variables which “are known at time of filing”…..since the courts WILL NOT UNDER ALMOST ANY CIRCUMSTANCES GIVE YOU ANY EQUIVALENTS.
    (Maybe in case of a first office action allowance…. a very small bite at that could be coaxed out of a court, IF you can show such variation was TOTALLY UNKNOWN at time of filing…)

    2. You have to fill in as many blanks as possible… the cases are most often decided by judges whose decisions telegraph to the world that they or their clerks don’t understand the technology. Insane examples in physics abound. Things are NOT “tangible” or “physical” to folks who don’t understand light, magnetism, optics, etc., etc. ….. …so how can they be “inventions”??? Yet “de novo” claim construction at the CAFC, without benefit of technological experts below, is the order of the day. Simply insane.

    3. US practice now requires “perimeter fence” claiming… with post to post support for all fencing. This is NUTS.

    4. US does NOT have “backbone” claiming as in Europe, where you can “claim the heart” of the technology and rely on technically literate decision making before BOTH the Patent Office (not true these days at the USPTO, in many cases) AND before the courts, especially, for example, in places like Germany.

    5. Even the Canadian courts have adopted a “purposive construction” approach that provides for some degree of flexibility in claim construction. In the US, lack of an enormous specification leaves you totally vulnerable compared to either of those jurisdictions.

    6. So, to get better and shorter applications…. GET SOME NEW JUDGES… with technically trained brains…and technically competent clerks. The current US Supreme Court shows how this is NOT likely to ever occur, at least at that level….

    7. So, where do you start, for reform ???
    Maybe by changing the STANDARDS for CLAIM CONSTRUCTION during LITIGATION….

    YES…. lets have some PUBLISHED CASES that institute “PURPOSIVE CONSTRUCTION” from a TECHNICALLY LITERATE PERSPECTIVE so that a patentee can protect a technology with a “backbone” claim…… THEN you can see some shorter US specifications…..

  32. 13

    Fantastic idea for a thread Dennis. The USPTO might not object, but the EPO does, more or less every time. Readers might be excused for thinking “so what”, but might still care to read the following points, offered to stimulate the thread:

    1. Cases originating in the US are (typically) twice as long as those written elsewhere.

    2. So, translation costs borne by US industry are (typically) twice as high as those borne by everybody else.

    3. And (typically) US cases have three independent claims where everybody else has but one, leading the EPO to object every time to a lack of “conciseness and clarity” in the app as filed. Either those several independent claims all recite (in substance) the same thing, in which case they are not concise, or they define different things, in which case it is not clear whether the app is directed to one invention (OK) or more than one (not OK).

    4. And, finaly, if the foreigners are prolix, it’s because they are doing their best to mimic American drafting for the US filing. After all, the US patent is more important to them than their patent in any other jurisdiction, right?

    5. Why are American cases relatively so long? Each will have his/her own view but here are some possibilities:

    I. Your litigator tells you it’s necessary.

    II. Your boss bills the client by the page.

    III. You don’t know what’s old and haven’t decided yet what the invention might be, so you describe exhaustively everything that lies in front of you.

    IV. Fear of a malpractice suit bears down on you (but not on those drafting outside the USA).

    I compliment Amy and Malcolm in raising the “Laundry List” point, which comes as nothing new for European readers. I hope to see that point explored further, in the context of “written description” and “new matter” (as it would be at the EPO).

    I have asserted in an earlier thread that cases drafted in the USA are characterized by a diamond (the illustrated embodiment) in the middle of a huge balloon (the independent claim) with a notable absence in between of the intermediate generalisations (contributions to the technical field, expressed in ever-increasing specificity of features, delivering ever better technical performance) that will eventually provide the basis in the app as filed for the claim that is actually allowable. So, me, I think US cases should actually be 50% longer than they are at the moment. But maybe your litigator wouldn’t like it that way?

  33. 12

    I recall an application having a Specification describing a large system in over 1000 pages. The Examiner required that about 950 pages and over 100 drawing figures be deleted as not relevant to the claimed invention. This was probably not an issue of confusion, as the Specification was well organized in sections.

  34. 11

    I think I owe Amy a coke for making more or less the same point prior to my 5:19 pm comment.

  35. 10

    But usually the source of this prolix is a foreign language application (translated) in which several words (usually “function…” + “part”) are used to name a component that could/should ordinarily take only one or two words to identify.

    Sort of like “individual discrete all-natural number geographic coordinate and measurement representations”.

  36. 9

    “since they contain long recitations or unimportant details which hide or obscure the invention.” I’ve seen similar language from examiners in several cases over the past few years.

    But usually the source of this prolix is a foreign language application (translated) in which several words (usually “function…” + “part”) are used to name a component that could/should ordinarily take only one or two words to identify. This may occur in the description or the claims. For example, “a check sum and communication protocol transfer and intermittently receiving part”. Using this same language repeatedly in the description or a claim tends to mix-up where the subject/verb/objects begin and end.

    Certainly tehre is a lack of “conciseness”, but such mixed-up or vague descriptions rarely fail under today’s lax indefiniteness standard, since it is rare that the courts find anything that is “insolubly ambiguous” (which itself is an insolubly ambiguous expression).

  37. 7

    In my brief search, I found no cases invalidating a patent because its description was not concise.

    I guess I assumed the “concise” requirement as one that would preclude an applicant from filing, literally, an encylopedia of machine parts or a massive chemical index, preceded by a statement that “the invention, in certain embodiments, comprises one or more of the following species combined in an operable manner consistent with the invention’s purpose.” You know, the sort of the thing that might allow an applicant to pick and choose from a gigantic grocery list of limitations until the Examiner tires of making obviousness arguments.

  38. 6

    This isn’t quite on point, but Fujikawa v. Wattanasin, 93 F.3d 1559 (Fed. Cir 1996) did seem concerned that the patentee had created a laundry list of subspecies. Because the description did not lead a PHOSITA to the subject matter, the court refused to permit the patentee to add a proposed claim during an interference. The court makes an analogy to providing to many trees in a forest, without marking any of them.

  39. 4

    “It is interesting that part of the court’s problem with the patents was that they had too many claims.”

    Having too many claims is quite distasteful.

  40. 3

    “Of course, the PTO regularly rejects individual claims that are overly wordy or ‘prolix.'”

    I don’t think in 20 plus years of private practice that I have ever seen that rejection

  41. 2

    I sense a fishing for (an open invitation to trainwreck from) a “software patent” discussion, in that the “not every bolt” equivalent is easily seen as an affirmative expression that there is no need for inclusion of explicit source code.

  42. 1

    “Of course, the PTO regularly rejects individual claims that are overly wordy or ‘prolix.'”

    I’m not so sure about “regularly”. Most of the time I think practitioners are given the benefit of the doubt, and are not subjected to such rejections. What does happen fairly often is indefiniteness based on “narrative” claims, usually for foreign apps.

Comments are closed.