Tuesday, March 19, 2013

Purposive Construction in the Patent Office

Examination Practice Respecting Purposive Construction PN2013-02

CIPO has issued new practice guidelines respecting purposive construction (PN2013-02) and computer-implemented inventions (PN2013-03). Both sets of guidelines strike me as basically sound. This post makes a few comments on the purposive construction guidelines. A subsequent post will comment on the guidelines regarding computer-implemented invention, which mark a radical departure from CIPO’s prior formal position.

The guidelines respecting purposive construction begin with a puzzling statement suggesting that while the approach to construction set out in Free World 2000 SCC 66 and Whirlpool 2000 SCC 67 will “continue to guide the courts,” these cases were not directed to patent examiners. The guideline relies for this on two paragraph from Gibson J’s decision in Genencor v Canada (Comm’r of Patents) 2008 FC 608, in which Gibson J at [61] quoted a passage from Whirlpool [52] to the effect that a “dictionary” approach to construction must be rejected and the claims must be construed with reference to the entire specification. It was in respect of this passage that Gibson J stated that “I am satisfied that the foregoing, in all its implications, was directed to trial judges and to judges of courts of appeal and not to patent examiners in the course of examinations.” To the extent that this suggests that a dictionary approach should be taken in examination, and the claims should not be read in the context of the patent as a whole, this statement is clearly wrong, and it would be disturbing if this approach were to be adopted in examination. But later on the guidelines emphasize that “it is critical that a purposive construction of the claim be performed considering the specification as a whole as read through the eyes of the person skilled in the art,” which is exactly what the SCC said in Whirlpool in the passage impugned by Gibson J. Also, the guidelines go on to note that “during examination of an application the language of the claim may change from that initially proposed by the applicant for a number of reasons,” and that construction of the patent “must take into account the role of the patent examiner and the purpose and context of examination.” This seems to me to be a reasonable proposition. For example, the courts are very reluctant to hold a patent invalid for ambiguity. In contrast, rejections in examination based on indefiniteness are relatively common. This difference seems reasonable, since the remedy is different. During examination, the patentee can amend the claims, and if the claims in an application are arguably ambiguous, they should be clarified before grant in order to provide clear notice to third parties. Once the patent has issued, it may be appropriate to give the best interpretation to those same claims, in light of the relatively draconian alternative of holding the claims invalid. More generally, the functional point the guidelines seem to be making is that the examiner is not obliged to engage in the same kind of detailed scrutiny and fine parsing of claim that might be undertaken at trial, and if the validity of a claim turns on its scope, the scope should by clarified be amending the claim rather than by engaging in a debate over claim construction. Genencor was presumably the best authority that could be found for this proposition, and that may well have been the real thrust of that decision, despite is peculiar treatment of Whirlpool.

Apart from that, the main feature of the guidelines is a focus throughout on identifying the problem and solution as part of the construction, relying on s 80(1)(d) of the Patent Rules, which require the description to “describe the invention in terms that allow the understanding of the technical problem, even if not expressly stated as such, and its solution.” The problem-and-solution approach is an accepted part of the obviousness analysis in the EPO, and while systematic focus on the problem and its solution is not similarly accepted in the Federal Courts, there are many cases which invoke similar language, as discussed here. My own view is that this is generally a useful approach, particularly in the obviousness context.

I also entirely agree with the statement that “Without having first considered the problem and solution, the identification of essential elements would be circular - it would begin and end with the language of the claim.” I make a similar point in my Essential Elements article, 22 I.P.J. 223 (2011).

The practice guidelines also invoke Rule 80(1)(d) as support for the promise of the patent doctrine for determining utility. My view is that the false promise doctrine is incompatible with the Canadian Act, but I can’t criticize CIPO for accepting it, given that the FCA has said it is the law.


  1. The fundamental difficulty with the problem-and-solution approach is that it can easily fail to acknowledge the degree to which recognizing that a problem exists is part of the inventive exercise. Often, the 'problem' that inventors overcome is that others don't see that there is any 'problem' with carrying on as before. For this reason, the problem- and-solution approach too often leads to the trap of 20/20 hindsight. Canadian courts have, as you say, not adopted this as our model, and the Patent Office is accordingly not free to impose it. The tenor of the guidelines as an effort to strain against the central features of our jurisprudence is given away in the opening lines

  2. I agree that that is the main difficulty with the problem-and-solution approach, but so long as this is acknowledged, then it can be dealt with (see eg Actavis v Novartis [2010] EWCA Civ 82 [35]). The problem-and-solution approach, just like Windsurfing / Pozzoli, is an approach, not a straightjacket. The main difficulty with the Windsurfing / Pozzoli approach is identifying the inventive concept, and that is where I think the problem-and-solution approach is helpful. So I see the two as complementary. (Also, I should say that the guidelines do not adopt the problem-and-solution approach to obviousness specifically, but talk more generally about the need to understand the problem and solution in order to construe the patent.)

    As to whether the Patent Office is free to impose this approach, the Patent Office is relying on Rule 84(1)(d), which requires a statement of the problem and solution. On a purposive interpretation of the Rules, it is reasonable to understand this as implying that the problem and solution should be considered in the construction of the patent. So the problem and solution approach has a stronger basis in the Act than does Windsurfing / Pozzoli.

    So far as the case law goes, Windsurfing / Pozzoli is certainly more prominent as an explicit approach, but identification of the problem and solution is also firmly grounded in the case law: see eg Wellcome / AZT ¶ 37 “A patent. . . is a method by which inventive solutions to practical problems are coaxed into the public domain.”; Uhlemann Optical [1952] 1 SCR 143 at 152 “We have it, therefore, in the present case that there was a problem to be solved and a want to be supplied. The 2-point Numont mounting made a substantial contribution to the solution of the problem.” AstraZeneca v 2010 FC 714 Hughes J at ¶ 33 “¶33 In order to address [sound prediction, anticipation and obviousness], the Court must first address the following: ...c) What was the problem that the patent addresses? d) What is the solution offered to that problem by the patent?”; Jay-Lor, 2007 FC 358 at ¶ 9, 79 “ 9 Every invention solves a problem. This is so even if the problem is not universally recognized or encountered by all users. Thus, I begin by identifying the problem. 79 The key question [regarding obviousness] is this: Would our mythical skilled technician directly and without difficulty have determined that an inclined upper surface on the auger would solve the problem of jamming or hung up hay bales?” Pfizer v Apotex 2007 FC 971 at ¶ 57 aff’d 2009 FCA 8 at ¶ 11 “More specifically, with respect to the issue of obviousness, which is the only one left to be considered in the present appeal, the Federal Court Judge held that the core question in the case was whether the person of ordinary skill in the art, in the light of the state of the art and of common general knowledge as at the claimed date of invention, would have found the solution taught by the patent. The "solution taught by the patent" that he used for this inquiry was consistent with his claim construction...” In Biogen Inc v Medeva Plc, [1997] RPC 1 at 45 the House of Lords went so far as to say that “A proper statement of the inventive concept needs to include some express or implied reference to the problem which it required invention to overcome.”

  3. I would agree that as part of a nuanced approach to construing a patent, the problem-and-solution approach is useful. This is not, however, the structure provided for examiners by the Guidelines, which mandate this as the central analytical exercise.

    The Guidelines get rapidly off on the wrong foot by suggesting that “Guidance in identifying the problem the inventors set out to address and its proposed solution should be found in the description, in accordance with paragraph 80(1)(d) of the Patent Rules, and not by reference to the closest prior art.” The Rule does indeed require the inventors to describe the problem, presumably as they appreciated it, and thereby give context for the solution which is claimed. Unfortunately, the Guidelines explicitly take the inventor’s apprehension of the problem as the starting point. What if the prior art has no intimation that the problem existed? The Guidelines instruct examiners to ignore this gaping hole in the art, and leap directly to the comforts of 20/20 hindsight based on the inventor’s own disclosure.

    Instead of an upfront acknowledgement that Free World Trust and Whirlpool continue to guide the Patent Office, in construing the claims in accordance with the principles of purposive construction, we get the fairly explicit suggestion that the Guidelines will chart another path. This is not a nuanced approach to construction; this is deliberate administrative insubordination flying in the face of Amazon.

    Once the problem-and-solution approach has been rigorously applied to construing the claims, in accordance with the Guidelines, it’s not going to be left behind as the examiner ponders obviousness. It will displace the framework of Windsurfing/Pozzoli/Free World Trust and Whirlpool, with the frequent pattern of objection being: now that you’ve told me what the problem is, I see that your solution was obvious.

    Although it’s true that the cases are replete with examples in which the problem-and-solution approach has been useful to construction, I don’t think there’s evidence that this approach is required if one is to properly identify essential elements. The case law reflects the nuanced application of the approach, but does not suggest that it is necessary and sufficient to arrive at a proper result.

    The Guidelines take us down the road to identifying the “inventive solution.” This is not much of an advance on the “inventive concept” approach that was so thoroughly repudiated in Amazon.

    The Guidelines are a necessary effort to give certainty to an intellectually difficult exercise. However, in the larger context of the overall patent system, it doesn’t serve the purpose of certainty for the Patent Office to be bucking the guidance that it’s been given by our courts. The Commissioner isn’t afforded that discretion by the Act.