Expert evidence in patent cases

by Bruce Stratton

Originally published February 2007 by Thomson Carswell on IPSource. Used with permission.

In the recent decision of Pfizer Canada Inc. v. Canada (Minister of Health) (2006), 2006 CarswellNat 4264, 2006 FC 1471 (F.C.) a typical analysis of expert evidence in patent cases is found:

• "Both parties called a host of experts."
• "The experts differed in their testimony. Particularly, their evidence differs on what a person skilled in the art would have known or assumed."
• "Their evidence will be extremely helpful..."
• "These issues, however, are ultimately questions for the Court to decide..."

In the Pfizer case, the Court turned to the decision of Campbell J. in AB Hassle v. Apotex Inc. (2003), 27 C.P.R. (4th) 465 (Fed. T.D.) at para. 16 as setting out guidelines for dealing with expert evidence:

Each of the expert witnesses to the present case have sworn that the evidence they have provided is true. On this basis, an evaluator of the evidence must start from the proposition that the witnesses are credible unless good cause is shown, and can be articulated, to the contrary (for an example of this general principle see: Maldonado v. Canada (Minister of Employment and Immigration), [1980] 2 F.C. 302 (C.A.). That is, while they might hold differing views on a given topic, it must be assumed that they are not just saying things to bestow a benefit on the party who is relying on their evidence. In my opinion, it is unfair to the witnesses and, accordingly, to each of the parties, to make negative credibility findings in the guise of findings of weight without seeing and hearing each witness testify.

It should be noted that the Pfizer proceeding was an application under the PM(NOC) Regulations and therefore there was no opportunity for the Court to see or hear each witness testify.

In this recent Pfizer decision the Court suggests both that the expert evidence will be helpful, despite the conclusions of the experts being contradictory, and that the Court will make its own decision regarding the subject matter of the expert evidence, despite a hesitancy to make findings as to weight of the evidence without seeing the witnesses testify.

If this seemingly contradictory approach to expert evidence is typical, why is expert evidence tendered in patent cases, and how should judges deal with such evidence? For it is a virtual certainty that in a patent case there will be expert evidence tendered by both parties.

The rationale for such evidence being tendered is set out by the Supreme Court of Canada in R. v. Abbey, [1982] 2 S.C.R. 24 (S.C.C.) at 42, per Dickson J.:

An expert's function is precisely this: to provide the judge and jury with a ready-made inference which the judge and jury, due to the technical nature of the facts, are unable to formulate.

In the same case, Justice Dickson, as he then was, set out that the value of such evidence is that it is able "to furnish the court with scientific information which is likely to be outside the experience and knowledge of a judge or jury."

The leading case regarding expert evidence in Canada is the Supreme Court decision of Sopinka J. in R. v. Mohan, [1994] 2 S.C.R. 9 (S.C.C.). In that case, four criteria for admitting expert evidence were set out:

• Relevance;

• Necessity in assisting the trier of fact;

• The absence of any exclusionary rule; and

• A properly qualified expert.

Evidence tendered to a court will be considered probative if it is "so related to a fact in issue that it tends to establish it." However, relevance is to be determined by weighing the impact of the evidence on the trial process against its probative value. The impact in terms of the prejudicial effect, the time to present the evidence at trial, and the effect on the trier of fact relative to reliability are all factors to be weighed against the "logical relevance" of the expert evidence (at S.C.R. p. 18).

As Mohan sets out, expert evidence is allowed only when "necessary". However, this necessity is not "too strict a standard." One way to formulate the standard is to state that the information in the expert's evidence should be something that is "outside the experience and knowledge of a judge or jury" (at S.C.R. p. 22). If, due to the technical nature of the evidence, the trier of fact is unlikely to otherwise form a correct judgment then the expert evidence will be considered "necessary".

In patent cases, as was the case in Pfizer, supra, experts are typically called by both plaintiff and defendant. Although it is, strictly speaking, a matter of law, often experts are asked to construe the terms in the patent claims in suit (but not always, see: Wessel v. Energy Rentals Inc. (2004), 32 C.P.R. (4th) 315 (F.C.) at para. 14). Such evidence may be properly admissible as it can provide "scientific information"(the meaning of a technical term well-understood in the art but not known to the judge) or the evidence may provide a "ready-made inference". An example of the latter type of expert evidence is often found when terms in a patent claim are suggested to be essential or non-essential by evidence in an expert report. The expert will typically give an opinion that goes beyond simple scientific information and will include reference to what a notional worker would have understood the patentee to have meant in the patent as a whole.

Similarly, issues of obviousness in patent cases call on experts to provide inferences leading to conclusions about what solutions would occur directly and without difficulty to a person skilled in the art. Technical inferences are also involved when prior art elements are compared with claim elements (anticipation) and when infringing device elements are compared with claim elements (infringement). Testing of products or devices is also carried out by experts, for example where infringement depends on the chemical composition product. Given the role of experts and the type of evidence in patent cases, it is not surprising that experts are called to give evidence both to set out scientific information outside the knowledge of the judge and to provide ready-made inferences. There are surprisingly few experts used in Federal Court patent trials who are qualified to give evidence due to their "hands on" experience in a field, as opposed to their theoretical or academic qualifications (the "artisan" is well received by the courts but rare, see: Unilever PLC v. Procter & Gamble Inc. (1995), 61 C.P.R. (3d) 499 (Fed. C.A.)).

Because the notional skilled worker can be a composite or a team, experts who may not be each themselves have the knowledge of a notional skilled worker will be accepted if they are each qualified as experts as to the knowledge of individual team members (Westaim Corp. v. Royal Canadian Mint (2002), 23 C.P.R. (4th) 9, 224 F.T.R. 184 (Fed. T.D.) at paras. 34-39).

When faced with contradictory expert evidence, as occurred in Pfizer, supra, the Court must make a choice. The role of the Judge in assessing expert evidence is closely tied to the rationale for admitting such evidence, as set out in Abbey and in Mohan. Usually, the role of the Judge in considering expert evidence is not, as may be the case with fact evidence, to determine the credibility of the witness on the basis of honesty or bias. Rather, the balancing of the expert evidence, and the weighing of the evidence, will be carried out by considering the value of the thesis set out in the evidence of one expert, as against the value of the competing thesis set out in the expert evidence filed in response (see Henkel Canada Corp. v. Conros Corp., 2004 FC 1747, per Hugessen J).