The Experimental-Use Defence to Patent Infringement: A Comparative Assessment
I Introduction
The tension between incentives to create and restrictions on dissemination runs throughout intellectual property law. A contentious issue in patent law, in particular, is the potential for patents to stifle improvements to a patented invention.1 While a pioneering breakthrough may be the basis for a new technology, subsequent incremental improvements normally yield the greatest direct social benefit attributable to that breakthrough. As Newton's famous phrase has it, 'If I have seen far, it is by standing on the shoulders of giants.'2 Patents, it is feared, might deprive today's Newton of a place to stand.
This article focuses on a doctrine that has the potential to mitigate this problem, namely the experimental use defence to patent infringement. This doctrine has gained recent prominence as a result of three controversial Federal Circuit decisions affirming that the defence is very limited in the United States, with no application when the experiment has commercial purposes.3 In contrast to the restrictive US approach, [End Page 333] European law generally provides a broad statutory defence for experiments on the subject matter of the patented invention that applies regardless of the commercial nature of the research. The time is now ripe to ask what lessons Europe and the United States might learn from their contrasting treatments of the experimental use defence, and what reform conclusions might be drawn in countries, such as Canada, where currently the defence is poorly defined.4
This article compares the US and European approaches to the experimental use defence in light of the experience reflected in the reported cases in both jurisdictions.5 The need for an experimental use defence is invoked in a variety of contexts, and the US and European approaches agree in a number of these areas: neither applies the experimental use defence to research tools, nor is there a general exemption for university-based research.6 While this cross-jurisdictional agreement on points controversial in the academic literature is interesting in its own right, we focus instead on the main point of difference between the two systems: their contrasting treatment of research on the patented subject matter itself. This research includes experiments to develop new indications for an existing drug and improvements on an existing invention. The European approach allows such experiments without licence from the patentee, while the US legal system does not.
Despite the striking formal difference between European and US law respecting improvements to patented inventions, we conclude that the European approach is not likely to have a major practical effect on the development of improvements. Other considerations, in particular sunk costs incurred in product development, encourage licensing by a potential [End Page 334] improver prior to experimentation in any event. What effect the European approach does have is of dubious merit. The European approach is likely to reduce the bargaining power of basic research institutions, such as universities or small start-up companies, seeking to license their fundamental inventions to developers of commercializable downstream products. This in turn may hamper start-ups, whose main asset is patented knowledge, in their efforts to raise financing. The European case law shows that the experimental use defence in Europe also creates significant legal uncertainty as to the scope of the patentee's rights. On the whole, the European experience with the experimental use defence provides a cautionary tale rather than an example to be followed.7
II The contrasting approaches to experimental use
The experimental use defence in the United States is entirely a creature of case law. Its origin lies in Justice Story's remark in Whittemore v. Cutter that 'it could never...
