Claim construction and the scope of the word ‘animal’

George Nathanael is a JD candidate at Osgoode Hall Law School

Are humans animals? The U.S. case of Martek Biosciences v. Nutrinova, which was recently decided by the Court of Appeals for the Federal Circuit, is an excellent example of how difficult patent claim construction can be sometimes. Interestingly, five judges sat on the panel (as opposed to having the usual three or a hearing en banc), and they were split 3-2 on the issue of whether the term “animal” used in the claims ought to be interpreted broadly enough so as to include humans, with the majority holding that it did.

Martek’s U.S. patent number 5,698,244 dealt with a method for increasing omega-3 highly unsaturated fatty acids within animals by feeding them particular microorganisms. Within the detailed description of the invention, it was stated that “[t]he term ‘animal’ means any organism belonging to the kingdom Animalia”, which includes humans. Despite this clear language, other parts of the specification seemed to indicate that humans were not to be included in this definition. First, there was a distinction between “human intake” and “animal feed” when discussing how omega-3 highly unsaturated fatty acids can be added to both. Additionally, the patent was concerned with food products derived from the ‘animals’ mentioned in the claim, but humans are not to be thought of as food products. Finally, though perhaps least convincing, the stated preferred animals for this process were “economic food animals”, and the list did not contain humans. Based on these facts, as well as an interpretation of relevant jurisprudence (i.e. Phillips v. AWH Corp.) to mean that the specification must be viewed in its totality, the minority believed that humans ought not to be included in the definition of ‘animal’.

Using Phillips themselves, the majority ruled that “[w]hen a patentee explicitly defines a claim term in the patent specification, the patentee’s definition controls”, and this applies even if the definition is contrary to the conventional use of the term. It was stated that “absent a clear intention to restrict the invention to particular members of the kingdom Animalia, [the majority] cannot limit the claims to the listed preferred embodiments”, such as food products. Similarly, it was believed that there was no language within the patent that can be interpreted as clearly intending to disclaim coverage of humans, and the dependant claims that were directed to specific types of animals should not be viewed as limiting the scope of the broader claims that applied to all animals. Finally, it was noted that just because an inventor may have believed that an invention would be used only in a particular way does not mean that the scope of the invention is limited to that context, unless the language of the patent clearly shows otherwise.

From the Patently-O blog, one commentator believed that “there is no suggestion [from the Phillips case] that the existence of an express definition means you can ignore the remainder of the specification”. This was replied to by another who said that it would be worse to ignore an express definition based on the remainder of the specification. However, if the remainder (specifically the distinction between “human intake” and “animal feed”) can be seen as an implicit definition (or implicit modification of the original express definition), as a third poster remarked, then what reason is there for ignoring such an implicit definition when it is compatible with the express one? It would also be interesting to note what the outcome of this case would be if an explicit definition of ‘animal’ was not given. Since the patent application would have to be viewed from the perspective of the person having ordinary skill in the art, how strongly would a typical biologist’s definition of ‘animal’ hold up in light of the suggestive language to the contrary? When it comes to patent specifications, where semantics are everything, it is the drafter who gets to decide to be or not to be an animal.