IP insider: A way with words: functional claiming and unbounded patents
Confusion in the courts over functional claiming has encouraged opportunistic behaviour – in particular, using functional language which is disguised as structural language
If Samuel Morse had had his way, he would be remembered today as the inventor of email. Fortunately for humanity and global communications technology, the US Supreme Court took a balanced approach in 1853, upholding Morse’s claims to telegraph technology while invalidating his claim to all “use of… electro-magnetism, however developed, for marking or printing intelligible characters, signs, or letters, at any distances”. The court protected Morse’s contribution to the knowledge pool while preserving the incentive for further innovation.
The 1853 court demonstrated a prudent aversion to unbounded patents and a specific wariness towards functional claiming – the practice of describing an invention according to what it does rather than what it is. Ensuring that an inventor is given exclusivity to only his or her actual invention is key to maintaining balance in the patent system. Functional claiming disrupts this balance.
The statutory provision for functional claiming, Section 112(f), limits the literal scope of patent claims to the structure, material or acts specifically disclosed in the patent specification. Articulating innovation is difficult; a system that rewarded only inventions that can be easily expressed in structural language would discourage investment in vital areas of technology, in which functional expression is indispensable. Appropriate treatment of so-called ‘means-plus-function’ claims bridges this gap, allowing claimed functional language to be read in conjunction with the associated patent specification in order to determine the proper scope of the claim’s limitations.
Confusion in the courts over functional claiming has encouraged opportunistic behaviour. By using amorphous language, which may be construed as a means-plus-function claim by a court, a patentee can effectively defer the choice as to whether and how Section 112(f) applies until a lawsuit arises and then choose a stance depending on the circumstances of the litigation.
Traditionally, patentees deliberately using means-plus-function claiming would expressly use the word ‘means’ in the claim element. In this way, the Federal Circuit established a semi-magic word whose inclusion in a claim creates a rebuttable presumption that Section 112(f) applies and whose absence creates a strong presumption that it does not apply.
A common tactic to evade the reach of Section 112(f) is the use of so-called ‘nonce’ words. These are substitutes for the word ‘means’ that facially suggest structure, but in fact merely describe function. For instance, ‘mechanism for’, ‘module for’, ‘device for’, ‘unit for’, ‘component for’, ‘element for’, ‘member for’, ‘apparatus for’, ‘machine for’ and ‘system for’ are all popular nonce words. Distinguishing a nonce word from a structurally informative word can be a nuanced endeavour. Thus, whether Section 112(f) will apply depends on whether a court finds that a person with ordinary skill in the relevant art would recognise a term as providing specific structural guidance or merely incanting a device’s function.
“There is a risk that the Federal Circuit – in its pursuit of predictability – has invited drafting gamesmanship”
In recent years, courts have demonstrated a formalistic tendency to find Section 112(f) inapplicable in the absence of the word ‘means’. As the strength of the presumption has solidified, there is a risk that the Federal Circuit – in its pursuit of predictability – has invited drafting gamesmanship, which is antithetical to the purpose of Section 112(f).
Because structural limitations from the specification are not read into the claim in situations where Section 112(f) does not apply, others innovating in close proximity to ambiguous functional claims must struggle to determine the outer bounds of minimally structural claim elements. Through borderline functional claiming, some patentees may achieve protections that outweigh their contributions to the knowledge pool – thus discouraging investment in neighbouring technologies which otherwise might have even greater appeal. By punting on close-call claims, the Federal Circuit shifts the burden of interpreting the ambiguity – and the risk of misinterpretation – onto would-be innovators.
When functional language is disguised as structural, the dutiful application of concurrent patent validity doctrines becomes key to avoiding damage. EnOcean GMH v Face Int’l acts as a cautionary tale: in 2014, the Federal Circuit found the term ‘receiver’ to have a structural meaning distinct from ‘means for receiving’ – a linguistically equivalent functional term which appeared elsewhere in the claims. The word ‘receiver’ was found not only to convey sufficient structure to avoid Section 112(f) when appearing in the patent’s claims, but also to be sufficiently structurally informative to limit a means-plus-function claim when appearing in the patent specification. In cases like this, it is vital that the remaining guardians of balance in the patent system – written description, enablement, novelty and non-obviousness analysis – be rigorously performed.
Scrupulous enforcement of the Section 112 disclosure requirements is equally important. Requiring that an inventor enable his or her invention strips amorphous functional language of much of its pernicious potential. Had Morse been allowed to claim all use of electromagnetic means for printing at a distance under modern-day Section 112(f), his claim would nonetheless have failed for lack of enablement. Morse would have enabled one species of electromagnetic means for printing at a distance (the telegraph), but not the entire genus (such as email). Thus, by closely adhering to the precepts of novelty, obviousness and disclosure doctrine, the courts can address the risk of amorphous functional language.