05 May 2017 in Newsletter
Functional Claim Language: The Indefiniteness Trap
Functional claim limitations define an element of an invention in terms of what it does rather than in terms of its structure. In the United States, 35 U.S.C. §112(f) (pre-A.I.A. 35 U.S.C. §112, 6th paragraph) provides that “means-plus-function” and “step-plus-function” limitations are interpreted to cover the structure or acts disclosed in the specification to perform the recited functions, and equivalents thereof. Thus, the scope of such limitations are narrower than the literal wording of the claim would otherwise indicate. For this reason, conventional wisdom among many practitioners has been to avoid means-plus-language in U.S. claims. However, recent trends at the U.S. Patent and Trademark Office (USPTO) and at the courts make it timely to consider whether this conventional wisdom is still valid.
What language is covered by §112(f)?
To fully consider this issue, it is first necessary to examine what types of language fall under the specific interpretation rule of 35 U.S.C. §112(f). When the words “means for” or “step for” are used in a claim followed by a function, there is a presumption that the interpretation of §112(f) applies. Conversely, when the words “means for” or “step for” are not used, there is a presumption that the interpretation of §112(f) does not apply. These presumptions, however, are now very weak and are readily set aside by both the USPTO and the courts.
In connection with the conventional wisdom of avoiding means-plus-function language in U.S. claims, applicants over the past two decades have increasingly used various claim wording strategies in an attempt to have the benefit of functional language without the “penalty” of the §112(f). Examples of this include replacing the word “means” with another word such as “mechanism” or “unit,” and/or replacing the word “for” with “configured to” or “arranged to.” Generally speaking, under the current practice of the USPTO and the current jurisprudence of the U.S. courts, these linguistic attempts to avoid application of §112(f) do not work. Both the USPTO and the courts will apply §112(f) to claim limitations that are purely functional and do not recite structure or acts in support thereof, regardless of whether the actual words “means for” or “step for” are used. It is now quite common to see USPTO examiners make a statement of interpretation under §112(f) at the beginning of Office Actions even when the claims are not written in traditional “means for” or “step for” format. The danger of this is that when claim limitations the patent drafter believe were not means-plus-function limitations are interpreted as such, the corresponding disclosure may be insufficient to support a reasonable breadth of claim coverage and, in certain situations, may be so insufficient as to lead to invalidity of the claim for indefiniteness. This is the “indefiniteness trap.” To consider this further, we will first examine the difference in scope between a means-plus-function limitation and a limitation drafted in structural terms.
What is the scope of a limitation interpreted under §112(f)?
Consider a claim reciting “a transmitter” versus a claim reciting “means for transmitting a signal.” In the first instance, the scope of “transmitter” in its noun form would be anything a person skilled in the art would consider a transmitter. Contrast this with the scope of “means for transmitting a signal,” which depends on what is disclosed in the specification. If the specification discloses only one specific way of transmitting a signal, the scope would be limited to that single way, plus equivalents of that way. This would be narrower than “transmitter.” Conversely, if the specification discloses multiple ways of transmitting a signal and suggests the intent is to cover all ways of doing so, the scope is essentially the same as “transmitter.” Thus, the widely-held view that means-plus-function limitations in US claims are always narrow is not necessarily correct; it depends on what is disclosed in the specification. From a practice standpoint, it follows that it is not necessary to avoid means-plus-function language in all cases. However, when means-plus-function language is used, it is vital to ensure there is clear structural support for this language in the specification and to ensure the example embodiments are as broad and inclusive as possible. When the requisite support is lacking, the claim might not only be narrow, it may be invalid as indefinite.
When does a limitation interpreted under §112(f) lead to indefiniteness?
35 U.S.C. §112 (b) requires that a claim particularly point out and distinctly claim the subject matter which the inventor or a joint inventor regards as the invention. When the claim includes a limitation interpreted under §112(f), part of that duty to “particularly point out and distinctly claim” the invention shifts to the specification. It is not sufficient in this situation that the specification merely enables (teach how to make and use) the invention. Rather, the disclosed structure must be sufficiently specific to be able to determine the metes and bounds of the claims. Thus, returning to the prior example, if the specification does not disclose any specific way of transmitting a signal a claim reciting a “means for transmitting” may be invalid as indefinite even though how to transmit a signal would be known to a person skilled in the art. Moreover, because new matter cannot be added to the patent application, there may be no way to remedy this problem. There is a strong tendency at the USPTO to apply 112(f) more often, and to find claims (which may be disguised, but are interpreted as claiming some form of means-plus-function language) as being indefinite for lacking sufficient structure in the specification. This is particularly true when the structure disclosed in the specification as performing the claimed function is highly general in nature, such as a “black box.”
How to avoid the §112(f) indefiniteness trap?
A first approach to avoiding the trap is clearly to avoid using functional language. As explained above, this means more than simply substituting a “nonce word” for the word “means.” It requires using a noun that has an understood structural meaning in the context of the invention (such as transmitter). Once the structure that performs the function is recited, the function may also be recited without risking §112(f) interpretation. It is not the recitation of a function that triggers §112(f), it is the recitation of a function without also reciting the structure that performs the function.
A second approach to avoiding the trap is to ensure the specification fully and broadly supports any claim limitations that might be subject to §112(f) interpretation. It is not always easy to tell which words will be considered structure, and which will be considered “nonce words.” If there is any doubt, an extra sentence or two in the specification giving examples of various structures that could be used to perform the structure will provide significant insurance against a finding of indefiniteness. The disclosure in the specification should be broad in the sense of giving or at least suggesting several actual examples, not broad in the sense of being vague. This distinction is essential.
A third approach is to embrace §112(f) and to draft functional limitations in means-plus-function form. An advantage of this is that there would be no fight (and litigation expense) as to whether §112(f) applies. Moreover, it provides a clear sign to the patent drafter that the specification must appropriately support this limitation. As noted above, when the supporting portion of the specification is drafted properly, the scope of a means-plus-function limitation can actually be quite broad. Thus, more than anything else, avoiding “accidental” interpretations under §112(f) where careful thought has not been given to the supporting disclosure is highly important.
Special problems for computer-implemented invention claims
Computer-implemented inventions present additional challenges to the question of what suffices as “the corresponding structure, material, or acts described in the specification,” for purposes of 35 U.S.C. § 112(f). Often, generic computer components are viewed as “black boxes,” and can lead to indefiniteness rejections. To ensure the specification has the proper structure, algorithms necessary for performing the claimed functions must be described. Further, when software is claimed as, for example, a “distributed learning control module” or similar phrases, the specification must describe how the claimed module functions and interacts with other components to qualify as sufficient disclosure and avoid indefiniteness. Mere disclosure of a general purpose computer, without disclosure of an algorithm (such as a flowchart or description in prose) is insufficient. Because many computer-implemented invention patents have been drafted with only a general purpose computer as the structural disclosure, the claims of patents have a heightened risk of falling into the indefiniteness trap. It is worthy of note, however, that courts have recently held that disclosure of only a general purpose computer may be adequate to avoid indefiniteness if the claimed function is one that could be performed by general purpose computer without an additional algorithm.
In summary, a proactive decision should be taken at the time of claiming an invention as to whether application of § 112(f) is desired. If so, multiple specific examples should be given of the corresponding structure, material or acts in the specification, and clear descriptions should be provided including hardware structures and algorithms performing the claimed functions linking the disclosure to the claimed functions. On the other hand, if § 112(f) interpretation is not desired, claim terms associated with functional recitations like “mechanism,” “module,” “device,” “unit,” should be avoided. Instead, structural terms like “circuit,” “detent mechanism,” “digital detector,” “reciprocating member,” “connector assembly,” “perforation” should be used. In cases where any doubt exists, multiple examples of structure that performs any functions recited in the claims should be provided.