Structural Equivalence 'Way' Must Not Add Functional Requirements

If you like means-plus-function claims, you'll just love Applied Medical Resources v. United States Surgical (May 15, 2006) where the Federal Circuit considered the structural equivalence of a means-plus-function limitation with regard to the "way" in which the claimed functions were performed by the two structures. The claim language at issue required:

"means disposed circumferentially outwardly of the valve portions for supporting the valve portions within the seal housing, the supporting means being movable relative to the housing to permit the valve portions to float relative to the axis of the cannula."

The corresponding structure from the patent was the so-called "ring-levers-teeth" embodiment shown to the left. In this embodiment, the valve 36 is supported by levers 54 which separate the valve into inner portions and outer portions. The levers (i.e., the "supporting means") grip the valve at the junction between the inner and outer portions. When an instrument moves off-axis (i.e., "floats"), the inner portion of the valve remains undeformed while the outer portion deforms.

In contrast, the accused device (below) grips the valve at its outer perimeter, and achieves off-axis movement without deformation The gimbal responds to off-axis forces from an instrument by rotating as a ball-in-socket joint.

In describing the way in which the ring-levers-teeth structure performs the supporting function, the district court stated:

Applied’s ring-levers-teeth structure supports the inner "valve portions" of a
two-portion septum valve by levers that are attached to a ring that moves from
side to side. The tooth members of the levers surround the inner valve portions,
holding them in the desired circular configuration and isolating them from the
outer portions that are intended to deform.

The district court then found that the accused structure performs the supporting function in a substantially different way because "the entire valve in [the accused structure] is supported . . . by a gimbal which holds the entire valve at its outer perimeter."

Since the appeal was from summary judgement, the Federal Circuit declined to independently construe the claim terms in the first instance. Instead, it merely adopted the two functions identified by the lower court as

1. holding the portions of the valve defining an orifice within the seal housing and
2. to permit the portions of the valve defining an orifice to move freely to and from the axis of the cannula.

The Federal Circuit then concluded that the district court improperly determined the way in which the disclosed ring-levers-teeth structure performed the supporting function:

In this step, the inquiry should be restricted to the way in which the structure
performs the properly-defined function and should not be influenced by the
manner in which the structure performs other, extraneous functions. . . .
In its description of the way in which the ring-levers-teeth structure performs
the "supporting" function, under its adopted construction of that function, the
district court improperly included the way in which the structure performed
extraneous functions.

Nothing in the court’s adopted construction of the supporting function, requires "holding [the valve portions] in the desired circular configuration," or "isolating them from outer portions that are intended to deform." Rather, the defined function only requires holding the valve portions defining the orifice within the housing. Therefore, the district court improperly imported unclaimed functions when analyzing the way in which the disclosed embodiment performed the claimed function.

In his dissent, Judge Dyk concluded that district court did not improperly redefine the function. Rather it simply concluded that the two devices did not perform the agreed function in the same way:

The analysis of the "way" a function is performed necessarily requires the
use of descriptive language that is not contained in the claim itself. Our
precedent reflects this reality. See, e.g., Kemco Sales, 208 F.3d 1352; Ishida,
221 F.3d at 1316-17. In Kemco We simply described the way in which the patented structure closes.3 The same is true here. Sales, for example, the disputed claim recited "a plastic envelope closing means." Id. at 1355. We construed the function "closing" to simply mean "sealing, such that entry or exit is prevented." 208 F.3d at 1361. We found that "both the accused and disclosed structures perform the identical function, which is to close the envelope." Id. at 1365. However, we held that the accused structure performed the function in a different way because, "unlike the disclosed flap, which closes by folding over the envelope, the dual-lip structure closes the accused envelope in a different way by meeting together and binding via the internal adhesive." We therefore affirmed the district court’s conclusion that no reasonable jury could have found that the accused structure infringed. Id. We did not impermissibly import an unclaimed function of "folding" into the closing function when we stated that the patented structure closes "by folding over the envelope." Id. We simply described the way in which the patented structure closes. The same is true here.

Professor Wegner describes the decision as just another example of our disastrous experiment in means claiming. "That no other country from amongst the more than one hundred Paris Convention member states has seen fit to emulate this odd and arcane practice should speak for itself'" he writes. "The time has come for the United States patent reformers to include the simple abolition of 35 USC § 112, ¶ 6, from the United States Code."

A less radical view of the decision might be that the lower court erred simply by using gerund terminology ("holding" and "isolating") terminology to describe structural aspects of the "supporting means." From that point of view, the last sentence in the dictrict court's "way" analysis could have just as easily been written in terms of structural elements and linkages as "The tooth members of the levers surround the inner valve portions, in the desired circular configuration," without referring to the problematic isolating effect.

Of course, whether the "way" analysis should consider structure alone, or some combination of structure and function (as suggested by the dissent), merely sidesteps the real problem of drawing a bright line bewteen structural and functional language. Perhaps a better way would be to eliminate the function-way-result equivalence test from means-plus-function interpretation and simply rely on the prior art to prevent overbroad functional claim language. The could be accomplished relatively easily by redefining the Section 112 "equivalents" of the corresponding structure, material, or acts described in the specification as "any structure, material, or acts for effecting the function described in the specification." Among other things, this approach would allow claiming a structure in purely functional terms and minimize the need to come up with crazy new structural names in order to broadly identify the function(s) associated with complex stuctures.

Here's an example of what I mean using "the spectrum of patent claiming" from structural to functional format for an application that discloses a switch arranged in series with a battery and a lamp:

• A switch connected to a lamp for signalling. (Pure Structure: flashing is just one inherent effect, allows any signalling function)
• A switch for flashing a lamp. (Structure Plus Function: broader structure, but must effect flashing function as signal)
• A signal lamp (Switch/connection structure is inherent to signal function: terms of art often use this format, neither switch, connection nor flash is required to infringe)
• A flasher (Signalling function is inherent to switch/lamp structure: is this a structural or a functional recitation? arguably a genus for signal lamp species, but when not a term of art, such "lexicography" may be confusing as to what structure is required; but avoids M-P-F structural equivalence limits at cost of clarity)
• Means for flashing (Means-Plus-Function: why should it be limited to "equivalent structure" disclosed in specification, should this be narrower than "a flasher?," no need to consider structure - all flashing structures should infringe)
• Flashing a lamp (Pure Function, but must be done with lamp)
• Step for flashing (Step-Plus-Function, no lamp structure required)

More examples of this "spectrum of patent claiming" for other "elements, linkages, and effects" in a variety of technological fields is available on page 7 of http://media.aplf.org/rm/20031120-public/rt2003-11-20.pps.