The doctrine of equivalents, explained

The doctrine of equivalents lets a patent owner prove infringement when an accused product does not meet a claim limitation literally but contains an equivalent of it. Equivalence is judged limitation by limitation, as of the time of infringement, usually by asking whether the accused feature performs substantially the same function in substantially the same way to obtain the same result. The all-elements rule and prosecution history estoppel set its outer limits.

Where the doctrine comes from

Literal infringement requires every limitation of a claim, as construed, to be found in the accused product. That leaves a gap: a competitor could copy an invention and swap one component for a substitute the claim language happens not to cover.

The Supreme Court addressed the gap in Graver Tank & Mfg. Co. v. Linde Air Products Co., 339 U.S. 605 (1950). An accused element is equivalent if it performs substantially the same function in substantially the same way to obtain the same result. The Court also treated known interchangeability as an important factor: whether persons skilled in the art knew that the substitute could replace the claimed element.

Almost fifty years later, Warner-Jenkinson Co. v. Hilton Davis Chemical Co., 520 U.S. 17 (1997) confirmed the doctrine and set how it is applied. Equivalence is measured against individual claim elements, not against the invention as a whole. It is assessed at the time of infringement, not at the time the patent issued, so a substitute that became known only later can still be an equivalent.

Two ways to ask the question

Warner-Jenkinson accepted both of the formulations below. They ask the same question in different words, and a chart can use either or both.

Function, way, result

Three separate questions about one limitation:

1. Function. What does the claimed element do in the claimed combination? Take the answer from the claim and the specification, not from the accused product's marketing.
2. Way. How does it do it: what structure, mechanism or sequence of steps?
3. Result. What does it produce?

The accused feature must match substantially on all three. Most disputes turn on "way". Two features that do the same job with the same outcome often do it differently, and that difference is what the defendant will point to.

Insubstantial differences

This test asks whether the differences between the accused feature and the claimed limitation are insubstantial to a person skilled in the art. Practitioners often present function-way-result in the chart because it breaks the question into parts a reader can check, and use insubstantial differences as a cross-check.

Known interchangeability supports either test. Useful evidence includes textbooks that list the two as alternatives, product documentation that offers both as options, and standards that treat them as substitutes.

The all-elements rule

Because equivalence applies element by element, every limitation needs a counterpart in the accused product, literal or equivalent. An argument that the product as a whole works like the invention cannot make up for a limitation with no counterpart at all. A strong row elsewhere in the chart does not fill a missing one.

A related limit is claim vitiation. An equivalence theory that would read a limitation out of the claim entirely is not permitted, because it does not find an equivalent of the limitation; it removes it. A theory that two actions are equivalent to a claimed "only a single action" would face exactly that objection.

Prosecution history estoppel

Under Festo Corp. v. Shoketsu Kinzoku Kogyo Kabushiki Co., 535 U.S. 722 (2002), a narrowing amendment made for a substantial reason related to patentability raises a presumption that the patentee surrendered equivalents for the amended limitation. The patentee can rebut the presumption by showing one of three things:

• the equivalent was unforeseeable at the time of the amendment;
• the rationale for the amendment was only tangential to the equivalent; or
• some other reason means the patentee could not reasonably have been expected to describe the equivalent.

In practice this means pulling the file history before arguing equivalents for any limitation. For each one, record whether it was added or narrowed by amendment, what the amendment was responding to, and whether any rebuttal ground is realistic. Arguments the applicant made to distinguish prior art can also narrow the range of equivalents, even where no amendment was made, so read the remarks as well as the claim changes.

Other limits worth checking

• Dedication. Subject matter disclosed in the specification but not claimed is generally treated as dedicated to the public. It cannot be recaptured through equivalents.
• Prior art. A range of equivalents cannot extend so far that the claim would cover what was already in the prior art.
• Construction first. The equivalents question arises only after the literal meaning of the claim is settled. If a gap disappears under a different construction, that is a construction dispute, not an equivalents argument. For how limitations should be split before any of this, see how to break a patent claim into limitations.

How equivalents appear in a claim chart

A common approach is to keep the equivalents analysis separate from the literal one, as its own column or its own block under each row:

1. Complete the literal analysis for every limitation first.
2. For each limitation that is not literally met, state the claimed function, way and result, citing the patent, and the accused feature's function, way and result, citing the product evidence.
3. Add evidence of known interchangeability.
4. Record the estoppel check: amended or not, why, and any rebuttal ground.
5. Record the vitiation check: does the theory leave the limitation any meaning?
6. Give the equivalents finding its own label, such as "equivalent", "arguably equivalent" or "not equivalent". Never fold it into "present".

The separation matters because a reader needs to see at a glance how many rows rest on equivalents rather than on literal evidence. A chart where one row in ten relies on equivalents is a different proposition from one where four do. The reading habits in how to read an element-by-element claim chart apply to the equivalents block too: evidence before finding.

A worked example

Take claim 1 of US 5,960,411 A, "Method and system for placing a purchase order via a communications network" (the 1-Click patent), which issued on 28 September 1999 and has expired. Chart it against a hypothetical one-tap ordering app. Assume the literal analysis found two limitations not met.

The two rows show why the analysis has to stay separate. Row 1[b][i] fails on vitiation, and no amount of function or result evidence rescues it. Row 1[d] has a real argument, but it stays provisional until the file history has been read. Whether the fingerprint step even counts as a second "action" is a construction question, and it should be settled before the equivalents column is filled in at all.

Try it

One Transform's Infringement search charts each candidate product limitation by limitation, with a finding of present, partially present or not found and the evidence for every element. The partially present and not found rows are where an equivalents analysis starts, and the evidence already attached to them is the first input to the function, way and result comparison.

Next step: run an Infringement search and review the rows that are not literally met.

This article is general information about patent practice, not legal advice. For advice on a specific matter, consult a registered patent attorney or agent.