“Double patenting,” one of the more arcane subjects in patent law, is based on a deceptively simple idea. A patent is a government grant that gives an inventor exclusive rights in his or her invention for a limited period. An inventor should not be allowed to circumvent that time limit by obtaining multiple patents for the same invention or for obvious variations of that invention. If an inventor obtains two patents for the same invention, or for the invention and an obvious variation, at least one of those patents will be invalid.

“Claims” and “continuations” are two concepts that are relevant to double patenting. In patent law, an invention is defined by a patent claim. Most patents have multiple claims. Although each of those claims effectively represents a different invention, this is allowed if those claims are drafted so that they are all sufficiently related to one another. However, double patenting is concerned with the existence of multiple claims for the same invention (or obvious variants) in multiple patents, and not with multiple claims within a single patent .

A patent, as well as the application from which a patent issues, includes a great deal more than claims. In particular, a patent includes a description of how the invention represented by the claims can be implemented. For many technologies, a patent also includes multiple drawings to explain the invention. The description and drawings often include many alternative elements and/or uses, a discussion of the relevant technology, and numerous other things that may not be recited by a claim. It is common for an inventor to file an application and obtain a patent with claims directed to certain aspects of what is set forth in the description and drawings, and to then file a continuation application to obtain a separate patent. The continuation application (and the resulting separate patent, which may also be called a “continuation”) normally has the same description and drawings as the first application and patent (the “parent” application/patent), but has different claims directed to different aspects of what is set forth in the description and drawings.

There are two kinds of double patenting. “Statutory” double patenting bars an inventor from having two patents with the same claim (or with claims that are effectively identical). The prohibition against statutory double patenting arises from 35 U.S.C. § 101, which allows an inventor to “obtain a patent.” Because it requires the same claim in two patents, statutory double patenting is relatively easy to avoid and is fairly uncommon.

“Obviousness-type” double patenting bars an inventor from having a patent with a claim that is obvious over a claim in another of the inventor’s patents. The prohibition against obviousness-type double patenting arises from case-law doctrine created by judges. The principle behind this doctrine is that an inventor should not be able to extend the life of a first patent by obtaining a second patent with a claim to an obvious variation of the invention claimed in the first patent. The doctrine is also designed to protect third parties from harassment by multiple patent owners in connection with the same invention.

The following example helps to explain obviousness-type double patenting. Assume that Jim, an employee of Tempus Timepieces, Ltd., has invented a mechanical clock. Jim’s invention is a system of gears that rotate in response to force from a spring. The gears are selected so that one gear rotates at 60 revolutions per minute (rpm), another rotates at 1 rpm, and another rotates at 1/60 rpm (or 1 revolution per hour). A patent application for Jim’s invention is prepared. That application has numerous drawings and an excruciating level of detail that only a patent lawyer or an insomniac could appreciate. Among the many embodiments and variations included in the description and drawings are the following: a free-standing clock, a clock sized and configured to be fastened to an adult wrist by a strap, clocks with three hands (hour, minute, and second), and clocks with only two hands (hour and minute). Jim assigns his invention and the patent application to Tempus. The application issues as patent A with the following claim:

A1. An apparatus comprising:


a windable spring mounted inside the housing;

an hour hand coupled to an hour gear configured to rotate at 1/60 revolutions per minute (rpm) in response to force from the spring;

a minute hand coupled to a minute gear configured to rotate at 1 rpm in response to force from the spring; and

a second hand coupled to a second gear configured to rotate at 60 rpm in response to force from the spring.

Just before patent A issues, Tempus instructs its patent lawyer to file a continuation application. Tempus’ main competitor is Acme Corp. Shortly after patent A issues, Acme begins selling a wrist watch with no second hand. Upon realizing that Acme’s wristwatch does not infringe patent A because it lacks a second hand, Tempus’ patent lawyer amends the continuation application to include the following claims:

B1. An apparatus comprising:


a windable spring mounted inside the housing;

an hour hand coupled to an hour gear configured to rotate at 1/60 revolutions per minute (rpm) in response to force from the spring; and

a minute hand coupled to a minute gear configured to rotate at 1 rpm in response to force from the spring.

B2. The apparatus of claim B1, further comprising a strap attached to the housing, and wherein the strap and housing are sized for fastening around an adult human wrist.

The U.S. Patent and Trademark Office promptly rejects claims B1 and B2 for obviousness-type double patenting over claim A1. Claim B1 is similar to claim A1, but omits the second gear and the second hand. In general, a claim that simply omits features of another claim will be considered obvious over that other claim. Claim B2 adds strap and size limitations not present in claim A1. Although the USPTO is not allowed to treat the description and drawings of patent A as part of the prior art, it is allowed to consider the prior art in a double patenting analysis. In this case, the examiner finds an historical document showing a picture of Fred Flintstone’s foreman wearing a sun dial on his wrist and using it to tell time. The examiner argues that wrist-borne timepieces were known, and that a person of ordinary skill would thus have had reason to modify the device of claim A1 to achieve the device of claim B2. Tempus is unable to present a credible counterargument. 

Fortunately, U.S. patent law offers a solution. Tempus can overcome the obviousness-type double patenting rejection of claims B1 and B2 by filing a “terminal disclaimer.” In that terminal disclaimer, Tempus agrees that the continuation patent with claims B1 and B2 will expire at the same time as the reference patent (parent patent A in the above example), and that the patent with the terminal disclaimer will only be enforceable if it and the reference patent are commonly owned. Terminal disclaimers are only available to avoid obviousness-type double patenting. As indicated above, however, statutory double patenting is easier to avoid (by slightly changing the claimed subject matter) and is not very common.

So if statutory double patenting is easily avoided and obviousness-type double patenting is easily overcome with a terminal disclaimer, what’s the big deal? Unfortunately, obviousness-type double patenting is not always as easy to detect as the above example suggests. Applicants often use different words for similar elements in claims of different applications, often arrange claim features in different ways, and otherwise draft claims so that similarities are less noticeable. In such situations, an examiner may simply miss the possible obviousness of one claim over another. This can be a more serious problem in large application families that may involve separate examiners for different applications.

If an examiner allows an application with a claim of a first patent that is obvious over a claim of a second patent, a defendant accused of infringing that first patent claim can assert invalidity because of double patenting as a defense. Although a patentee can submit a terminal disclaimer during litigation, this is only available under certain circumstances. If the reference patent (the patent with the claim over which an asserted claim is obvious) has expired, a terminal disclaimer is not available.

Moreover, a terminal disclaimer will not be helpful if the owner of an asserted patent does not also own the reference patent. For example, the original owner of the asserted and reference patents may have sold one of those patents and retained the other patent. As another example, the same inventor may have obtained one of the patents while working for a different employer. Returning to the previous fact pattern, assume that inventor Jim worked for National Time Devices, Inc., before joining Tempus. While at National, Jim developed a clock that used a rubber band instead of a spring. National filed an application for Jim’s rubber band clock and obtained a patent C with the following claim:

C1. An apparatus comprising:

an elongate cabinet having an interior cavity defined therein;

a flattened elastomeric element in the form of a band, the elastomeric element attached to a twistable fixture within the cavity, the elastomeric element being configured to store energy in response to twisting of the twistable fixture and to controllably release said stored energy to turn a drive sprocket;

a first time indicating member attached to a first time cog, wherein the first time cog is positioned within the cavity and is configured to interact with the drive sprocket via multiple intervening cogs and to rotate, in response to a drive force from the drive sprocket, once per hour; and

a second time indicating member attached to a second time cog, wherein the second time cog is positioned within the cavity and is configured to interact with the drive sprocket via the multiple intervening cogs and one or more additional intervening cogs and to rotate, in response to the drive force from the drive sprocket, once per minute.

The application that became patent C was filed before the application that became patent A. Because of its earlier filing date, patent C expires before patent A. Patent C also expires before patent B, which issued on the continuation of the patent A application after Tempus filed a terminal disclaimer over patent A. Moreover, the patent C application was examined by Examiner Sally, while the patent A and patent B applications were examined by Examiner Bob. Because of this, and because of different terminology used in claim C1 relative to claims A1, B1, and B2, Examiner Bob did not notice the similarity of claims A1, B1, and B2 to claim C1.

When Tempus tries to assert claim B1 against Acme, Acme could argue that claim B1 is invalid for obviousness-type double patenting over claim C1. “Housing” (claim B1) is a more generic term for “cabinet” (claim C1). “Gear” (claim B1) is generally synonymous with “cog” (claim C1), and there is no apparent difference between a “hand” (claim B1) and an “indicating member” (claim C1). A “spring” (claim B1) is different from a rubber band (i.e., a “flattened elastomeric element in the form of a band,” as recited in claim C1). However, a spring and a rubber band are known equivalents for at least some purposes and are used in similar ways in claims B1 and C1. Claim C1 recites more details than claim B1, but the features of claim B1 and the relationships between those features are nonetheless present in claim C1.

If presented with the above argument, Tempus would need to show how a spring is not an obvious replacement for a rubber band, or otherwise show an aspect of claim B1 to be a non-obvious change from claim C1. If Tempus is unable to do so, claim B1 would likely be found invalid. Tempus would not be able to avoid invalidation of claim B1 with a terminal disclaimer over patent C, as Tempus does not own patent C. Acme would thus be able to invalidate claim B1 based on claim C1, even though patent C may not be prior art to claim B1. For example, assume Jim was the sole inventor named in patent C and in patent B and that patent C was not issued (or otherwise published) more than a year before the application for patent A (the parent of patent B). Under those facts, which are quite plausible, patent C may not be prior art to patent B. Nevertheless, a claim in patent C can still be used to invalidate claim B1.

The above discussion only includes some of the problems that can result from double patenting. There are numerous other situations in which double patenting can raise issues. Accordingly, and regardless of whether it is raised by an examiner during prosecution of a patent application, double patenting should always be a consideration for a patent applicant, patent owner, or a party accused of infringement.