Brown & Michaels
How do I read a patent? - the Claims
In a printed patent, the claims usually come last. In the USPTO database, they
are first, immediately after the bibliographic information and abstract. In
either case, the claims are usually preceded by the phrase "What is claimed
is..." or simply "I claim...".
The claims are the heart of a patent, in that they define the limits of exactly what the patent does, and does not, cover. That is, the patentee has the right to exclude others from making, using or selling, only those things which are described by the claims. Reading and understanding the claims of a patent is the key to determining if a given product or process infringes the patent.
This is not an easy thing to do, for the following reasons:
This page will attempt to introduce the subject of reading claims, but if your concern about the claims of a patent is more than idle curiosity, it should be understood that claim interpretation is an arcane art best left to patent attorneys, after analysis of the prosecution history, review of recent court cases, and a search of the prior art. In other words, "Professional Driver on Closed Course - Kids, don't try this at home."
Claims 101 - The Basics
Special note on Design Patents and Plant Patents: Since they cover only the appearance of an object, Design Patents have only one claim, which reads, "A design as shown in the figures." Plant patent claims are similar: "A new and distinct Chrysanthemum plant named Balada, as described and illustrated." (from PP8009). The discussion on the rest of this page will deal only with utility patent claims.
A claim is a sentence. As such, it starts with a capital letter and ends with a period. In many cases, claims are typographically broken out in outline form, to make reading the claim easier - although it is not uncommon to find a claim which is presented as a run-on sentence half a page long.
Claims are numbered sequentially, with claim 1 usually being the broadest claim in the patent (but not always), and the sub-parts of a claim are often identified by an outline letter or number. .
Claims are of two basic types: Independent and Dependent.
An independent claim stands alone, for example:
This claim has a preamble ("A motor vehicle..."), the connecting word (or "transitional phrase") "comprising", and a list of four elements - wheels, a transmission, a motor and a fluid supply.
The preamble usually sets out the name of the product being claimed, and perhaps a use for the product. It will also list anything which is not part of the invention, but which is needed to define or supply support for the elements of the claim (for example, the body and chassis of the motor vehicle are included so that we can later on say the wheels, motor and transmission are mounted to it). Normally, the preamble should not be used to limit the claim, unless it is necessary to make sense of the claim.
Sometimes there will be a series of independent claims which differ only by the preamble - for example, if the novelty in the example above was in the fluid supply coupling, the patent might have one claim starting "A motor vehicle...", as above, and then a number of successively narrower claims starting "A power source for a motor vehicle...", "A fluid supply for a motor in a motor vehicle..." and "A coupling for a fluid supply for a motor in a motor vehicle...". In each case, more of the items listed in the elements a-d would be moved into the preamble, so that the last of these claims would read:
The reason for having a series of claims like this is that if the patent owner ever sues for infringement, he wants to be able to use the biggest base for calculating royalties, and the hope is that since motor vehicles cost more than fluid couplings, that the damages will be larger for infringing a claim directed to "a motor vehicle" than one directed to "a coupling".
That said, it seems unlikely that a court would award royalties based on the price of a car if someone merely produces an infringing fluid coupling. At least, if the court does use the cost of the car as a base, it seems likely that they will then use a lower percentage rate than if they based the damages on just the coupling, so that, in the end, the result will be more or less the same. Still, there's little harm in the practice, other than the cost of some extra claims.
By use of the connecting or transition word "comprising" ("open" language), the claim says, "the device has at least these things, but might have other things also." Other connecting words have other meanings - for example, if what was claimed was a motor vehicle "consisting of" the list ("closed" transition language), then the motor vehicle would have that list and nothing else. A car having the listed elements and seats would not infringe.
Notice that each element of the claim is tied into something which went before - the wheels are under the body, supporting the chassis, and so on. Claim elements are supposed to be related, so that each element builds on the preceding elements. Any element which is described as "the ..." should have been introduced previously as "a ..." - this is called "antecedent basis". Thus, the first time we see the transmission it is called "a transmission" (claim 1(b)). Then, in claim 1(c), we can say that the motor is coupled to "the transmission".
Claims to methods would be written in the same way as the apparatus claim above, except that each of the elements would be a step, usually starting with a verb:
A dependent claim, in contrast, only has meaning when combined with a preceding claim. Here are a few dependent claims which depend upon claim 1, above:
Thus, claims 2 and 4 depend on claim 1, and claim 3 depends on claim 2, which in turn depends on claim 1. What this means is that each of these dependent claims cannot stand on its own (claim 3 does not cover "air", as such) but rather must be read with one or more of the earlier claims to make sense. Put another way, each dependent claim has the same coverage as an independent claim which is made up of the claim plus all of the claims upon which it depends, back to an independent claim. So, claim 3 really means,
Each element in a dependent claim must have "antecedent basis", as discussed above for the elements of an independent claim. That is, we couldn't have "the gas" in claim 3, unless there was "a gas" defined in claim 1 or 2.
Claims 2 and 3 narrow claim 1 by limiting what kind of "fluid supply" and "motor" is covered - so, while claim 1 would cover any fluid (say, gasoline) supplying any kind of motor (a gas turbine), claim 3 only covers compressed air supplying a variable-displacement fluid motor. Claim 4 narrows claim 1 by adding another two elements - the steering wheel and seat, so that in order to infringe claim 4 you must have at least six elements, rather than the four needed to infringe claim 1.
Each dependent claim must be narrower than the claim upon which it depends. Thus, it would not be proper claim format to have a dependent claim in this example which says,
This is improper form because claim 1 requires a motor, and claim 5 says there is no motor. Thus claim 5 would be broader than its parent claim. Having a dependent claim like this would allow a product which does not infringe claim 1, because it has no motor, to infringe claim 5.
The claims rules introduced in 2007 (rescinded in 2009) said that a dependent claim must claim an invention which is in the same class (method, apparatus, etc.) as the claim upon which it depends, or it is considered an independent claim. Thus, a claim which reads "A device manufactured by the method of claim 1...", or "A method of chopping wood comprising using the device of claim 1..." would be considered independent, even though it refers back to an earlier claim. While these rules were rescinded, clearly the USPTO does not like this sort of claim and intends to do something about them at some point. It's best to avoid them.
Many foreign countries have a special form of dependent claim, called "multiple dependent claims". A claim of this form might be:
This kind of claim is allowed in the USPTO, but discouraged - there is a large additional fee if there are any multiple dependent claims at all in the application, and each multiple dependent claim counts as more than one claim for the purpose of determining the basic filing fee (in this example, claim 6 would count as five claims). Also, although other countries allow multiple dependent claims to depend on other multiple dependent claims, this form is forbidden in the US. We do not recommend the use of multiple dependent claims.
Special Claim forms
Jepson claims are a special type of claim which are most often used when the invention is an improvement on an existing product. Here is claim 1, written as a single Jepson-style claim:
In the "ordinary" claim form of claim 1, above, the body of the claim lists elements which are both old and new. In this form, everything before the "in which the improvement comprises" is admitted to be prior art for the purposes of this claim, and everything after the phrase are considered to be the novel elements of the invention. Although this kind of claim is not often used in US practice, it can be very effective in separating the old and new parts of an invention.
This style of claim is the preferred form in European practice, where they are called "two-part format" claims, and the transitional phrase is "characterised by..." or "characterised in that..."
Markush groups are a way of writing a claim to cover an element which is selected from a list of elements. For example,
In terms of claim coverage, this would be the same as five dependent claims, each claiming the fluid is one of the gasses, but by using this form we only have one claim. This can save money on the excess claim filing fees if there are more than 20 claims. The risk of a Markush claim form is that if any one of the group is found to be old, the entire claim is invalid. If the claim were written as separate claims, the invalidity of one of the claims would not affect the others.
Note that on August 10, 2007, the USPTO issued a "Notice of Proposed Rulemaking" which proposes changes to how claims presenting alternatives (which includes Markush groups) would be considered for examination and claim-counting purposes. If enacted, the rules would be changed to provide, among other things:
1.75(j) A claim that reads on multiple species by using alternative language must meet the following conditions:(1) The number and presentation of alternatives in the claim does not make the claim difficult to construe;
For more details, see the Federal Register notice (72 FR 44992).
Named after in re Beauregard, the 1995 which approved this form of claim, a "Beauregard Claim" is really a claim to a computer-implemented method, but it is written to claim the computer-readable medium on which the program which is executed to implement the method is stored. The original idea was based on the then-current uncertainty about whether a method which was performed internally in a computer was patentable subject matter. By claiming the method as a physical object (the floppy disk or CDROM upon which the program was written which, if executed, would perform the method), the argument ran that this must be patentable subject matter, since floppy disks are tangible things. Since then, Beauregard claims have gone into and out of style, as the USTPO and the courts struggle with the situation regarding computer implemented methods. The present status of Beauregard claims is unsettled - they are usually accepted by the USPTO, but a recent case (Digital-Vending Services International, LLC v. The University of Phoenix, Inc. (Fed. Cir. 2012), among others) stated what many practitioners have long believed: what really matters is the underlying method - if a method isn't patentable in the first place, it doesn't become patentable because you claim it as a CDROM.
An example of a Beauregard Claim (from patent 6,282,573, the patent involved in the Digital-Vending case) is as follows:
Now that you know the basics of reading claims, we can go into a bit more detail. However, what you need (or want) to know will depend on why you need (or want) to know it. Therefore, we have provided three ways to proceed from here:
<- Back to the Specification
Back to "How to read a patent" main page
400 M & T Bank Building
© Copyright 2004-2012 Brown & Michaels, PC