Independent claim and dependent claim

The scope of a patent right is defined by an item called a "claim".
For example, assume patent right P1, which contains the following two claims.

[Claim 1]
A device that produces a suitable amount of drink when button 1 is pressed.
[Claim 2]
The device according to claim1, when button 2 is pressed, also brings out ice.


Claim 1 means "Do not make or sell without permission a device that produces a suitable amount of drink when a button is pressed (feature A)”.
Claim 2 means "Do not make or sell without permission a device that produces a suitable amount of drink when one button is pressed and also provides ice when another button is pressed (feature A + feature B)”.

Claim 1 is called an "Independent Claim", and a claim, such as claim 2 that defines the scope of rights by relying on another claim (in this case, claim 1) is called a "Dependent Claim".

A product that infringes claim 1 does not necessarily infringe claim 2 (e.g., a cup vending machine that dispenses drinks but not ice (feature A only)). However, a product that infringes claim 2 will always also infringe claim 1.

The scope of rights in the independent claim is always broader than that in the dependent claim (with rare exceptions).

This raises the question of whether a dependent claim, which is less powerful than an independent claim, is unnecessary.

Reasons for making dependent claims

To estimate the possibility of patenting

Even if the applicant believes that feature A is patentable, feature A alone may not be granted a patent. However, if feature B is added to feature A, the possibility of patentability will have a higher chance of being granted. This is because a device that provides ice in addition to a drink is considered more convenient and more ingenious than a device that simply provides a drink.

If only claim 1 (independent claim: feature A) is described, it will be difficult to determine the pass line for how much claim 1 needs to be amended to be patentable if the patentability of claim 1 is rejected.

If claim 2 (dependent claim: feature A + feature B) is listed and this claim 2 is found to be patentable, you can be confident that you can obtain a patent by giving up claim 1 and compromising on claim 2.

It is possible to make a claim that is narrower than claim 1 (non-patentable) and broader than claim 2 (patentable). For example, it may be possible to patent a device at the level of "a device that provides a suitable amount of drink when button 1 is pressed, but also provides materials (ice, hot water, etc.) to adjust the temperature or concentration of this drink when button 2 is pressed," without having to specify even "ice".

To avoid missing points that deserve to be patented

If it is an idea described in the patent specification, it can be moved up to a claim at a later date, or it can be used to amend an existing claim ("modification/additional matter").

However, the searching for ideas in the patent specification that could be made into claims at a later date can be a bothersome process. It is also possible to miss something.
Rather than spending time on this search process, it makes it easier later to list any ideas recognized at the time of filing as dependent claims without omission. Even if it costs a little more, it is better to do so if it makes the work more efficient.

Also, describing an idea in the patent specification does not necessarily mean that it is written precisely enough to make it rise to the level of a claim. If ideas are listed as dependent claims in the patent application, the ideas will be explained in detail in the patent specification in anticipation of examination, which may make it easier to obtain a patent.

Reasons for leaving dependent claims

A patent right may be invalid even after it is granted.
For example, suppose that patent right P1 contains only claim 1 (independent claim: feature A). If at a later date, a reference D1 is newly discovered that discloses feature A (but does not describe feature B), then this patent right P1 will be invalidated.

However, if claim 2 (dependent claim: feature A + feature B) is included in patent right P1, claim 2 will survive even if claim 1 is invalidated. With dependent claims, the risk of invalidation of the entire patent right P1 can be reduced.

As an example, suppose company X sells product Y with features A and B.

(Case 1)
Suppose that patent P1 lists only claim 1 (independent claim: feature A). Since product Y contains feature A, it infringes patent P1.
Suppose that company X finds reference D1 (prior art) that discloses feature A. Then the only claim 1 of patent right P1 becomes invalid.
Since patent P1 is invalidated, product Y is no longer an infringement of the patent.

(Case 2)
Suppose patent right P1 lists claim 1 (independent claim: feature A) and claim 2 (dependent claim: feature A + feature B).
If you can find reference D1 (feature A), claim 1 of patent right P1 becomes invalid, but claim 2 (feature A + feature B) survives. Product Y falls within the scope of claim 2, so it is still a patent infringement.

It is more difficult to find a reference that discloses both feature A and feature B than a reference that discloses only feature A, so patent right P1 is more likely to survive.

If claim 1 may be able to be invalidated, but claim 2 seems difficult to invalidate, it may discourage Company X's motivation to fight back (i.e., to search for prior art).

(Case 3)
Suppose that claim 1 (independent claim: feature A) and claim 2 (dependent claim: feature A + feature C) are listed in patent right P1. Claim 1 is invalidated by reference D1 (feature A), but claim 2 (feature A + feature C) survives.

Since product Y does not contain feature C, patent right P1 survives but is no longer infringement. In case 3, claim 2 (the dependent claim) is " missing" as far as product Y is concerned.

When making dependent claims, it is important to ability to imagine the features that a product containing feature A would want to implement. In the above case, feature B was meaningful as a dependent claim as far as product Y was concerned, but feature C was meaningless.

The cost performance of dependent claims is determined in light of the above. A dependent claim without a clear rationale has no reason to exist.

Dependent claims for divisional applications

Dependent claims are also useful in considering the possibility of filing a divisional application.
For example, assume the following
Claim 1 (independent claim) = A
Claim 2 (dependent claim) = A + B (= claim 1 + B)
Claim 3 (dependent claim) = A + B + C (= claim 2 + C)
Claim 4 (dependent claim) = A + B + D (= claim 2 + D)

Suppose that, as a result of the examination, claims 1 and 2 are rejected, but claims 3 and 4 are determined to be patentable.

In this case,
Claim 1 (independent term) = A + B + C
Claim 2 (independent clause) = A + B + D
then the patent will be granted after the amendment.

The fact that claims 3 and 4 are allowed can be interpret as that features C and D are patentable. Therefore, it may be possible to patent the claims "A+C" and "A+D".
It may be possible to patent "B+C," "B+D," "C+D," or just "C" or "D" alone.

Making dependent claims makes it easier to determine whether, once a patent is granted, another patent can be obtained by filing a divisional application.

In particular, patent specifications for software inventions are often studded with small ideas, and it is often the case that additional ideas that are different from the baseline idea that was originally targeted turn out to be rather more valuable.

There are many situations where the patent specification of another company describes the features of our product, but it was not claimed, so we were saved.

Even after obtaining a patent, it is important to consider whether it is possible to obtain additional patents from different angles.