There is no statutory requirement that formal logic be adhered
to during examination of patent claims. Examiners and patent
practitioners are free to use a wide variety of discussion and
argument styles and topics, and these tend to be specific to the
subject matter at hand and more informal than reliance on pure,
formal logic. Yet there are times when a formal logic fallacy
is committed in the rejection of a claim, and formal logic tools
can be applied to the specifics to analyze and then formulate a
response to refute the rejection. The following
(fictionalized) example is given in generalized specifics, and
followed by a formal logic discussion.
Suppose the independent claim under examination has, as an intended
point of novelty, claim limitations involving moving an object (if
this is a physical machine) or some specific data (if this is a
processor-based invention) to a specified location responsive to a
condition that the object or the data has some specified property,
and leaving the object or the data where it was if the object or
the data does not have the specified property. The Examiner,
in an Office action rejecting the claims, has set forth a first
reference that shows two possible locations for an object or data,
and moving the object or data from the one location to the other.
The Examiner has set forth a second reference that shows
various conditions and properties the object or the data could
have, and various operations that could be performed on the object
or the data responsive to the object or data either having or not
having the various conditions and properties. The Office
action asserts that the claim limitations of "leaving the
object or data where it was, and moving the object or data from
where it was to the new location" are shown in the first
reference, and the claim limitations of "responsive to the
object or data having or not having the various conditions and
properties" are shown in the second reference, therefore all
of the limitations are shown in the combination of the references.
Can you spot the logical fallacy in this argument?
In terms of formal logic, there is a logical fallacy in the
proposition set forth in the rejection of the claims. The
proposition is of the form of showing A in a first reference, and
"responsive to condition B" in a second reference, with
that assertion that this proves "A is responsive to condition
B". But, neither reference teaches a logical connection
between A and B. The second reference teaches actions, in the
second reference, that are responsive to condition B. The
first reference teaches action A, possibly responsive to conditions
taught in the first reference. But, the conditions in the
first reference are not equivalent to the conditions in the second
reference, and the actions taught in the first reference are not
equivalent to the actions taught in the second reference. The
two teachings cannot be logically combined to prove that A is
responsive to condition B. That this is a logical fallacy can
be shown by reductio ad absurdum, a formal logic proof style.
Suppose that the proposition that A shown in the first
reference, and "responsive to condition B" shown in the
second reference, does show that A is responsive to condition B, is
true. But, while the second reference teaches that action C
is responsive to condition B, the second reference also teaches
that action D is responsive to not condition B, which could be
called action D is responsive to condition E. By the logical
proposition, the combination of the first reference and the second
reference therefore teaches that A is responsive to condition E,
which is equivalent to A is responsive to not condition B.
But, this contradicts the supposition that the combination of
the references teaches that A is responsive to condition B.
The two are in logical contradiction, which now disproves the
proposition, by reduction to a contradiction or reductio ad
absurdum.
In terms of arguments to be made during examination of patent
claims, there is no teaching to combine the references. That
the second reference teaches some other actions responsive to the
condition that the object or the data has or doesn't have some
specified property has no bearing on whether or not the actions of
leaving the object or data where it was, or moving the object or
data from where it was to a new location as taught in the first
reference are to be performed responsive to the conditions in the
second reference. They are independent of one another, and
neither reference teaches the connection between the actions in the
first reference and the conditions in the second reference.
Therefore there is no teaching to combine. Any
Office-cited general-purpose motivation, such as that the two
references are in the same or related fields, are general in
nature, and do not direct or reveal that purported specific
connection between the actions in the first reference and the
conditions in the second reference. A proper combination of
the two references would teach that the actions of the first
reference are performed responsive to the conditions in the first
reference, and the actions in the second reference, unrelated to
those in the first, are performed responsive to the conditions in
the second reference, unrelated to the conditions in the first
reference. Only if one or both references teach a connection
between actions in the first reference and conditions in the second
reference would they be properly combined to show the claim
limitations.
Returning to formal logic, the logical deduction from the proof by
contradiction or reductio ad absurdum is that the combination of
reference one and reference two shows that the actions A shown in
the first reference are not dependent on the conditions B shown in
the second reference, since the (fallacious) logical proposition
could be used to show that action A is responsive to both B and not
B. In other words, condition B is a "don't
care" condition for action A. Formal logic shows that
the two references do not establish a connection between action A
and condition B (or not B). In formal logic terms, according
to the (fallacious) logical proposition and what is shown in the
two references, A is dependent upon (B or not B), which logically
resolves to just A (independent of or not dependent on any state of
condition B).
Thus, the formal logic proves the invalidity of the proposition in
the Office action, and the validity of arguing that the references
do not teach or establish any connection between the actions of the
first reference and the conditions in the second reference.
Practically speaking, where formal logic is a useful tool is
when it is not immediately clear how to argue against a claim
rejection that somehow feels fallacious or just plain wrong.
Formal logic can point to a logical fallacy advanced in the
rejection, which can then point to what specific claim terms and
what specific factual findings to apply and arguments to refute the
rejection. For the patent practitioner, abstracting the
rejection to formal logic terms clears away the specific details of
the subject matter, and lets us look at whether or not the argument
proposition or format itself, advanced in the rejection, is based
on sound reasoning. If the argument itself is employing
fallacious logic, the patent practitioner can then resolve this to
the specific details of the subject matter, and develop a response
strategy with reference to those details. While formal
logical will not necessarily develop every possible response or
argument (see, e.g., the incompleteness proof by Goedel), it can be
used as good guidance and is a useful tool to have in the toolkit.
The above examples are just one application of one part of
formal logic, and there are many more constructs in formal logic
and many more places where it can be applicable in reasoning during
patent claim examination. This is all part of the art of
patenting.
The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.