They admit nobody did this before, so how can it be obvious?
That is the right instinct, and it is exactly the argument §103 exists to overcome. Obviousness is not about whether anyone actually did it. It is about whether a skilled person, with the earlier documents in front of them, would have found it obvious to put them together.
Which is why the examiner’s reason for combining matters so much. A §103 rejection stands on that reason. If the only thing linking the references is your own invention showing the way, the rejection is vulnerable.
What the examiner has to establish
- The scope and content of the prior art — the specific references, identified by number and date.
- Where each element of your claim appears, with a citation to a column, paragraph or figure in a reference.
- A reason the skilled person would have combined those references, articulated in the action rather than assumed.
- A reasonable expectation that the combination would work.
- The level of ordinary skill in the field at the relevant time.
This structure is not a formality. Since KSR v. Teleflex, an examiner may rely on common sense and market forces rather than an explicit teaching to combine, but the reasoning still has to be stated on the record so it can be answered. An action that lists two references and asserts the combination without explaining why is a weaker rejection than one that identifies a genuine reason in the art itself.
The four ways of answering
An element is genuinely missing
If no cited reference discloses a claim element, the combination cannot produce the claim. This is the cleanest answer and the first thing to check, element by element, against the actual text of each reference.
There was no reason to combine
If the references are from unrelated fields, or one teaches away from the other, or the only reason to combine them is hindsight from reading your application, the motivation fails.
The reference does not say what the action says it says
Examiners work at speed across large volumes of art. Reading the cited passage in context, rather than as quoted, sometimes shows it teaches something different.
Amend, deliberately
Adding a limitation the references do not disclose usually works, and permanently narrows the patent. It is a trade, and it should be made knowingly rather than as the first move.
An element is genuinely missing
If no cited reference discloses a claim element, the combination cannot produce the claim. This is the cleanest answer and the first thing to check, element by element, against the actual text of each reference.
There was no reason to combine
If the references are from unrelated fields, or one teaches away from the other, or the only reason to combine them is hindsight from reading your application, the motivation fails.
The reference does not say what the action says it says
Examiners work at speed across large volumes of art. Reading the cited passage in context, rather than as quoted, sometimes shows it teaches something different.
Amend, deliberately
Adding a limitation the references do not disclose usually works, and permanently narrows the patent. It is a trade, and it should be made knowingly rather than as the first move.
Secondary considerations
Evidence of unexpected results, commercial success traceable to the claimed features, long-felt need, or the failure of others can rebut an obviousness case. It has to be real evidence with a nexus to what is claimed, usually in a declaration, and it is more work than an argument — but in the right case it is decisive.
35 U.S.C. 103
KSR International Co. v. Teleflex Inc., 550 U.S. 398 (2007)
MPEP 2141–2145
Three months from the mailing date
Answering the references and ignoring the stated motivation to combine