Graham Test for Obviousness


Graham 4-part test for obviousness

During patent prosecution, the claims are often rejected by the examiner as being “obvious” in view of the prior art.  In addition to being new, an invention must be not obvious in order to be patented.  The test for nonobviousness is set forth in a 1966 Supreme Court case called Graham v. John Deere.  Subsequent cases have affirmed that the Graham factual inquires must be used for evaluating the obviousness of a claim under 35 U.S.C. 103.  This mandatory 4-part test provides the objective framework for evaluating obviousness.  The four parts to this test are as follows:

  1. Determining the scope and contents of the art when the invention was made;
  2. Ascertaining the differences between that art and the claim(s) at issue;
  3. Resolving the level of ordinary skill in the pertinent art when the invention was made; and
  4. Considering objective evidence present in the application indicating obviousness or non-obviousness when the invention was made.

The Graham Test is described in detail in the Manual of Patent Examining Procedure (MPEP) at section 2141 (II).

The Graham Test appears deceptively simple.

At first glance, the Graham test appears easy.  The examiner studies the prior art.  Next, the examiner determines how the invention differs from it.  Then, the examiner determines whether an ordinary artisan would have found the invention obvious despite that prior art.
While this test may appear easy, applying the Graham Test often proves extremely difficult.  These difficulties arise from problems viewing the state of the art at the time the invention was made.  Inventors should draft patent applications at the time the invention is made.  But, those applications are examined several years later.  Accordingly, by the time the applications are examined, the public may consider them to be old news.  With the benefit of the inventor’s disclosure, a patent examiner may not appreciate the genius involved in creating the invention from scratch.
The answer to a problem often seems obvious after looking at the solution.  (Consider picture search puzzles and word search puzzles where the answer leaps off the page after seeing it the for the first time).  A patent examiner necessarily approaches a patent application by looking at the solution first.  Then, that examiner must attempt to ignore that understanding and appreciate how the problem would have appeared just before the invention was made.

The Graham Test attempts to curb hindsight.

The Graham Test attempts to ensure that the examiner evaluates obviousness “at the time the invention was made” and not at the time the invention is examined.  The goal of the Graham Test is to prevent the examiner from using the inventor’s own disclosure as a blueprint (or roadmap) for piecing together the prior art in a manner that makes the invention appear obvious.
Practically speaking, hindsight cannot be prevented.  Accordingly, it becomes extremely important for inventors to use legal arguments to protect themselves against losing property on account of a patent examiner’s hindsight.  During patent prosecution, the applicant will inevitably argue with the examiner over whether the invention would have been obvious.  Those arguments will often converge on the fact that the examiner is approaching the problems solved by the inventor with the benefit of the inventor’s own disclosure.  This leaves the inventor with the task of defending the intellectual contribution.  Fortunately, given the lengthy historical development of the obviousness doctrine, a patent applicant has many opportunities for preventing the examiner from drawing conclusions based on impermissible hindsight.  Nevertheless, all off these opportunities will eventually follow the above listed 4-part Graham test.  Accordingly, while the Graham test may appear simple, it should be regarded as one of the patent applicant’s most powerful tools.

One Response to “Graham Test for Obviousness”

  1. David says:

    Thanks very much Andrew, for explaining the Graham Test.
    However, regarding step 4:
    “Considering objective evidence *present in the application* indicating obviousness or non-obviousness when the invention was made.”
    Is that evidence really to be present *in the application*??
    The Graham Test then would go like so:
    1. Determine the the (closest) prior art document [D1)
    2. Determine the difference between the claim(s) and D1
    3. Discuss the level of ordinary skill in the pertinent art
    4. Look for evidence present *in the application* [!] that makes the difference of step 2 obvious.
    Hence, if the application *itself* [!] does not contain anything that makes the difference between the claim(s) and D1 obvious, there is inventive step and the claim is patentable — correct?
    Thank you very much for any further remarks to possibly clear this up!

Leave a Reply

Your email address will not be published. Required fields are marked *