On May 4, the USPTO issued a new memorandum for patent examiners, “Formulating a Subject Matter Eligibility Rejection and Evaluating the Applicant's Response to a Subject Matter Eligibility Rejection” (“Examiner Instructions”)  along with a new set of five example claims, this time in the life sciences and chemistry arts.   The Examiner Instructions are a positive step forward in refining the examination process, but leave open many questions.  

The Examiner Instructions are primarily procedural in nature—how the examiners are to draft § 101 rejections and respond to arguments—rather than addressing substantive issues such as the scope of abstract ideas, laws of nature or natural products  This comes, I believe, in response to many of the public comments to the Interim Guidance and the July Update that focused on having examiners provide more complete and thorough rejections.  There is a widespread perception, backed up by real evidence, that a large percentage of examiners issue boilerplate § 101 rejections that simply identify some alleged abstract idea, and then formulaically assert that there is no inventive concept because all of the claims limitations recite well-known, routine and conventional steps or the functions of a generic computer.  

The Examiner Instructions address this concern in a number of ways.  First, to discourage the boilerplate approach of simply rejecting an entire block of claims in one fell swoop, the Instructions require the examiner to address each claim individually (all bold and underlining in the original):

After determining what the applicant invented and establishing the broadest reasonable interpretation of the claimed invention, the eligibility of each claim should be evaluated as a whole using the two-step analysis detailed in the Interim Eligibility Guidance.

When making the rejection, the Office action must provide an explanation as to why each claim is unpatentable, which must be sufficiently clear and specific to provide applicant sufficient notice of the reasons for ineligibility and enable the applicant to effectively respond.

There is emphasis that each claim must be analyzed at the level of specifics, not hand-waving generalities:

identify the judicial exception by referring to what is recited (i.e., set forth or described) in the claim and explain why it is considered an exception;

identify any additional elements (specifically point to claim features/limitations/steps) recited in the claim beyond the identified judicial exception; and

explain the reason(s) that the additional elements taken individually, and also taken as a combination, do not result in the claim as a whole amounting to significantly more than the judicial exception.

In regards to Step 2A, identification of the judicial exception, the Examiner Instructions hit a keynote that is regularly ignored by examiners: namely, that they must both 1) identify the judicial exception and 2) explain why the claims limitations are considered as such:

The rejection must identify the specific claim limitations and explain why those claim limitations set forth a judicial exception (e.g., an abstract idea). Where the claim describes, but does not expressly set forth, the judicial exception, the rejection must also explain what subject matter those limitations describe, and why the described subject matter is a judicial exception.

This statement is interesting because it draws for the first time a distinction between a claim that “sets forth” and one that “describes” a judicial exception, and how examiners are expected to address this distinction.  

These two steps are emphasized for every type of judicial exception:

When the examiner has determined the claim recites an abstract idea, the rejection should identify the abstract idea as it is recited (i.e., set forth or described) in the claim, and explain why it corresponds to a concept that the courts have identified as an abstract idea.

When the examiner has determined the claim recites a law of nature or a natural phenomenon, the rejection should identify the law of nature or natural phenomenon as it is recited (i.e., set forth or described) in the claim and explain using a reasoned rationale why it is considered a law of nature or natural phenomenon.

When the examiner has determined the claim recites a product of nature, the rejection should identify the exception as it is recited (i.e., set forth or described) in the claim, and explain using a reasoned rationale why the product does not have markedly different characteristics from its naturally occurring counterpart in its natural state.

Ideally, this should prevent two common types of erroneous Step 2A rejections.  Some examiners simply gloss the entire claim, treating it as a very broad category, completely divorced from the context and details.  Other examiners cannot figure out what the generalization should be, so they quote the entire claim and assert that it’s an abstract idea.  In neither case is any real explanation given as to why the claim is abstract.

When “explaining” why something is abstract idea, the examiner typically cites Alice or Bilski, as well as buySafe, Smartgene or Planet Bingo, often when the claims have little similarity to the subject matter of those cases.  The Examiner Instructions note that citation of cases is a “best practice,” but provide an important caution:

Examiners should be familiar with any cited decision relied upon in making or maintaining a rejection to ensure that the rejection is reasonably tied to the facts of the case and to avoid relying upon language taken out of context.

This statement appears to be a response to public comments that examiners routinely take language out of context from cited cases in support of their rejections. To be fair, the examiners only do this because the USPTO itself did this in the Interim Guidance by glossing these decisions very broadly.  After all, the vast majority of examiners are not reading Federal Circuit cases—they are simply quoting from the Interim Guidance when making a rejection.

The best (worst?) example of this is SmartGene, Inc. v Advanced Biological Labs. In the Interim Guidance, the USPTO used the claim in this case as an example of an abstract idea, glossing it simply as “comparing new and stored information and using rules to identify options.”  This is precisely how the case is quoted by examiners.  The problem is two-fold. First, Smartgene is a non-precedential case—on that basis alone, the case should not be relied upon by examiners or even mentioned in the Interim Guidance.

Second, the USPTO’s gloss is a significant overreach. The SmartGene court stated that the

[w]hatever the boundaries of the ‘abstract ideas’ category, the claim at issue here involves a mental process excluded from section 101: the mental steps of comparing new and stored information and using rules to identify medical options. (emphasis added). 

The court itself unambiguously limited the scope of its decision: “[o]ur ruling is limited to the circumstances presented here, in which every step is a familiar part of the conscious process that doctors can and do perform in their heads.”  These two statements make it clear that this case was narrowly decided on the facts and the specific claims.   Nonetheless, the USPTO ignored this express statement limiting the holding, as well as the express qualifications of "the mental steps of" and "to identify medical options" and improperly generalized the abstract idea as simply “comparing new and stored information and using rules to identify options.”  In essence, the USPTO created a new "category" abstract ideas out of a very specific and narrowly tailored non-precedential opinion.

The impact of this manipulation has been dramatic. SmartGene has been cited over 8,8801 times by examiners since it was first mentioned in the Interim Guidance—the vast majority of cases having nothing whatsoever to with the medical decision-making as in SmartGene.  See for example, “Method for Calculating Feed Value of Alfalfa Hay Using Information Available at the Time of Baling” (13/573,461), and “Method of Operating an Internal Combustion Engine (14/138,367).  Most amazingly, even though SmartGene is non-precedential and expressly limited to its facts, since the July Update last year, SmartGene has been cited more times (7315) than Mayo (4703).  That's hard evidence that the case is being relied upon far beyond it's very limited holding.

Of course, this is entirely expected, since it hard to imagine a data processing method that cannot be characterized as “comparing new and stored information and using rules to identify options.”  This broad generalization covers from the most simple case of storing a single number X and comparing that number with a new value Y (“If Y > X then X=Y”), to the most complex data operations using hundreds, or even millions of input values, and any number of computations. If taken literally, this statement can be used to find any and every claim to data processing an abstract idea.  

SmartGene is not the only non-precedential case relied up on by examiners.  Here are further statistics showing the frequency of citations in office actions.

  • Cyberfone Systems, cited more than 6,600 times
  • Planet Bingo,cited more than 2,870 times
  • Gametek, cited more than 520 times
  • Dietgoal, cited more than 440 times
  • Fuzzysharp, cited more than 300 times

This reliance on non-precedential cases in part responsible for a second serious problem: the creation of new categories of abstract ideas beyond what the courts have identified. Hence the Examiner Instructions state:

Examiners should not go beyond those concepts that are similar to what the courts have identified as abstract ideas.

and

If applicant challenges the identification of an abstract idea that was based on a court case and the challenge is not persuasive, an appropriate response would be an explanation as to why the abstract idea identified in the claim is similar to the concept in the cited case. If the original rejection did not identify a Supreme Court or Federal Circuit decision in which a similar abstract idea was found and applicant challenges identification of the abstract idea, the examiner would need to point to a case in which a similar abstract idea was identified and explain why the abstract idea recited in the claim corresponds to the abstract idea identified in the case to maintain the rejection.

Prosecutors regularly face rejections in which the examiners push the envelope of judicial exceptions by identifying something as abstract when there has been no precedential Federal Circuit case on point.  In one application, we argued this point and the examiner replied that she was not limited to existing examples from the courts, arguing that how else would the list of abstract ideas be expanded?  This examiner did not appreciate that the exceptions are judicial, not administrative, in nature: it is up to the courts, not the examiners to expand the list. The limiting statement in the Examiner Instructions should restrain these more aggressive attempts to enlarge the scope of abstract ideas.  

The Examiner Instructions say that appropriate response to the applicant challenging the abstract idea is to find a case in support of the rejection.  This is incomplete, as there is another obvious option: withdraw the rejection.  If there is no precedential case with an abstract idea that is similar to the claims, then the rejection should be withdrawn.  Of course, that devolves into a question of what is similar, and therein lies a world of differences between what examiners and applicants think.  Any claim that has any hint of a commercial flavor is said by examiners to be “not meaningfully different” from Alice or Bilski, regardless of how different in fact it is: to examiners, hedging is the same as payment processing is the same as advertising is the same as insurance pricing is the same as inventory management and so on.  The Alice test may be well call Death by Analogy.   To those of skill in the art–our clients in particular–the difference between these fields is blindingly clear.  

So while the Examiner Instructions do caution examiners here on how to set forth a rejection under Step 2A, they do not remedy the underlying problem: the inclusion of non-precedential cases in the Eligibility Guidance. Either the cases should be removed, or the gloss of the abstract ideas in those cases should be–as the Examiner Instructions state–"reasonably tied to the facts of the case and to avoid relying upon language taken out of context."   Further, the examiners should be further instructed that any cited case has to be similar not just in a general sense, but similar in context and content as well.

In future posts I’ll review the other aspects of the Examiner Instructions dealing with Step 2B and the claim examples. 

Footnotes:

1.  The statistics here regarding the number of citations for various Federal Circuit cases come from full text searches of office actions since January 24, 2014 (the publication date of SmartGene).  The searches were conducted on May 6, 2016 using Patent Advisor.  All numbers are cited are minimum counts, since Patent Advisor only has access to published applications in PAIR.  Published applications account for approximately 60% of all applications filed with the USPTO. Hence the actual number of citations of these cases is likely to be 66% higher; i.e., it's likely that SmartGene has been cited about 14,800 times by now.