Skip to content

The Power of the Pen: The Role of Examiners’ Discretion in Patent Examination

The Power of the Pen: The Role of Examiners’ Discretion in Patent Examination published on 7 Comments on The Power of the Pen: The Role of Examiners’ Discretion in Patent Examination

Imagine a federal taxation scheme where your taxes are determined by an IRS agent, who is empowered to use “common sense” in setting your taxes at a “fair” rate. Or a Transportation Security Authority where security guards are allowed to admit or exclude passengers from flights based on whether or not they look “suspicious.” Or a Bureau of Motor Vehicles where your driving rights are issued or withheld based on whether or not the officer thinks you’re a “safe” driver.

Government agencies that allocate legal rights are never based on such discretionary mandates, for a large number of plain and obvious reasons. Yet, recent trends in patent law have elevated examiners’ discretion to dominate the formerly rule-based, evidentiary process – such that the outcome is primarily driven by the personal philosophy and temperament of the patent examiner.

Many practitioners have noted the existence of this trend, and some of its effects. The purpose of this article is to explore the sources of such discretion – the particular opportunities in the examination process that are amenable to arbitrary decisions – and that are routinely exploited to compel the outcome toward the examiner’s desired result.

In the Patent Act of 1952, U.S. patent examination was formulated as an evidence-driven process. The patentability decision was centralized in 35 U.S.C. 102-103, as a comparison of the claimed technique with the text and substance of specific prior art references. Additionally, today’s U.S. patent examination process is constrained by an increasingly voluminous set of rules and policies. 1

Of course, these facets of U.S. patent examination are intended to confer consistency upon the examination process – to ensure that if you give the same application to examiner (A) and examiner (B), the examination processes will reach the approximately the same result. And for some aspects of patent examination that are measurable, such as docketing, the USPTO closely monitors its metrics and places great stock in the results, with announcements such as a “Final Disposition Compliance Rate” of 96%.

Despite these constraints, per-art-unit examination statistics reveal wide discrepancies in the results of different art units. Much has been written about the extreme variance in allowance rates by examiner, art unit, and technology center.

These observations are clearly at odds: How can an examination system that is based on objective evidence, and that is regulated by an intricate administrative apparatus, produce such vastly divergent results? The rules of patent examination are highly specific about procedure, but much less specific about substance. Many process-oriented rules are subject to interpretation. Examiners utilize these loopholes as they wish. Also, the violently shifting court cases about patent law have left the USPTO in the unenviable position of preparing rules that correctly apply the courts’ holdings; and absent any centralized guidance, art units and examiners are formulating their own policies.

To this end, examiners find opportunities to apply their discretion to the examination process in the following circumstances:

  1. Prior Art Search

    An examiner rejects a patent application under 35 USC 102, and the applicant provides arguments and/or amendments that distinguish the claims from the reference. The claims are allowable over all of the prior art of record. The examiner’s next step is based on a critical question: How much more must the examiner search? 35 USC, 37 CFR, and the MPEP, together, provide no rules whatsoever about this determination. It is completely within the examiner’s discretion.

    Accordingly, if you hand the same application to three different examiners:

    • Examiner #1 may opt to send a first-action allowance.
    • Examiner #2 may perform some searching, reject the application over the closest references in the first search, and then allow the application once the applicant has traversed those references.
    • Examiner #3 may perform some searching, reject the application over references (A + B) in the first search; perform a second search and reject the application under new references (C + D); perform a third search and reject the applications under new references (E + F)… For some examiners, there may literally be no end to the number of searches performed, and the application may never be allowed.

    Notably, according to the rules of U.S. patent examination, all three examiners are correct, because each case was properly allowed or rejected based on the prior art known to the examiner. As long as each rejection and allowance is properly based on the references of record, the USPTO approves of any of these three results. 2

  2. Broadest Reasonable Interpretation

    Does the term “network,” in the field of computers, include two devices talking together directly over a cable, or a computer talking to itself through a network interface or stack?

    Does the term “surface” require a tangible physical structure, or can it be an imaginary boundary existing in a user’s mind, like a perceived area of “personal space,” or by an intangible physical property, like the “surface” of a cloud?

    Is the term “camera” limited to devices that capture images, or can it also include a microphone that captures sound, or a photosensor that senses (but does not capture or store) a light level?

    These types of questions raises questions of the “broadest reasonable interpretation” of the claim language. 3 Nothing in the patent act places any limit on the extent to which an examiner may construe the words of a patent claim beyond their intended context and plain meaning.

    A frequent complaint about the patent system is that claim language is abstruse – that inventors can’t even recognize their inventions in the allowed claim language. But the examiner’s “broadest reasonable interpretation” standard is a direct cause. The original claim may have provided:

    A golf club comprising: a shaft connected to a club head at a 40-degree angle.

    …but during prosecution, the examiner rejected the original claims by citing a fishing pole, an umbrella, a snow shovel, etc., and required the applicant to provide explicit claim limitations in order to exclude these unrelated implements from the “broadest reasonable interpretation” of the claim, until arriving at length upon the allowed claim:

    A golf club comprising: a substantially cylindrical shaft oriented along a shaft axis and having a hollow shaft thickness not exceeding approximately 0.5 millimeters and a terminating in a club head comprising a forward club head surface that is rigidly oriented at a 40-degree inclination angle with respect to the shaft axis.

  3. Obviousness Combinations

    Before the landmark case of KSR Int’l v. Teleflex, Inc., an examiner could only combine references to present an obviousness rejection if there was some objective reason to do so: a “teaching, suggestion, or motivation” demonstrated in the art to combine the references. In KSR, the Supreme Court held that requiring such an explicit statement or suggestion to combine references would “confine” the obviousness determination to “a formalistic conception of the words teaching, suggestion and motivation,” and encouraged the use of “common sense.”

    And in applying KSR, the USPTO shifted from an obviousness standard that required a “teaching, suggestion, or motivation,” to… no standard at all. An examiner who rejects a claim by combining twelve references in completely unrelated fields is in full compliance with the standard of obviousness in patent examination. 4

  4. “Abstract”

    The Alice Corp. v. CLS Bank case presents the latest area of expanded deference to examiners’ discretion, where an examiner can dissect a claim into its inventive concept (to be called an “abstract concept”) and its implementation (to be called “routine” or “conventional”), and then simply discard the dissected claim.

    As I have noted, this decision provides a roadmap for rejecting any claim – irrespective of the area of technology, the novelty of the invention, or the clarity or specificity of the claim language. It can simply be dissected and discarded. We are seeing only the tip of the iceberg in terms of the application of Alice to our body of patents, and already it is growing out of control.

  5. Content of Office Actions

    The MPEP provides specific instructions about the content of office actions, requiring examiners to “address all patent owner responses to previous actions,” to “respond to the substance of each argument raised by the patent owner and requester,” and to specify “the grounds of rejection to such an extent that the patent owner may readily judge the advisability of an appeal.”

    In theory, these clear instructions should compel office actions that are clear, specific, and complete. In practice, these requirements are meaningless.

    Some examiners issue office actions that clearly address the substance of every single claim, in nearly painful detail; extensively discuss the examiner’s interpretation of the references, and analogies with the claimed techniques; and provide full, point-by-point responses to the applicant’s arguments.

    Other examiners issue office actions that provide a blanket rejection of the claims under huge swaths of a reference (“claim 1 is rejected under the Woo reference (see paragraphs 2 through 46)”). The applicants’ arguments are acknowledged, but dismissed as “not persuasive, because” (copy-and-paste the same rejection).

    These latter office actions entirely fail to inform the applicant of the examiner’s reasoning, and strongly suggest that the examiner has not bothered to consider the arguments. But these tactics are widespread, and are apparently tolerated by supervisors. There is no mechanism by which an applicant can compel a more detailed explanation or consideration by the examiner.

  6. Restriction Requirements

    In keeping with the requirement that each patent cover only one invention, 37 CFR 1.142 requires that “if two or more independent and distinct inventions are claimed in a single application, the examiner will require the applicant to elect an invention to which the claims will be restricted.”

    However, the USPTO provides no centralized guidance about restriction practice; rather, the actual details of restriction practice are instead formulated by technology center and art unit, which guarantees variable and inconsistent standards.

    Unsurprisingly, the result is variance and inconsistency. Restrictions requirements are issued for claims that are identical but for the statutory class; for different sides of the same process (e.g., a communication protocol used by a server and a client may be restricted between claims directed to the server, and claims directed to the client); for sets of dependent claims that the examiner just doesn’t want to bother examining; for claim styles that examiners don’t like. The lack of centralized guidance about the proper application of restriction requirements gives examiners the discretion to constrain the claims to a more limited set, simply by claiming that conducting a complete examination is “a burden.”

  7. Claim Type Characterization

    Over the years, the courts and USPTO have crafted certain rules for particular types of claims. Often, these rules become so onerous and restrictive as to strangle an entire claim type. 5 This tactic is why means-plus-function claims are no longer used: although explicitly authorized by 35 USC 112(f), the means-plus-function claim has been subjected to such onerous claim construction requirements by the courts as to damage it beyond use.

    The desolation around these condemned claim styles has given examiners an opportunity to strangle undesirable claims by dragging them into one of these claim styles. Whereas a claim used to be construed as invoking means-plus-function only if those words were explicitly used – e.g., “compression means for compressing the data” – the 112p6 claim style has been expanded to include a much larger scope of claims. And the case of In re Nuijten, identifying a nebulous class of non-patentable subject matter known as “signals,” has led to seven years of shifting patent policy by the patent office over what types of claims do or don’t include a “signal,” providing more opportunities for examiners to create trouble for undesirable claims.

  8. The “RCE Hole”

    The USPTO is extremely deadline-conscious, and has established strict rules requiring examiners to issue office actions within three months of receiving a reply, in order to reduce pendency and the backlog of unexamined applications. These rules are strictly enforced: examiners are financially rewarded with bonuses for maintaining a high compliance with deadlines, so allowing a case to languish past its deadline has a direct financial penalty on the examiner.

    However, there is one loophole in this process: Requests for Continued Examination. When an RCE is filed, the application is placed on the examiner’s “special docket” – where it may languish without attention for several years. This process, called the “RCE hole,” allows an examiner to simply disregard an application, and take no action on it, for a completely undefined period. And since filing an RCE cuts off the patent term extension period, every day that the application spends in the “RCE hole” is a day of lost patent term – indeed, the entire term of the patent might vanish before the patent issues.

    This loophole is not accidental. The USPTO’s formal announcement of the docketing change that created the “RCE hole” provides:

    Despite the placement of RCEs on the examiner’s Special New docket, the USPTO believes these initiatives and the new examiner production system will work together to reduce the overall pendency of applications. Applicants seeking early allowance of an application can help eliminate the need for the filing of an RCE by addressing all issues in the application early in the prosecution. If an RCE is needed, cooperative efforts between the examiner and the applicant will go a long way toward obtaining early action of the RCE. Of course, for those applicants who do not desire fast action of their applications, the placement of RCEs on the examiner’s Special New docket will afford them the flexibility to better control the rate at which their cases are handled by the USPTO.

    The tone of the USPTO’s own description of the purpose of the “RCE hole” speaks volumes. On its face, this announcement classifies applicants into two categories: those who “cooperate” with the examiner… and those who “do not desire fast action of their applications.”

  9. Written Description Requirements

    For many years, patent examiners have used the label of “patent informalities” to cover a wide range of minor objections to a patent application:

    • Unusual antecedent basis conclusions
    • Unusual mechanics of non-substantive claim elements such as “wherein,” “respective,” and “comprising”
    • Objections to the title as not “descriptive”
    • Objections to the abstract as not “descriptive” or “narrative”
    • Objections to the figures based solely on examiner’s preferences, such as the directions of arrows or the positions of elements
    • Objections to trivial typographical or clerical errors

    Such objections used to be rather trivial, and were resolved wither by trivial amendments, simple statements on the record, or appeals to the supervisor or the ombudsman’s office.

    However, examiners have also begun asserting “written description” rejections in a more aggressive manner:

    • Rejecting claims under 35 USC 112, first paragraph, as not adequately supported by the specification – despite claim language that plainly matches an example presented in the specification or depicted in a figure
    • Rejecting claims under 35 USC 112, second paragraph, as including “indefinite” claim terms that are readily known in the field of technology (e.g., “the term ‘data store’ is indefinite”)
    • Rejecting claims under 35 USC 112, second paragraph, because not every claim term is illustrated in the figure

    These rejections closely resemble the “abstractness” decision in the Alice opinion: the examiner asserts the rejection as a conclusory statement, often using standard boilerplate, and the burden shifts to the applicant to make a lengthy, persuasive argument on the record to traverse it. These bases of rejection provide opportunities for examiners to throw obstacles in the path to patentability – obstacles that require much less effort from the examiner to raise, than for the applicant to overcome.

  10. Examiner “Cooperation” and Compact Prosecution

    As director of the USPTO, David Kappos made a key effort to shift the patent office from the applicant-hostile tone of the preceding Dudas administration to a more cooperative environment. With programs such as the First Action Interview Pilot Program and the After Final Consideration Pilot Program, the USPTO has definitely improved communication and cooperation between patent examiners and patent applicants. And in general, this trend has been extremely positive and productive, as evidenced by reduced pendency and backlog and tremendously improved examiner morale.

    However, this mandate is not uniformly honored. Some examiners still approach the patent process with hostility and mistrust. Some examiners still pride themselves on low allowance rates. Some examiners still refuse to discuss the references or the field of art, simply content with pointing at the cited portions. Some examiners still refuse interviews, or participate in interviews grudgingly, saying little beyond the text of the last office action. While the USPTO is to be commended for its improved climate, work remains to be done in this area.

The sum of these results is indisputable: examiners have tremendous opportunity to force a particular result of the examination process – for reasons of expedience, like or dislike for the applicant or counsel, or even personal philosophies about technology or patents.

Most government agencies that allocate legal rights do so according to a rigid process, with little latitude for discretion. The small number of government agencies that place great discretionary power in the hands of individuals, such as the judiciary and security agencies, also subject these individuals to rigorous vetting and oversight – and even so, the track record for these processes is not encouraging. But our current model of patent examination gives examiners the authority to allow or reject patent applications with complete discretion, and with little oversight into the merit of their decisions. As long as examiners follow the administrative process – as long as they use the right forms, and file their office actions on time – they can grant or deny patent applications at will.

Moreover, the shift of patent examination from an evidentiary process to a discretionary process is still quite new, stemming from the 2007 KSR decision. Because patent law and business communities respond to change fairly slowly, the consequences have not been studied or evaluated – but are already emerging, and will continue to emerge. And once the extent of the damage wrought by this trend is fully appreciated, the repair process will be protracted and onerous.

As I wrote about Alice, I believe that the overwhelming opportunities for examiners’ discretion in the patent process should be alarming to the entire patent community. And I do not understand why the patent community, as a whole, is not speaking out about the dangers of distributing critically important legal rights by an unconstrained and capricious process.


  1. Title 37 of the Code of Federal Regulations is frequently updated by notices posted in the Federal Register; the Manual of Patent Examining Procedure (MPEP) is not only compendious, but is frequently revised at irregular intervals; and the USPTO frequently publishes examiner training materials with updates to examination policy.
  2. The obvious exception is the superseding opinion of the examiner’s primary or supervisor, or the Office of Patent Quality Assurance. But these decisions are often made on a more cursory basis, based on a glance at the claim scope. Moreover, these processes are strongly biased toward reviewing the substance of allowances, rather than reviewing the substance of rejections – hence, improper rejections occur with much higher frequency than improper allowances.
  3. Of course, the full statement of this principle, under Phillips v. AWH Corp., is that the examiner is entitled to the “broadest reasonable interpretation consistent with the specification.” However, examiners pay no attention to this latter section. If you argue that the examiner’s interpretation of the claims is not “consistent with the specification,” the examiner will immediately accuse you of trying to “import limitations from the specification into the claims,” citing In re Prater. This stalemate cannot be easily broken without a trip to the PTAB.
  4. There only clear exceptions to this rule are (1) technical incompatibility, where the techniques of the references actually cannot be used together; and (2) teaching-away, where reference (A) explicitly states: “Don’t do (B).” But examiners are often unpersuaded even by clear and unequivocal teachings in the references for either aspect, and will require a trip to the PTAB to force the examiner to abide by such rules.
  5. An earlier example of a claim style that was choked out of existence is the “Jepson” claim style, which states a combination of parts, and then specifically calls out the improvement of the invention. Jepson claims are tremendously more clear about the nature and scope of the invention than current claim types, which simply provide a list of elements (ABCDEFG), and leave the determination of what is “new” – e.g., element E – as an “exercise for the reader.” Unfortunately, the increasingly lax and lazy rules of obviousness make the task of rejecting the claim as obvious far too easy for the examiner. Jepson claims have therefore been abandoned as good practice.


Our patent system is now on the way to becoming completely inequitable and anarchic! I believe our patent laws now need to be entirely challenged on Constitutionality grounds, since they discriminate against and place undue (financial) burdens on an entire class of citizens – the 99%! When such a large class of citizens are discriminated against and have their rights unfairly limited in arbitrary, subjective and even capricious (or vindictive) ways, then clearly their rights are also being denied! This then moves the issue from one that should solely be considered under the commerce clauses, and becomes one that should also be considered as a ‘civil rights’ issue. When only the rich and super rich can afford to get broad patents (or even worse, are ‘preferentially’ awarded broad patents), then it becomes possible and even fairly easy to ‘prove’ that a clear pattern of discrimination has been created, that actually prevents ordinary people from exercising their rights to apply for and be granted a patent, purely as a matter of merit and equity. If Google, Apple, Facebook, Tesla, etc. are able to fairly easily and quickly get broad patent coverage, but ordinary people, medium size companies, or new start-ups cannot, then that’s just plain discrimination based on class and income!

I am actually ‘gung-ho’ about eventually challenging the constitutionality of our present (as opposed to past) patent laws and their present implementation and interpretation (not to mention fees), using a pure Civil Rights argument and strategy, and would only hope that such a question might be taken up by the SCOTUS (since it would be an important and valid one)! If only 1% of people can now afford to get any kind of a patent (much less a good and enforceable patent), then what is going on now is not in anyway in keeping with what the Founders clearly intended (which was to allow a means for all creative people to protect their inventions and writings – in a financially rewarding way – rather than a financially ‘ruinous’ way). The Founders wanted to encourage more invention by all citizens and wanted to allow ANY citizen to move up the economic ladder, as a result of their inventing new and useful things. They never intended the exclusionary rights provided by patent law, to only apply to large corporations (which didn’t even exist) or merely wealthy individuals. How we came to what we have now (without much resistance), I don’t even understand, since it’s completely ‘obvious’ what is now going on and who will be the ONLY ONES to benefit from it! The AIA and along with other recent changes to tried and true prior patent laws, must be challenged and overturned as violating the civil rights of a majority of our society – based only on their class and financial standing! Without repeal, our current patent laws will severely harm the future of innovation, which in turn will cause an eventual economic crisis (while at the same time fueling the creation of new monopolies – owned by the ‘usual suspects’).

Thanks for the publication of this article. It is extremely difficult to try to protect the right ownership of intellectual property with a novel invention. One hard work and process of thinking is let in the hands of a capricious person that has the authority to do whatever he/she wants. It seems that the system is left to power corporations to take advantage of the system. Righteousness and moral balance is lost. None one is perfect, but the judicial system is just the law of man.

“Moreover, the shift of patent examination from an evidentiary process to a discretionary process is still quite new, stemming from the 2007 KSR decision.”

Nothing in the KSR decision shifted patent examination from an evidentiary process to a discretionary process. Some (many?) examiners may think so, but it did not. Examiners are still required to resolve all required factual inquiries by a preponderance of the evidence and all legal conclusions must be based on facts and correct.

Thanks for the excellent article. I have one disagreement and that is that Jepson style claims are more clear. By definition each element of an invention is old (conservation of matter and energy, plus 112 requirements) as a result an invention is always the unique combination. I think Jepson claims suggest the old point of novelty idea, which is just wrong. I agree that some practitioners bury the invention under a pile of well known elements and connections. One reason for this is that some examiners and supervisors refuse to allow claims that are ‘too short’ or ‘have to few elements.’ For example, image drafting the Jepson claim for the laser. What is the part that is new?

@Dale: Well, let me explain a problem that I typically encounter with examiners.

I draft patent applications that read: “Many computing scenarios involve Task (T). People often perform (T) through Solution (S1), which involves steps (ABCD); but (S1) often exhibits Problem (P). The technique presented herein is Solution (S2), which instead involves steps (AB_EFG_D), which avoids Problem (P) and may also confer other advantages (A1, A2, A3…). Therefore, the claimed invention is: (ABEFGD).”

You’d be surprised how many examiners don’t initially follow that train of reasoning. During the first interview, the examiner often says: “Can you explain your invention to me, and how it’s different from well-known Solution (S1)?”

Of course, I respond by telling them about Problem (P); by explaining how (S2) avoids (P); by pointing to the figures, and the parts of the specification that explain it all.

The examiner then says: “Okay, I understand, but your claims don’t recite that feature. All I see is: ABEFGD, which together looks really similar to ABCD.” I then point them to claim elements (EFG).

Sometimes, the examiner gets it, and prosecution can begin in earnest. And sometimes the examiner doesn’t, and we need another go-around, often in front of the examiner’s SPE.

This happens all them time – and I really mean, ALL THE TIME. I am now conducting First Action Interview Pilot (FAIP) interviews in virtually every case to expedite this spinning-up process. It is eminently frustrating that all of this could be avoided if the examiners simply read the specification. But if that’s not possible… Jepson claims might achieve the same result, although their pariah status makes them unusable in practice.

Leave a Reply

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