(Presented during the AIPLA Annual Meeting for the Patent Law Committee/Mentoring Committee session on Friday October 20, 2017 )
The title of my presentation is a play on words with the concept of “wisdom of the crowds,” which had its origins over a century ago. In 1907, at a lifestock annual show in England, a contest to estimate the weight of an ox attracted 800 participants. It was observed that the median guess of these participants, 1207 pounds, was accurate within 1% of the true weight of 1198 pounds. This finding spawned the wisdom of the crowds concept—that a median of a crowd’s individual judgments can be used to reflect a true center of a quantity to be estimated. This concept has infiltrated many disciplines and studies. In short, aggregate data can be a powerful thing.
Can Board decisions help in knowing how to overcome obviousness rejection? I think so. There may not be a one true obviousness response, but looking at the frequency of arguments can provide a framework and see the most effectively used. In appeal briefs, the Board entertains a lot of arguments and they coalesce around a finite set of them in overturning an obviousness rejection. Board decisions can definitely help organize these arguments. Plus, knowing the most frequent arguments can help in knowing how to draft appeal briefs. But more generally, it helps to align response arguments for proper prosecution. Because if you are following the way that judges overturn rejections, that information can be useful.
Laying the Foundation for an Obviousness Framework
Stepping back a little bit to talk about appeals more generally, it’s an interesting data point. One reason why I like looking at appeals data is you’re only looking at arguments of two sides who have vested interests in the case. And obviousness is the most common ground of rejection that gets appealed (by far).
With anticipat.com, we’ve found that there is quite a lot of disparity in how grounds of rejection are handled by the Board. Section 102 and 112 rejections are consistently reversed at a greater frequency than Section 101 and double patenting, for example.
Obviousness is in the middle of the pack. Since there are so many obviousness appeals, this means that there is a lot to learn from. Going through the decisions gave us order to what is going on with obviousness. There are a lot of possible arguments, but seeing how the judges apply the terms gives us a nice framework.
First, you have the prima facie case, which the examiner must establish for a proper obviousness rejection. This prima facie case first requires that claim features be disclosed or suggested in the prior art. CFMT, Inc. v. Yieldup Int’l Corp., 349 F.3d 1333, 1342 (Fed. Cir. 2003) (“requires a suggestion of all limitations in a claim”).
The second requirement is that there must be an articulated reason for obviousness. KSR Int’l Co. v. Teleflex Inc., 550 U.S. 398, 418 (2007) (“a reason that would have prompted a person of ordinary skill in the relevant field to combine the elements in the way the claimed new invention does.”).
After creating the above diagram, it appeared something like a brain with a left hemisphere and right hemisphere. After some thought, I realized that’s actually a good way of thinking about these two elements. As you may be aware, the left side of the brain is responsible for the logical functions such as science and math and the right side of the brain is responsible for the creativity functions, such as the arts. And that fits with what these elements do. The teaching or suggestion element is more of the logical “is it in the reference or not?” analysis. Whereas the second element (why would it have been obvious?) involves more creativity.
So starting with the left side, we touched on the highest level “are the features disclosed or suggested in the prior art.” But it breaks down into several categories.
On this left side, claim interpretation is a big aspect of whether the prior art discloses or suggests the claim features. This includes whether features are properly given patentable weight, how the claim term is construed, and whether the broadest reasonable interpretation is correctly applied. There is also whether inherency is properly applied if a reference does not explicitly disclose a feature. Then there are arguments about the references themselves, whether they qualify as prior art and whether they are non-analogous art.
On the right side, a reason for obviousness must be clear and articulated—not conclusory. Usually Examiners use the rationales listed in KSR (e.g., obvious to try, etc.), but common sense can also be used. Thus, the Examiners can run afoul by incorrectly applying an established or made-up rationale. But beyond this global argument, it breaks down as well.
If combining references, one skilled in the art must have a reasonable expectation of success in combining the references and must be motivated to combine them. Kinetic Concepts, Inc. v. Smith & Nephew, Inc., 688 F.3d 1342, 1360 (Fed. Cir. 2012). The references cannot teach away, a reference cannot be modified to render it unsatisfactory for its intended purpose, and cannot be modified to change the principal mode of operation. Also, the Examiner cannot rely on improper hindsight bias.
In sum, the left side of the arguments involve more technical thinking while the right side involves some creativity—just like our brains.
Finally, once a prima facie case of obviousness has been made, the applicant can rebut that with objective evidence.
A lot of components go into objective evidence such as, for example, secondary considerations and unexpected results.
Here is the complete framework of arguments that are used in overturning an obviousness rejection.
Which arguments are most frequently used by the Board?
Now that we have the framework of arguments that the Board uses to overturn, we drill into the data. I used Anticipat.com Research database for a window of two months’ worth of recent decisions: July 1, 2017 – August 31, 2017. There was a total of 1607 decisions. Of these, 1345 included obviousness rejections (84%). And of these, 446 were wholly reversed on obviousness (33%). I looked at each of these manually to confirm accuracy. Any of the data discussed can be accessed using the research tool.
1. Must disclose or suggest (63.9%)
By far, the most frequent way that the Board overturns the Examiner’s obviousness rejection is on the reference(s) not disclosing or suggesting the claim features. Sometimes this comes out as “the examiner didn’t make his case clearly enough so we can’t uphold”, but other times it is more confrontational by saying “the examiner is incorrect; the claim features are not shown or suggested in the cited references”.
2. Clear and factually-supported articulation of reasons for obviousness (35.9%)
The reason for obviousness (often times based on KSR language) was found to be faulty or not good enough.
3. Examiner Bears Initial Burden (Prima Facie Case) (12.6%)
So far it is to be expected that these arguments are the most frequent because they are required for a prima facie case so it makes sense that board uses these to attack. The next is the argument that the Examiner has not met his/her burden for a prima facie case. This gets used as a conclusion at the end or as a introduction. It never is the only argument used, but is coupled with either element of the prima facia case.
4. Scope and Content of Prior Art – Claim Construction (10.8%)
Often times the Board couples a generic argument (such as listed above in 1-3) with another argument nested inside. That is most true of the fourth most frequent argument to overturn obviousness: claim construction. This is always tied with “disclose or suggest” because the argument is that the prior art is not disclosing the claim language by focusing on what the claim language actually means.
5. Scope and Content of Prior Art – Broadest Reasonable Interpretation (6.7%)
Next is broadest reasonable interpretation. This is also almost always tied with “discloses or suggests” focusing on where the Examiner is interpreting too broadly. Suzanna Sundby gave a fantastic talk on obviousness on broadest reasonable interpretation. This can be a sore point of contention between an applicant and the examiner. The Board will step in and overturn unreasonably broad interpretations.
6. Inherency (5.8%)
Next is inherency. This is a relatively common argument that is tied to left side “discloses or suggests”. The Examiner may assert that the claim feature is inherently in the reference and the judge panel disagrees or holds that the examiner’s showing was insufficient.
7. Hindsight Reasoning (5.2%)
Hindsight is the most frequent argument that is usually tied to the right side, although it is not exclusively used here. In other words, the Examiner may propose a reason for why it would have been obvious and the panel can say that the reason was flawed and also on top of that the Examiner relies on hindsight. Sometimes in the combination discloses analysis the judges can say that the references do not disclose a particular feature and thrown in that the Examiner relies on hindsight.
8. Positively recited structure required for patentable weight (3.4%)
This is another left side argument. Here the Examiner may not be giving patentable weight to claim language and the judge panel overturns because the Examiner improperly did not give the claim patentable weight.
9. Proposed modification cannot render the prior art unsatisfactory for its intended purpose (2.5%)
The first of two classic rebuttal arguments is used by the Board in overturning obviousness rejection, but perhaps at a lower rate than one might expect. This gets coupled with the right side or reason for obviousness.
10. Proposed modification cannot change the principal mode of operation of a reference (1.8%)
The second of the two classic motivation rebuttal arguments. This is also tied to the right side.
11. Non-Analogous Art (1.3%)
This gets used usually with the left hand side of the logical argumentation. But sometimes when the Board overturns by attacking the Examiner’s reason for obviousness, the Examiner will conclude that the one reason why the reason for obviousness is flawed is because the Examiner improperly used a non-analogous reference.
12. Reasonable Expectation of Success (1.1%)
This is a right side argument that is highly tied to the motivation to combine. This argument will be thrown in as a small companion to the bigger motivation to combine argument.
13. Objective Evidence (0.9%)
Yesterday a presenter talked about how objective indicia are gaining in popularity since KSR. But it seems like they are not used very much here likely because objective indicia are more persuasive over time (after grant of patent) as more evidence becomes available with time.
14. What is Prior Art? (0.7%)
Whether a reference is prior art or not is rarely applied by the Board. One possible reason for this is that the appeal conference may kick out improper rejections by having the SPE and another primary analyze the contested prior art reference.
15. Teaching Away (0.2%)
The Board only used “teaching away” in one decision, showing how difficult of an argument it is to make. If the frequency of arguments by the Board is in any way correlated with effective arguments, teaching away is a losing argument.
Looking across tech centers, it is interesting to see some anomalies. Biotech (tech center 1600) strongly overrepresents “articulated reason for obviousness” and “what is prior art?” and underrepresents “features not disclosed or suggested”. This suggests that it is best to focus on attacking the reason for obviousness instead of attacking the features not being disclosed or suggested.
Chemical (tech center 1700) strongly overrepresents inherency and hindsight. This suggests that Examiners in this space excessively use inherency.
Electrical/Computer (tech centers 2100, 2400, 2600, and 2800) overrepresents “must disclose or suggest,” and “broadest reasonable interpretation”. This shows that Examiners in this space may interpret claim terms too broadly and that it is worth fighting back on it. It also shows that it is effective to point out gaps in the applied art for the claim elements.
Business methods/software (tech center 3600) underrepresents “articulated reason for obviousness”. This may be because Examiners are given more leeway in this art in crafting creative reasons for obviousness since software is easy to plug in different modules or features.
Medical Device/mechanical (tech center 3700) overrepresents “reasonable expectation of success” and “claim construction” and “non-analogous art”. This suggests that it can be helpful to rely on the interpretation of a claim term. It also shows that non-analogous art can be effective in this space.
The Board generally tends to use the left-side arguments in overturning obviousness rejections, but there are exceptions as noted in specific tech centers. Some arguments are much more frequent than others.
Why is this useful? Have you struggled with a particular point of contention with an Examiner (for example thinking that the Examiner’s reason for obviousness is inadequate), wondering how this specific point of contention compares to others. Anticipat Analytics provides you with the organized and personalized information from the Board.
You can see a ranked listing of all the arguments that were found to be successful at the Board. For example, if you believe that your Office Action has an unreasonably broad interpretation, you can see where that argument falls in relation to others. Plus, you can access the Board decisions for your particular Examiner on that issue (if it exists). If it does not exist, you can go up to the art unit level or tech center for greater coverage of arguments. Seeing how often these arguments get used compared to others gives you a better sense of the likelihood of success in appealing or in working with the Examiner.
Consider using the most relevant arguments, but also find the persuasive arguments to support your particular application’s case. You might be surprised at how much the Board borrows arguments from a good appeal or reply brief. The tool allows for you to easily find the language that the Board uses, which can be a model for you to incorporate into your own filings.