THE MANY LOGICAL FALLACIES OF JUDGE HARRY T. EDWARDS’  COVERED-BY DOCTRINE

Federal Circuit Court Judge Edwards and I both took our high school lessons in logic, reasoning, proof, and bias in the little town of Uniondale, NY—named after a union training center once located there.  There we learned about fallacies, i.e., the use of invalid or otherwise faulty reasoning in the construction of an argument or conclusion, e.g., all dogs are dangerous because I was bitten by one. They are used to persuade by deception, even though they can appear quite valid at first glance. So, I am at a loss to understand how the Judge so recklessly built his defense of the covered-by (CB) doctrine on a bundle of fallacies. Given that his fellow judges are currently considering the NLRB’s rejection of the covered-by doctrine in a case known as Endurance Environmental Solution v. NLRB, it seems like a good time to shine a little light on the loads of illogical babble the CB doctrine rests atop. Consequently, I have listed below the many fallacies in his latest covered-by decision (DOJ v. FLRA, 875 F.3d 667 (2017), with an occasional reference to his 1992 one (Dep’t. of the Navy v. FLRA, 962 F.2d 48 (D.C.Cir. 1992), which first introduced the federal sector to the Harry T. Edwards covered-by doctrine.

Let’s begin with an example of the bargaining situation where Judge Edwards’ covered by concept might be raised.

A Typical Situation

Suppose that the parties negotiated an agreement providing that if the agency moved to a new building “the agency could replace the current open parking system with assigned spaces.” This agreement followed bargaining table discussions where the agency explained how it would assign spaces. But when move occurred about a year later, employees quickly started flooding the union office to complain  that they were getting tickets.  Without any specific notice to the union, the agency had started using parking monitors to impose fines and threatening loss of any right to park on agency property if there is a second violation. Does the union have the right to bargain over the implementation of an enforcement system since the agency never mentioned it in the original notice and negotiations?  Stated differently, did the union clearly and unmistakably waive its right to bargain over the enforcement system or is the topic covered by the agreement because they already bargained over it? Those who adhere to the decades old waiver standard would likely conclude that the agency had a bargaining obligation because it never gave the union specific notice of the enforcement procedures it would use. Judge Edwards, on the other hand, would likely rule the surprise change in working conditions is covered by the previous agreement.

The Harry T. Edwards Fallacies

#1. The Ad Hominem Fallacy– attacks the character, motive and intelligence of those who disagree with him to make his  own arguments appear valid.

The judge didn’t just disagree with the opponents of his covered by 1992 and 2017 decisions. He felt a need to name-call and deride them, which is almost universally seen as an admission that the name-caller has no valid case. The judge wrote they are naïve, simplistic, illogical and even perverse. I am surprised he stopped there when he could have  added that they eat our cats and dogs as they wander among us gifting satchels of fentanyl to our kids. It worked great for Trump.

Who are these people Edwards is calling naïve, stupid, and perverse? They are all the judges who rejected his view of labor agreements by using the “clear and unmistakable waiver” doctrine which was adopted decades before he came along. It is also all the Congressional reps who failed to adopt his perspective and years of legal precedent when updating statutes, and all the labor relations practitioners who disagree with him.

2. The Fallacy of Bare Assertions– claims something is true without providing any supporting evidence, facts, or reasoning.

The judge could barely restrain himself declaring this and that true without any foundation.  For example, before the judge came along with his CB brainchild LR practitioners and judges relied on five reasonably tangible pieces of documentary evidence to determine whether the union had clearly and unmistakably waived its right to bargain over agency proposed mid-term changes. They were:

(1) an “exact congruence” between existing contract provisions and the matter in dispute,

(2) evidence that the parties “specifically discussed” a subject,

(3) settlement agreements authorizing the mid-term bargaining,

(4) a prior practice of bargaining the issue mid-term, or

(5) even  indications that the matter was “contemplated” during bargaining.

But the judge dismissed them as irrelevant in favor of ordering the parties to look to the undefined, totally subjective “compass of the provisions of the agreement.”   Why?  Because he said so without any legal foundation. In fact, he tossed aside the extensive legal precedent that was already there. The clear and unmistakable waiver standard was adopted in the federal sector as far back as A/SLMR No. 223 (1972), before there was a FLRA or even a Judge Edwards, giving Congress seven years to consider changing it before it passed the CSRA.

Ironically, Judge Edwards has been a leading critique of “junk science” being used to support legal positions.[1] But apparently, when it comes to a conclusion he is trying to sell, the judge just wants us to take his word for it.

Not content with that bald assertion, the judge went on to make others in support his conclusion that his covered by creation is valid.  For example, he claimed that unions “… would almost invariably prevail in duty to bargain cases, because it almost always could find some ambiguity in the relevant contractual language.”[2] He offered no data to back that up. While unions have won the majority of the failure to bargain cases under the pre-Harry Edwards era, agencies won several.  But that is typical of all union filed ULPs that are decided by the FLRA for a reason.  There is a system in place that weeds out cases unions are likely to lose before they get to FLRA.  To begin, unions themselves tend not to file real losers.  When they do file, the FLRA regional office will not move forward on cases unless the evidence and odds favor the union. Then the ALJ’s get a chance to pressure unions to withdraw or settle cases they are likely to lose.  The result is that the FLRA generally only sees those cases where the odds are high that the union was right about the employer being obligated to bargain.

Besides, it is illogical, if not boldly biased, to demand a change the law because one party wins the majority of the cases.  For example, the Dep’t. of Justice annual report shows that in 2024 federal prosecutors won 48,666 criminal  prosecutions and lost fewer than 3,000.  Should we change the laws so that criminals have a better chance to win? Of course not!

His third grandiose claim without foundation was that a management covered-by argument that the union did not have the right to negotiate over a management-initiated mid-term claim is not an assertion that the union waived its right to bargain.  Rather, it is merely  a contention the union has already bargained over the matter. Hence, rather than management having the burden to prove the union clearly and unmistakably waived its right to bargain, the burden shifts to the union to prove the negative that it did not—a seismic shift.

The judge justified this by writing, “The covered-by doctrine is analytically distinct from waiver. A waiver occurs when a party knowingly and voluntarily relinquishes its right to bargain over a subject; when a disputed subject is covered by the parties’ agreement, however, the parties have exercised their rights to bargain over that subject.”  That begs the question of how can the situation in our opening example about employee parking not involve a waiver when management never told the union that as part of its assigned spaces policy it would also be implementing a system of tickets, fines, etc.  The union never had any knowledge that was part of the plan or to use more traditional labor law terminology it never had “specific notice” from management of the plan. And without that it could not “knowingly and voluntarily” waive bargaining over enforcement techniques.

The judge correctly explained what a waiver is; he just drew a distinction that is not real.  It is like trying to defend cheating on your wife by claiming it was not cheating, but a loving effort to learn new techniques that the marriage could benefit from. Courts, including the Supreme Court, have held that rights created in the public interest cannot normally be waived by an individual party because they are the public’s property.[3] But even in situations where an individual waiver is permissible it must be a “clear and unmistakable” (C&U) waiver. The C&U standard appears throughout labor and civil rights law. You can find it to be the pivotal question in FLSA claims (Wright v. Universal Maritime Service Corp., 525 US 70 (1998), FMLA disputes (Rogers v. New York University, 220 F. 3d 73 (2nd Cir. 2000), and civil rights matters (14 Penn Plaza LLC v. Pyett, 556 U.S. 247 (2009))  Why the judge is so eager to stamp it out is a mystery.

Frankly, as someone who spent over 40 years at bargaining tables, this should all come down to whether the union had specific notice of the change. If management gave the union specific notice it was going to enforce parking assignments (or it was reasonably foreseeable from management’s specific notice) and the union did not address it in the contract, then it waived the right to bargain over its implementation. The “specific notice” concept has been in place for roughly 70 years of labor law without much opposition.  Judge Edward’s covered-by approach quietly guts it without ever saying that.

3. The Fallacy of Ambiguity – uses unclear, vague, nebulous, or shifting language an argument.

After tossing out the five tangible, reasonably objective, long-used evidentiary factors parties could look to, the judge decided that “What matters is whether a subject is within the compass of the provisions in the parties’ agreement.” COMPASS? If you are wondering what that is, so is everyone else on planet Earth. Judge Edwards never defines what the compass includes or excludes. Using the word “compass” does have a ring of erudition and might titillate legal linguists, but it is of absolutely no value to the thousands of LR advocates on both sides of the bargaining table. Zero! Zilch! Zippo! That is important to keep in mind because those are the only beings in the known universe who will be using the concept and who need a reasonably objective or tangible standard to apply.

Why did the judge choose such an indefinable term? He did it for the same reasons anyone else does. By choosing a word impossible to tightly define he makes it nearly impossible to show he is wrong. After all, who can say someone is wrong if they have no idea what is being said? He also rendered the hundreds of legal decisions using the old criteria void, wiping the case law slate clean for him and others of his ilk to create anew.

Ironically, elsewhere in his decision he condemned ambiguity when practitioners use it to draft contracts, “The [complaining party] would almost invariably prevail in duty to bargain cases, because it almost always could find some ambiguity in the relevant contractual language.” In other words, when the judge uses ambiguity, it is good, but when the practitioners do it causes problems.

Not content to dip into the fallacy of ambiguity once, the judge also advised the parties to consider the analysis of whether there is a bargaining obligation to be a matter of contract construction, not interpretation. “The Authority’s position confuses issues of contract “interpretation” with issues of contract “construction.” “Application of the ‘covered by’ doctrine is an exercise in construction….”  “‘[C]onstruction’ ‘determines [the] legal operation’ of agreement; ‘interpretation’ of agreement resolves any ambiguity in terms used.”

Not aware of the distinction between the two, I went in search of an explanation of it. That led me to a piece entitled, “Interpretation and Construction in Contract Law,” by Gregory Klass and published by the Georgetown University Law Center in 2018. He wrote,

Although the distinction between interpretation and construction is easy to state in the abstract, a complete account of the two activities and the relationship between the two is no easy thing….”

There may not be a greater understatement in the history of legal scholarship given that it took Klass 48 pages to explain the difference, including charts, a deep dive into philosophy, and 155 footnotes. It is a mushroom-induced fantasy to think that thousands of labor-management practitioners will be able to use the distinction. Raising this merely further confuses the practitioners who are responsible for government running smoothly without running up litigation and financial liabilities.

Finally, I cannot leave Edwards’ efforts to impose a haze over the bargaining obligation analysis without reaching back to what he wrote elsewhere when condemning junk science, i.e., “The rule of law embraces the quest for constancy and predictability, as well as a determination to treat like cases alike.” How are we to get constancy and predictability if we are to toss the tangible criteria used by others for decades, that rest atop hundreds of legal decisions to guide practitioners, in favor of a search for the “compass” of an agreement? The word’s synonyms only highlight how absurd that search will be, e.g., remit, breadth, scope, ambit, orbit, grasp, etc. Can you imagine getting caught cheating by a spouse and arguing that it was not prohibited within the compass of your “love, honor and cherish” contractual vow?

(Edwards is using the old Justice Potter Stewart’s trick. Potter demanded that Americans avoid pornography, but admitted he could not define what it was and was not. Nonetheless, he assured us that, “I know it when I see it.” (Jacobellis v. Ohio, 378 U.S. 184 (1964)) Just check the Internet to see how helpful that was in stopping the spread of pornography.)

4. The Fallacy of the False Cause– claims that simply because two things have a relationship to one another that one causes the other.

The judge wrote,

If the obligation to bargain could be imposed whenever a party insisted upon reopening bargaining because it did not understand the full reach of the parties’ agreement when it was executed, this would wreak havoc in bargaining relationships.

To begin, every contract dispute in the known universe, whether it deals with labor-management matters, commercial sales, partnership formation, or a wedding prenup flows from the fact that at least one of the parties claims it had a different understanding of its reach. Yet, you don’t see the judge claiming those disputes create havoc. Actually, they create very nice incomes for lawyers like himself.

Second, merely claiming the right to bargain does not create an “obligation to bargain” despite the judge’s  suggestion that they are one in the same. The claim creates a disagreement, although not always. And bargaining claim disagreements are worked out by the parties or the federal dispute process law created. It does not create mayhem, destruction, disaster, devastation or any of the other synonyms of havoc. Claiming it does is a near hysterical overreaction, if not outright disingenuous.

Third, the only thing that potentially triggers “an obligation to bargain” is a management decision to make a mid-term change in employment conditions—with very small exception.  If management  makes no changes that it has not given the union specific notice are coming, there likely will be no bargaining obligations.

5. The Fallacy of the Switcheroo – involves shifting the focus of an argument to an unrelated topic.  

The judge further supports his CB theory by arguing that parties are entitled to “stability and repose with respect to matters reduced to writing in the agreement.” I agree with that statement. If it is in written contract, the parties must live by it. But watch what Edwards then does with the stability and repose (S&R) issue elsewhere in his decision: “The Statute …also seeks to ensure repose and stability in bargaining relationships.” Suddenly the concept applies not just to negotiated written agreements, but to the entire “bargaining relationship.” The judge not only is trying to piggyback on a widely accepted precedent applying the stability and repose concept “to the specific conditions of employment mutually established in their agreement.”  (IRS and NTEU, 17 FLRA 735 (1985)). But he is also hoping that the reader will miss the switch and not pause to realize that contracts are different from relationships. One is normally committed to writing to fix things in place while the other is shaped by circumstances of the moment, personalities, and other constantly shifting elements. Neighbors and co-workers become closer or more distant as time goes on. A married couple’s relationship evolves greatly from the day they exchange vows. The same goes for a parent’s relationship with a child. You just can’t enter a contract with a child that obligates him to always listen to or love you. Frankly, a good argument can be made that relationships need a dynamic element to mature and renew.

The judge’s switcheroo is critical because it enables the judge to argue that bargaining itself destabilizes relationships. However, it is collective bargaining that is repeatedly endorsed by the statute as good for the labor-management relationship whereas the S&R concept is mentioned nowhere in it. There is a very good reason for that.  If either party was allowed to change working conditions without bargaining with the other, it would create resistance, if not  retaliation, and government would suffer. If I was the labor rep in the situation described at the outset and management refused to bargain over parking enforcement measures that it never told the union were coming, I would try to get every unit employee to park in an unassigned space, park over the lines of their space, and file a grievance over their assigned space.  In short, I would “encourage’ management to sit down and work something out with the employees using my right to engage in concerted activity—in addition to trying to ruin the perception of management’s integrity.

A second, even more disingenuous use of the switcheroo ruse, is the judge’s statement that,

This seems to suggest that the crucial question is whether a party contemplated every scenario that might arise under the terms of the Master Agreement. This makes no sense because it would effectively eviscerate the covered-by doctrine.

He wants us all to think that the goal of the statute is to preserve his covered-by doctrine rather than collective bargaining and smoothly operating workplaces. That is simply not true. The judge’s doctrine is not mentioned anywhere in the statute nor very voluminous legislative history. Yet, the judge wants us to administer the collective bargaining process so as to promote and serve his doctrine above all else. I believe that is what we also call hubris.

6. The Fallacy of the Hasty Generalization – involves making a claim on the basis of insufficient evidence. 

The judge wrote, “…construing collective bargaining agreements as covering only those outcomes the parties concretely foresaw would make extensive future bargaining inevitable, removing the parties’ incentive to try to comprehensively bargain in the first place.”

But, what about the historical fact that construing collective bargaining agreements that way has been the legal practice under the clear and unmistakable waiver doctrine for decades without unions seeking to avoid comprehensive agreements? Noticeably, the judge never did point to actual examples of where that happened.

Moreover, what about all the reasons why unions want comprehensive agreements, e.g., it makes them look stronger before the membership; it minimizes the need to bargain throughout the contract; it avoids the need to engage in costly and lengthy litigation to enforce a right to bargain over a change; it creates clarity, objectivity, and predictability in the workplace; etc.?

Claiming that bargaining over every or most management-initiated changes is inevitable is another substantial overstatement.  For example, if the mid-term change does not upset any members, the odds of a union pursuing negotiations are small. At other times the impact of the change is unclear, prompting the union to sit back and watch what happens before deciding how to react. Perhaps they wait until the contract reopens to deal with it at the term table. Finally, if they think the contract does address the situation, the union will generally prefer a grievance to enforce its interpretation. After all, demanding to bargain is an admission that the contract does not address the matter. That is not something a union would admit unless it is certain the contract does not.

7. Bias– while it is not a fallacy, it is just as deceptive and disingenuous reasoning error. I am including it here because it potentially reveals a motive for such a disingenuously reasoned decision.

The judge made the following observation: “…it would have required near-supernatural prescience for the parties to have foreseen, at the time of drafting the MLA-, what implementation issues would arise with respect to ‘specific individual details’ that had not even been conceived, much less implemented, at the time….”  Even if that is true, why is the judge’s remedy to deny the union the right to ever negotiate while vesting management with the right to move full steam ahead with a change whenever it did not tell the union about it during term bargaining?  Is he so tied to protecting management that he cannot find some balanced rule or even see the employee resistance and resentment that will arise toward management if it is given carte blanche?

The judge further revealed his predisposition to help management at the expense of unions with this statement, “…the scope of what is covered must be construed to give the parties the benefit of their bargain.” That sounds quite legalistic, judge-like, and reasonable, but Edward’s solution is to only give management the benefit when it claims ambiguity.  They get to rush some unannounced change into place while employees are denied any opportunity to bargain over the substance, impact of implementing procedures. Leaving the union to try to prove that a particular change in working conditions was not within the compass of the agreement, leaves them with the age-old dilemma of trying to prove a negative.

*          *          *

The judge obviously has a big problem with the concept of mid-term bargaining. He sees it only producing mayhem, stagnation, and costs injurious to American society. Having spent 40 years doing it with about three dozen employers, I saw it work repeatedly. By that I mean I saw disputes be resolved quickly, employee resistance to change reduced, and employee input improve the original management plan.

[1]   https://www.washingtonpost.com/opinions/a-wake-up-call-on-the-junk-science-infesting-our-courtrooms/2016/09/19/85b6eb22-7e90-11e6-8d13-d7c704ef9fd9_story.html

[2]   I am not  going to dwell on it here, but “ambiguity” had little to do with the prior approach to cases because case law turned on the five evidentiary documents Judge Edwards dismissed as irrelevant.

[3]   “The enactment of such provisions by the Congress is an expression of general public policy creating rights in members of labor organizations which cannot be waived or contracted away.” Brooklyn Savings Bank v. O’Neil, 324 U.S. 697(1945)