« Stone On "Perilous Times" | Main | The November 28 hearing »

November 30, 2005


Feed You can follow this conversation by subscribing to the comment feed for this post.

Carl Sagan

"In science it often happens that scientists say, "You know, that's a really good argument; my position is mistaken," and then they actually change their minds and you never hear that old view from them again. They really do it. It doesn't happen as often as it should, because scientists are human and change is sometimes painful. But it happens every day. I cannot recall the last time someting like that happened in politics or religion."


Interesting post. There's a recent update to the NFL story, which is that the league is now denying having admitted that the referrees blew the call. See this link: http://sports.espn.go.com/nfl/news/story?id=2241325

Here's one alternative theory. Mistakes are more likely to be admitted when, ex post, they didn't determine outcomes. Maybe NFL officials were more likely to be forthcoming (albeit off the record, and I'm guessing from the second story not via official channels) to Coach Holmgren because the Seahawks won the game despite the blown calls. Maybe appellate judges are more likely to find trial court errors in criminal proceedings when those errors can be deemed "harmless," eliminating the need for a new trial. Admitting mistakes that didn't alter outcomes are a win-win: an exhibition of humility that doesn't cost the mistake-maker all that much.


I agree that harmless confession is more likely, but there is no denying that judges and legislatures also change course when there is no doubt but that the earlier decisions had consequences. Moreover, the NFL is a modest example of this point because so many fans care about the point spread rather than the identity of the winner. All the more interesting that presidents do not confess harmless error more often.

That scientists confess error supports the transparency thesis. And that the NFL prefers not to be quoted on its error confession is, I agree inconvenient. Giving feedback to referees is obviously a good thing, but perhaps the optimal rate of open confession is much less than the rate of constructive criticism.

joe p

There have been other, higher profile, instances of the NFL admitting mistakes that would have changed the outcome of a game. Interestingly, the admission occurred in the midst of a broader discussion of improving the quality of officiating, and certain officials were fired.

1) The 49ers-NY Giants playoff game from 1/2003, with the botched field goal and no call on the resulting pass interference at the end of the game.

2) Also from the 2002 season, after the Dec. 2nd Vikings-Packers game the NFL admitted the refs made 9 mistakes, all in favor of the Packers, with 8 blown calls in the fourth quarter alone. These blown calls certainly handed the game to the Packers, and the NFL admitted it.


Here's another possible explanation. People will be more willing to admit error when they have committed ex ante to the idea that the choice is hard or complicated. Scientists and legislator can quite honestly say, "this issue is complicated," without looking like someone trying to avoid a hard question. It's a positive sign of humility and appreciation for subtle question that one admits error in complicated cases.

If that's true, we would expect to see high rates of error denial in cases where the choices are hard, but the actor depends on the illusion that the calls are easy, or at least have only one plausible answer. Presidents and executives (at least in this country, recently) seem to have little to gain by admitting that a question is hard and might not have a clear cut answer. Ref's might not make all easy calls, but every call should have only one plausibly correct answer.

I suppose that just brings up the prior question: why do we demand that some choices be presented as easy one?

Michael Martin

"Transparency" facilitates later reconsideration by decisionmakers -- without review by another, they may never realize that there was a "mistake." But shortness of time is another -- in every case mentioned above (except possibly that of religious leaders), the decisionmakers that either never admit mistakes or rely on changing circumstances as an excuse are also pressed for time.

So an alternative explanation for this phenomenon is that the opportunity cost for reconsidering decisions is higher for those who have to make a higher volume of more important decisions. Executives are at one end of the spectrum, legislators intermediate, and a committee of Tuesday-morning referrees at the other. (I'll leave it to Judge Easterbrook to tell us where appellate judges should fit in to that scheme.)

But the "transparency" point is very interesting with respect to religious leaders.


Another (albeit not fully thought out) distinction between the actors you mention is the question of whether or not the actors are, or are perceived to be, personally responsible for the change in direction. Appellate judges may find it relatively easy to reverse precedent because no one will hold them personally accountable. Similarly for the NFL executive, who can pass the buck to the ref who blew the call, and the public may be more likely to blame the ref rather than the commissioner or the "league" writ large. Perhaps the executive branch official and religious leader are more likely to deny mistakes because they are more likely to be deemed personally responsible - we, the voting/praying/paying public, somehow hold them more personally responsible.

The flip side, somewhat cynical, is that we may be less likely to credit the president or religious leader who owns up to error. Perhaps we think it makes them look weak, as you suggest, or even out of touch with God. (Posner's recent intro to the HLR notes that religious leaders are careful never to make claims that can be empirically falsified on just these grounds.)

Craig Tindall

Very interest post and worthy of study. Referees and the NFL present a good case study but I'd not overlook the umpiring of baseball. Frankly, I think there is an amazing degree of accuracy that is actually displayed. But there have been some fairly significant errors, this year’s World Series faux pas is a good example. Baseball presents a very long history, much longer than football, and a wealth of information by virtue of the available sports writing that has been maintained down through the decades. As far judicial decision-making, I'd second anon's mention of Judge Posner's Supreme Court Forward for the last term. As with most of Posner's writings, its certain worth a read, providing an interesting discussion of the political aspects surrounding constitutional decisions. Errors made in the arena might even be more significant than umpiring mistakes in the World Series . . . okay; they might at least be equal . . . sometimes.

Seth Ayarza

Might the structural role be something to consider? For instance: A Scientist is to a computer model as a Democracy is to a President. But, unlike computer models, presidents don't appreciate being deleted due to bad internal programming.
Scientists do often say "there is no such thing as a bad hypothesis", similarly "an act of parliament can do no wrong, though it may do things that look odd."

In addition, are you implying that if we had access to the voice recordings of a Presidents war room, it would be far more possible to empirically determine whether an executive order was a bad decision?

I wonder if doing so might neutralize healthy criticism of a president by changing the role into a position where mistakes become not only predictable, but acceptable. (I find the typical Referree 'it was the right call from my position' defense preety strong)

This could be especially problematic in a democracy, given that citizens might concieve of governmental failures as technical problems to be resolved. Thus leading to 'activist' style democracy.

Nevertheless, more transparency sounds like good tv, Tuesday at 8pm Eastern, white house conference replay on the disbanding of the iraqi army...


Following up on my previous thought (Levmore's transparency thesis is thought-provoking), I really do wonder about the relevance of public cynicism. One question that would be interesting to test: is the decision to fess up to error proportionate or inversely proportionate to the availablity or salience of a predecessor's mistakes? On the one hand, I could see that (say) a president would be eager to distinguish herself from her predecessor; on the other hand, she might worry that should she fess up, the public will all too quickly slap her with _______gate.


Isn't the common thread here hubris? Appellate courts, while their work may be more transparent than other institutions and, therefore, subject to outside criticism from legal observers and scholars, still are largely insulated from ridicule inasmuch as the critical observers are a miniscule portion of the public and have no vote with respect to rehearings or future cases. They seem to be able to handle it and, to the extent that they are sensitive about it, they have constructed rules allowing them to reach opinions which are unpublished and unable to be used as precedent. Does "ad hoc" ring a bell?


It seems to me that the costs and benefits of admitting error may vary widely depending upon the context. In some instances, refusing to admit error until doing so becomes "transparently" absurd may convey a message that one is resolute and unswerving in one's beliefs. There may be some perceived benefits in such signaling. For example, the president may refuse to admit error in connection with the war in Iraq out of concern that doing so would demoralize the troops, encourage the insurgents, etc. Others may disagree with this assessment, but this is at least a plausible explanation for publicly refusing to admit error. Similarly, a religious leader may be reluctant to admit error out of concern that doing so will undercut the tenets of the faith, e.g., that the proposition at issue was divinely inspired. Occasionally religious groups do give way (e.g., the Catholic Church's relaxation of the ban on usury and its rehabilitation of Galileo; Joseph Smith's renunciation of polygamy), but generally only when refusing to do so costs more in terms of loss of credibility (and even then, the renunciation may be characterized as a new revelation, as error on the part of previous interpreters, etc.). Moreover, the political or religious leader may sincerely believe the view he or she expresses. I seem to recall that the literature on the evolutionary psychology of religion postulates that the best way to convey the sincerity of one's beliefs is to sincerely believe them, i.e., internalize them. As for the NFL, the benefits of refusing to admit error seem to be less substantial and the cost of admitting error relatively small. Fans know that refs can make mistakes, and from behind the veil of ignorance fans might prefer a system where obvious errors are corrected. Transparency factors in because it makes the cost of refusing to admit error high; denial seems absurd. As another poster noted, however, transparency in the executive and religious contexts is often low, and hence there is often a lower cost of refusing to admit error, at least until the proposition at issue (e.g., the sun revolves around the earth) reaches a sufficiently widespread level of incredibility. As for legislators and judges, I would be less inclined to make sweeping generalizations; I think the specific context and psychological makeup of the actor might predominate.


Consider the amount of information available to the decision maker vs the reviewer vs the public. The referee who reviews the on-field call in the replay booth seemingly has the same information that the NFL has the next day when it reviews the tape on Monday. As Dean Levmore suggests, it would be silly for the reviewer to deny the decision maker's error to the public when the public has the same information as the reviewer and the decision maker. Legislators reviewing their own decisions ex post have the same information that they had at the time of decision, but also have empirical information gleaned from observing the effects of their decision and also have the the public's reaction to the decision to factor into their review of the decision. The same is true of prime ministers and presidents. The public has a limited ability to analyze the legislator's/executive's self-review because the public doesn't have the same information as the legislator/executive at either the time of the decision or the time of review.

The difference seems to be that in the NFL context the public can analyze the decision maker as a reviewer because the public has the same information as the decision maker at the time of the decision and as the reviewer at the time that the decision is reviewed. In the legislative/executive context, the public's ability to analyze the decision maker at the time of the decision is the same as the public's ability to analyze the review at the time the decision maker reviews his/her initial decision-in both instances the public has to take the legislator/executive at his/her word. The lack of transparency at both the time of the decision and the time of review means that the public takes the executive's/legislator's word at both the time of the decision and the time of review.

In the judicial context, the decision presented to a judge as reviewer is nearly as transparent as in the NFL. The information that was available to the decision maker at the time of the decision is available to the reviewer at the time of the review and also available to the audience, meaning that the judge-reviewer's ability to deny errors has to pass the 'laugh-test.' Legislators and Executives conducting self-review have the luxury of asserting better information than the audience at both the decision-making and self-review stage...

Craig Tindall

It seems that some degree of willingness to admit an error may depend on the indeterminacy, or perhaps more correctly the underdeterminacy, of the issue. Some decisions are binary, so to speak: Was the pitch within the strike zone? Were both of the receiver’s feet inbounds when the ball was caught? Did the defendant commit the crime? Many decisions are not; instead there is a range decisions that can be legitimately justified as “correct.” Depending on the facts, a higher authority may find one decision option to be more “correct” than another, but the original decision-maker who might have chosen another option need not always be considered “incorrect,” and consequently, need not admit error. One might argue this is quite often the case with respect to appellate review. In a realm where there is no authority that is deemed to be “higher”, e.g., political leaders who ostensibly base their decision on indeterminate policy issues or religious leaders who base their decisions on equally indeterminate religious matters, it is difficult to definitively state that a decision was “incorrect”, only, perhaps at best, that the consequences of the decision were misjudged. Even then, arguably the full consequence of a decision may only be determined by a historical view, at which time it either the decision maker cannot admit a mistake (the historical view may be outside the decision maker’s life span, for example) or there remain significant indeterminacy with respect to the consequences such that the “incorrectness” of the decision remains arguable.

The comments to this entry are closed.