« "What Do You Mean, 'We?'" | Main | Gershon, Antigone, Madoff, and Punishing Family Status »

Tuesday, December 23, 2008

Is an Argument Any Less Valid if You're "Lying?"

Jay's posts have been very enjoyable, and I was intrigued by his post asking how people would feel if a professor simultaneously published "two articles at the same time taking diametrically opposed questions on the same question or issue or set of problems." 

Most of the early commenters on the post were simply perplexed: why would someone do it?*  One commenter wrote, "I think that when an author puts his name to a piece of scholarship, it's an implicit statement that the author actually subscribes to the conclusions stated in the article, for the reasons stated in the analysis part. And as a general matter, I think it's unethical to publish an article where the author doesn't actually subscribe to the article's conclusions."  Our prized commenter Orin writes in part that the professor would be "a bit of a loon. If you're torn on an issue, the best way to communicate that is to engage the difficulty and explain why you're torn: the worst way is to pretend no difficulty exists by writing two separate articles."

I wonder why it matters. 

 Let's assume that both articles are compellingly argued and persuasive.  Maybe they take similar normative and methodological starting points and just diverge on particularly difficult issues; or maybe they examine the same issue but with different normative or methodological starting points (ie., one proceeds from an efficiency-oriented standpoint and the other doesn't; or one is originalist and the other is not).  But they are both strong pieces.

Why, then, should we care that the author believes that one article is right and the other isn't, or indeed that she may not particularly share either view?  I understand that people may have different emotional reactions to being kicked and being stumbled over.  Physically speaking, however, the result is the same.  Now, in my supposition and Jay's, both articles are compellingly argued and persusasive.  If read without regard to the author's motives, they would both be seen as at least potentially valid.  So why should it matter that the author actually believes only one -- or neither, or both -- arguments?  Why should we care why the author did what she did?

One commenter, as I said, suggests that such an action would be unethical because authorship implies agreement with the arguments contained in the article.  I'm not sure that's right, and I'm not sure how much motive matters provided the article is written with scholarly integrity with respect to the arguments contained therein.  Let's say the author acted as she did out of perversity -- a valuable motivation, in my view; would that there were more of it!  Is it not possible to be a scholar who is both possessed of integrity as a scholar and perverse in motivation?

I suppose I find this especially interesting because the same commenter also suggests that all this is different from the case of a lawyer writing a brief, who is simply making the best possible argument on behalf of her client.**  What I think this fails to capture is the fact that most law-review writing, whether or not it is true "scholarship," is highly argumentative and normative; as many people have observed, countless law review articles resemble nothing so much as briefs written to a court urging a particular result.  Often enough, the convictions moving the author are ideological, not just the "search for truth" that is the scholar's province in the traditional view.  To that end, I would suggest that law review authors (too) frequently make compelling arguments that they do not believe in -- using a methodology (say, originalism) even though they think it is illegitimate, or making efforts to distinguish cases that, in their heart of hearts, they think are indistinguishable, or arguing that a particular position does not carry certain implications when, logically, it does.  They arewriting briefs. 

I am reminded of the efforts of many academic lawyers to turn headstands arguing that their position in favor of the law school plaintiffs in the Solomon Amendment case in no way required reexamining any of the civil rights cases, like Runyon v. McCrary.  (I won't name names here, but one example can be found in a very long footnote in this article.)  In that case, the arguments were, in my view, unpersuasive in part because they were internally inconclusive; the degree to which the arguments were internally flawed, in my view, also called into question whether the authors had acted with full scholarly integrity or whether they were just acting as advocates, and unpersuasive ones at that.  But if a legal academic makes an argument that is not internally flawed, that is in fact persuasive, should we care if she doesn't necessarily believe it?

Many people seem to have less of a problem with the ideologically driven professor than the hypothetical perverse writer that Jay posits.  But I wonder: aren't authors who fall in the former category as dangerous to the cause of scholarship, if not more so, than someone who writes a wholly compelling argument (or, in Jay's example, two arguments) that she doesn't necessarily personally believe in, for no better reason than perversity?

* In the comments section, Jay provides two possible reasons: the author might be completely torn on the issue, or she might be doing it as, in effect, a piece of performance art.  As I suggest above, I think there are other reasons one might do such a thing.

** I wonder in which category the commenter would put the anecdote about Justice Douglas writing an opinion and then, because the Justice on the other side of the issue was having trouble, drafting the contrary opinion as well.  Or the one about Judge Posner as a law clerk, mishearing Justice Brennan's instructions and drafting an opinion that reached the opposite result of the one he had been instructed to reach -- an opinion that Brennan and the Court then signed onto.          

Posted by Paul Horwitz on December 23, 2008 at 01:21 PM in Life of Law Schools | Permalink


TrackBack URL for this entry:

Listed below are links to weblogs that reference Is an Argument Any Less Valid if You're "Lying?":


To be sincere about an ultimate assertion means simply this: that you believe your assertion is true. (For normative assertions, we could say that to speak insincerely is to say that we ought to do that which you do not believe we should do.) Following Paul Grice, I think that such an assertion of belief is an implied feature of virtually every utterance, including scholarly arguments.

You make an important point when you note that motivations to persuade can motivate authors to mislead regarding the strength of their argumentative support. But the norms of scholarly discourse allow for this: we expect, to a certain degree, that authors will tend to characterize evidence in a way consistent with their asserted beliefs. Having that expectation, we can discount our degree of trust in their conclusions. This is why, for instance, the same assertion can be more persuasive when it is made by a scholar who would normally tend to disagree with it, than when it is made by a scholar who we think is motivated to agree with it.

But note that this effect only works to the degree that the first scholar is actually convinced that the unappealing proposition is correct. If she is convinced, despite her best efforts to believe otherwise, that is powerful evidence that the existing counterarguments are insufficient to defeat the assertion. But her insincere assertion would carry little weight; we have no reason to think that the same process of truth-testing underlays it.

So the short answer is: it is easy to adjust our beliefs to correct for the bias you describe when we evaluate scholarship; it has a predictable direction and is an acknowledged feature of this type of discourse. But when authors start misrepresenting what they believe, how do we decide which of their conflicting statements to discount? By the very act of concealing their actual belief states, such authors are concealing important evaluative information from their readers.

Posted by: Mark Spottswood | Dec 24, 2008 9:33:59 PM

Interesting Mark. But your theory rests on the assumption that an author's disagreement with his ultimate assertion (which I don't think is the same thing as "being insincere") necessarily calls into question the underlying data. It seems more plausible (and many examples in this post support it) that it is the person with a personal interest in the ultimate assertion who is most likely to mess with the underlying data.

What does it really mean to "be sincere" about an ultimate assertion?

Posted by: anon | Dec 24, 2008 7:48:23 PM

Two points:

(1) You often see repeated some variation of anon's "How does honest belief add any merit to a particular argument?" There is a simple reason that it does: the speaker's credibility is at stake in nearly every argument he or she might make.

Imagine that a scholarly argument looks a pyramid. At the pinnacle is its conclusion -- that we should or should not amend the law in a certain way. At the next level there are several supporting normative reasons to do as the author suggests. At the level below, each supporting reason has a combination of normative assumptions and empirical assertions. (The regress, in fact, continues for several levels farther down, eventually bottoming out on either citations to authority or unsupported factual claims.)

In principle, we could check every bit of empirical data ourselves before we decide whether to find the author persuasive. We could investigate in detail each source the author relies upon, and regress backwards, giving each source the same exhaustive level of scrutiny. We could review the raw data of every empirical study ourselves, run our own regressions, etc. But we don't do this, because it is enormously time consuming. It is in fact bad for the growth of social knowledge to read scholarship in this way, because if we spend this much time on one new article, we neglect many others that could give us contrasting views, and also neglect producing our own new work.

Rather, we depend upon credibility to take the place of independent verification. We trust the author to accurately represent his source material, to address and respond in good faith to flaws in his argument of which he is aware, etc. If the author isn't doing these things, the utility of his scholarship falls drastically. And when we know an author is being insincere with regards to his ultimate assertion, we properly lose trust that he is being candid and sincere with respect to his treatment of the underlying data.

(2) Which brings me to my second point: the search space of possible things we could believe is very large. We have to have some way of winnowing down what claims are plausible enough to be worth investigating. Reading only sincere scholarship constrains our truth-seeking in a useful way: If we assume that most of the scholarship we read is sincere, we know that at least one truth-seeking person has investigated a topic and found it to be persuasive. Even if we do want to carefully fact-check every claim made in scholarship, we have to be selective in choosing which articles are worth such scrutiny. And if we further limit our search space to sincere claims made by experts with a track record of accuracy, then we are really getting somewhere.

So the moral of this story is that good Bayesians update on others' beliefs. Accordingly, we harm others when we misrepresent our belief states to them.

I explore these issues in a bit more depth here. And for discussions on the virtues of updating our beliefs based solely on the informed beliefs of other people, see Alvin Goldman's work here and here.

Posted by: Mark Spottswood | Dec 24, 2008 5:05:57 PM

I think anon 1 has it exactly right. Universalize that maxim, and watch the castle crumble.

Posted by: Paul Gowder | Dec 24, 2008 3:00:05 PM

Paul, thanks for the post.

Perhaps the original claim about what it means to sign your name to a piece was lacking in nuance. The basic point I think I'm after, though, is that I think there's more to intellectual dishonesty than just plagiarism or misrepresenting facts. I think that when someone calls an academic a "hack" or a "shill," that person feels insulted. Don't they? My gut reaction is that the answer is yes, and rightfully so, and I'm trying to make some sense of it.

I thought this claim was interesting:
law review authors (too) frequently make compelling arguments that they do not believe in -- using a methodology (say, originalism) even though they think it is illegitimate, or making efforts to distinguish cases that, in their heart of hearts, they think are indistinguishable, or arguing that a particular position does not carry certain implications when, logically, it does. They are writing briefs.

The bit about originalism makes perfect sense to me. If someone believes: (1) that gun control laws are excellent policy, (2) that originalism is illegitimate, and (3) that the main obstacle to implementing good gun control policies is the objection of originalists, then it stands to reason that the author might want to write something called An Originalist Defense of Gun Control Laws in an effort to remove the main obstacle to his preferred policy.

It's also possible that someone could write the same piece simply because no one else had done it before--i.e., "no one's defended gun control on originalist grounds, so I'll take a crack at it, not because I believe originalism supports gun control but because 'the debate' would be enhanced by having that defense out there."

In each of those cases, though, I think the author can describe the enterprise in a way I'd call "honest." Someone who is not an originalist can say "accepting the assumptions and methods of originalists, I intend to show that these gun control laws are perfectly constitutional." Or that person could say "no one has defended this law on originalist grounds, and this article is meant to explore the question whether such a defense is possible or plausible."

Maybe what I'm getting at is that one can accept another person's assumptions that one doesn't necessarily agree with and explore what conclusions might result without being accused of being dishonest, but I think that one should be at least a little bit careful to explain precisely what the enterprise is.

To move away from constitutional stuff, is it not dishonest to "spin" the conclusions to be drawn from an empirical study? Is it not dishonest for an academic in the sciences to argue that the facts support a conclusion he doesn't believe in?

And to respond to the anecdotes,

The Justice Douglas one is bizarre. Part of me still believes that judges are fundamentally not results-oriented and try to decide cases according to neutral principles. Perhaps it's different at the Supreme Court than at the courts of appeals in that at the Supreme Court, as Judge Posner writes, there are no clear answers and the Justices have a ton of discretion. I can't decide if it's ethical or not, but I'm inclined to say not. Presumably the reason judges write opinions is to explain how they reached their decisions. Apply the anecdote to your favorite controversial Supreme Court case... if the majority reached the result they did in Bush v. Gore because they didn't want Al Gore to be president (and not because their views on the Constitution led them to that answer), they should either have said so forthrightly or simply affirmed the Florida Supreme Court... not ask the best writer you have to provide a legalistic cover story.

As for the Justice Brennan anecdote, again, maybe it's different at the Supreme Court, but I don't think judges should give their clerks marching orders to write the best opinion they can in support of result X. Of course, if Justice Brennan was persuaded by actual legal reasoning to change his mind, that restores my faith in the judiciary a bit.

To get back to Jay's example, why would someone write an article that he thought was wrong? Why do professors write in the first place if they're not willing to stand behind their own words in some sense? Maybe I'm missing something, but if it means nothing to sign your name to a piece of scholarship beyond ("I didn't plagiarize"), then it seems to me that academic writing is largely a useless exercise. There are only 24 hours in a day, and it seems to me to be a waste of time to read something that a professor wrote just for the sake of putting some words on a page.

Thanks again for the post. It's an interesting question, and my views on the answer are obviously still in flux and a bit underdeveloped, but it's very interesting to see everyone's thoughts on it.

Posted by: krs | Dec 24, 2008 2:53:38 PM

Sincerity is overrated.

Posted by: anon | Dec 24, 2008 2:53:30 PM

How does honest belief add any merit to a particular argument? (Assuming the reader has the ability to evaluate the merits and is not relying wholly on the reputation of the writer)

How could this possibly be a concern anyway? How often, outside of the apparently unavoidable instances discussed, do people really present positions or arguments that they personally do not believe in?

Is it possible that people are imagining ways that such a "performance" could embarrass others?

Posted by: anon | Dec 24, 2008 2:16:17 PM

I see this as a very interesting debate, but let me ask this:

- if an author has re-examined a position taken in a prior work, and has determined that the other position than the one originally argued would be the better one, what then? Correct the record? Write the second work and note the first? Because while I see that genuine intellectual development is important (even expected?), doesn't the questioning of a prior assumption merely point out to the tenure committee where your prior articles were weakest? If so, why/can you do it prior to tenure, in a realistic sense?

I suggest you would be crazy to do so, and that is disincentivizing an intellectual process we wish to foster.


Posted by: anon | Dec 24, 2008 1:59:11 PM

An extension of Paul Gowder's point above (though he's obviously free to disagree): imagine you thought everyone writing in law journals was lying about their views. How seriously would take articles published in law journals? Similarly, if you thought judges never (or rarely) believed what they say in judicial opinions, how seriously would you take published opinions, say, as guides for future decisions, or as warrants (or justifications) for the law?

The examples Paul (Horwitz) gives above draw much of their strength from being isolated instances. We might not care if we stumble over a persuasive argument and later learn that the author disbelieves it. The same with the Douglas and Posner examples. (Although query whether the Posner example is illustrative here, since there's no indication that Brennan disbelieved the final product. But then, on the Horwitz/Kerr view: would it matter if he didn't believe the opinion he signed? Just muddling through and ... lying through his teeth?)

The larger issue here is whether a system of communication and reason-giving would function well if we held the rational expectation that writers and speakers didn't believe what they were saying. A system set against the background of that expectation would be costly and unstable. Statements would be viewed as cheap talk -- noise without any epistemic value. We wouldn't trust the truth or accuracy of any information given to us. Perhaps we could sort through the resulting mess, but we could avoid that expense by supporting a system in which there is widespread compliance with norms of truth-telling.

Paul (Horwitz) is clearly right that an argument can be valid even if the person making it is lying. And just because an author believes an argument doesn't (usually) make that argument valid. But that doesn't make it ethical to lie. Lying has costs, some of which are on display in the adversarial system. The academic enterprise shares some of those costs, but not to the same extent, and certainly not to the extent it would if we expected that no one believed what they were saying.

Posted by: anon | Dec 24, 2008 12:04:34 PM

Fascinating. Merit v. personality/reputation seems to be the theme.

You should have to look at the merit to decide what you think of the person. And many say the person doesn't matter at all. But some merit is assumed based on reputation. We assume senior faculty have wisdom in their critiques. This puts the person above the merit - unless you also assume that junior just can't appreciate the wisdom.

Those who dislike the idea of the two articles seem to place a greater emphasis on the reputation and thus, in this case, *false merit* that will adhere to the arguments.

Posted by: Craig | Dec 24, 2008 10:32:50 AM

I have no feelings about the person who wrote the piece. Who cares? Your emotions, reasons for writings, and ear infection are unimportant. Heidegger's politics---i don't care. His Sein und Zeit--let's take a look. Hilary desire to stay with her philandering man--i don't care. Her voting record and policies---that's what matters. Simone de Beauvoir as Sartre's sexual submissive--i don't care. Her Second Sex--let's take a look. Too often scholars confuse biography with argument.

Posted by: jake | Dec 24, 2008 1:04:01 AM

Oh, and I should add to my addition that I realize perceptions of the merits are subjective: I am not imagining that I or any person has a pure insight into the objective merits. As an Oakeshottian myself, I think we're all just muddling through as best we can: the goal is to aim for the merits as best we can, even if we generally stink at the job.

Posted by: Orin Kerr | Dec 24, 2008 12:10:37 AM

Oh, and seriously, I should say that I tend to agree with Paul: My point was about what you would think of the person who wrote the pieces, not what you would think about the pieces themselves, which is I thought what Jay had in mind. Arguments stand on their own; who wrote them or why isn't that important. Also, I agree with Paul on this:

"To that end, I would suggest that law review authors (too) frequently make compelling arguments that they do not believe in -- using a methodology (say, originalism) even though they think it is illegitimate, or making efforts to distinguish cases that, in their heart of hearts, they think are indistinguishable, or arguing that a particular position does not carry certain implications when, logically, it does. They *are* writing briefs."

Yup. In my view, being a law professor means never having to make an argument you don't actually agree with. It is quite a luxury to go where the merits take you.

Posted by: Orin Kerr | Dec 24, 2008 12:04:34 AM

"Prized commenter"? Suh-weet!

Posted by: Orin Kerr | Dec 23, 2008 11:55:38 PM

You should be focused on the argument regardless of the personal affections, beliefs, or hair color of the writer. Giving weight to any of the latter is tantamount to an argumentum ad hominem.

Posted by: jake | Dec 23, 2008 10:57:08 PM

I said "in principle." In practice, I am a softie, although I don't think that's entirely a good thing; and I would rather blame senior colleagues who use their power over a junior in this manner (assuming, that is, that the changes really are bad ones; perhaps they actually are offering wisdom) than punish only the junior colleague. But the junior colleague shares some responsibility for this too, don't you think? It may well be that this kind of thing happens all the time; I can't really say for sure. But it didn't happen to me, and I also think there are gentle ways of ignoring bad advice and going one's own way.

Posted by: paul Horwitz | Dec 23, 2008 10:42:24 PM

Glad to discomfort you, Paul -- even if I thought we were on the same side. I'm wondering though if you really mean what you said. Let's say you write an article, and your tenure committee says you have to change it in X, Y, and Z way. You think X, Y, and Z are stupid and would make your piece a worse one in light of what you were trying to accomplish. Do you make the changes, or take your chances? Would adding X, Y, and Z simply be chasing petty coin? It seems to me that this kind of thing happens every year, across the professorial land. And most people make the required changes. Should we vote against them, for doing what their seniors told them to do? I know it's a lesser problem than the one I posed earlier. But it's a question of degree, not kind, isn't it?

Posted by: Vladimir | Dec 23, 2008 8:58:25 PM

I'm more bothered by the idea of writing an article that expresses an position the author actually does not believe. I think my take is different than Paul's because I don't buy the analogy between lawyers' and academics' work. When a lawyer submits a brief, they are not professing the truth of the arguments it contains, but rather are saying "This is the best argument that can be made on behalf of my client." They may believe it, or not, but either way, that is not the representation they make to the court when they submit it. For academics, there are no formal rules of professional responsibility, so it's harder to parse out the social and professional meaning of producing an article, but from my perspective at least, when I submit an article I'm saying "Here's a thesis that I think it is right, along with my explanation for why I think so." This is different; it's not a representation in the interest of a third party but an indication that you've come up with something that is worth saying, and also--for lack of a less loaded word--right.

So. If this account of the social/professional meaning of producing law articles is right, then there's something very problematic about writing an article with which you disagree. In a nutshell, it's dishonest to profess to the world the validity of a thesis that you think is wrong.

Of course, this all presumes that you think there is some notion of absolute right and wrong (or better and worse arguments in favor of competing theses), and there's certainly much disagreement about that. As Vlad interestingly points out, there could be truth value in writing two mutually contradictory articles if you're a crit and the point you are seeking to make is that law is indeterminate. I also understand the professional self-interest explanation that says "write something safe regardless of whether you think it's right, then when you get tenure, go nuts" (didn't Richard Delgado own up to doing something like this?). But that's a justification for being dishonest, not an argument that there's no dishonesty at play.

What I'm saying here is certainly consistent with Paul's point. If you read law articles because you think there is some truth out there that we can all get closer to by engaging in the project of scholarship, then certainly the notion that people commonly make claims that they think are wrong undermines the value of the endeavor.

Posted by: Dave | Dec 23, 2008 4:57:40 PM

If the logic is good, then I personally think the concept is a delightful one. Scholarly writing should explore, and just because there are two different explorations, especially in legal analysis, isn't a bad thing -- it is perhaps a better thing that writing two articles that agreed.

Posted by: Stephen M (Ethesis) | Dec 23, 2008 4:50:42 PM

Nobody publishes everything they know and have thought about the topic of a publication. It's just impossible. (Unless you're Rawls.) One always eliminates obviously implausible arguments, minor supporting points that aren't even important enough to make it into a footnote, etc.

Given this, a Bayesian member of the author's intellectual community would update his probability in any given proposition P based on two things: 1) the strength of the argument for P as presented, and 2) his estimate of the general ability and thoughtfulness of the author to have considered all those things that don't show up in the publication.

Dishonest arguments undermine 2), and hence the epistemic utility of any given argument.

Similar points can be made for the utility of articles to another audience -- people who aren't members of the author's intellectual community, for example, people from other disciplines or laypeople. For those people, they're going to have to place a lot of weight on things other than the argument. If Leonard Susskind publishes something about cosmology, I'm going to put a lot of probability in it whether or not I understand the argument, because it's fucking Leonard Susskind, man. And I'll be rational to do so ... unless Susskind turns around and publishes something else in the other side, making it impossible for me to update on what he says at all.

Posted by: Paul Gowder | Dec 23, 2008 3:15:57 PM

Vladimir, thanks for the interesting comment. It's funny, but I agree on your first example but am having a hard time swallowing the second one. I don't know whether that was your intention, and I don't know whether it means I am in conflict with my own post. Perhaps the difference is that I find perversity a perfectly fine motive and one that is not in direct conflict with academic integrity, but as an academic I reject as lacking in academic integrity someone who alters what he or she would otherwise write merely to win the petty coin of tenure. However often it may go on, I worry that such individuals will show no more independence of mind once tenure is in hand, and would in principle support denying them tenure on that basis. Still, it is a fair question to ask -- if the piece this calculating person writes is itself a good one, whether or not it is also insincere, should we look askance at the piece itself, even if we find its author lacking in courage? Maybe others can tell me whether there's a way to disapprove of this individual consistent with my arguments above. In any case, thanks for posing the question.

Posted by: Paul Horwitz | Dec 23, 2008 3:09:14 PM

Paul, I think you're spot on to question those who question the ethics of writing a piece you don't believe in. Why, indeed, should it matter what you subjectively believe, as long as the piece is well done? I can think of two perfectly good reasons why one would write a piece that is, deep down, from the author's point of view, either contradictory or a lie. First, doing what Jay did would be a perfectly valid crit move -- I'm surprised someone didn't try it in the 80s, in fact -- to emphasize the indeterminacy of the topic?

Second, suppose that you, a junior faculty member, believe X. Most of your faculty, however, despise X and believe in rival theory/ideology Y. Would there be anything wrong with writing a piece using ideology or methodology Y and vigorously condemning X in the process? Your motivation would be to get tenure, so you could then write what you truly believe later. It seems to me that the tenure system not only allows, but in fact compels such behavior, and that there is nothing unethical about it. Here's an example. Let's say that you are an economist. Your scholarly goal is to revive say, the theories of John Kenneth Galbraith or maybe Karl Marx. However, you have noticed that no serious economics department would hire someone who rejects neo-classical, mathematical economics in favor of Galbraith's or Marx's theories. (For convenience, let's assume that this is happening a good year ago, before every economists started writing mash notes to Keynes.) So you embrace positions and methods that your department will like, and give you tenure on the basis of, so that you can then lead the Galbraith or Marx revival once the University President has signed your lifetime contract. Surely no one would think that unethical, right? Isn't that one of the ways that new fields and movements get going, under the current tenure system we have?

Posted by: Vladimir | Dec 23, 2008 2:51:48 PM

Post a comment