Art Carden  

Giving Writers the Benefit of the Doubt

PRINT
Henderson on Fama, Shiller, an... Against crony capitalism, Ital...

It's been said that libertarian philosophers love Murray Rothbard as an economist and historian, but not as a philosopher. Libertarian historians love Murray Rothbard as an economist and philosopher, but not as an historian. Libertarian economists love Murray Rothbard as an historian and philosopher, but not as an economist.

I'm not picking on Rothbard at all; this is just an example. I might love what a writer says about areas outside my narrow field of expertise but dismiss what the writer says in an area where I can tell when he or she is making a mistake. This leaves me in an uncomfortable position: if Walter Writer is unreliable on economics (where I'm an expert), why should I think he is reliable on philosophy or theology (where I'm not)? If Walter Writer is an expert on philosophy or theology, then it is appropriate to give him the benefit of the doubt, but when Walter Writer is writing in a field outside his area of expertise and mine, I can't help but wonder where I should be most skeptical. If he gets the economics right, I'm more likely to believe him on philosophy or theology. If he doesn't, I'm less likely to believe him on areas outside his area of specialization.

This makes me lose sleep at night, so I'll turn it over to EconLog readers. When are you most and least willing to give a writer the benefit of the doubt?


Comments and Sharing





COMMENTS (21 to date)
Daublin writes:

I think you've nailed it, really. If you detect someone making major mistakes on stuff that you know, then you should downgrade them on the parts you don't know, too. At the very least, you can't rely on someone who is not gracious enough to admit when they just don't know.

That much said, it's worth trying to mine out from people the things they really are experts on. Even in that case, you should have some sort of reason to believe they are experts, though. Just that you enjoy what they say is not sufficient reason.

Likewise the other way around. You shouldn't dismiss someone just because you don't like their conclusion. On two separate occasions in the last few months, I have encountered a domain expert being dismissed because, "oh, they're a libertarian, of course they say such and such".

Ted Levy writes:

Art, there's an old H L Mencken quote to the effect: If you read a newspaper article about a topic in which you are an expert, you invariably find numerous and major errors. There is no reason to assume, as you read a newspaper article on a topic in which you are NOT an expert, that anything has changed.

Your point is similar, and equally well taken. Journalists are just people who, for a living, often write outside their area of expertise.

Tony N writes:

The category of error committed is an important consideration. Some errors expose bad information, others expose bad thinking. There's an important difference there, I think.

Emily writes:

I'd argue that the type of errors someone is making matters. For instance, if the strategies they seem to be using outside their area of expertise are reasonable ones, I'd vote for giving them more of the benefit of the doubt even if they happened to get things wrong.

For instance: you're referencing a paper, you don't have a statistical background, you take the word of someone who legitimately seems credible that this coefficient means a certain thing, and this turns out to be wrong. That's not great - maybe you just shouldn't be writing about this topic until you learn how to read the research. But I understand what your process was and it's not a terrible one, even though it turned out to be wrong.

By contrast, if you're directly referencing a paper but you misrepresent the study population or say there's causality when the author said in plain language that there wasn't in order to make your point, that's a bigger problem. That's not a reasonable mistake, that's you knowing better and misrepresenting something, and so there's no reason to think that won't carry over into your own field.

Greg G writes:

Excellent post Art. We are all most tolerant of errors by people we tend to agree with on most things. We should remember that more. But we probably won't.

Foobarista writes:

In a world of subject-matter experts, particularly now that the world is networked, journalistic lack of subject-matter expertise becomes rapidly apparent. This weakens the whole journalistic enterprise as currently constructed, made up as it is by people whose main skill is "journalism" and writing skills.

This same problem extends to public intellectuals in general.

While better reporting and informed analysis is a good thing, this does mean it's hard to get credible "big-picture guys" who can do high-level, cross-discipline analysis without getting bogged down in subject-matter pedantry. If you have to have 50 PhD's to be a credible "Renaissance man", we aren't going to have many of them.

Martin writes:

It depends on the field I guess. To go with your example, I think that Rothbard is a solid economist, so I will trust him on philosophy and history were I to read him.

My impression of history and philosophy however is that there is more disagreement there than that there is disagreement within economics. So how much should I trust someone who is a solid philosopher or a solid historian?

To illustrate if there are ten big questions in a field and economist agree on nine of those, while historians and philosophers agree on six and three of those questions, then a mistake in history seems less forgivable than a mistake in philosophy but more so than in economics.

The trouble though is that if you are in history and philosophy you're a worse judge of someones quality as an economist, than if you're in economics and you're judging someones competence in philosophy and history. The reason is that four out of ten or seven out of ten times your disagreement is probably due to your own bias.

Brian writes:

I don't think it matters whether someone is considered an expert in a given field. My approach is always the old Russian saying "trust but verify." If authors don't back up their claims with evidence sufficient for the typical reader to understand, I assume they're either pulling a fast one or do not properly understand what they're talking about. If they do provide the evidence, then I can evaluate the claims myself, leaving no room for doubt either way.

Tom E. Snyder writes:

What makes one an expert? A PhD? A Nobel prize? Do you really trust Krugman's economics?

MingoV writes:

I'm skeptical all the time. I never give writers the benefit of the doubt, even if they're considered to be experts on the subject matter.

However, I am less skeptical when the writer's previous works seemed to be correct. This reduced skepticism saves me some work but increases the likelihood that I'll accept incorrect conclusions.

~FR writes:

Google "Gell-Mann Amnesia Effect" as described by Michael Chrichton for this phenomenon wrt newspapers.

I believe Feynman said that outside of his area of expertise, his opinions were no better informed than those of any random person.

I think I am more likely to trust someone who takes care to state provable basic facts that underlie their argument. Then it's "falsum in uno, falsum in omnibus..."

Glen S. McGhee writes:

I am with MingoV on this.

For example, libertarian economists seem to presuppose markets and market behavior as a given rather than recognize that this is what calls for explanation.

This happens when markets are used as a way to bridge self and society. A conceptual chasm exists between them (i.e., the micro-macro problem), and wholly locating discourse on one side does an injustice to the other, particularly in modern circumstances dominated by bureaucratic administration.

Neo-institutional economics, as I understand it, addresses this problem within the context of economics the discipline, as does political economics.

Tom West writes:

In any area involving human beings, the concept of *the* right answer seems almost ludicrous. There's obviously different degrees of right-ness and wrong-ness, but I as I grow older, I assume that people who disagree with me are probably right, but with respect to a different segment of the population.

Thus I fight for what I believe is right, without believing that those who oppose me are wrong.

Lawrence D'Anna writes:

I recently read something Eliezer Yudkowsky wrote where he made some points that sound a lot like things Scott Summner might say. The fact that they agree in this area makes me think they're both more likely to be right in other areas.

Martin writes:

I have given it some more thought.

If we assume that you get pleasure from writing for an audience, but that to write well you need to collect information (i.e. research) which is costly, then the marginal benefit of writing well for an audience will equal the marginal cost of collecting the information to do so.

If you write for several audiences and audiences are like apples and beer, then you will equate the marginal benefit from writing for audience A with the marginal benefit from writing for audience B. Provided the marginal cost schedule for collecting information is one and the same, and provided how well you need to write to be able or allowed to write for an audience is the same across audiences.

In this case how well someone writes for audience A will tell you something about how well someone writes for audience B.

Change the standards per audience, allow for economies of scope and scale when it comes to collecting information, etc, and it becomes a lot more complicated to say anything at all.

UnlearningEcon writes:

I guess it depends on whether the error is substantially related to their central point.

One fairly recent example I can think of is David Graeber's book Debt: the first 5000 years. There was controversy concerning Graeber's claim that Apple was:

..founded by (mostly Republican) computer engineers who broke from IBM in Silicon Valley in the 1980s, forming little democratic circles of twenty to forty people with their laptops in each other's garages

Brad Delong has chided Graeber for this repeatedly, and also pointed out a lot of other factual errors in Graeber's book.

However, despite this, even Delong acknowledged to me that many of Graeber's central points were sound, most notably the idea that moneyless barter as envisaged by economists is a myth. Obviously the Apple story is basically irrelevant to Graeber's book, as were many of the other errors Delong highlighted. This doesn't mean they were unimportant, but on the hierarchy of disagreement, these criticisms were probably on the second level or below, hence they shouldn't be taken as cause to dismiss Graeber's work outright.

Matt Skene writes:

Most things written by academics within their fields of expertise are wrong. This is despite the fact that people in those fields are as smart as people working in other fields, spend far more time thinking about the topics in their fields, have tons of training in the nuances of evidence evaluation needed to do good work in those fields, and have lots and lots of extra knowledge about the relevant facts and the history of ideas within that field. Thinking someone without all of those advantages can make significant contributions to a field is like thinking that smart undergraduates should regularly publish valuable academic research, possibly even in areas outside their own majors. If the answer to these questions could be settled by thinking about them a bit after reading a few things in your spare time, then people in the field would have already learned it.

Anyone who has obtained expertise in a field should know that there are lots of subtle mistakes in evaluating the importance of data that amateurs are subject to, even really smart ones. There's a good chance that something sounds plausible to the writer only because they don't have enough training to avoid the problems that make it seem correct. Since I'm probably subject to the same shortcomings, the fact that it sounds good to me when I read what that person had to say isn't good reason to think it is correct. This strikes me as a very good reason not to read work outside of the field of expertise of the writer.

LD Bottorff writes:

Thank you, ~FR, for introducing me to the Gell-Mann Amnesia effect. I have been skeptical of the press ever since the 1979 reporting on Three Mile Island.

I am older now and I think I understand that reporters have to deal not just with their own biases, but with the biases of their audience. If what you write is way out of sync with common beliefs, no matter how wrong they are, you will not be a successful journalist. Add to that the constraints of time, print space, and biased editors, I'm amazed that the truth ever gets published.

Jason Brennan writes:

[Comment removed pending confirmation of email address and for name-calling. Email the webmaster@econlib.org to request restoring your comment privileges. A valid email address is required to post comments on EconLog and EconTalk.--Econlib Ed.]

Matt Skene writes:

Three points worth mentioning:

1-If it really was a comment by Jason Brennan, I'm pretty sure he's qualified to make such a judgment. He literally wrote the book on what people need to know about libertarianism.

2-Jason Brennan is a professional philosopher (and a good one), and so his evaluation of Rothbard is relevant to the worry in the original post about whether or not we should trust Rothbard in other fields.

3-The original post claimed that historians don't think much of Rothbard as a historian, philosophers don't think much of him as a philosopher, and economists don't think much of him as a economist. Assuming this is all correct, that seems to be some pretty good evidence in favor of Brennan's conclusion.

[comment edited--Econlib Ed.]

Bear writes:

I think that this can be generalized into a caution against arguments from authority or the more extreme but essentially identical adopting of a person as Messiah and his/her words as revelations of the One Truth.

Test whatever ideas you like, or dislike, against reality. "Rothbard/Rand/Marx/Marcuse said it, I believe it, that settles it" is silly, and unworthy of anyone with intelligence and education.

While I do enjoy some quotes, particularly those of a good wordsmith like Heinlein, it's important not to make them the secular equivalent of some Holy Scripture.

If we don't expect someone to tell us The One Truth, we can learn from anyone -- even if what we learn is just how wrong they are.

Comments for this entry have been closed
Return to top