Some friends and I got interested in the controversy about Nancy MacLean‘s book *Democracy in Chains: The Deep History of the Radical Right’s Stealth Plan for America. If you’re a libertarian, historian or political scientist, then you’ve heard about this. If not, the short version is that it’s a typical entry in the Koch Brothers Conspiracy Theory genre, arguing that Charles Koch has been trying for decades to slip policies past the electorate that it would never vote for if they were presented honestly. The book’s particular contribution to this genre is painting James Buchanan and Tyler Cowen as complicit. Critics are mostly liberterians who take exception to the extremely uncharitable view of Buchanan and Cowen the book takes, as well as the general weakness of the evidence for the case against them. Maclean uses the full bevy of misleading debate team tricks: insinuation, guilt by association, selective quotation, context dropping, and strawmanning. (For the full case, read David Bernstein, Michael Munger, Russ Roberts, and Phil Magness)
The defenses against these criticisms have tended to seem a little disingenuous. Two in particular that my friends and I discussed were one from John Jackson and another from Andrew Seal. You can follow the links and see for yourself, but both Seal and Jackson – Jackson much more than Seal – use defenses that they’d hardly endorse for a book they disagreed with.
Their basic counter is that libertarian critics are misreading MacLean, and it centers mostly around how to take the charge that Buchanan and others saw an opportunity in popular Southern opposition to Brown v. Board of Education to pick up some fellow travellers for their free-market causes. MacLean’s overall point seems to be that leveraging opposition to a case like Brown – now universally seen as a critical turning point in the Civil Rights movement – for political gain amounts to selling black people down the river (literally?). Libertarians object that MacLean is trying to paint Buchanan as a crypto-segregationist by baseless insinuation. In Jackson and Seal’s version, MacLean’s point is that it doesn’t matter whether he’s a crypto-segregationist, building on opposition to an important case like Brown – even for totally unrelated purposes – is already sin enough.
To some degree, the two sides really are talking past each other. But an honest read of both pieces – along with the back-and-forth in the comment sections – can’t help but paint the libertarians as the more honest side in the exchange. Basic objections go unanswered, fended off with technical disqualifiers, MacLean’s underhanded techniques go unacknowledged (Jackson) or hand-waved away as beside the point (Seal), and Seal even compounds MacLean’s egregious selective quotation of Cowen by excising and distorting its context in his own way. Damningly, neither Jackson nor Seal seems to care about any of the multitude of positive mainstream media reviews of MacLean that interpret her in precisely the way Munger and Bernstein and Magness do. When George Monbiot sees MacLean as making Buchanan the central figure in a probably racist conspiracy to implement "totalitarian capitalism," for example, neither Jackson nor Seal feels the need to write any posts correcting him. One assumes this is because neither Monbiot nor Magness misread MacLean at all: MacLean said exactly what they both claim, and Jackson’s objection isn’t to Magness’ interpretation but rather to its negative spin. It’s a depressing irony in the context of a book that’s grounded in the idea that failing to lay all one’s cards on the table is a Crime Against Democracy.
Which is why I was rather surprised to hear one of my friends describe Jackson and Seal as trying to read MacLean "as charitably as possible." Is that what charity means? To me, they looked more like defense attorneys, ethically comfortable with any legal evasion that served to weaken the State’s case against their client. And yet, as a fan of Arnold Kling, "charity" has a lot of persuasive power for me. So it’s worth thinking about: what exactly is it I value in "the Principle of Charity?"
I think it means: "resolve any ambiguity in a way favorable to your target." Further, I think it’s useful to distinguish charity from Less Wrong’s steelmanning and what I’ll call "running interference."
"Steelmanning" is meant to be a kind of opposite to strawmanning. If a straw man is an informal fallacy where you give the impression of refuting someone’s argument by arguing against version of it so weak or caricatured that it just isn’t the same point your opponent is trying to make, then a steel man is … well, the "let me fix that for you" version. It’s where you take someone’s bad argument and make it better for them, and then proceed to discuss or argue against that. What they both have in common is that you’re not really addressing the issue on the table. Once you steelman someone’s argument, it’s in some important sense no longer their argument.
"Running interference" is when you’re trying to obscure someone else’s argument, because if presented accurately it would function like a straw man for the opponent. It’s: someone makes exactly the argument I can comfortably knock down, so his allies try to distort the argument he made to prevent my doing that. What makes this different from steelmanning is that it’s solely a PR move. Steelmanning is about getting at the truth. You do it to compensate for incidental communicative (or cognitive) weaknesses. Running interference wants an argument to succeed whether or not it’s sound. It tries to help an argument win – for some value of "win" – not by improving it but merely by shielding it from criticism.
Neither steelmanning nor running interference is a form of charity. Steelmanning because it replaces the argument with another, superior, one. Running interference because it prevents engagement with the argument. In different ways and for different motives, both sideline the original argument and engage an opponent with something else.
Charity, by contrast, tries to get at the "real" argument by filtering out noise. It can be hard to distinguish from steelmanning, but the difference is important. When we’re being charitable to someone’s argument, we’re still engaging with that person’s argument, as that person (we surmise) intended to make it, or would have made it under better circumstances.
To my eyes, it looks like what Seal and Jackson are doing isn’t that. What they’re doing is running interference when they should be steelmanning. Magness and Bernstein aren’t wrong that MacLean lacks sources for a lot of her claims. They aren’t wrong that she’s doctored quotes to obscure their original meaning. They aren’t wrong that she’s highly selective about who she presents as influences on Buchanan – including some based purely on speculation whilst leaving out others whose influence is well-documented – and that the most likely explanation for this otherwise puzzling selectivity is that she’s trying to insinuate a racism she can’t document. They aren’t wrong that she ignores all kinds of exculpatory evidence – including evidence that Buchanan’s opposition to Brown was motivated in part by a sincere belief that Brown would damage educational opportunities for blacks. And they certainly aren’t wrong that she adopts a cheap, sensationalistic presentational approach that discourages, rather than encourages, critical engagement with the subject by poisoning the well. The case against MacLean’s particular book, in other words, is pretty solid. Jackson and Seal and similar should therefore stop trying to defend it as such.
If they want to push back at libertarian criticism, steelmanning would be a more effective approach. Just admit that the book itself is unfortunate, and then move on to note that that doesn’t mean that MacLean hasn’t identified a rich vein for research. Her book is what the statisitcs crowd would call a pilot study: it doesn’t prove anything on its own, but it does provide a prima facie case that there’s something worth investigating here. MacLean isn’t the person to do that investigating, but that doesn’t mean no one should. In fact, several people should, both those suspicious that MacLean may be right, and those convinced she is wrong. That’s how scholarship is, after all, meant to work.