Jeffrey Friedman on Persuasion Strategy

From a very interesting comment thread at The Austrian Economists blog:

I've found that the very brightest Ivy League students become suddenly receptive to Austrian economics if I posit the complexity of the world, and present market mechanisms as making our cognitive task easier. When I point out that there are no equivalent political mechanisms, they are sold--on libertarianism!--but without any invocations of "rights," "liberty," and all the rest. If you start insisting on those concepts, not to mention "universal laws," Homo economicus, etc., all you do is make Austrian economics look like a cog in an ideological contraption. (I can't help but wonder sometimes whether that's all that it is.)

Share this

Ivy Leaguers

I've found that the very brightest Ivy League students become suddenly receptive to Austrian economics if I posit the complexity of the world, and present market mechanisms as making our cognitive task easier.

It's not just the very brightest Ivy Leaguers who you might want to convince. Sometimes one argument works, sometimes another. Sometimes the moral argument really is the most effective one. It depends on the audience.

Moral argument?

It might be the most effective, but it is never much of an argument.

Persuasion

The blog entry is about what does or does not persuade (some people), not what is or is not "much of an argument" in the eyes of Eelco. I'm not even sure what the latter can possibly mean, other than that it does not persuade Eelco. To which the reply is the same: it depends on the audience.

It means that a moral

It means that a moral 'argument' always rests directly on some moral assertions. Not even 645,000 words will do anything to hide them for anyone willing to see them.

In other words, a moral 'argument' always can be trivially decomposed to 'because i say so'. If you want to include that in the definition of 'argument', fine with me, but i dont think you are accomplishing anything but dilluting its meaning.

Crude caricature of moral arguments

It means that a moral 'argument' always rests directly on some moral assertions.

Perhaps. Meanwhile, a factual argument surely always rests directly on some factual assertions.

In other words, a moral 'argument' always can be trivially decomposed to 'because i say so'.

Same could be said of factual arguments, with equal reason (or lack thereof). Moral persuasion can take the following forms (don't take this as an exhaustive list):

1) Argue from moral common ground between speaker and listener.

2) Lead the listener through e.g. storytelling to gain new or reinforce old moral intuitions.

2b) A variant of (2) is to argue morality starting from simple, agreed-on examples, not necessarily stories, and generalize from those examples by e.g. analogy.

3) Persuade the listener of a validity of a moral theory, and then use that theory to derive results (e.g. utilitarianism).

4) Attack certain moral presuppositions of the listener which are standing in the way of clear thinking on a topic. You yourself may be susceptible to such an attack and may in fact have been successfully brought around (to something) by such an attack, since you think morality reduces to "because I say so", a pretty corrosive view of morality. The explication of (some) morality as veiled self-interest and/or mere arbitrary assertion can carve someone's moral views into a new shape by removing chunks here and there.

5) Challenge some of the listener's key morally relevant categories without directly questioning his core sense of morality. For example, of someone thinks of the state as somehow morally special, get him to think of the state as fundamentally a group of individuals which happened to subjugate a territory - which, of course, factually, they are. Without changing his fundamental moral views, he may nevertheless change his views about certain things in the world such as the state.

Scott agreed with me

When he noticed we agreed about something he took down his comment. But I can prove it was there. I have pictures.

Ha. Your comment said

Ha. Your comment said everything I said and more, with more flair, rendering mine pointless. So I deleted it.

Very true, all arguments

Very true, all arguments rest on some fundamental assumptions.

The interesting question is: are they shared assumption? Everybody on this side of certain institutional walls believes in fundamental things such as consistency, observation, and some other things by assumption.

Insofar as moral arguments are based on common moral assumptions, they are perfectly useful. For instance, if you are against stealing, it is fair to ask: why not be against stealing when the state does so?

However, these discussions usually revolve around different definitions of stealing, or different views on property rights. Even if the defendent isnt aware of his own underlying assumptions, you can construct any moral system by picking the right set of axioms. Hence, given skillfull enough participants to a debate, it will always be narrowed down to 'because i feel this way', and 'i feel this other way'.

'real' arguments distinguish themselves, in that they are held on common axiomatic ground. The argument is about settling which consequences follow from a given set of axioms, rather than which axioms to pick.

Baring the moral presuppositions is useful

Everybody on this side of certain institutional walls believes in fundamental things such as consistency, observation, and some other things by assumption.

In the abstract. In principle. But in the concrete, the world is complex enough, there is a sufficiently rich network of causes and effects, and the potential evidence is sufficiently hidden (an individual person can only witness a pinprick portion of the whole world), that people can easily argue on the basis of a massive body of evidence that everything evil is caused by capitalism, or by the CIA, or by the Freemasons or the Jews. Science has advanced through peculiar means: e.g., controlled laboratory experiments, where the environment can be controlled to the nth degree. People have had evidence in front of their eyes through all of history, but science has not been rapidly advancing through all of history, and even at any given point in history only a tiny fraction of a percent of humans have managed to advance science, despite an ocean of evidence being in front of everyone's eyes every second of every day of their lives.

However, these discussions usually revolve around different definitions of stealing, or different views on property rights.

If you can get people to lay bare their moral presuppositions, that's useful in itself.

Getting the definitions right

However, these discussions usually revolve around different definitions of stealing, or different views on property rights.

To continue the discussion.

A key part of the advance of science (or of any knowledge, whether or not it is labeled 'science') is getting the definitions right. Once you have the concepts of mass, of force, of velocity, and so forth, then you can start hypothesizing that force equals mass times acceleration, and so on, and then testing your hypothesis. Without the terms in which to say it, there is little you can say.

A lot of purely factual arguments revolve around, or boil down to, disagreeing definitions. For example, what is capitalism, the market, the state, terrorism, racism, democracy - a lot of key terms are defined differently by different people, and this is in no way limited to definitions about moral categories. So the supposed distinction you're making between moral categories (where different definitions abound) and factual categories (where your argument implies that everyone agrees on definitions) simply does not exist.

Definitions are important because they are key tools in making generalizations. In a sense, to define is implicitly to generalize, because it is to imply that the category being defined is a useful one, and this is to create the expectation that discoveries about one XXX are useful in making predictions about another XXX.

Here's a made-up example. Suppose you define "square" to mean anything square-shaped or anything red that's circular. Then you look at a bunch of squares (none of them happen to be red) and notice that the area is always the square of the shortest diameter. You make the quite reasonable generalization: "the area of a square is the square of the shortest diameter". Because of the way the word "square" was initially defined, the generalization implies that the "shortest diameter" (the only diameter) of red circles is the square root of the area.

This is a simple example where the problem is easy to see. But that is only because I deliberately made it so. Don't imagine that in real world arguments it is so easy to identify bad definitions.

We think that people who are merely defining terms are not actually, yet, making any assertions, and so we tend to let our guard down. But definitions are often where the most fundamental generalizations are being asserted (if only implicitly). Experience has shown me that once you let Marxists get away with defining the terms of an argument, you've made your own task ten times harder (or ten times easier if you are the Marxist).

This is a simple example

This is a simple example where the problem is easy to see. But that is only because I deliberately made it so. Don't imagine that in real world arguments it is so easy to identify bad definitions.

Im not saying its easy, but at least it might get you somewhere.

Whereas arguing natural rights with a communist isnt going to get you anywhere: he simply has different moral axioms.

We think that people who are merely defining terms are not actually, yet, making any assertions, and so we tend to let our guard down. But definitions are often where the most fundamental generalizations are being asserted (if only implicitly). Experience has shown me that once you let Marxists get away with defining the terms of an argument, you've made your own task ten times harder (or ten times easier if you are the Marxist).

Certainly. Social constructivism itself is not bullshit: its just that the people applying it have been nearly consistently full of shit.

I do certainly not intend to let marxists get away with defining the terms of an argument. I realize the most effective way to stop that is by posing a frame of your own, but at least among likeminded people we can be intellectually torough, right?

Experience is the ultimate teacher

We discover bad factual definitions when our predictions go astray. Grue and bleen turn out to be bad concepts when the switchover moment passes and grue things suddenly all become bleen and bleen things suddenly all become grue. (Google grue and bleen for background.)

I would argue that we also discover bad moral definitions when our predictions go astray. Moral categories, properly defined, have predictive power. We should probably avoid the company of murderers. If our definitions of right and wrong are bad ones, we are in danger of associating with murderers and possibly winding up dead. A lot of leftists found this out the hard way - associates of Stalin or of Pol Pot.

Thoughts

* I have found that "natural rights" arguments work well with conservatives. They already believe in natural right-ish types of things. Then it's just a matter of showing them the logical extensions of their core beliefs.

* I'll grant that just about every type of moral argument, including consequential arguments, rest on something like, "because I say so". However, everyone, even psychopaths, has some sort belief and practice about morality in their day-to-day lives that can never be proven as correct or incorrect, but also reduces to "because I say so". Everyone teaches their kids not to steal, hit, etc. We can still make an argument from consistency even if the bedrock is "because I say so":

"If we teach our kids not to steal, why is it okay for people we vote into office to steal?"

It may not convince everyone, but it's powerful.

Moral Arguing

But isn't a moral argument about what is good for you? Which ultimately is going to require you to present facts about the world?

Certainly if it is factual that hell, heaven, and god don't exist, and it is, then sitting in church every Sunday praying for good things isn't an effective way of achieving what's "good for you". That is, unless some maniac named Mohammed is out their chopping the heads off people who don't show up in the mosque. Then it might just be "good for you" to at least make an attempt at pretending you had faith.

Certainly one can approach economic issues from the "what's good for you angle". That's all moral arguing is about.

Good for whom? Various

Good for whom?

Various variants of collectivists might argue for optimizing the minimum of a 'utility vector', or minimizing its standard deviation. A utilitarian might want to maximize the l-n norm, and someone like me might reject all of these approaches.

My point being, you cna only have an argument in so far as you share some axiomatic grounds. If you swear a pink elephant is standing in front of us, yet i cannot see it, there is little we can do but agree to disagree.