A Philosophical Inquiry

On another blog, someone recommended this piece here.  It argues that it would be very difficult for a modern nation to undergo a national mobilization on the scale of WW2.

I wrote the following, not about the piece, but about the philosophy of science:

I don't like the Elhefnawy article. The reason isn't the thesis: he may, in fact, be correct. The reason is that he pulls a rhetorical bait-and-switch in which he assumes his conclusion and dresses up a series of hypotheses as a logical argument building to a conclusion.

I really hate this form of rhetoric, and it's one that I see far too much of in the military journals that I read.

A convincing way to argue --- and one that is far more likely to lead to, well, correct arguments --- is to argue like a social scientist, and not like a lawyer.

(1) present a series of logical if-then statements (a "theory");

(2) falsify them ("if I see this out there in the world, then the theory must be false);

(3) show that the data does not contradict them ("I don't see this out there in the world");

     --and with more difficulty--

(4) divide the world into A and not-A ("either my theory is correct or this counter-theory is correct");

(5) falsify the counter-theory ("if I see that out in the world, then the counter-theory must be incorrect");

(6) show that that is true. So A is not contradicted, and not-A is contradicted --- and you have a conclusion.

Better yet, you have a conclusion tied to a set of explicit assumptions, meaning you have at least some idea of the circumstances under which it does not apply. And even more better yet, you have a conclusion that can be shown to be wrong if your facts are wrong, or new facts come to light, or the facts change.

I actually think that Elhefnawy is wrong (he makes one very simple false comparison --- although I suspect that he understands it and chose deliberately to obscure it --- and his thesis shifts halfway through the article) but that's not the particular point I'm trying to make.

My point is that he not only didn't prove his point, he didn't really try ... and this problem affects a lot of military writing.

Abstracting from the point about military writing, am I right, or is the above horse manure?

Comments

[Carlos]

Um. Your third quoted paragraph has a conceptual flaw: "A convincing way to argue [...] is to argue like a social scientist, and not like a lawyer."

[Noel Maurer]

I am not smart enough to see it. So here are two guesses:

The missing "to whom," to which I'm implicitly saying, "to everyone, in a perfect world." Including in a court of law, although I understand the practical reasons why that isn't feasible.

The missing "about social reality," in which case, yeah.

But if it's something else, like I said, I'm not quick enough to see it.

[Carlos]

The purpose of convincing isn't to make a correct argument. The purpose of convincing is to make someone else believe your conclusion. Accuracy, quality, and logical rigor of the argument are often incidental or even detrimental to this purpose.

(I could make a crack about lawyers here, but that's half our readership.)

Unless you're seeking to convince other social scientists, a social scientific argument may not be the most efficient way to do it.

In this particular case, I would show the original writer's false comparison, and play up his rhetorical shift. *Then* I'd bring in the logical argument.

[Noel Maurer]

Yes, I misspoke. That's because, I suppose, at the end of the day I really am a social scientist.

My point is indeed about "finding the correct argument as best we can," not "convincing people that my argument is correct regardless of the underlying reality."

In other words, I should have said, "The way to make an argument convincing to people who care about the underlying reality is ..."

[Carlos]

Not if you wanted to convince them! People might passionately care about the underlying reality, and not be familiar with social science methodology.

The logic looks easy. It's not. You're using a Boolean framework, but that wasn't invented until the 1850s. Falsifiability wasn't even a criterion until 1934. (Somehow science, even social science, got done before this.)

There's also the unspoken assumption that people will be able to recognize and judge the quality of confirming or falsifying observations. A single damning fact which falsifies a theory to *you* might look like argument by anecdote to someone of goodwill who doesn't understand the methodology behind that judgment.

In the same way, the division between A and not-A is also often a matter of judgment. Hard cases, edge cases, outliers.

So, what looks like a simple six-point plan is actually replete at the microlevel with a need for a lot of specialized skills.

Upshot: in practice -- and science is a social practice (and policy even more so) -- using this method of argument limits the participants in an argument to a very small group of the people who might be interested in the argument -- people who might, for instance, have falsifying observations of their own. I myself consider this socially problematic.

But your six-point plan is philosophically correct.

[lala]

hey, Carlos, your first & second offerings are, yk, ++zzz, and then all of a sudden, zzz to 120 in the third. think about that (haha, are you 6? 8?), and I will tell you what I think later (not holding out, don't exactly know now).

[Michael]

Noel, are you back in T&T? That little map thingy they've put on the front page shows me in Puerto Rico, somebody (whose name I'm forgetting) in Bermuda, and somebody down there in the Antilles, looks like T&T to me, but given that the map dots are, like 1500 miles on a side, it's hard to tell for sure.

Just curious. And no, not germane to this post, because frankly, I don't understand diddly about what all you people here do. And this is an unfamiliar condition for me. So I try to ignore it.

[Noel Maurer]

Hey, Michael: Nope, I'm currently in Boston. Welcome, T&T readers!

As for what Carlos and I do, the post above is probably the easiest way of describing it in the broadest possible terms. We find problems, no, call them puzzles, and then use the above technique to come up with a plausible explanation.

As Carlos pointed out, points (3) and (4) can be a judgment call ... point (4) more than (3), but the principle holds.

If that doesn't help, I'm happy to go on at length. Heck, if people are interested, I'll start posting about work-in-progress. Bueller?

[Michael]

Always interested in work in progress on things I don't understand written by people whose style I like. So that's one vote yea.