On fallacies

This article is mainly intended as a placeholder for some links I've had as tabs in my browser for way too long. I'll add a few thoughts to try to connect them together. Basically the question I've been pondering is a continuation of my meanderings in On the dimensions of the noosphere. In this world awash in information with the deluge increasing seemingly exponentially, what structures will help us to find those pieces of knowledge that are true and useful? What elements of trust are needed, what processes of understanding? As all the new fact-checking websites out there attest, we no longer trust the traditional media as gatekeepers - perhaps we never should have, and the internet just makes their failings more obvious.

Facts and pure logic are essentially matters of mathematics. The physical sciences encompass a broader spectrum of checks - it is not a matter of things being wrong, but quantitatively how wrong something is, because there is always some residual error in observations or some missing element in theories or models. Discussions of physical science are conducted by human beings who themselves will make logical as well as measurement errors. A taxonomy of types of wrongness, fallacies in general, is an extremely helpful element in understanding.

I made some attempt at classifying the logical errors in Monckton's climate claims a couple of years ago - my list of types consisted of:

  • Errors of fact
  • Irrelevant conclusions and non sequiturs
  • Other errors of logic
  • Errors of interpretation or misunderstanding
  • Arguments that only work for specially selected data
  • Other arguments that have no scientific validity
  • Statements that contradict or conflict with other statements in the text

I recently discovered a site that does a much more complete job of classifying logical problems of this sort - The Fallacy Files. This includes a detailed taxonomy with the following broad classifications:

  • Formal Fallacies
    • Propositional Fallacy
    • Masked Man Fallacy
    • Probabilistic Fallacy
    • Syllogistic Fallacy
    • Quantificational Fallacy
    • Modal Fallacy
    • Bad Reasons Fallacy
  • Loaded Question
  • Informal Fallacies
    • One-Sidedness
    • Ambiguity
    • Accident
    • Appeal to Ignorance
    • Red Herring
    • Composition
    • Division
    • Non Causa Pro Causa
    • Black-or-White Fallacy
    • Vagueness
    • Begging the Question
    • Special Pleading
    • Weak Analogy

If I ever do a Monckton-style response again I could have a lot more fun with all these to choose from (Monckton himself uses quite a few of them that I lumped together under the "other" classes).

A related website I've had on my list for a while is Fallacy Detective - interestingly it seems to be geared to the "conservative Christian" set, but has some good reviews, quotes, and analyses. My favorite item there is the Mob Appeal fallacy - where the issue is mainly the performance of the speaker, not the logical reasoning, if any. Whenever I watch a talk by Christopher Monckton, Bjorn Lomborg or some of the other favorites on that side of the "debate" I feel reminded of good old Harold Hill from The Music Man. To quote from "Fallacy Detective"'s conclusions on that one:

Good arguments usually involve emotion, and they probably should, but that isn't their main purpose. If emotion is used, it should be to convey some of the intensity of the author's zeal for his subject - not to manipulate the audience. In arguments, emotion should be the slave and logical reasoning should be the master.

How do we deal with emotional appeals? Unfortunately, logical arguments do not work very well when everybody is caught up in the drama of an event. It is difficult to bring people back without resorting to similar charismatic tactics. Dash them over the head with a cold bucket of reason. Point out the loose connections, assumptions and exaggerated rhetoric. Explain the long term consequences of their actions. Sober them up.

This is essentially the Emotional Appeal (one of the Red Herring fallacies) from the Fallacy Files. But I like the connection to "Music Man" for some reason.

All this is getting into questions of how we know what we know, and language itself. I've read bits of Locke in the past on general human understanding, and Popper and Kuhn on philosophy of science - I found a book by Philip Kitcher rather enlightening on that topic about 15 years ago as well, particularly on how consensus forms from lines of evidence. There's a book by Polanyi in my bedside pile still waiting to be read, and I've long been meaning to read more of the writings of Charles S. Peirce, recently featured in Physics Today. A few years back I was particularly struck by Wittgenstein's "Tractatus", which at first seemed like a bit of a joke; but I think I understand what one of his main points was, at least for me: the tautologies that mathematics and pure logic provide aren't really interesting, what's interesting is the connections between symbols and the real world. Meaning. A few wikipedia pages I've been meaning to think about some more, to finish off this list here (but I won't try to explain the connections I'm seeing here yet):