A Structured Argument for Traditions, and Notes on the Form

An argument for traditions:

  1. The space of possible human cultural practices is large.

  2. The subspace of 1 which is well tuned, robust, compatible with human flourishing, and generally good is small.

  3. Humans generally have limited ability to distinguish between good cultural practices and bad ones without direct experience.

  4. Therefore, by 1, 2, and 3, cultural practices constructed ex-nihilo are not likely to be good.

  5. Given experience with a specific bad cultural practice, humans are generally able to critique it and improve upon it.

  6. Given experience with a specific good cultural practice, humans will generally believe it to be good.

  7. Barring cultural shocks, humans will mostly successfully pass on to their descendants those cultural practices which they believe to be good.

  8. Therefore, by 5, 6, and 7, barring cultural shocks, bad cultural practices that cause specific bad experience will tend to wipe themselves out across generations.

  9. Therefore, by 3, 4, and 8, a mature and established cultural practice is likely to be better than a new one, in ways that are not obvious before experience.

  10. Therefore, by 9, we should bias our lifestyles towards mature cultural practices, and away from newly invented ones, even sometimes against our own judgement. QED

This is a structured argument. The premises and inferences are all made explicit, labeled, and stripped of confusing rhetorical device. This lends an appearance of formality, but we have to be careful to not mistake this kind of argument for a formal proof. It's not. Any one of the premises might be nonsense, and any of the inferences may not follow. Arguments cast in this form are often presented as being inescapable logic, when they are just as often incoherent nonsense. The above argument is not inherently all that different from the more compact "traditions have proved themselves over the generations, so are more likely to be good than newly made up practices", just taken apart and labeled for easy reference and analysis.

For example, instead of a vague sense of "that doesn't quite make sense", the structured form invites the criticism that "#3 was true in the past, but the modern intellectual environment makes us more able to design cultural practices than our predecessors". We may then proceed to restructure the argument to see what that actually implies, or to argue whether that is actually true. I don't think it is true, but at least we have a more specific disagreement.

Note that if we disbelieve #3, that does not imply a simple "humans are naturally good at cultural design", but includes all statements that contradict the premise as given, including the above more nuanced objection, and "that whole way of framing it is stupid or otherwise misguided". This is again something to watch out for; the implicit frame of the premises is part of the argument that needs to be justified like the rest. This is often ignored or papered over by people trying to convince you of something.

The other thing to note is that you should not (and won't) find the above structured argument fully convincing on its own. It's deceptively phrased as a proof, but even if sound, it's not a proof. It's only one argument and one line of evidence of what should be many. It should cause a small update, possibly inspire some other lines of reasoning, and help contextualize later evidence in this direction, but not much more.

With a few caveats like the above, I think structured arguments like this are a valuable way to communicate arguments among sincere collaborators, though obviously most useful for disagreements and uncertainty. When you share most of your models, the compact form usually communicates the new idea more efficiently. But if you're going to argue about it, or check your work in more detail to make sure it makes real sense, a structured argument like the above is helpful. A structured argument is for when extra clarity is required.

The example argument above is the culmination of arguments I've about tradition with rationalists of the constructivist "but I don't see any reason not to tear down the Patriarchy and eliminate gender" variety. I hope the structured argument helps them to understand part of my reasoning. In future posts, we will explore the other empirical and theoretical justifications for specific traditions, and tradition in general.