This is a form of assessment I use that I think it makes sense to call:
If God is not taken as given, then any theoretical idea has to hold true to darwinistic principles or else it is false by definition.
Explanation of terms:
This still leaves a lot of wriggle-room
in how we apply darwinistic principles to a theory. It might be as simple as saying any theory that is also a successful meme
because it encourages its own survival.
I think it is more productive to apply the darwinistic principles to the subject of the theory rather than to the theory itself. Otherwise we could result in true theories that are scientifically false.2
- Does the action encouraged lead to the continued or increased well-being of that which processes or executes the action?
- Does the description aid or hinder that which assesses it?
Note by this application of our Razor there is still no requirement regarding the accuracy of the content of the theory, only its outcome. This new application still implicitly contains the previous one. It could rule out any unsuccessful meme, by virtue of such a poor meme's lack of any outcome at all.
Finally we can further apply Darwin's Razor to include the accuracy of the content of a theory. Just as life has a dual role: Immediate survival and relatively infinite survival; So too, theories need to be viable in the short run and in the long run. If the theory, by its nature, prevents the development of derivitive theories, it has no survivability. Such a theory will die a big crunch as those who find use for the theory dwindle, or a heat death as the theory becomes so prevalent that it is taken for granted and forgotten. Example:
- The world is flat. True, the Flat Earth Society exists, but the theory that the world is flat has suffered a Big Crunch, brought on by the inaccuracy of its content preventing the derivation of new theories.
- Different things are sometimes different colors. This could have been a very deep theory at one time, but now it is taken so much for granted that no such theory really exists. Such a theory could even be very true, for the most simplistic vision algorithms, but for the most part it fails Darwin's Razor.
My goal in formulating Darwin's Razor:
To be able to quickly and simply assess the overall relevance of a theory. Through application of Darwin's Razor I can eliminate many theories without having to delve too deeply into their own individual subtleties.
Inevitably someone will apply Darwin's Razor to Darwin's Razor, and rather than take the first swing at it, I will acknowledge that there are many ways to disprove Darwin's Razor, but would the disproof survive Darwin's Razor?3
1. By unfounded I do not mean that there is no basis, but that there is no undeniable basis. There is no infinitely concrete foundation upon which to base such a claim. Unfounded as I use it here means that there might be reasonable doubt. (This is not a personal belief, it is a definition for the purpose of this wu.)
2. An example would be: People are good at generating random numbers. Most people think they can, so it is a fairly accepted belief. Meme-wise it is doing just fine. However, studies show that we are very bad random number generators in practice.
3. I couldn't resist that last point, it was too amusing not to type once I had thought it.
(For an illustration of applying Darwin's Razor, see my (unpopular) wu on the Saving the Environment, One (Fewer) Child at a Time
node. I wrote it prior to clarifying how I test theories for relevance but I think it does a fairly good job of illustrating what I mean.)