Commentary on the so-called Creation/Evolution/Intelligent Design Debate and Right-Wing nuttery in general - and please ignore the typos (I make lots!)

Monday, August 07, 2006

Is writing a computer program really a scientific theory???

If you have no understanding of what an actual scientific theory is, and all you did for a living was write computer programs and you have an odd, almost obsessive vendetta against evolution and all things in the scientific mainstream, then I guess so...


FreezBee said...

Or turning the question around, is a scientific theory really like writing a computer program?

To some extent, I would say "yes". Often a computer program models part of the world just as a scientific theory would. For natural science, a scientific theory is not considered to be part of the natural world itself, whereas many computer programs are integrated into the world they model and therefore actually are a part of that world. In psychology and sociology, we might also say that a scientific theory becomes integrated into its own object world and therefore actually should explain itself as well. This is even more clear with economics and politics.

Theories in natural do not directly affect the natural world, but they affect how we view that world and how we interact with it, so there isn't a too clear subject-object distinction there either.

Doppelganger said...

Hi Freezbee,

I'm not sure I agree entirely. In this discussion (re: anti-Darwinist Bergerson's use of terminology and application of concepts), a computer program/model, in my view, is merely a tool with which a 'standard' theory can be tested or otherwise analyzed. The program/model is no more a 'theory' on its own than would be a mathematical equation used in population genetics. The equation is not the theory, it is a tool through which theory is quantified.

I think Bergerson's use of the equity of 'scientific theory' with what he refers to as 'hard science predictive theories' is fairly demonstrative of his personal subjective opinion that all endeavors, in order to be considered 'scientific', must perform and operate under the same criteria that producing computer programs/models do.

As a programmer/modeller, Bergerson's use of 'predictive theories' is irrelevant to nearly all other fields of science, for he has claimed, for example, that his 'theories' do not and should not be required to actually explain anything.

FreezBee said...

Hi doppelganger,

Ok, the intention of my post was merely to say that a computer program is analogous to a scientific theory in the sense that both may interact with the world they describe.

As for Bergerson, you are right that a scientific theory is supposed to explain something. For instance, Charles Darwin's theory of evolution was supposed to explain, why species are so well adapted to their environments - not by design, but by natural selection.

In that way, a computer program doesn't explain anything. It may be written based on a theory and used for prediction, but it isn't the theory itself.

However, assume we have to theories with exactly the same predictions, but different explanations. Why choose the one over the other?

What Bergerson might be referring to could be the more-or-less metaphysical assumptions in theories. Two theories with the same predictions can only differ in something that is not observable and therefore metaphysical, right?

I cannot say that's what Bergerson means, but he may not be completely out in the dark.