On 03 Aug 2012, at 17:34, John Clark wrote:

On Fri, Aug 3, 2012  Bruno Marchal <marc...@ulb.ac.be> wrote:

> Define "God"

The God I don't believe in is a omniscient omnipotent being who created the universe. If you define God, as so many fans of the word but not the idea do, as "a force greater than myself" then I am a devout believer because I believe in gravity, electromagnetism, and the strong nuclear force. I believe in bulldozers too.

I knew you were a believer. Good to acknowledge the fact. Indeed, if you believe in primary force and primary gravity, you are an aristotelian believer, like christians and unlike many platonists. I am agnostic on this, but comp itself is atheist with respect of those aristotelian gods. Of course I already knew that because you did say that you are atheists in the christian sense of the word.

> "theology"

The study of something that does not exist.

I don't know about theologians interested in square circles or unicorns. That definition is much to vast.

> and "crackpot idea".

Examples work better than definitions in this case, in most cases actually. There are 3 types:

1) A minor crackpot is someone who works very hard on a problem and produces absolutely nothing of value, and there is little or no hope of him or anybody else doing better in the immediate future. Part of genius is knowing what problems have the potential to be solved with existing intellectual tools and which do not. In 1859 Darwin realized there was no hope of him figuring out how life started, but if he worked very hard he might figure out the origin of species. And he did, and he left the origin of life to later generations, Darwin knew that if he tried in his day he'd just be spinning his wheels. This type of crackpot is the most interesting and in some ways is almost heroic, but at the end of the day they are just wasting their time. You might even say that Einstein turned into this sort of crackpot during the last 20 years of his life with his doomed attempts to develop a unified field theory uniting electromagnetism and gravity, if he had died in 1935 instead of 1955 physics would have been changed very little despite the herculean amount of work he put in during those two decades.

2) A mid-level crackpot is someone who advocates ideas that have already been proven wrong.

3) The least interesting crackpot is the major crackpot, he advocates "ideas" that are so bad they are not even wrong.

All that is a bit cliché. The "not even wrong" is also an easy way to dismiss an argument without taking the time to study it. This does not mean that in some case a proposition cannot be indeed "not even wrong".

> There is no recipe for intelligence.

Prove that and you will have made a major advance in the field.

See "Conscience et mecanisme". It is easy (but in french). Of course it depends on accepting some definitions, and it needs some background in recursion theory/computer science.

> Only for domain competence.

OK, so give me a recipe for a competent mind in the domain of understanding how biology works, or meteorology, or how to write funny jokes.

Sorry, I have been unclear. There are no general recipe for arbitrary competence in arbitrary domain.

> Intelligence can "diagonalize" again all recipes.

I don't see how the diagonal argument can work if you include things like induction, statistical laws, and if X and Y then PROBABLY Z. I don't know the recipe for intelligence but I am certain these things are some of the ingredients.

Diagonalization works well in the inductive inference field. For example Royer extended the whole speed-up theorem (which works for proof (Gödel) and computations (Blum)) to inference inductive, probabilistic or not. The construct are based on diagonalization.

You seem to be not aware of the field of computational learning theory, which is as much based on diagonalization than provability and computability theory.



You received this message because you are subscribed to the Google Groups 
"Everything List" group.
To post to this group, send email to everything-list@googlegroups.com.
To unsubscribe from this group, send email to 
For more options, visit this group at 

Reply via email to