Fairy tales for adults (long)

Jeremy Larner jeremy.larner at dunelm.org.uk
Sat Feb 15 21:18:12 PST 2003


Björn Lindberg wrote:
> Jeremy Larner <jeremy.larner at dunelm.org.uk> writes:
> 
>>Yes, but in this case there is a state which has a very large
>>possibility of occurring, precisely nothing.  The fact that that
>>didn't occur could be argued to be evidence against the model
> 
> 
> Why do you think that 'nothing' would have a large probability of
> occuring? I would say the opposite. The laws of thermodynamics states
> that everything moves towards increased entropy, which pretty much
> guarantees that 'something' will happen, rather than nothing.

Sorry, it depends which model we're talking about.  If we're talking 
about the chance of the universe producing something interesting based 
on random settings of the laws of physics (which I think is the model of 
the statistic I quoted) then the two most likely states following a big 
bang are that the universe contracts upon itself shortly after 
exploding, or that it expands too quickly for gravity to pull it back 
together, and spreads itself over two wide an area for the essential 
elements of life to come about.  At least, that's my (possibly wrong) 
understanding of it.  From what I understand, the possibility of not 
doing either of those is hundreds of orders of magnitude smaller than 
doing either of them.  My point was that it's not really a model where 
there are a huge number of possible outcomes, all with a similar (very 
small) probability of occurrence.

When I said nothing, I was actually thinking more about models which 
have tried to predict the possibility of DNA forming from random 
collision of protein molecules.  In that model the probability of 
'nothing' (i.e. DNA not forming) is much higher than what actually 
happened (if indeed it happened at all...)

Apologies, I was unclear.

>>
>>I can do my best.  I sounds like you have a far better idea of what
>>I'm talking about than I do, so don't be to surprised if I'm way off.
>>I'm applying things I know a little about (econometrics) to things I
>>know almost nothing about (theory of origins).  If I'm trying to build
>>a model relating inflation to currency rates, I formulate a model,
>>assume my model is correct, and try to fit it to historical data.  If
>>my statistical analysis tells me that with my model, the data I get
>>fairly probable, then I'm very happy.  If it tells me that it might
>>happen, but it's not likely, then I have no real idea about the
>>validity of my model.  If it tells me that, given my model, the
>>historical data has an extremely small probability of happening, then
>>I throw out the model and start again.
> 
> 
> I see what you mean. I haven't seen any evidence suggesting that the
> current established scientific theories have a low probability of
> being true though. The scientific method (which Richard Lightman has
> explained very nicely in several of his posts) gurantees that if a
> 'better' theory is discovered, one which better matches data and one
> which makes better prediction, that one will replace the old and
> 'poor' theory.

I think that depends what you mean by evidence.  Because we are talking 
about models and theories which can't really be tested in the 
laboratory, in a strictly scientific sense, there is no evidence for or 
against evolution (to the best of my knowledge).  It is generally 
recognized as the best theory to explain what happened.  I'm sure that 
if a better one was devised, people would start using it.  In a 
non-strictly scientific sense, my understanding is that there is a 
reasonable amount of evidence against it, and certainly some prominent 
biologists have recently written criticisms of it, based on this 
evidence, but in the scientific community it still seems widely accepted.

> 
>>Certainly if I have two
>>models, and one seems to explain the data very well, and the other has
>>a very small chance of producing the data I can observe, I'll
>>generally choose the first.
> 
> 
> That is reasonable. I can assure you that scientists do just that --
> they choose the best theory.
> 
> 
>>My point (originally) was that for
>>someone who believes in God, that's effectively the choice they have.
>>If you don't believe in God, evolution becomes the best available
>>model, and so most people go with it.
> 
> 
> True, but then the believer puts his faith and beliefs before the
> science, which is not a good scientist should do. The non-believer
> arrives at evolution by going at it with an open mind. Science shows
> no evidence of a god (or several), so god cannot be used in scientific
> models. God has no place in science.

It's all assumptions though, isn't it?  Scientists have no evidence that 
there senses present them with an accurate view of the real world (if 
such a thing exists at all).  Yet they assume that in their experiments. 
  You would probably find that believers would say they 'go at it' with 
a far more open mind than the scientists do, because they are prepared 
to allow for the existence of God, which the scientists don't (or won't).


>>>>As far as I am aware, this is one of the reasons why
>>>>so many biologists are moving away from evolution as a theory (another
>>>>being the major gaps in the fossil record...)
>>>
>>>I have heard people say that before, but I have never seen it. I work
>>>at a bioinformatics department at a university, and I can't think of
>>>anybody working there who has any trouble accepting the current theory
>>>of evolution.
>>>Björn
>>>
>>
>>Ok, I'm just going on what I've read.  I could well be wrong.
> 
> 
> I have read it too, but I have never seen any evidence of it. I have
> most often seen it mentioned by people who are critizing evolution
> thus have an agenda (not referring to you here). That is why I tend to
> be sceptic about it until I see some evidence.
> 
> 
> Björn

Yes that's the only case that I've seen it in too, so you are probably 
right to be skeptical.

Jeremy



-- 
Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-chat' in the subject header of the message



More information about the lfs-chat mailing list