🚀 go-pugleaf

Marc A. Belanger <3914MABE@vm1.ulaval.ca> says:
>In article <30vca1$8g9@Jester.CC.MsState.Edu>
>fwp@CC.MsState.Edu (Frank Peters) writes:
>>No.  Does he suggest any replacement for falsifiability?  Or does he
>>recommend that we just accept any theory that sounds reasonable without
>>testing?
>>
>I does not really argue about replacing or not falsifiability but says that
>it is not the way things go. 

I'm afraid that I don't understand this sentence.  Not the way what
things go?  Not the way theories are tested in science?

>>                                               I'd have to think about
>>whether I would call it a science.
>>
>If mathematics is not a science then half of physics (which is maths) is
>not either. But it still remains that predictions does not work in maths.

Relevance?  Washing beakers is a major part of chemistry and it isn't
a science.  Math is a tool.  The field of mathematics need not be a
science in order to be a useful tool in science.

Not, mind you, that I've decided that math isn't a science.

>>Fine.  I never claimed that linguistics is unscientific (that was Mark
>>Israel I think).  I just gave the most accepted (I believe) definition
>>of what "scientific" is.
>>--
>Sorry to have thought this.
>But I would like to add that science is not (all) predictions. 

Your example doesn't address this.  It demonstrates that all predicitions
are not science.  Not at all the same thing.

>                                                                If I get
>into a car, put the key in and turn it, the motor, if all goes well, will
>start. This is a prediction. It is not scientific (unless you place common
>sense into science). What would be scientific would be to posit some
>explanation. 

That still isn't science.  Suppose I posit elves pulling belts in response
to electrical stimulation?  And happily accept that theory without testing
it?

Science is the process of developing theories (usually called hypotheses
in text books) and then testing them.  And the mechanism for testing
them is most often falsifiability.

Theory and Test.  Test and Theory.  The rest is just details. :-)

-- 
Frank Peters  -  UNIX Systems Group Leader  -  Mississippi State University
Internet: fwp@CC.MsState.Edu  -  Phone: 601-325-7030  -  FAX: 601-325-8921
             WWW Home Page:  http://www.msstate.edu/~fwp