Below is the post I was replying to, and following that is my reply.

 From: Bowen Simmons
 To: alt.books.cs-lewis
 Date: 11/12/1999
 Subject: Re:faith

 First, because nothing can formally be deduced from what is observed. The
 problem of philosophical skepticism is a thorny one. Ignoring it does not
 in any way reduce the seriousness of the challenge it poses.
  
 Second, because most observations are not useful scientifically.
 Observations that don't fit the mold of scientific acceptability are
 ignored by science because it doesn't know what to do with them. For 
 example, once you start talking about such things as "repeatability" 
 or "measurability" as being requirements, you set
 aside all non-repeatable, non-measurable observations, which happen to be
 the overwhelming majority of observations people have ever made, continue
 to make, and are ever likely to make. 
  
 > Regardless of what Hume or Berkeley or Popper may say, or Feuerabend
 > for that matter, scientists stubbornly go on doing their work, and
 > requiring observations before they think a theory is true,
  
 And this is where the faith part comes in, by setting aside the very
 reasonable objections to this method and proceeding anyway. At least
 theologians have a claim of revealed truth to buttress their belief. 
 Scientists in this matter don't even have that - in fact they have nothing 
 at all.
  
 One cannot reasonably defend a challenge to induction with defenses that
 are themselves only inductive - it is induction itself that is under
 attack. 
  
 >> I would characterize science as dealing with mechanical modeling. 
 >> What cannot be modeled mechanically is not amenable to the scientific 
 >> method.
  
 > An odd way of describing it.  _Mechanical_ models went out with a
 > whimper after the turn of the century...
  
 I only mean mechanical in the sense of predictability - that for some set
 of inputs some set of outputs must result - the view of the universe as
 machine. This can be either a strong form (fully deterministic) or somewhat 
 weaker form (quantum-mechanical) or very, very weak form (social "science"). 
  
 > Science is significant in our lives and intellects because of all the
 > observations that it predicts correctly, starting from that
 > assumption.
  
 It certainly does predict correctly, especially if you ignore all the
 times that it doesn't predict correctly, and the even more numerous times
 that it doesn't predict at all.
  
 > This is what makes it feel kind of un-arbitrary to the people who pursue it.
  
 It is arbitrary in the sense that it isn't logically necessary. There is
 no logical reason to make the assumptions that science makes.

 > Their real faith is in a universe that behaves
 > uniformly, and to some degree understandably, subject to local
 > differences that can be understood.
 >
  
 Exactly. The faith is that reality is such that the scientific method is
 capable of understanding it. This is a belief that is logically
 indefensible; to hold it is an act of faith, as we both now seem to agree.

And my response:

 From: brothers benjamin j
 To: alt.books.cs-lewis
 Date: 11/13/1999
 Subject: Re:faith

 "Bowen Simmons"  writes:
  
 >And this is where the faith part comes in, by setting aside the very
 >reasonable objections to this method and proceeding anyway. At least
 >theologians have a claim of revealed truth to buttress their belief.
 >Scientists in this matter don't even have that - in fact they have nothing
 >at all.
  
 But scientists have something better than the mere *claim* of revealed truth. 
 They have a model for interpreting the world, which *works*.  We can use our 
 observations to make new theories, and to gain new insights. 
  
 For example, we have a theory that certain materials act like semiconductors. 
 Based on this belief, we can develop transistors.  And the transistors work 
 exactly like we thought they would.  That's why we can put millions of them 
 together to form micro-processors.  And the micro-processors work just like 
 we thought they would. If some semiconductors could arbitrarily decide to 
 not follow the laws of physics, none of this would be possible.  It's not a 
 matter of pure logical induction.  It's a matter of having a model for the 
 universe which allows us to successfully interpret it. 
  
 >One cannot reasonably defend a challenge to induction with defenses that are
 >themselves only inductive - it is induction itself that is under attack.
  
 But science isn't simply induction.  It's the combination of both induction 
 and deduction.  One gathers facts and observations about the issue in dispute,
 and develops a general hypothesis.  This is induction.  But that's not where 
 the real work of science lays.
  
 Once one has a working hypothesis, one must test that hypothesis.  The 
 scientist develops control experiments, makes predictions based on his 
 hypothesis, and sees if that hypothesis works or not.  The real test of the 
 theory is in deducing correct observations from the more general hypothesis.  
 The hypothesis that fails to do this correctly is quickly abandoned. 
  
 >It certainly does predict correctly, especially if you ignore all the times
 >that it doesn't predict correctly, and the even more numerous times that it
 >doesn't predict at all.
  
 You seem to be confusing "science", that magical idea with all the answers, 
 with "science", a collection of theories made by inquisitive minds, which 
 try to explain things, and which can be modified or changed to explain them 
 better.  There's nothing many scientists would like better than to find out 
 that, for example, general relativity doesn't quite explain things to perfect 
 accuracy, and to come up with a better working model. 
  
 Some people think that science's greatest flaw is that it's sometimes wrong.  
 I think this is it's greatest strength; it is self-correcting.  Theories are 
 refined and changed until they offer the best explanation of available 
 evidence.
  
 >Exactly. The faith is that reality is such that the scientific method is
 >capable of understanding it. This is a belief that is logically
 >indefensible; to hold it is an act of faith, as we both now seem to agree.
  
 Even if reality is such that we can't understand it, it it definitely such 
 that we *think* we can understand it, and it definitely consistent in this
 sense.  Thus, the idea that reality is capable of being understood is 
 clearly *not* logically indefensible.  
  
 Cheers,
 --
 Ben Brothers 
 ---------------------------
 bbrother@ews.uiuc.edu
 www.ews.uiuc.edu/~bbrother/