[extropy-chat] Smalley, Drexler and the monster in Lake Michigan

Chris Phoenix cphoenix at best.com
Sat Dec 6 21:55:52 UTC 2003


This was CC'd to me, so I'll respond.  Remember that I won't see any
answers unless they are also CC'd to me.

The debate is all screwed up.  Each side thinks they have made a strong
case and are obviously right and the other side is obviously wrong. 
Onlookers generally have strong emotional reasons to pick one side or
the other.  The situtation is a lot more symmetrical than it appears to
either side.  It's like a messy divorce: a lot of "he said she said,"
and no one can take a step back to realize that what's really important
is that the lawyers are getting all the money.

If you want to know who's right, you have to strip away the debate and
look at the scientific claims.  Most of the trouble here comes from the
context and structure of the debate, not the scientific claims
themselves.

It's easy to reach an impasse in this discussion.  You see the claims of
cryonics, assemblers, doom-and-gloom scenarios, and you say, "These
claims are extraordinary.  Therefore I won't believe them without very
strong evidence."  But what evidence is possible?  Note that these
claims are not about scientific theory; they're about the projected
consequences of one side's understanding of the theory.  

Drexler says that we have to prepare for MNT and we should be
researching it further.  This assumes there's a reasonable chance of MNT
working.  Which, when you take a step back, is at least plausible;
there's a lot of careful theory that says it should, and no careful
theory that says it can't.  Drexler isn't (or at least shouldn't be)
insisting that all scientists agree with him.  It's just that MNT should
be given a fair study, so that if it does work we won't be caught by
surprise.  

To argue with this, a critic doesn't have to *prove* MNT can't
work--just show that it's so unlikely there's no point in wasting time
on it.  This is a practial, not a theoretical issue.  And it doesn't
seem unfair to me.  It's like an epidemiologist saying, "SARS has killed
eighty people, but reasonable extrapolation shows it could kill ten
million unless we stop it soon."  Is this putting an unfair burden of
proof on the Chinese government to open up unless they can show that
SARS isn't a threat?  No, it's a heavy burden, but it's not unfair.

Hal complains about a lack of specifics on Drexler's side, and asks,
"What 'proposed assemblers'?"  Ironically, the specifics are there, but
they're *too* specific.  There's been a substantial amount of work in
the past decade on mechanochemical systems, both the mechanics and the
chemistry.  But it's buried in a thick book with lots of equations, and
a number of academic papers from several sources.  One would have to
spend substantial effort to find the details, and more effort to
evaluate them.  I've spent a decade doing this, and it turns out that
building a mechanochemical system without "fingers" is not a big deal at
all.  That part of the claim is simply not extraordinary.  

Let me repeat that.  The claim that mechanochemistry can be a useful
manufacturing technique is not extraordinary.  That doesn't mean it's
right.  But it's fairly simple to decide whether it's right or not.  

We already know mechanochemistry works in a few cases.  The question is
whether it can be used to build useful diamondoid shapes.  So far,
there's a significant amount of careful evidence that it can, from basic
theory, chemical simulation, and experiment.  And there's no careful
evidence that it can't.  (Whatever you may think about Smalley's
argument, you must admit that he was not careful; his statements about
enzymes were shockingly wrong.)  Mechanochemistry is no stranger than
surface catalysis or flame chemistry.  It only seems weird and
extraordinary because of its association with weird insect-like
assemblers.  

On this particular claim, there should be no extremism or requirement
for extraordinary proof on either side.  It looks plausible and useful,
but hasn't been demonstrated in detail.  Fine--let's look at it more
closely.

The question may have some urgency, because if it works it may be the
foundation of a very powerful manufacturing capability with lots of
implications.  But why does that imply that Drexler has to provide
extraordinary evidence that it works?  Seems to me just the opposite.  

Actually, I know why.  If I tell you, "Lend me $10 and tomorrow I'll
give you $15," you may do it.  But if I say, "Lend me $10 and tomorrow
I'll give you $100," you know something must be wrong somewhere.  I
think this is one reason why Drexler's claims, taken in aggregate, are
not acceptable.  It probably would've been better if Drexler had cloned
himself, and Eric Drexler only talked about the science (which is pretty
mundane and should have been respectable) while Drew Ericson talked
about the policy implications if Drexler was right.  It also would've
been better if the description of assemblers did not map so closely to
the archetype of bugs (in the sense of both insect and germ).  

But that's hindsight.  The point is, the debate got off to a very
unfortunate start.  But if you want to understand the science, the only
way to do that at this point is to strip away the debate history and
just look at the scientific claims.  Smalley says that precise
positional chemistry can only work underwater.  Is he right?  No. 
Meanwhile, Drexler says NEMS can do carbon deposition on diamond to
build NEMS.  Is he right?  So far, the evidence is on his side.  Let's
go get more evidence.

Chris

--
Chris Phoenix                                  cphoenix at CRNano.org
Director of Research
Center for Responsible Nanotechnology          http://CRNano.org



More information about the extropy-chat mailing list