[extropy-chat] Wired article on Drexler

Hal Finney hal at finney.org
Sun Sep 26 06:34:38 UTC 2004


The October issue of Wired magazine has a fascinating article on
Eric Drexler, by Ed Regis, author of Nano: The Emerging Science of
Nanotechnology.  Based on their previous publishing habits, the article
will probably be available on wired.com in a few days or weeks.

It is a sad article on the decline in the status of Drexler and his ideas for
molecular manufacturing.  The recent debate with Richard Smalley was one of
the most visible signs:

   ... a series of letters between Drexler and Smalley in which
   the Nobelist made his position clear: Molecular assembly is
   impossible. "Chemistry of the complexity, richness, and precision
   needed to come anywhere close to making a molecular assembler -
   let alone a self-replicating assembler - cannot be done simply by
   mushing two molecular objects together," Smalley wrote.

   It was a public takedown from the man fast replacing Drexler as nano's
   leading light.  But Smalley wasn't done. In remarks so overheated
   that they bordered on bizarre, he accused Drexler of terrorizing the
   world with the prospect that self-reproducing assemblers might escape
   the lab and devour everything in their path, turning the Earth into
   an inert, undifferentiated blob of gray goo.

   "You and people around you have scared our children," Smalley fairly
   shouted in print.

The other recent defeat was the deletion of the proposal for research
on molecular manufacturing in the recent nanotech funding bill.
Regis describes Drexler as "bloodied but combative":

   He speaks in apocalyptic terms. "In a competitive world," Drexler says,
   conjuring the frightening prospect of hostile forces wielding gray
   goo nanoweapons, "suppression of research in molecular nanotechnology
   is the equivalent of unilateral disarmament." The outcome, he claims,
   could be nothing less than "the destruction of the United States as
   a world power."

I don't think Regis is quite right here in describing gray goo as a
nanoweapon; it sounds like Drexler is speaking more generally of the
enhanced effectiveness of military forces equipped with nanotechnology.

Drexler's personal situation is described as quite shaky:

   Drexler's rejection by the scientific and political establishments
   comes at a particularly bad moment. Last year, he divorced Christine
   Peterson, his wife of 21 years and president of his nonprofit think
   tank, the Foresight Institute.... Never a rich man, Drexler is barely
   solvent. He recently moved from his three-bedroom ranch house in
   Silicon Valley into a modest apartment.

It's unfortunate that he is dealing with personal and professional crises
at the same time, but keep in mind that it's common for divorced men to
move into smaller apartments at first. I am confident that a man with
Drexler's talents is not going to end up as a pauper, the way Regis
seems to paint him.

Of course the big question, which dwarfs issues of Drexler's personal life,
is whether nanotech will work.  Regis has some interesting quotes:

   Even at MIT [where he got his PhD in nanotechnology], his work divided
   the faculty.

   "It's very impressive, there is no question," said MIT chemist Rick
   Danheiser, who served as Drexler's thesis adviser, in 1992. "I couldn't
   have done a better job."

   "It showed utter contempt for chemistry," countered Danheiser's
   colleague Julius Rebek. "And the mechanosynthesis stuff I saw in
   that thesis might as well have been written by somebody on controlled
   substances."

Regis has some quotes from Smalley which offer more insight than I
have seen elsewhere into the thought processes that turned him from an
enthusiastic supporter of the molecular assembly concept into its most
visible opponent.

   Sitting in his hotel room at June's Nano-Summit, Smalley explains
   that he was once captivated by Drexler's notions. "I was enchanted by
   Engines of Creation," he says. "I read it in a single sitting, and
   then I reread it."  As late as 1999, he testified to Congress about
   "what will be possible when we learn to build things at the ultimate
   level of control, one atom at a time."

   But doubts crept in as Smalley pondered the theory. "For months I said
   to myself, How could we have missed this? Is it really possible to
   do chemistry in this way? After a while, I thought I saw what might
   be some problems. The more I thought about it, the more troublesome
   they appeared.  Finally I ended up thinking, it's just hopeless."

   Smalley recalls a meeting he arranged with Drexler at Stanford about
   a decade ago. "I wanted to talk about the tip," he says, referring
   to the business end of Drexler's machines. "I love the idea of the
   assembler.  So I tried to drag him into a conversation about the tip,
   and he stonewalled.  It was as if the tip was a job for later."

   Drexler remembers the same meeting with no less frustration. "I found
   it very hard to explain things to him," he says. "He was asking for
   an irrelevant and impossible control of the motion of every atom. The
   question isn't, Are there some things that won't work? The question is,
   Are there enough things that will?"

I find that this exchange encapsulates the problem with this argument.
Smalley is saying, show me that it will work.  Drexler replies, show me
that it won't. I can't count how many times I have seen this same dynamic
at work in the ongoing feasibility debate.

My opinion, as I have stated here before, is that when reduced to these
terms, Drexler loses.  He has the burden of proof here.  I know that some
disagree and argue that the power of the technology demands investigation
if there is even a chance that Drexler is right.

Regis goes on to criticize Drexler's use of the picture which was on the
cover of the debate issue of Chemical and Engineering News.  I can only
find a small version of it right now,
<http://pubs.acs.org/cen/images/8148/cencover8148.jpg>.

   On a screen, Drexler projects an illustration of a molecular mill,
   a wheel that transfers one atom at a time to a succession of molecules
   as they move past on a conveyer belt. The image, the latest expression
   of Drexler's vision, is absurdly oversimplified. The machinery is
   rendered with solid surfaces rather than the atoms they must actually
   comprise - a choice that plays into the hands of his critics."

I had a similar reaction when I saw the diagram. I thought the magazine
came up with it on its own and that it was a very poor representation of
the molecular assembly concept. But apparently Drexler either created it
or at least likes it enough to use it. Probably it's because it represents
the "mill" concept which Drexler developed in Nanosystems rather than the
"robot arm" which is widely associated with assemblers. The robot arm
more nearly matches Smalley's description of a "finger" for positional
assembly, making it vulnerable to his "fat fingers" and "sticky fingers"
arguments.  Mills have no fingers, at least not to the same degree
(there are some stubby projections holding the atoms in the diagram). My
reading of the Smalley-Drexler debate was that Drexler was talking about
mill systems to escape these critiques.

Regis closes on a somewhat more positive note:

   Yet there are indications that Drexler wants to remain a player -
   even if that means backpedaling and retrenching. First, he has
   abandoned his staple rhetoric. "Self-replicating nanomachines are not
   necessary for molecular manufacturing and should be de-emphasized as
   a goal," he wrote in the January 2004 issue of Foresight's newsletter.
   Instead, they would be replaced by "desktop nanofactories" conceived as
   "general purposes manufacturing systems." Second, he now regularly
   invokes Richard Feynman's name, calling his claim that molecules
   can be positioned mechanically "the Feynman thesis." Finally, he has
   proposed renaming his vision of molecular manipulation zettatechnology,
   a tacit acknowledgement that he has lost the tug-of-war over the term
   he coined.

Well, "zettatechnology" isn't going to fly, and bandying Feynman's name
about is pretty doubtful too.  Eliminating the gray goo threat however
may be politically useful, although the meme may have enough momentum to
be self perpetuating.  Regis argues that one reason for avoiding Drexler's
ideas is the fear that nanotech will come to be seen as an evil technology
and be tainted at birth like biotech was.  But after all the years of
gray goo stories, after Bill Joy and Michael Crichton have scared us all,
I'm doubtful about how much it helps to have Drexler saying, in effect,
"never mind".

Hal



More information about the extropy-chat mailing list