- About Us
- Career Center
- Nano-Social Network
- Nano Consulting
- My Account
Home > The Atkinson-Phoenix Nanotech Debate -- Page 4
The Atkinson-Phoenix Nanotech Debate - Page 4 - EmailsCommentary text by Chris Phoenix, except where otherwise noted. [Bracked Italics] indicates a correction. Italics used on text indicates a comment taken from an earlier email, written by the other person.
Last update: August 04, 2003.
Page 1: Overview
Page 2: Emails through June 27, 2003
Page 3: Emails through July 14
Page 4: Emails through August 4
Page 5: The wrap-up
July 17, from Bill Atkinson to Chris Phoenix
July 22, from Bill Atkinson to Chris Phoenix
Subject: Summary To Date
Thought it might be useful to give you a quick snapshot of where I am now. My latest entry in The Well dialogue was as follows. I was responding to a query from a reader asking me who had the final say in the dialogue that Rocky is posting, i.e. whom I could consider to have "won." I said:
So here's where I stand. Kicking, screaming, and biting, I have found myself moved to admit - Let the trumpets sound! - that the molecular assembler has not been conclusively demonstrated to be impossible. Further: that the difficulties I have with the MAM, if addressed, may make me not only suspend my opposition to said MAM, but even (mirabile dictu!) support it. If the Calpurnia Institute (or the same thing by another name) rigorously explores the science behind the molecular assembler; and transparently posts its findings; and cuts itself adrift from the MAM's fringe elements; and generally conducts itself like a dull, staid province of mainstream science; - then it (and you, Chris) can count on me as an ally.
This may not sound like much, so I'll offer a koan. "Mount Everest was moved a mere six inches to the west by the latest intervention of the Applied Tectonics engineers."
August 4, from Chris Phoenix to Bill Atkinson
Subject: Re: Dialog page...
I apologize for taking so long to reply to this, and also for taking so long to forward your letter to Rocky to post on this site. I promise I won't make a habit of it. I'm moving cross-country from the 11th to the 18th and will be completely off-line, but we can work something out with Rocky if you don't have a response before, say, the 8th.
This time, the delay was caused by me flogging myself to finish the rewrite of the Nanofactory paper--55 pages of architecture and calculations that say that *if* an assembler works, an integrated tabletop nanofactory will be pretty easy by comparison. (The nanofactory paper is linked from http://CRNano.org/bootstrap.htm .) This means that the final stages of MNT (assuming they happen at all) could happen quite quickly--very important if you're planning MNT policy. It also goes a long way toward answering your concerns about control and heat problems in MNT manufacturing.
Are you now a nanobooster? That would be very good news indeed. You're certainly talking like you think TCI might succeed at making enough breakthroughs to legitimize MNT. I hope you think it might, because I have a question/request. I don't know very many scientists, and those I do know are generally busy with careers already. Whereas it's your job to meet nanoscientists (or at least it was while writing Nanocosm). If TCI is a good idea--and you've got me mostly convinced that it is--would you like to be the one to recruit its leader? If you're right about its public success, and I'm right about its possibility for scientific and technological breakthroughs, it shouldn't be too hard to find someone who wants to take it on.
I agree that a top priority of TCI would be to gain the trust of the scientific establishment. I'm a little surprised that if all they're doing is stolid experimental work, they'd need written permission from other scientists to cite them in scientific papers. But I'll assume you're right. But that leads to another, more immediate question: Does CRN, as a non-scientific organization, need to ask permission of scientists we mention in our publications?
Let me give you an example. We're working now on a big paper, one part of which explains the various kinds of information available about MNT. I've come up with five classifications: demonstration, theory, projection, fiction, and inaccuracy.
- Demonstration includes the silicon mechanochemistry, as well as the Fe-CO mechanochemistry, and a benzene-joining demonstration (or something similar).
- Theory is stuff that hasn't been demonstrated, but follows by clear and mostly reliable reasoning. This includes Nanosystems. It also includes Smalley's "fat fingers" argument, but only because he's a Nobelist in chemistry; if it came from anyone else, I'd file it under projection because there's been very little formal work done to back it up.
- Projection is a reasonable guesstimate based on ideas that are more or less plausible. This includes some of the stuff in Engines of Creation (some should count as theory, though Drexler didn't show his work in the book), and your opinions in Nanocosm about why nanobots can't be controlled effectively.
- Fiction is stuff like Prey or Star Trek: the author feels free to violate physical laws.
- Inaccuracy is stuff that's claimed to be non-fiction, but is simply wrong. Smalley's "sticky fingers" argument is inaccurate: sticky fingers grabbing atoms is an incorrect characterization of Drexler's proposals and is completely unnecessary to them. Arguments based on thermal noise and Heisenberg uncertainty used to be projection, but are now also inaccuracy (though it's an easy mistake to make, so I'm not blaming proponents of these arguments unless they've been in the discussion for a while).
Obviously, these form a hierarchy: demonstration usually trumps theory, theory usually trumps projection, fiction can be ignored, and inaccuracy should be actively disputed. I'd like your opinion of this taxonomy and of whether I've been too dismissive of anything I've mentioned.
But the other question is: if CRN is writing this kind of overview of what's known about MNT and how to classify each piece of information, do we need to ask permission of scientists before we mention their work?
I also agree that neither Foresight nor CRN is appropriate to take on the role of TCI. My reasoning is simpler than yours: Foresight and CRN are not scientific organizations. You said, "TFI, in other words, is stuck with a rep that has it being far more about PR than it does about science." What I don't understand is why anyone expected them to be about science. That's what the Institute for Molecular Manufacturing was supposed to be for. Foresight has always been about activism and education. I think it comes back to perception: Drexler was trying to wear two hats and was continually being taken out of context. IMM might be a good choice for TCI. Except that they've done mostly theoretical work, and they haven't shunned Drexler, so by now they're wearing a big scarlet D and can never be accepted in polite company again.
Furthermore, I agree that TCI can easily separate itself from cryonics and from making projections about MNT. As you've described it, its goal is simply to do research, and that's a worthy goal and a good focus. But I wonder: if they're not supposed to associate themselves with anything Drexlerian, what keeps them from being just one institute among thousands? With so many "nanotech" groups already formed and suckling from NNI, how would TCI distinguish itself without ever talking about the applications of its work? Is this something that they'd gain the right to do over time, so that eventually they could republish the relevant chapters of Nanosystems and get them accepted? Or would they be swallowed by the peer pressure and focus more and more narrowly on short-term nanostuff, which will be safer to talk about even (perhaps especially!) after TCI gets accepted and funded?
Of course, original research being what it is - adventuring into The Undiscovered Country - there's always the chance that you'll stumble onto wild new effects that can support some truly revolutionary engineering. But that will happen in a way the SE knows and appreciates: first the science, then the technical applications. Again as you say, MAM has to date been driven by the engineers, who have imagined cool things and then set about trying to find ways to make 'em happen. From the SE's viewpoint, that's ass-backwards. And you know, they may be right. Try it their way: for all their rust, fust and harrumphy stodge, the SE do have several centuries of successes they can point to. What has MAM got to lose?
What have we got to lose? Time. There comes a point when theory, demonstration, and supporting technology are all sufficiently developed that a lot of lab work will do the trick. We can call this the Manhattan Point, or TMP since you seem to like TLA's. I think we may be at TMP already. Five years from now, I'm quite sure we will be--though obviously, I'm less sure that the establishment will realize it. And with my CRN hat on, I'm worried that when we reach TMP (whether or not the establishment realizes it), some country will fund a secret assembler project and a few years later we'll all get an unpleasant surprise with no chance to make sensible policy.
As for your lament that the Pentagon does not seem to be examining the possible effects of the molecular assembler, why should it? It is at least possible that, say, a workers' revolution will engulf parts of Eastern Europe, or that looming Latino majorities in the US Southwest should radically destabilize societies there. But it has yet to be demonstrated that the MA is even possible.
Wait a minute. You're using "possible" in two different senses here. I think you've agreed that "it is at least possible" that an assembler and a nanofactory could work. And certainly "it has yet to be demonstrated that" a workers' revolution will engulf anything.
Going back to my taxonomy, workers' revolutions and looming majorities are *projections*. And molecular manufacturing (positional chemistry fabricators integrated into manufacturing systems), I claim, is *theory*--and many projections of its effects (such as unstable arms race and economic disruption) are at least as well grounded as projections of social trends.
I wouldn't ask policy people to address anything in Engines. But Nanosystems has been out for over a decade. Now, I agree it's not their fault that Nanosystems has been invisible to them. But it certainly has been demonstrated that the MA is plausible. At this point, as we've both said, the problem is one of perception and frustration: getting the word out in a way that policy people can hear, past scientists who don't want to admit the new paradigm.
TCI is your golden bridge to acceptability; thus to funding; thus to success. As it does its work, and assuming it does demonstrate MA feasibility, the policy discussions must follow as the night the day. I even foresee a letter to the White House, much like Einstein's famous note to FDR: "Mr President: Recent scientific data now indicate that..." And there's your Manhattan Project, right off.
Who (what position and reputation) would be able to write such a letter and have it listened to? What other conditions would be required?
But without clear and replicable data in support of the MA, that will never occur. Others will say and write, as they have written to me, that MAM may even be the unintended result of a CIA-sponsored disinformation project, attempting to lure America's commercial rivals into pursuing an unattainable piece of science fiction! So that foreign Manhattan Project you fret over may actually be a US Government aim! I'm not that paranoid: but there is now a clear dearth of sound data indicating an MA can be built. TCI will address this void. (I discount the datum that a Japanese speaker at a Foresight conference expressed confidence in the MA concept; TFI would hardly invite, say, me to address them - part of that perceived, and I think real, TFI bias.)
Lots of points to answer here. TFI invited Smalley to address them at a Foresight Technical Conference a few years ago, at a point where he was speaking quite negatively about the idea of "universal assemblers." And the speaker I mentioned wasn't Japanese; she was an American who'd been studying Japanese opinions on MNT. She didn't express confidence in MA herself; she was reporting her understanding of the general Japanese opinion.
If MAM is CIA-sponsored, who does the rumor suggest is in the pay of the CIA, and who is a dupe? Who other than Drexler is doing *anything* to export the idea that MAM is possible? And--this seems actually plausible--is it possible that this rumor is an effort on the part of the other countries to make us think they're not taking MAM seriously, while in reality they already have a Manhattan Project underway?
Data derived from theory can be sound data; I agree it's not always sound, but it's certainly better than nothing. So far, there's been lots of pretty solid theory indicating that an MA can be built, and virtually no theory indicating it can't. Smalley's "fat fingers" are rapidly losing credibility, and they're the only counter-theory I know of. And the pro-MAM theory destroys in detail most if not all of the anti-MAM projection. Wearing my CRN hat again, if people keep saying "I won't believe it till I see it," they could get a very nasty surprise. How would you characterize the situation at the time of the "Mr. President" letter that started the Manhattan Project, before anyone had demonstrated a chain reaction? Were there sound data, or not? If so, I claim that data supporting MA are equally sound in every way (except that we don't have an Einstein backing them--but we do have a Feynman). If not, then we don't necessarily need sound data to act on the MA possibility. Please answer this paragraph in detail; it really cuts to the heart of the frustration of MAM policy people, including me.
Thanks for sending the summary of your position. It's always encouraging to see mountains move. Perhaps, we can but hope, your example will encourage others to take an unprejudiced look past the MAM at the theory behind it. And perhaps, we can but hope, something respectable but MAM-inspired can arise and provide a foundation for useful policy discussion.