Foresight Nanotech Institute Logo

« Go Back

You are viewing
Foresight Archives

Image of nano

Foresight Debate with Scientific American

Foresight Institute vs Scientific American
Debate on nanotechnology: Round 3

Overview of the Debate

Round 1 | Round 2 | Round 3 | Round 4 | Epilogue

Round 3 -- Part 1 | Part 2

Email from Scientific American, with Foresight responses

In responding to Round 2, Scientific American has again sent email rather than a URL. Those wishing to see their response together with our remarks can read the following; for those wishing to experience the full intended impact of their response alone, with its rhetorical flow unimpaired, we have provided space for the unannotated version.

To summarize the case made in the annotations below: Scientific American's response displays a striking lack of scientific substance; it substitutes an equally striking reliance on repetition, appeals to unnamed authorities, and innuendo. We like it because it's so easy and entertaining to explode, but they should find it embarrassing.

In the following, the text of the Scientific American response is indented, and the remarks from Foresight are full screen width.

From Scientific American, 10 May 1996

Dear Foresight Institute:

Ralph Merkle and you were kind enough to invite us to respond to your comments concerning Gary Stix's nanotechnology article, "Waiting for Breakthroughs," in the April 1996 issue of Scientific American. We have followed with interest the considerable feverish discussion about it and would like to answer as follows:

Scientific American characterizes the discussion as "feverish", slipping a bit of name-calling into the first paragraph. (Metaphorically, feverish "can refer to haste and confusion"; S. I. Hayakawa, Modern Guide to Synonyms). Scientific American is, however, now quite new to the Web, having opened their site in April 1996. They may find the speed and volume of exchange a bit unnerving.
[Note: numbers in square brackets in the following paragraphs from Scientific American were added by Foresight to facilitate specific reply.]

Dr. Merkle and other commentators make a number of points, but none of them persuasive. Scientific American stands by the article and by its major conclusion: that although nanotechnology attracts ardent support among bright, creative people [1], most researchers [2] working in allied areas [3]--including ones embraced by the nanoists [4] themselves--think that the actual science in nanotechnology has gaping holes [5], and that there are few chances those holes will ever be filled [6]. Drexler and like-minded individuals engage in what many materials scientists feel is wild speculation about the future of technology.

[1] Nanotechnology attracts not only "ardent support" from "bright, creative people" in general, but reasoned approval from bright, creative scientists and technologists in related fields. Scientific American has obscured this.

[2] This would require polling data to support, but would still be only statistics on opinions, not an analysis. We welcome technical criticism from Scientific American. Remarkably, given their thundering presentation, they have neither offered nor cited any.

[3] Molecular manufacturing as a whole is more a problem of software and systems engineering than it is a problem of chemistry. Accordingly, major "allied areas" include mechanical engineering and computer science, and chemists aren't trained to pose the right system-level questions and find answers. Scientific American, though, cites opinions drawn almost exclusively from a handful of researchers involved in laboratory work based on existing chemical and microtechnological techniques. Their unease, as one might expect, centers more on the mysteries of engineering systems than on any obstacles they see in their own field. One might as well ask 1950's metallurgists and machinists about the prospects for a moon landing, on the grounds that the proposed rockets were to be built of machined aluminum, then report their unease about guiding a projectile over so vast a gulf.

[4] Name-calling -- whatever a "nanoist" may be, it presumably isn't a researcher.

[5] Technical professionals developing careers in this nascent field are alert to holes and gaps in the science underlying these proposals. Discovery of a damaging hole would spur at least some of them to abandon the field. With the Internet, news of a hole would spread to the ends of the Earth overnight. If such "gaping holes" had been identified, Scientific American might be expected to offer at least a brief description (Are they chemical, physical, or computational? Bigger than a breadbox?) and a citation or URL for a more extensive discussion. Instead, it seems that they are either cruelly toying with us and their public by feigning weakness while holding a killer argument in reserve, or (more straightforwardly) merely bluffing.

[6] It is hard to fill holes that no one can identify. Over the last decade, Drexler and other researchers have worked with and presented to cross-disciplinary groups of physicists, biologists, chemists, computer scientists, and others capable of evaluating the technical issues. Many researchers in a field A have initially raised objections in their own fields and had them answered, but then continued to suspect a big hole in some other field B -- while researchers in B have had a similar experience with reversed roles. As a thought experiment, one can easily see that if everyone in every field had reacted this way, the resulting pattern would constitute good evidence for an absence of holes, despite everyone in every field expressing major (but insubstantial) doubts.

In the real world, some researchers have noticed a rough approximation to this pattern and have drawn the obvious conclusion. Others can describe genuine, fatal flaws, but the flaws are in muddled proposals (e.g., nanomachines running without an energy source) that no serious researcher ever advanced. (This absurd "proposal" was duly attacked in Scientific American's original article.) So long as none of these researchers describes a gaping hole related both to their area of expertise and to actual proposals, it is misleading to say that they, as scientists, "think" there are gaping holes -- "feel" or "hope" or "assert to journalists," perhaps.

The field's proponents do much more than calculate thermal effects on little gears [1]. The most prominent scientists at the Foresight conference--Whitesides, Smalley, Fraser Stoddart--have little faith that a self-replicating assembler will ever be made [2]. Yes, they attended the conference, but ask them directly about what they think about Drexler's vision [3] and you'll hear another story [4]. Smalley talked explicitly about his doubts during a presentation at the conference [5].

[1] Yes, there is now a substantial literature on a wide range of topics.

[2] Judging the feasibility of a technological proposal is not a matter of "faith." Such proposals can be analyzed based on our current understanding of physical law. Whether a feasible proposal will ever be implemented is a separate question, involving many non-technical factors. Scientific American has discussed neither the technical analysis nor the non-technical factors.

[3] In contrast to his proposals and analyses, presumably.

[4] If such "stories" comprise technical objections to any part of the proposal for nanotechnology, they should be presented. Saying that researchers spin tales to journalists that they (apparently) won't state before their colleagues is hardly evidence of anything substantive.

[5] As Drexler's letter notes (and as he pointed out at the conference), Smalley addressed an outlandish proposal that no one has made, and is thus irrelevant here.

One of the article's central points is that these laboratory researchers have reported a number of advances in their field, many of which suggest real applications [1]. None of them [2], however, sees their work as moving toward making assemblers [3].

[1] Which is why continued progress does not require that laboratory researchers examine and judge long-term engineering goals. They do excellent scientific work in the lab.

[2] None of them? The survey costs required to obtain this (false) piece of knowledge would be staggering, unless this refers merely to the few researchers Scientific American chose to quote.

[3] James Burke's Connections series documents how disparate developments are drawn together into inventions utterly unforeseen by their originators. Point [1] (real, short-term applications) explains why progress doesn't require anything other than experimental progress from experimentalists. There is no strong reason to expect (or even to desire) that most experimentalists contemplate the long-term applications of their work. Some do, of course, and have been very helpful as critics and teachers in this interdisciplinary field.

Drexler's vision [1] of manipulating matter at the molecular scale had a wonderful futuristic allure [2] when Engines of Creation first appeared in the mid-1980s [3]. But it hasn't moved very far beyond this original statement [4] (theory and computer models are insufficient proof [5] of the ultimate feasibility of these concepts).

[1] Mild name-calling.

[2] Stronger name-calling.

[3] It has inspired a new generation of researchers, having steadily growing impact over the years, but this doesn't fit the tale.

[4] The "original statement" (five years before Engines) was a refereed article in the Proceedings of the National Academy of Sciences. On the experimental side, this article is at the root of the citation tree in the field of de novo protein design (of the first two papers in Science and Nature, one cites the other and both cite Drexler). A body of citations to more recent work is developing in the proximal probe literature. On the theoretical side, Nanosystems presents an analysis that builds from fundamental principles and atomistic modeling to systems design. Merkle's work goes beyond Nanosystems in several directions, and others are entering the field.

[5] Proof is not a matter of degree, sufficient or "insufficient," and proof isn't how science works anyway (see Drexler's letter). One should instead ask, "What is the weight of the evidence?", or at this point perhaps, "Can Scientific American actually produce any evidence?" Demanding that an opponent give proof is useful chiefly as a rhetorical tactic, though it tends to lose effectiveness after high school.

Because of this stagnation [1], there is a chance that the Engines of Creation version of nanotechnology will become increasingly irrelevant--even though some other implementation of nanotechnology may indeed take off [2].

[1] Experimental work on nanotechnologies has advanced. Design and modeling work on longer-term proposals has advanced. Scientific American acknowledges both. Is "this stagnation" intended to describe something about the field?

[2a] This statement misrepresents the purpose of the book and of exploratory engineering in general. Work of this sort aims not to predict the future -- an impossible goal -- but to identify technological possibilities. Unless a fatal or serious defect can be found in the case for nanotechnology as proposed, it will retain and solidify its status as a class of systems and capabilities that can reasonably be presumed feasible. In that event, it will remain relevant to future projections until either (1) it is implemented, or (2) some other proposed (or implemented) technology is shown to be equal (or superior) in all respects -- that is, delivers (or credibly promises to deliver) as much or more than the kind of nanotechnology described in Engines of Creation. Either case leaves the consequences we have described intact.

[2b] Here we see Scientific American and Editor-in-Chief Rennie beginning to "cover their behinds," adjusting their position such that as nanotechnology takes off, they can claim to have expected it. And since future systems will differ from anything we can design today, they will be able to say that current researchers failed to offer impossibly accurate foresight.

Perhaps the best way to rebut the accusations of incompetent journalism is to go through general categories of them one by one.

* 1) The article lacked technical depth: As should be unnecessary to explain, an overview article in a printed magazine for the general reading public is not the same as a literature review in a journal [1]. To fault it for lack of technical depth is like faulting a swimming pool for lack of trout [2].

[1] Yes, but this is no excuse for misleading the public by implying that technical objections exist in the journals or anywhere else.

[2] The problem isn't lack of technical depth, but bias and lack of genuine technical content in a purported review of a technology. As Foresight's publications (and many Scientific American articles) demonstrate, it is possible to describe science accurately without plunging into the depths of a specialist journal, and without washing away fact in a flood of opinion. Regarding trout, our analogy would be that Scientific American has presented a picture of a pond labeled "Great big trout here!" when the picture actually is a retouched image of a chlorinated swimming pool. The article pictures a scientific community swimming with big, juicy refutations of our position; yet despite all its fish stories, it seems they all got away.

A more telling criticism would be that the technical summaries in the article are factually wrong [1], which brings us to....

* 2) The article is riddled with errors.

The Foresight Institute and Merkle certainly have the right to disagree with the article's conclusions. But they have set forth few specifics of how facts went awry.

[1] Nonexistent technical summaries can't be factually wrong. Speaking of technical summaries when they don't exist is another matter.

The rebuttal does claim [1] that Edward Reifman, the dentist who appears in the article's lead paragraph, was quoted out of context, that he made the remark in jest [2]. But [3] when Reifman was contacted during a routine fact-checking call prior to publication, he did not deny the quote [4], nor did he say that it was made in jest [5]. In fact, he acknowledged what he had said before [6] and even went on to remark that Drexler now has amassed a significant group of followers [7].

Oh my. This looks bad.

[1] What they term a "claim" by us is supported by Dr. Reifman's letter, which Scientific American should have received long before they wrote this.

[2] Yes, he states these points in his letter: "I was quite upset that a reference made in jest to their reporter, Gary Stix, was used out of context to ridicule nanotechnology and the conference that we both attended."

[3] "But" is generally used to introduce a contrast, but here is used to produce the appearance of a contrast. Better uses of "But" follow --

[4] -- But no one said he denied the words. We cited his letter acknowledging them.

[5] -- But no one claimed that he both anticipated and denied Mr. Stix's misuse of his words.

[6] -- But why is Scientific American again saying that he said the words he said, when what Dr. Reifman objects to is the misrepresentation of their meaning?

[7] -- But this adds further insult by implying that Dr. Reifman supports their view of the matter, after he has presented a written objection. An apology to him would seem more fitting than twisting his remarks yet again. Followers? Since "followers" appears above without attributive quotes and has negative connotations here, we can presumably chalk it up as a form of name-calling. The substantial support among researchers for the goal of molecular manufacturing was visible at the conference, though invisible in the report on it. Perhaps Dr. Reifman was attempting to bring this to the writer's attention. If so, it had no noticeable effect.

In the conclusion of his commentary, Merkle writes, "A few people were quoted in a way which suggested they had a technical criticism of the feasibility of nanotechnology. The quotes were rebutted." Actually, Merkle has only disagreed with them [1]. Given that the people in question are the likes of George Whitesides, Julius Rebek, Jane Alexander and others, they can muster technical arguments in support of their position [2], much as Merkle has. To imply otherwise [3], in Merkle's words again, "is not only insulting, it's... very... very... dumb! " [4]

[1] He noted that the quotes contained no technical criticisms, a statement which Scientific American, wisely, has not disputed. We regard this as an adequate rebuttal.

[2] As pointed out in Drexler's letter, this is but a hope. We have been fishing for a telling technical criticism of these technical proposals for almost twenty years, from a world-wide pool of researchers at the highest level in many disciplines. No such criticism has surfaced. If Scientific American thinks that a particular researcher has such a criticism, they should get it written up and published. If they can offer only appeals to nameless, imaginary, inarticulate, or mistargeted authority, then perhaps they should fold their tent and sneak off.

[3] We aren't implying, but stating, that Scientific American has repeatedly failed to present a credible technical objection to molecular nanotechnology, and that at this late date the odds of their succeeding appear vanishingly small. But we urge them to try; their public failure is informative and hence valuable.

[4] Boomerang name-calling -- what can we say? At least it's a direct assertion of incompetence, rather than innuendo.

* 3) The article sets forth an erroneous definition of nanotechnology. Many definitions of nanotechnology were put forth at the Foresight conference in November. The article makes the point that people define it in different ways. Neither Drexler nor Merkle had much of a problem with the definition of nanotechnology in the article when the text was fact-checked with them before publication [1].

[1] Definitional arguments can continue indefinitely. Those interested can pursue it elsewhere. By the way, having fact-checking done by the author, as Scientific American did with Gary Stix's article omits a safeguard for objectivity that is widespread in the press (even OMNI had fact-checking done by someone not immersed in the project).

* 4) The article is a personal attack on Drexler. Where do we suggest that the man's personal traits invalidate his technical arguments? [1] Scientists are real people with both strengths and weaknesses [2]; good science journalism depicts them as such. In recent years, Scientific American has profiled dozens of major scientists, including Nobel Prize winners, while describing their opinions, prejudices, personal foibles and mannerisms [3]. Drexler received no different treatment [4]. (For better or worse, even if the foundation of nanotechnology seemed impeccable, Drexler would still sound like Mr. Peabody to Gary Stix [5].)

[1] This is disingenuous. Numerous literary techniques are used for precisely this purpose in the story, and they are indeed suggestions, not statements. For an analysis, see Will Ware's essay.

[2] Scientific American did a thorough and imaginative job on the weaknesses, but the strengths seem missing. For example, the article describes Drexler's Nanosystems as an "attempt" to provide scientific grounding (merely an attempt, implying a weakness), and as a "plea for respectability" (very weak), from someone who is, basically, regarded as a "flake" (practically dead). A genuinely balanced article might have mentioned one of the following strengths: The flake was awarded the first doctoral degree in molecular nanotechnology, from MIT, and his plea is not only an extension of his doctoral thesis, but was named by the Association of American Publishers as the Outstanding Computer Science Book of 1992. Instead, the steep slant continues. Drexler's "ruminations" elsewhere provoke "guffaws"; Nanosystems is a mere "object of curiosity," and the whole body of research is just "show business."

Scientific American has announced its negative view of his work, and readers with any knowledge of the field generally perceive the article as an attack. Why pretend that it was balanced?

[3] Yes, but aside from making a buck, why?

[4] "We're OK -- we do shameful things like this habitually, honest!"

[5] And this perception is important news about science.

Also, though Merkle asserts that the article "is not about nanotechnology so much as it is about Drexler," that is wishful thinking on his part. Any fair-minded reading of the article shows that the many technical criticisms [1] leveled against nanotechnology by various scientists are not merely ad hominem [2].

[1] But if there are many, then it follows logically that there must be at least one. Where is it? Is it worth repeating, or has it been withdrawn or disappeared, as has their claim that David Jones made a substantive criticism in Nature? Merely chanting "many technical criticisms" won't draw down bales of thoughtful analysis from the sky. Expecting that these empty, decoy criticisms will draw down real ones amounts to "cargo cult journalism."

[2] Some are ad hominem, none are substantive.

* 5) The article is tabloid journalism, dwelling on gossip rather than science [1]. Were we wrong to point out the hype rampant in nanotechnology? [2] What continues to attract most people to the field is not the technical challenge, but rather the promise of rectifying the most basic social ills, such as death and poverty [3]. At the Foresight conference in November, Drexler talked about how a nation's capital stock might be doubled in an hour using nanotechnology [4]. An hour? [5]

[1] Yes, relative to anything that could pass for responsible and informative.

[2] Quickly wrapping the rags of tabloid gossip in the mantle of honest skepticism: very smooth.

[2b] If the story had been balanced -- covering the actual progress in the field, not pretending to see stagnation -- then describing the healthy skepticism and enthusiasm of both technical and non-technical observers, along with the problems of over-enthusiasm and know-nothing-skepticism, would have been entirely appropriate. A focus on hype piled on top of assertions of no substance is another matter.

[3] Altruistic motives bring many into science, especially medical research. Some researchers maintain these goals throughout their careers. Surely Scientific American isn't suggesting that only "technical challenge" is a legitimate motivation for research?

[4] This is somewhat sloppy reporting. Drexler described advanced molecular manufacturing systems as a form of flexible capital, able to produce almost any product that fits in a box (as distinct from, say, a nation's road system). He stated that this form of capital stock could, with appropriate inputs of energy and raw materials, be made to double in an hour.

[5] Yes, "an hour" was accurately reported. Is the skeptical query "An hour?" a compressed description of a problem someone has identified in Drexler's analysis? Scientific American's Editor-in-Chief, John Rennie, is said to have a background in biology, and presumably understands how fast molecular machinery can work. Bacteria can double their "capital stock" in a quarter of an hour, which makes "an hour" at least unastonishing. Drexler has offered a number backed by a detailed, quantitative analysis; Scientific American offers a raised eyebrow, apparently with nothing behind it.

In an interview with Scientific American, Ralph Merkle said he didn't want to be a member of the last generation to die [1], a reference to how he believes nanotechnology might one day allow his head to be thawed so he could live on forever [2].

[1] Presumably, we are expected to dismiss this as silly. To actually address the issue behind this comment, we would need to explore the prospects for significant advances in life extension techniques, first within the next forty years, then afterward. This might be interesting and valuable; a better understanding could save lives. We invite Scientific American to notify us when they are ready to undertake a serious exploration of this topic, weighing the evidence for various views of the future of medicine. They could prepare by reading Dr. Merkle's publications.

[2] Cryonics has received a staggering amount of name-calling and misrepresentation. "Surgeons plan to cut the beating hearts out of two people, declare one dead, and make the other immortal by stitching it into his chest," would be a roughly equivalent description of a routine heart transplant. In short, Scientific American presents a crude misrepresentation of Dr. Merkle's views on the future of medicine. Genuine criticism is more careful; misrepresentations occur, but they tend to be milder, less frequent, and are always accidental. With genuine fact-checking, the residuum can be quite small.

These remarks go beyond a mere assessment of technical feasibility [1]. Rather they seem indicative of the chief reason that people are attracted to the vision of nanotechnology put forward by Drexler and Merkle [2]. Everyone wants to be rich and live forever [3]. Given these factors, covering nanotechnology as a techno-sociological phenomenon is wholly appropriate [4].

[1] There would be little point in discussing projected technical advances that lacked potential practical applications. The point of such discussions is precisely to go beyond an assessment of basic technical feasibility. Scientific American seems to be arguing, in effect, that the case for technical feasibility should be doubted because the magnitude of the alleged human consequences is so large. But feasibility depends, ultimately, on physics; the perceived magnitude of human consequences depends, ultimately, on human judgment. Reasoning from a (visceral?) rejection of consequences back to a rejection of causes gets causality backward. (Unless, of course, molecular manufacturing actually entailed a known impossibility -- but "impossible" doesn't mean "susceptible to ridicule.")

[2] Scientific American here suggests that any support for these ideas can be attributed chiefly to wishful thinking. Dismissing arguments on the basis of motivations is a classic ad hominem attack.

[3] This misrepresents Drexler's and Merkle's views. Connecting Drexler with "living forever" is again disingenuous: Scientific American has shown a familiarity with Drexler's book Engines of Creation. In the index, they will find an entry "Immortality, impossibility of," referencing the statement: "To speak of 'immortality' when the prospect is only long life would be to ignore the facts or misuse words." And it is, of course, false that "Everyone wants to be rich and live forever."

[4] Discussing nanotechnology as a "techno-sociological phenomenon" would be fine idea, after presenting it as a technology.

* 6) The article misuses the name of Richard Feynman [1].

[1] This has been publicly established by a letter from his son, Carl Feynman, which has been available to Scientific American since about March 25.

Molecular nanotechnologists tend to use Feynman's name as evidence [1] for their claims only when it suits them. In his book Unbounding the Future, Drexler quotes Feynman as saying that the ability to move atoms one by one would ultimately be "useless" because chemists would come up with traditional methods of making new substances [2]. Would Richard Feynman have bought into a technological vision that purports to solve the problems of death and poverty, with little experimental evidence to support these assertions? [3] Would he have called it cargo cult science? [4] Interesting questions--but also irrelevant [5]: Feynman's ideas have lasting value [6], but his opinions would not change the technical shortcomings of the field an iota [7].

[1] We do not "use Feynman's name as evidence" for anything. We occasionally cite his actual statements and his scientific work, properly, as evidence of the insights of a remarkably knowledgeable and foresighted scientist. As Scientific American's response to Carl Feynman's letter makes clear, tactics based on the misuse of Feynman's name can force quick dodging.

[2] Yes, Richard Feynman said that in his first talk on the subject in 1959 -- but surely he was joking. Fortunately for all of us, he lived decades longer, and his views evolved and clarified.

[3] No, presumably he wouldn't buy Scientific American's version of a vision. It's hard to imagine that he had much use for straw men.

[4] His son gave a clear answer: No, he wouldn't. But that's inconvenient, so let's ignore it.

[5] Let's get this straight: (1) The speculation that Richard Feynman would have regarded the goal of molecular manufacturing as cargo cult science (false, says his son) was important for Scientific American to "report," or whatever this sort of activity is called. (2) Now that his son has made it quite clear that Feynman should be counted on our side of the basic, scientific debate, his views are declared irrelevant. How convenient.

[6] As a biologist turned editor, Mr. Rennie is kind to acknowledge that some of Feynman's ideas have lasting value, but perhaps not wise in broadcasting his own, contrary intuitions regarding what physics says about the limits of technology. If one weighs Feynman's reputation and his clear, positive, published statements against the chemical knowledge and vague, negative, offhand remarks of Mr. Rennie's much-mentioned quotees, it seems that we would win a game of appeal to authority.

[7] But in the end, evidence and substantive arguments outweigh any number of grey heads and Nobel prizes. Scientific American's reference to "the [still nameless] technical shortcomings of the field," highlights again their failure to articulate one, despite having been challenged to do so.

* 7) The article was biased [1].

Scientific American is, if anything, biased in favor of new technologies and scientific concepts. Nanotechnology is an exciting idea. We entirely understand why so many people are drawn to it -- when Engines of Creation came out, some of us were very enthusiastic about the promise of nanotechnology [2]. And who would not like to live in a world where technology makes it possible for virtually any dream to come true? [3] But one of the jobs of good journalism is to determine whether some dreams are simply that [4].

[1] Biased? If Mr. Rennie is sincere in thinking it wasn't, then many readers can assure him that his editorial judgment of tone and content is seriously out of adjustment.

[2] Rumors leaking out of the Scientific American offices over the years, however, suggest that the atmosphere has long been poisonous (we protect our sources). A short, nasty piece on our 1989 conference managed to associate it with rock stars, as an aid to ridicule. How? Well, a cosponsor of the conference was the Global Business Network, which consults to huge corporations regarding both technology and society; GBN, as one might expect, follows cultural trends; they do this in part by having advisory members in the entertainment industry, some of whom -- Yes! Are rock stars! This certainly sheds light on that silly nanotechnology stuff, and working in both a rock-music theme and a nanosex theme in such a short piece was ingenious. Our belated compliments to author (and current Scientific American editor) Tim Beardsley.

[3] Who wouldn't want such a future? Most of us at Foresight would answer, "Anyone who begins to think through what 'virtually any dream' really means." The notion that unleashing unlimited technological power would be an obvious net benefit is naive to the point of being ludicrous; we've met a few people who would endorse it, but it's hard to believe that many members of the Scientific American staff ever held such views. From where we sit, the idea that technology won't turn the world inside-out seems quite comforting, a homey, familiar sort of wishful thinking not supported by any evidence. Countering arguments by citing motives is a dubious practice, but Scientific American hasn't actually made an argument, so perhaps we can be excused for a glance at their hypothetical psychology.

[4] Yes, and a serious effort along these lines would be a real contribution. A balanced sketch of the demographics of supporters and opponents, examining field, age, and institution, would tell an interesting story while placing no significant strain on actual scientific reasoning or judgment.

The article contains extensive passages in which Drexler and other proponents make their case for nanotechnology [1]. The fact remains, however: many of the leading researchers in materials science and chemistry do not place much store in it [2]. None of the chemists who gave presentations at the Foresight conference wrote to protest the portrayal of their views in the article [3]. In fact, responses to the article from chemists and other professionals have been positive, congratulating us on pointing out shortcomings that more congratulatory articles about nanotechnology generally overlook [4].

[1] Most proponents are ignored; the passages are brief; no decent explanation of the field is presented.

[2] Opinions of materials scientists and chemists regarding the prospects for systems engineering based on atomically precise fabrication can at best be expected to shed light on the basic constraints of materials properties and chemical reactions. One might as well ask a practitioner of semiconductor physics, circa 1940, about the prospects for word-processing software -- better (somewhat) to ask von Neumann or Turing. What is important is not a tally of laboratory scientist's opinions about the future, but their progress in the laboratory. This has been impressive.

[3] Selected chemists didn't protest the article, and may actually have the views attributed to them. Good work.

[4] Uh-oh. We are here confronted with another report of important but nameless shortcomings described somewhere in the original Scientific American article, this time found to be giving pleasure to nameless chemists (and other professionals of some sort) whose implied independent recognition of these shortcomings -- and further implied triumph in actually finding them in the article -- adds immeasurable credibility to something or other (possibly trout).

* 8) The article fails to show that nanotechnology is impossible [1].

The article is not trying to show that nanotechnology is impossible [2]. Showing that any speculative technology is impossible is itself impossible--we can't prove a negative proposition [3].

[1] Actually, we'd say: "The article gives no real hint that an effective argument against the feasibility of nanotechnology has been or can be made."

[2] Indeed, that seems to have been assumed. Otherwise, why try to show that the whole field is a misleading pile of rubbish to be hauled off to the dump and (as Scientific American suggests below) buried with the Star Trek warp drives? If their staff had asked why it found itself forced to deal in ridicule to pursue their goal -- if they had wondered why they could find nothing that even a decent instructor of rhetoric would accept as an argument supporting their side -- then they might have noticed that their basic assumption is wrong. Instead, they marched off to launch a war of extermination, believing so deeply in their cause that they felt justified in engaging in flagrant violations of the General Convention on Intellectual Warfare, not excepting attempted dastardly attacks with asphyxiating repetition gases and cerebral blister agents, which turned out mostly to smell like old cabbage. Proposed excuse for the record: "We had to stink up science to save it."

[3] At last, a clear-cut statement, not about the topic directly, but about reasoning about the topic. Aside from the cheat of declaring proof as a standard, it is a nice try at finding an excuse from the burden of actually saying something. But in fact, both scientists and engineers accept many negative propositions regarding technology (no perpetual motion machines, no faster-than-light starships, etc.), and use them as a basis for reasoning and as weapons in argumentation. More generally, the familiar process of evaluating proposals before doing designs, and designs before building hardware is imperfect, but hardly exotic. The present instance has unusual features, but is the same in principle. Sorry, but effective criticism is possible. Indeed, many proposals in the field have already been shot down -- usually by their originators, during the design process -- which is why the survivors are so tough.

As for proofs of impossibility, at this point in the exchange our expectations are quite modest. We'd be relatively impressed if Scientific American could find someone who could point out a necessary part of the proposal that is not, by design, both obviously and completely in accord with physical principles.

On the other hand, it is entirely incumbent on nanotech's proponents to show that what they hope to achieve is possible. [1] Stix's article reports that they have not yet done this [2], and that many technical experts doubt they ever will [3].

[1] Since Scientific American appears to reject as evidence not only theoretical analyses and computational experiments, but laboratory demonstrations of molecular machines and atomic manipulation, what can they want? Their notion of showing possibility seems to require completing the research and development program and delivering working molecular manufacturing systems -- but then it will be too late to make effective preparations. Do they oppose all attempts to prepare for future technologies, or is this one special?

[2] Done what?

[3] Yes, and we've made our case to the satisfaction of many other technical experts. But only our side publishes refereed articles on the subject and backs opinions with careers.

Does the fact that Drexler's goal may not be impossible mean that it should be pursued? [1] A recent book, The Physics of Star Trek, claims that the laws of physics do not preclude something resembling a warp drive, which the Enterprise uses to travel faster than the speed of light. Is that a technological problem to which we should therefore devote significant financial and intellectual resources? [2]

[1] A moot question; appropriate steps are being pursued with vigor, regardless of the number or percentage of researchers who think seriously about the long-term results of their work.

[2] The White House Science Advisor has called explicitly for the development of molecular manufacturing, and many students and researchers are aiming their careers at it. In contrast, textbooks explain why faster-than-light spaceflight is inconsistent with physical law, and no serious effort is being spent on it. If they wish to end the small but serious efforts directed toward molecular manufacturing, Scientific American might try to find a less absurd and insulting comparison.

As proof of the validity of nanotechnology, some advocates (including Merkle) like to point out that our world is thoroughly populated with self-replicating, molecule-manipulating entities [1]. True, but none of them is based on Drexler's notion of an assembler, a molecular construction mimicking the mechanics of factory equipment [2]. The molecular technology of cells is sui generis [3]: a ribosome is nothing like a robot. [4]

[1] Yes, this is called an existence proof.

[2] And this is called repeating the observation that nanotechnology has not yet been developed.

[3] "Sui generis: without a counterpart or equal; unique." Scientific American here is either repeating the observation that nanotechnology has not yet been developed (enough: mercy, please...), or saying that biological molecular machines are fundamentally and in principle unique. This smacks of vitalism -- at heart, a religious argument. We'll count this one as an attempt at persuasion by repeated repetition.

[4] Let's see. A ribosome is a precisely formed three-dimensional system that works together with other devices to read digital instruction sequences that direct the assembly of parts to form other precise, three-dimensional devices. Right -- nothing at all like a robot. How silly of biologists to call it a molecular machine.

Some definitions of nanotechnology may embrace such solutions, but if so, how do they differentiate nanotechnology from more advanced biotechnology? [1] Why does there not seem to be more interest among nanoists [2] in modifying cell organelles rather than reinventing the wheel at the submicron scale? [3]

[1] We could draw some useful distinctions here, but matters of definition aren't the issue. Much of the theoretical work on pathway technologies looks to molecular biology for inspirations on how to build and use self-assembling molecular systems; long-range goals are another matter. (We would count the majority of contributing experimentalists as working on self-assembly.)

[2] A term invented by ink-oriented paperists of ill repute.

[3] At last, a substantive question, not about feasibility, but about competing approaches. If Scientific American were genuinely interested, they could request an article from Drexler or Merkle. A few hints: human beings didn't reinvent the macroscopic wheel, because biology hadn't done it -- the horseless carriage looks nothing like a horse. Jet aircraft have wings and fly, but don't have muscles or feathers, and weren't made by modifying birds. To an engineer, cells above the molecular level look suspiciously like jellied crud reengineered from bags of seawater. Strong materials like diamond and graphite look much better on the nanoscale for the same reasons they do on the macroscale.

And finally...

* 9) Scientific American is trying to repress debate on the article [1].

This is not a criticism of the article, but it is a charge leveled at the magazine by some participants in this discussion, and it deserves to be addressed. In a letter to the Foresight Institute, the magazine pointed out that Dr. Merkle's quotation of our entire article--even with his own commentary interspersed throughout it--is frankly a violation of our copyright [2], and asked that the article's text be removed from the Web site [3]. In its reply, the institute has argued that this republication of the article is allowed under the "Fair Use" provisions of the copyright law.

[1] Well, yes. Scientific American sent email to Foresight stating that our use of quotations from Scientific American requires prior permission.

[2] We responded that a balanced consideration of the facts indicated that our Web article falls under fair use provisions. If we had offered Scientific American's article to the public as something of value, they might have had a case.

[3] We offered to transfer the Scientific American article's text to their own Web site, including anchors on each paragraph to facilitate fine-grained criticism from our site. This drew no response.

Some people who have learned of Scientific American's request apparently equate it with the Church of Scientology's efforts to curb the Web publication of its secret writings. [1]

[1] The analogy is obvious, but inappropriate, and we neither suggest nor endorse it. These complaints would likely vanish, however, if Scientific American were to move more briskly toward open, Web-based publication on matters of science and public policy. We urge them to do this, rather than hiding (relatively speaking) in the current cellulose-based medium, telling critics that the "proper way" to respond is to send a letter in the faint hope that the offending editors themselves will choose to revise some portion of it for distribution months later in linkless bundles of paper.

How are we squelching debate? [1] We're not trying to stop anyone from criticizing the article--frankly [2], we appreciate the extra attention [3]. We only asked that no one infringe on our copyrighted text in the process by posting the entire thing online [4]. Scholars don't need to reprint War and Peace to comment on it, after all [5]. But in making that request, we're not drawing a veil of secrecy around anything. We've published the article, for heaven's sake [6].

[1] Ineffectively.

[2] Explicit frankness -- excellent.

[3] In the short term, the discovery of an axe-murderer on their staff would probably boost circulation.

[4] Agreed. We urge all Web-based publishers be sure of their ground, both legally and ethically, before reviewing a whole news article.

[5] But when a work is genuinely rich in nonsense, both subtle and blatant, the astonished commentary can easily grow far larger than the document. It is surely a courtesy to readers to let them see the actual text under dissection. Better yet, having the text immediately accessible to readers helps to discipline the critic, removing the temptation to abuse straw men.

[6] Pretending that anyone charged secrecy sends the straw flying. Bias, bluff, content-free but verbally slick responses, even the sin of evading criticism, yes. Secrecy, of course not -- it's the act of publishing the article that was offensive, for heaven's sake.

Anyone who wants to read it in its entirety [1] can buy the magazine, look at it in a library, or read it on America Online. As to whether we are entitled to make that request [2]: Fair use, as defined in section 107 of the Copyright Act of 1976, does allow critics and scholars to quote from published works. Yet nothing in the law automatically entitles anyone to quote a text in its entirety [3]. In fact the law says that the amount of quoted material and the degree to which the republication of the work may interfere with the author's ability to derive income from it both bear on the question of infringement [4]. Merkle quoted the entire article and put it on the Web at the same time the magazine was selling on the stands, thus creating a disincentive for people interested in his criticisms to check them against Stix's original [5].

[1] With no distracting shrieks or belly-laughs from the intended victims.

[2] A request which, if "entitled," would be an enforceable demand.

[3] Automatic entitlement to quote a text in its entirety? Mind the flying straw.

[4] Among other factors to be weighed, such as the nature of the use.

[5] Their earlier statements suggest that the criticisms boosted circulation; this is probably true, in the short term. Many readers were awed and fascinated by Scientific American's willingness to publish such an article. Subscription renewals are another matter.

Our lawyers and the institute's lawyers [1] could fight out the merits of our respective arguments, but they are not going to. Why? Because we in Editorial have asked our business colleagues not to press the case, though they would surely win it [2]. We've decided that the ongoing discussion of Stix's article is more to the magazine's benefit than to its detriment [3].

[1] Gee, do we look that big? We've spent substantially over $100 on legal expenses so far, but we look to our membership to make good the loss.

[2] A pre-retracted bluff? Innovative, and cheap for everyone.

[3] But this suggests that the income-interference legal argument they propose above is, in their own stated judgment, false. Odd. Do they believe that they "would surely win" their case through yet another secret-but-powerful argument, or did they anticipate success based on driving home an admittedly false argument? But perhaps there is another interpretation -- that the benefit is not a matter of controversy boosting circulation and short-term income, falsifying their legal argument, but is instead a matter of painful external pressures raising editorial standards, to the long-run benefit of Scientific American as an institution in world society. Yes, it would be grand to have the old Scientific American back again.

So where does this leave us? In his conclusion, Merkle writes, "Scientific American should stop evading the fundamental technical question: given the currently accepted understanding of natural law, is nanotechnology feasible or is it not?" It's hard for us to believe anyone who has read the article would think we have sidestepped that question, but we don't mind answering it again: sorry, but far too many serious scientists say it is not.

Their article evaded the question. Their response runs from it at a gallop, leaving an extraordinary smokescreen of unstated statements by nameless authorities. Perhaps the Web is not suited to a style of argumentation employed by editors grown soft through years of answering only those criticisms they approve in advance.


John Rennie
Editor in Chief
Scientific American

Gary Stix
Staff writer

We suspect that most readers with the strong stomach and taste for grim humor required to wade through to this point will agree that, unless and until Scientific American can muster a response of genuine substance -- a criticism that, for example, refers both to actual proposals and to physical laws -- they can reasonably be scored as having lost the debate. Next round?

Chris Peterson and Gayle Pergamit
for Foresight Institute
(with Eric Drexler)

Round 3 -- Part 1 | Part 2

Round 1 | Round 2 | Round 3 | Round 4 | Epilogue

Overview of the Debate


Foresight Programs


Home About Foresight Blog News & Events Roadmap About Nanotechnology Resources Facebook Contact Privacy Policy

Foresight materials on the Web are ©1986–2020 Foresight Institute. All rights reserved. Legal Notices.

Web site developed by Stephan Spencer and Netconcepts; maintained by James B. Lewis Enterprises.