The Reluctant Visionary

"Nanotechnology-driven manufacturing will change our world in fundamental ways—but we shouldn’t get too worked up about it."

In 1959, Richard Feynman delivered a lecture with the provocative title “There’s Plenty of Room at the Bottom." Speaking at a meeting of the American Physical Society at Caltech, the Nobel-laureate-to-be speculated about the possibility of manipulating matter at the atomic level via exquisitely small machines. Would it be possible, Feynman asked, for such machinery to configure atoms themselves, producing atomically precise outputs? Might we one day have billions of submicroscopic factories working in parallel to produce anything and everything we need?

It was a profound and exciting idea, and yet one that received very little serious attention in the years that followed, until an MIT student named K. Eric Drexler took up the cause in the 1980s. Working within Marvin Minsky’s MIT Media Lab, Drexler earned a Ph.D. in molecular nanotechnology—the first such degree ever awarded anywhere. Along the way he wrote the bestselling Engines of Creation (1986), which outlined his vision of nanotechnology for non-technical audiences, and the technical treatise Nanosystems (1991), which got into the nuts and bolts of nanotech.

Engines of Creation kicked off a worldwide nanotechnology craze. Corporations and universities began sponsoring research. Governments formed committees to develop technology roadmaps. Speculation in the media and popular culture grew ever wilder and more colorful, promoting images of tiny robots that could keep our clothes stain-free and our arteries unclogged, provided they didn’t go into an unstoppable feeding frenzy and reduce the entire world to a quivering mass of goo. Along with this buzz grew skepticism as to when and if we would ever see such technology, and whether molecular nanotechnology as described by Drexler was even possible.

Now, more than 25 years after the publication of Engines, Drexler returns to the subject of nanotechnology with Radical Abundance. Eschewing as tainted both by hype and bureaucratic mismanagement the word he introduced to the world, Drexler refers in his new work to “atomically precise manufacturing” (APM), which he says reflects the concepts he originally introduced.

Drexler devotes an early chapter to the functioning of a typical APM environment, a small factory roughly the size of a garage that produces, appropriately enough, automobiles. At the top or front of this fully automated factory, full-size automobile parts are assembled to produce a finished product. One step below or behind this level, smaller components that make up the auto parts are assembled from still smaller components. And so the system regresses all the way to the molecular scale. Each preceding level produces components of roughly half the size of the next and, because of the tremendous advantages of scale, operates at about twice the speed.

This small factory can produce a car in a matter of minutes, which doesn’t sound all that extraordinary when compared to today’s fully automated assembly lines. But there is really no comparison. Today’s assembly lines can produce a finished car from premanufactured parts in a relatively compact space and in an impressively short period of time, but where did those parts come from? How long did it take to make them, and the materials they were made from? And what is the origin of those materials?

In his classic essay “I, Pencil,” economist Leonard E. Read outlines the unexpectedly widespread origins of a humble wooden pencil. Trees from Oregon, graphite from Sri Lanka, clay from Mississippi, factice (the eraser) from Indonesia, and many other components come together to provide this simple everyday object. Imagine conducting such an analysis for something as complex as a modern automobile. A car that takes a few minutes to assemble actually takes years to build if we add together all the effort required to produce the (finally) ready-to-assemble parts from earlier components traced all the way back to raw materials.

But Drexler’s APM factory produces a finished car directly from raw materials, cutting years down to minutes and shrinking a globe-spanning supply chain to the size of the (remarkably small) factory. In his essay, Read notes that the knowledge required to make a pencil is distributed as widely as its constituent parts. In a strangely prophetic passage, he writes (speaking as the pencil):

"Since only God can make a tree, I insist that only God could make me. Man can no more direct these millions of know-hows to bring me into being than he can put molecules together to create a tree."

In Drexler’s vision of atomically precise manufacturing, the production of material goods becomes an instance of information technology: The finished car is a digital product comparable to a movie burned onto a DVD. All of the know-how required to turn a few basic materials into a working automobile is written into the software that governs the operation of the APM factory, which begins its assembly process by quite literally putting molecules together.

It is that first step of the APM process, molecular assembly, that is by far the hardest to pull off. The question of whether and how molecular assembly could be accomplished is at the crux of the ongoing controversy concerning nanotechnology. There is little dispute that a very small factory can be built that operates in essentially the same way as a full-sized factory, or even that a microscopic factory can be built to operate essentially the same way as the very small one. But as Feynman pointed out all the way back in 1959, and as Drexler goes to some length to explain, once we begin to approach the atomic scale, the rules are quite different. Gravity becomes much less of a factor, surface tension and friction become much more significant factors, and something has to be done about the fact that molecules are always vibrating. The portion of the APM system that operates at the molecular scale would therefore have to be very different from the rest of the system.

That first step has had no shortage of detractors, including the late Richard Smalley, himself a Nobel laureate for his discovery of buckminsterfullerene (“bucky balls”), one of the top scientific contributions to the field of nanotechnology. Drexler describes Smalley as “the leading critic of what were wrongly said to be my views,” citing multiple examples of inconsistency on Smalley’s part concerning both Drexler’s ideas and Drexler himself.

The two men famously debated the issue of molecular assembly in the pages of Scientific American and Chemical and Engineering News. As recounted in the footnotes to Radical Abundance, Drexler portrays Smalley as a primary contributor to many prevalent misunderstandings that surround nanotechnology, in particular the fear of deadly swarms of “nanobots.” Concerning molecular assembly, Drexler notes that Smalley’s major objection was the so-called “fat-finger” argument, which states that it would be impossible to make a stable and usable pair of molecular fingers (or pincers) that would be able to grasp a single atom in order to put it into place.

This argument is a straw man, says Drexler, with little bearing on anything that he has ever proposed or any of the likely paths to atomically precise manufacturing. He devotes a chapter to cataloging the different disciplines that currently achieve atomic precision. These include chemistry, genetic engineering, materials processing methods, and work that is being done with crystals. While skeptics argue that we are no closer today to nanotechnology than we were when Drexler wrote Engines of Creation, contributors to these fields—none of which is considered to be part of “nanotechnology” per se—are rapidly, if quietly, laying the groundwork for that first step of the APM process.

The significance of turning the production of physical goods into an information technology would be difficult to overstate. Drexler puts APM in context as the fourth major revolution after agriculture, the Industrial Revolution, and the digital revolution. APM borrows from and builds upon each of its predecessors, and has the potential to be as disruptive as each of them.

Consider how disruptive the move to the digital realm was for the music industry. In the analog world, recorded music was relatively scarce. Although the means existed by which we could produce our own copies of commercially manufactured recordings—remember the mix tape?—those technologies weren’t much of a threat to the recording industry. Most of the music people owned, they had purchased at a record store or other retail outlet.

Then along comes digital. Suddenly, creating a perfect copy of a commercially produced recording is as easy as copying and pasting text in an email. Music becomes “free” to anyone who has a Napster account. The music industry is shaken to its core and, although it fights back against the new model with some success, ultimately its survival requires that it morph into something very much like the model that is killing it.

Where music is concerned, we already live in an age of radical abundance. Similar transformations have occurred in book publishing and film and video production. But those transformations are nothing compared to what will happen when that same “copy and paste” paradigm can be applied to essentially any manufactured good. As with recorded music, the cost of producing such goods will drop to a fraction of what it currently is, while much of the infrastructure currently required to produce these goods will become obsolete.

But in this case, that obsolete “infrastructure” is, essentially, the entire world economy of physical goods, from the extraction of raw materials to the production of precise machine tools to the manufacture of finished products. So we have, on the one hand, a superabundance of everything we could want or need, and on the other hand, the complete destruction—it might be fair to call it the “creative destruction”—of the economy as we have known it. Drexler describes this scenario as one of “catastrophic success.”

That same catastrophic success is what hit the music industry a few years back. In the end, we can expect a worldwide physical infrastructure for the production and distribution of goods as different from what we currently have as iTunes is from the old record-store model. Of course, as painful as that transition may be, there is no doubt that we would be immensely better off for having made it, enjoying the same kinds of economic benefits that we gained in moving from an agrarian society to an industrial one.

In fact, we should expect those benefits to be significantly greater than the ones provided by the previous revolutions, seeing as this revolution is effectively the culmination of all of them. We are talking about a world where people can make their own stuff, anything they want or need, and even produce their own energy. Drexler doesn’t get into many specifics about how very bright that future might look, however. On the contrary, at this point he issues an unexpected warning about abundance of a particular kind. He sees little advantage to an abundance of enthusiasm.

"There’s something that I feel I must say to some of my readers, and I hope that they will understand a somewhat counterintuitive message and take it to heart. If you find these ideas about prospective technologies compelling, convincing, and exciting—if you imagine vistas far beyond any I’ve outlined, or see solutions to urgent global problems and feel the urge to share the full measure of your excitement—then please lie down until the urge passes. In the world as it is, this kind of excitement triggers a negative response, and for reasons that usually make sense; almost all grand ideas proclaimed by excited proponents turn out to be wrong and are generally discounted without consideration. If you want to make a positive difference, please help to keep fundamentals first, help to correct mistaken ideas, and join the conversation without shouting."

It seems that decades of clearing up misconceptions about fat fingers and swarms of lethal nanobots have taken their toll. Drexler is apparently tired of those arguments, tired of the hype, and tired of the true potential of this technology being, in his view, overlooked. He makes a sober and articulate case for why we should expect to see APM technologies become a reality in the near future. The impact of those technologies will be enormous.

It will be interesting to see whether he gets his wish. It is possible that APM will arrive in full force after we have had the chance to deliberate, to plan, to prepare ourselves for the shock. But if the previous revolutions are any indication, we can expect the real dialog about catastrophic success and radical abundance to take place even as we are being overwhelmed by those changes.

In the not so distant past, the concept of "Santa Claus Machines" and the disruption they would cause in society were explored in a variety of Sci Fi stories, without worrying about how they actually worked.

Lots of skilled trades would wither. Why learn how to operate a lathe or become an accomplished cook when a machine can just punt out a turned part or pumpkin pie?

The cost and VALUE of such goods would diminish, and when the high tech gizmos stopped producing, the population left would be ill equipped to rebuild a modern society.

“Drexler goes to some length to explain, once we begin to approach the atomic scale, the rules are quite different. Gravity becomes much less of a factor, surface tension and friction become much more significant factors, and something has to be done about the fact that molecules are always vibrating. The portion of the APM system that operates at the molecular scale would therefore have to be very different from the rest of the system.”

I read a few nanotech books about a decade ago, and Drexler’s was one of them. I remember this being one of the major obstacle to the kind of disruptive impact being predicted. Some proposed the need to build a nanoscopic equivalent to the machines we use for normal manufacturing. For example, we need nanoscopic wheels, rotary motors, levers, fulcrums built up into conveyor belts, transporting vehicles, etc. The reality is that those paradigms do not fit the behavior of things on a nanoscopic level. For example a cup is useful for holding water in our world, but in the nano realm water molecules hold themselves together and wouldn’t “pour” out of a cup even if you could make one that size.

It seems like these things may eventually become disruptive, but I think it will start a lot earlier than nanoscopic manufacturing processes and programmable matter. If we can just figure out how to make long carbon nanotube cables efficiently and inexpensively while preserving their incredible tensile strength to mass ratio, we can easily begin colonizing outerspace. (Structures like steel cables cannot support their own weight from the earth’s surf to low earth orbit as needed. Carbon nanotubes have over a hundred times the tensile strength by weight as steel.)

So, all that is needed is a supply of elements and a software app? No infrastructure needed at all? Where do the replicator machines come from?

I’ve read a lot of science fiction but I’ve never really understood the background of replication machines. They remind me of Li’l Abner’s Schmoos.

You’ll still need, for lack of a better term, a Molecular Printer. But once you have one, you can print a BIGGER one. All you need is feedstock, designs, and energy.

And it does NOT mean the end of craftsmen: we still have skilled craftsmen today. It DOES, however, mean the end of mass production. . .

So, all that is needed is a supply of elements and a software app? No infrastructure needed at all? Where do the replicator machines come from?

In effect, from other replicator machines. In other words, the first nano-robot would need to be assembled by non-nano means, but it would then build the second robot. Repeat as necessary.

The idea is that these molecular-scale robots could manipulate matter at the atomic scale, combining elements together mechanically without a chemical reaction based on a prerecorded plan.

Want a t-bone steak? It would become software. One t-bone steak would be disassembled by these robots while they recorded the location and type of every atom. It could then be reassembled at will. Just shovel in a load of dirt from your back yard.

Just dirt? Really?! I’ve always thought you would at least need chemically pure elements in precise amounts. Would any old dirt have enough of whatever elements are needed for, say that T-bone? Could you use the same dirt for everything? I suppose you could recycle garbage into useful &/or edible stuff?

It is close to magic, for me.

Just dirt? Really?! I’ve always thought you would at least need chemically pure elements in precise amounts.

Well, "chemically pure" and "precise amounts" take on new meanings when you can manipulate matter at the atomic scale. Each robot selects the atoms it needs one at a time - multiplied by trillions of robots, each running at incedible speeds.

Consider what a cow is made of - the plants that it eats! And the plants? Add some carbon dioxide and nitrogen from the air and yes, dirt.

We are mainly carbon and water, after all. Now there may be some trace elements missing from your shovelful of dirt, so a few miniscule vials and that should be it.

Disclaimer: Opinions posted on Free Republic are those of the individual posters and do not necessarily represent the opinion of Free Republic or its management. All materials posted herein are protected by copyright law and the exemption for fair use of copyrighted works.