Small talk with Ralph C. Merkle.
Ralph C. Merkle is a key advocate of molecular engineering (nanotechnology), a cutting-edge science that involves rearranging molecules in order to create self-replicating manufacturing systems. Merkle was a research scientist at Xerox PARC from 1988 - 1999. He is currently an advisor to the Foresight Institute and a principal fellow at Zyvex.
UBIQUITY: Bill Joy's recent Wired article on the perils of today's advanced technologies -- including nanotechnology -- has certainly received a lot of attention, and we did a follow-up interview with him in Ubiquity. What are your thoughts on that subject?
RALPH C. MERKLE: Well, certainly the idea that nanotechnology would raise concerns is something that actually was a major impetus for the founding of the Foresight Institute back in 1986 -- and by 1989, Foresight had its first technical conference on nanotechnology, and in fact Bill Joy spoke at that meeting. So one of the things that's a bit surprising is that Bill's concerns about nanotechnology seem to be quite recent -- just the last year or two -- even though the understanding that this particular technology was going to be very powerful and would raise significant concerns has been around for at least a couple of decades.
UBIQUITY: Why don't you take a moment now to tell us about the Foresight Institute?
MERKLE: The Foresight Institute ( http://www.foresight.org/guidelines/index.html) was created primarily to guide the development of nanotechnology, and it was founded in large part because, when you look at where the technology is going, you reach the conclusion that, though it has great potential for good, there are also some concerns which need to be addressed. We've been having a series of gatherings at Foresight now for some years where Senior Associates (people who have pledged to support the Foresight Institute) can get together informally and off the record and discuss the various issues.
UBIQUITY: What are the meetings like?
MERKLE: The most recent gathering had over 250 people -- including Bill Joy, as a matter of fact -- and one of the sessions was a discussion of the Foresight guidelines for safe development of nanotechnology. A year and a half ago we had a workshop where we discussed the guidelines and worked out an initial draft, which was discussed at the 1999 gathering at Foresight, and then further modified and updated and then discussed again at the most recent gathering.
UBIQUITY: What do you think would explain the sudden increase of concern about this?
MERKLE: Well, I can't really address the specifics of Bill Joy's situation. I do know that nanotechnology is an idea that most people simply didn't believe, even though the roots of it go back to a lecture by Richard Feynman in 1959 ( http://www.zyvex.com/nanotech/feynman.html). That was a very famous talk in which he basically said the laws of physics should allow us to arrange things molecule by molecule and even atom by atom, and that at some point it was inevitable that we would develop a technology that would let us do this. I don't think that it was taken very seriously at that time, but as the years progressed it gradually began to be more accepted. If you think the technology is infeasible, you don't worry about what it might do and what its potential is. However, as you begin to internalize the fact that this technology is going to arrive and that we are going to have a very powerful manufacturing technology that will let us build a wide range of remarkable new products, then one of the things that arises is a concern that this new set of capabilities could create new problems, new concerns, and that these should be addressed.
UBIQUITY: But not the way Bill Joy is addressing them?
MERKLE: One of the things about Bill Joy's original article that concerned me is that he was calling for a relinquishment, as he put it, of research -- and I think that's a very foolish strategy. If you look at the various strategies available for dealing with a new technology, sticking your head in the sand is not the most plausible strategy and in fact actually makes the situation more dangerous.
UBIQUITY: Why so?
MERKLE: For at least three reasons. The first, of course, is that we need to have a collective understanding of the new technology in order to ensure that we develop it appropriately. The second reason is that the new technologies that we see coming will have major benefits, and will greatly alleviate human suffering. The third reason is that, if we attempt to block the development of new technology, if we collectively try and say, "These technologies are technologies that are not meant for humans to understand," and we try to back away from them, what we effectively have done is not to block the technologies, we have simply ensured that the most responsible parties will not develop them.
UBIQUITY: So you think "relinquishment" is exactly the wrong strategy.
MERKLE: Right. Those people who pay no attention to a call for relinquishment, and in particular those people who are least inclined to be concerned about safe development will, in fact, be the groups that eventually develop the technology. In other words, a relinquishment of the new technology, unless it is absolutely 100 percent effective, is not effective at all. If it's 99.99 percent effective, then you simply ensure that the .01 percent who pays no attention to such calls for relinquishment is the group that will develop it. And that actually creates a worse outcome than if the responsible players move forward and develop the technology with the best understanding that they have and the best efforts to ensure that the technology is developed in a safe and responsible fashion.
UBIQUITY: Let's go back to the second reason and expand on that to the extent of enumerating what you consider are the most prominent hopes that it offers.
MERKLE: Well, certainly what we see today is an entire planet, which has many limitations. I'm not quite sure how to express it, but certainly if you look at the human condition today, not everyone is well fed. Not everyone has access to good medical care. Not everyone has the basics -- the physical basics that provide for a healthy and a happy life. And clearly, if you have a lower cost manufacturing technology, which can build a wide range of products less expensively, it can build, among other things, better medical products. Disease and ill health are caused largely by damage at the molecular and cellular level, yet today's surgical tools are too large to deal with that kind of problem. A molecular manufacturing technology will let us build molecular surgical tools, and those tools will, for the first time, let us directly address the problems at the very root level. So today we see a human population of over six billion people, many of whom have serious medical conditions, which either can't be treated or cannot be treated economically. In other words, we don't have the resources to effectively treat all the conditions that we see. If we can reduce the cost and improve the quality of medical technology through advances in nanotechnology, then we can more widely address the medical conditions that are prevalent and reduce the level of human suffering. (See http://www.foresight.org/Nanomedicine for more information about medical applications.)
UBIQUITY: And besides the opportunities in medicine? What else?
MERKLE: On another level, food; the simple process of feeding the human population. Today because of technological limits there is a certain amount of food that we can produce per acre. If we were to have intensive greenhouse agriculture, which would be something we could do economically, if we could economically manufacture the appropriate computer controlled enclosures that would provide protection and would provide a very controlled environment for the growth of food we could have much higher production. It looks as though yields of over 10 times what we can currently grow per acre are feasible if you control, for example, the CO2 concentration, the humidity, the temperature, all the various factors that plants depend on to grow rapidly. If we control those, if we make those optimal for the growth of various crops then we can grow more per acre. And furthermore, we can grow it less expensively because molecular manufacturing technology is inherently low cost, and therefore it will let us grow more food more easily.
UBIQUITY: What are the implications?
MERKLE: The first is that it makes food less expensive. The second is that many of the people in the world today who are starving are not starving because there is an inherent inability to produce food, they are starving because they are caught in the middle of political fights and blockades that have been used as political weapons. As a consequence, food is available but it cannot be shipped into an area and so the people in that area suffer the consequences. However, if you have a distributed manufacturing technology, one of the great advantages is that it should let us have a much lower cost infrastructure. In other words, today manufacturing takes place in very large facilities. If you want to build, for example, a computer chip, you need a giant semiconductor fabrication facility. But if you look at nature, nature can grow complex molecular machines using nothing more than a plant.
MERKLE: Well, a potato, for example, can grow quite easily on a very small plot of land. With molecular manufacturing, in a similar fashion, we'll be able to have distributed manufacturing, which will permit manufacturing at the site using technologies that are low cost and easily available once the core technology has been developed. And as a consequence, you would have people able to build low-cost greenhouse agriculture tools even if there were a blockade because the manufacturing facilities would be widely distributed, and therefore they could avoid the blockade by simply making what they need inside the blockaded region using cheap raw materials and sunlight.
UBIQUITY: And if nanotechnology did so much for people's health and food production, what would it do, do you suppose, for their current economic institution? Would it transition to large-scale nanotechnology? Have unintended consequences in terms of disrupting the economy?
MERKLE: I think we would see changes in the economy. Previous technologies have made major changes. Old companies that have had major advantages in the past certainly find those advantages go away. Certainly as manufacturing becomes less expensive, then today's major manufacturing companies would find that they would be at a disadvantage in the future. Other companies that are producing intellectual products, software companies or companies that are not dealing with material objects -- banks and financial institutions, for example -- fundamentally are dealing with a flow of information so would be relatively less affected. I think you would see some major shifts in the economy in that manufacturing companies would find that what they were doing was either greatly changed or outright replaced. As in any technological revolution, there will be winners and losers. On balance, everyone will come out ahead, although there will be specific instances where particular companies will have major problems, and in fact, will simply not be able to cope with a new environment and presumably suffer the consequences.
UBIQUITY: What about the competition between different countries? Would, for example, the severely underdeveloped countries have an ability to do very rapid catch-up?
MERKLE: Yes. I think they would. Also, you have to remember that we are looking at a future where to a first approximation everyone is wealthy. Now, there are certain things that are inherently scarce. For example, there is only a certain amount of beachfront property in California. It is going to be scarce, it is going to be expensive, and only a small percentage of the population will be able to afford it. But if you look at other material possessions housing or electronics -- you find that the only limitation is our ability to manufacture them inexpensively. So the first approximation in this future that we're looking at is that everyone will be physically well off. They will have a great abundance in material goods, and as a consequence, I think that will soften and ease some of the conflicts that we see now. One of the issues facing us today is that there are countries where there is a serious lack of resources, the standards of living are very low, and as a consequence this creates a fundamental unease and discomfort in entire populations. If you have a higher standard of living, at least that source of conflict will be greatly reduced. Now, as we all know, there are many potential sources of conflict in the world, but even easing some of them will be very helpful.
UBIQUITY: Does nanotechnology apply to just about any industry, to just about any kind of manufacturing?
MERKLE: If there is a manufacturing component, then pretty much it applies. I think there'll be some instances where existing bulk chemical manufacturing techniques will continue to be used because they prove to be efficient.
UBIQUITY: What are the implications for the future?
MERKLE: One of the things that we really have to adjust to as we start thinking about this future is that, if there is a great quantity of material goods, then the primary questions no longer are what can we make, what can we afford? The primary question shifts to, what do we want? What is it that we find valuable intrinsically rather than what is it that we can afford to make from a manufacturing sense?
UBIQUITY: Give us a sense of the rate of growth in the field.
MERKLE: Certainly we are having more interest. The recent announcement by President Clinton of the National Nanotechnology initiative in a speech at Cal Tech gives an indication that the interest in this area is growing. We're also seeing a wide interest on the part of private industry. We're seeing a number of companies beginning to move into this area. My own company, Zyvex, (http://www.zyvex.com) is very interested in the development of molecular manufacturing because we feel that this is going to be a major advance not only in the human condition and in manufacturing, but also a major advance that we can help create. It's still early and obviously when you have startup companies entering into a new area, some of them will be successful and some won't. But we're now seeing that process begin, which I think is a very interesting development.
UBIQUITY: Is nanotechnology something that is now covered reasonably well in institutions of higher education?
MERKLE: Well, certainly there are a number of nanotechnology research centers around the country. Obviously Rice University with Rick Smalley's Center for Nanoscale Science and Technology is one. Bill Goddard has a molecular simulation center down at CalTech. There are a number of other nanotechnology centers springing up. One of the things that interests me is that some of the concepts and some of the ideas involved in nanotechnology have been more rapidly accepted, and other ideas are still being absorbed. The concepts that are most rapidly accepted are those that have built on previous research in a fairly direct fashion. For example, we now have the ability to manipulate matter at a smaller scale and do so somewhat precisely.
UBIQUITY: What's the core idea here?
MERKLE: One of the concepts that is essential to molecular manufacturing is that of a self-replicating manufacturing system. And that concept has lagged behind in its acceptance even though it's fairly obvious that such things are feasible. So in order to produce the economies that we're talking about, in order to economically produce complex products we're basically going to adopt a strategy which has been demonstrated by agricultural products. As I said, potatoes are a miracle of biology, and yet they're so inexpensive that we can mash them with butter and have them for dinner. The reason for this, of course, is that the potato can grow more potatoes. Potatoes are a self-replicating manufacturing technology. And this demonstrates the feasibility of a basic capability, which is to have a self-replicating manufacturing base. Now, the moment I say that, the moment I say we will build artificial, programmable molecular manufacturing systems, this creates a world of confusion.
UBIQUITY: How does it create confusion?
MERKLE: It creates a world of confusion for a very simple reason, which is that, when we think of self-replicating systems, we think of living systems. So if someone in the 1800s had pointed at birds and said, "Birds fly, therefore we know heavier-than-air-flight is feasible," that would be accepted as an argument, because obviously birds fly. But it would also create terrible confusion because people would say "Does that mean we're going to build flying devices that flap their wings and have feathers and hunt for worms in the woods?" And the answer is no, even though the biological example demonstrates the feasibility of the basic concept, the image of a modern airplane is very, very different from that of a bird. A 747 bears no resemblance to a hawk, and the image of a 747 swooping down out of the sky and clutching a hapless cow in its landing gear simply doesn't make sense.
UBIQUITY: How will artificial molecular manufacturing systems be unlike living systems?
MERKLE: They will not have the marvelous adaptability that living systems have. They will, like a 747, use a refined source of energy -- 747s use fuel. If you cut them off from fuel, 747s do not fly. They cannot suck sap from trees in the forest and continue to function. In the same way, if we talk about molecular manufacturing systems, they will use a specific source of energy, a specific fuel, and cut off from that source of fuel, they won't function. So I think that's one of the inherent issues that needs to be addressed up front, which is that when we talk about artificial self-replicating manufacturing systems, they will not be living systems. They will not be biological in their design. They will be very artificial. They will be very brittle. Like machines built by human beings they will function correctly if we provide them with a very specific environment. But if that environment is changed they won't function at all.
UBIQUITY: Any predictions about the future?
MERKLE: From a fundamental point of view we don't attempt to predict what will happen. What we can do is describe what is possible within the laws of known physics. Physics is well understood. Within that framework of well understood physical law we can describe some of the capabilities that we could develop, some of the things we could do, and we can describe systems that would be feasible. Quite likely by the time we actually develop such a system, there will be alternative designs that will be better.
UBIQUITY: Give us an example of a product that would be improved using molecular manufacturing?
MERKLE: The answer that comes most readily to mind is diamonds. Diamond has a better strength-to-weight ratio than steel or aluminum. Its strength-to-weight ratio is more than 50 times that of steel or aluminum alloy. So, it's much stronger and much lighter. If we had a shatterproof variant of diamond, we would have a remarkably light and strong material from which to make all of the products in the world around us. In particular, aerospace products -- airplanes or rockets -- would benefit immensely from having lighter, stronger materials. So one of the things that we can say with confidence is that we will have much lighter, much stronger materials, and this will reduce the cost of air flight, and it will reduce the cost of rockets. It will let us go into space for literally orders of magnitude lower cost.
UBIQUITY: Has NASA shown any interest in this?
MERKLE: Needless to say, they are pursuing research in nanotechnology with the idea of having lighter, stronger materials as one of the significant objectives. There is a whole range of other capabilities, of course, that would be of interest in NASA. For example, lighter computers and lighter sensors would let you have more function in a given weight, which is very important if you are launching things into space, and you have to pay by the pound to put things there.
UBIQUITY: Are there any other areas that would be significantly affected by nanotechnology?
MERKLE: The other area is in advanced computer technology. The computer hardware revolution has been continuing with remarkable steadiness over the last few decades. If you extrapolate into the future you find that, in the coming decades, we'll have to build molecular computers to keep the computer hardware revolution on track. Nanotechnology will let us do that, and it will let us build computers that are incredibly powerful. We'll have more power in the volume of a sugar cube than exists in the entire world today.
UBIQUITY: Can you put any kind of timeframe on that?
MERKLE: We're talking about decades. We're not talking about years; we're not talking about a century. We're talking decades -- and probably not many decades.
Note: For a good introduction to nanotechnology, see http://www.zyvex.com/nano.