University of California, San Francisco

Presented at NASA Goddard Space Flight Center

14 April 2000 Science Colloquium

Several weeks ago I was having lunch with two old friends, both senior scientists at the Lawrence Livermore National Laboratory. I told them that I had been invited to give this talk and was close to deciding how to explore its main theme. One of them had already seen the abstract posted on your website and told me he was pretty sure the title would leave people completely mystified and would probably persuade at least some of you to catch the early liberty boat today. I told him that I had chosen "preternatural pentadactyly" and "planetary probes" not simply because I have a fondness for alliteration but because this title afforded the maneuvering room I felt I needed to prepare a talk about the possible value of astronauts to a Mars mission. Based on what I had read in Jerome Groopman’s superb piece, Medicine on Mars, in the February 14 issue of the New Yorker, it was obvious that safely transporting even a small crew to Mars and back would be a long and extremely expensive undertaking, so taking a fresh look at the pro’s and con’s of that decision was appealing to me.

My Livermore friends, both of whom apparently know something about the history of internal debates about manned space missions, quickly took opposite sides on the astronaut-to-Mars question, and both argued the case purely on financial grounds. Peter, I will call him, said NASA was afraid to spend the money it needs to do the job right. Paul, I will call him, said it would be far less costly to complete the mission without humans aboard and the results would not be compromised.

After listening to them for a while I told them I didn’t think cost was really the core issue. Obviously, the expense associated with a manned mission could easily be so great as to force a rejection purely on financial grounds. But placing a crew on board necessarily alters the potential payoff in some way B quite possibly in a fundamental way B and even if a human passenger were treated as nothing but extremely fragile cargo, the public at large would attach a very different meaning to a manned mission than it would to an unmanned mission, no matter what the cost.

So I told them I thought we have to ask what we think can be accomplished under Case One B computers and robots only B and how Case One goals would be modified or even transformed under Case Two B humans aboard. I see no way to decide the question, nor any basis for making cost-benefit or risk-reward projections, until we recognize that an inaugural manned mission to Mars could never have the same goals as those set for an unmanned mission.

How exactly would these goals differ? What an interesting question that is! What is it, exactly, that a physiologically vulnerable human knows and could do on Mars that a mechanically robust and maximally computerized surrogate would not know or could not do? Any answer would have to apply not only now but twenty years from now, when we expect to be ready and able to send a crew to Mars. Surely you can see that this debate is not simply about Mars but about ourselves: What, if anything, is distinctive about human intelligence, and what is the prospect that computers will have made that distinction irrelevant within the next twenty years?

So that’s what I want to talk about this afternoon. I will begin by suggesting that it would not be a waste of your time to look very closely at the debate that has been going on between the Artificial Intelligence theorists and more traditional psychologists and cognitive scientists, and to hear what a neuro-Luddite like myself thinks about that discussion C well, you are going to hear at least a few words about that. Mostly, I want to introduce you to a way of thinking about intelligence that gets you away from the university and out into the world where people actually live, scheme, make discoveries, and take actions. Having done that, I will take you back a few million years to the time when human ancestors stood up and began to use their upper limbs in ways that had not been possible prior to that.

Then I will walk you very quickly through the thesis that grew out of the decade of work that led to my book on the hand, that thesis being that the human hand and brain became an integrated system for perception, cognition, and action through a process of co-evolution, so that what we think of as human intelligence is as deeply imbedded in the hand as it is in the brain. Hopefully, without my having to explain this, you will leave here understanding why it didn’t surprise me at all to read in the Groopman article that Dr. John Charles at NASA’s Space Biomedicine Reseaarch Institute says everyone agrees that the most important person on a Mars mission will be a handyman. How interesting that the first hominid with a brain large enough to compare to Homo sapiens was the first hominid to manufacture stone tools. Homo habilis lived two million years ago. The name means, literally, handyman. By the end of the hour, in other words, you will have a few anthopologic ideas to consider throwing into the mix as you approach the moment when a manned Mars mission has itself evolved from being a distant goal to that all too familiar go/no go decision that must be made, with real lives riding on it.

For those of you who have to leave early, let me tell you how I would decide the question about including astronauts or leaving them behind. The answer actually becomes trivial once you provide just one additional bit of information. What I need to know is, do we explore Mars because we know exactly what we are looking for and that we must go there to find it; or are we going because we have no idea what we might find nor any idea how to look for it? Putting the proposition this way makes it sound like a no-brainer. Really! Who’s going to spend a nickel on an age-of-Aquarius cosmic fishing expedition? But if we really are sure we know what we will be looking for on Mars in 2020, then this really is a no-brainer. There would be no way to justify sending a live crew along, because the AI and robotics people are unquestionably right on this point: within twenty years we really will be able to find anything we’re looking for on Mars, and we will be able to do this without exposing human passengers to the inherent risks of that trip C everything from the invisible high-speed iron filings to correspondence-course neurosurgeons C nor will we need to expose ourselves to the burdensome expense and worry of a manned mission.

Does this mean I think you should drop your plans for a Mars astronaut program? Not at all. Let me go back to the question I said I wanted answered before deciding whether or not to go with the high-price option. But this time I will rephrase it in the form of a calculation I’d like you to make, or at least try to make. What are the probabilities that the principal, enduring accomplishment of a manned mission to Mars will prove to be something completely unexpected? Mind you, I am not referring here to the familiar cornucopia of research-driven transferrable technology. What I am referring to is a result that depends on a live crew actually traveling to Mars (and hopefully returning), a result that was not only completely unplanned but for which no specific preparations could have been made, and a result that either immediately or eventually provides a benefit to life on Earth of historic proportions.

The probability of such a result, I would like to suggest to you, is not only not zero, but may not even be close to zero. In order to help you construct such a calculation, let me give you two imperfect but suggestive examples of the sort of result I think might qualify. The first example is the Apollo 11 mission that began on July 11, 1969 and that was undertaken as much for geopolitical as for scientific reasons. However much that mission achieved its planned objectives, I think most people would agree its most important and enduring result will prove to have been neither political nor scientific, but emotional. Within an extremely short span of time that mission created an essentially worldwide consensus, born of both sensation and imagination, that humans occupy a small and exquisitely beautiful but fragile planet that must be protected. There is no way to know whether any visual image alone, either from 11 or from any of the other Apollo flights, could have produced the same effect, but I strongly doubt it. The spontaneous reactions, movements, and words of Neil Armstrong and Buzz Aldrin made us all the earth-struck companions of men whose playful romping and cosmic contemplating came at us all in one piece B as if we had gone there ourselves B and gave us our first taste of membership in a planetary community. No one who participated in that experience will ever be the same, nor can life on Earth ever be the same.

The second example of a long voyage that completely transformed the world in a way that was not only unintended but that no one could have predicted is that of H.M.S. Beagle, a small British frigate that sailed from Plymouth, England on December 7, 1831 and returned to Falmouth, England on October 2, 1836. On board at the beginning, as you all know, was Charles Darwin, a 22-year old rich-kid just out of Cambridge who had been recruited by influential friends and former teachers and then approved by the ship’s captain to serve as his mealtime companion. Robert FitzRoy, although just a few years older than Darwin, had already been captain of the Beagle for three years and already had one long coastal surveying mission in Tierra del Fuego under his belt. FitzRoy had sought a companion B a scientifically-inclined gentleman, to be specific B because he felt himself at risk for the same suicidal depression that 3 years earlier had taken the life of Pringle Stokes, the previous skipper of the same vessel. When Darwin accepted, his hope was to have a bit of travel and adventure before settling down to life in the English countryside as a gentleman-minister. But things did not turn out as expected. A planned 2 or 3 year South American coastal mapping expedition stretched into 5 years C during which time Darwin turned himself into a first-rate revolutionary geologist, by the way C and the single most significant revolution in the history of biology, before or since, resulted.

It is worth noting, by the way, that the unexpected and unplannable outcomes of these two trips emerged on very different time scales (roughly 5 days compared to 5 years), involved complex interactions of two radically different pairs of explorers aboard two radically different vehicles, whose thoughts and experiences were disseminated by radically different means under radically different social conditions. The emotional consequences of the first manned moon mission were immediate, highly public, ignited and globally propelled by electronic media; by contrast, the revolution not only in biology but in religious and social institutions sparked by the voyage of the Beagle was a slow-motion global event that began not merely in private but in closely guarded secrecy, defining itself and acquiring importance over the span of several decades. Indeed, the influence of Darwin’s work has grown over the past century and is as vital a force in biology and as controversial a concept in religious doctrine today as it was at the outset.

With these two examples as background, I put it to you that the only reason to risk the lives of astronauts on an inaugural mission to Mars is that there is an uncertain but real prospect of a result that is utterly beyond the power of anyone anywhere to specifically predict or prepare for. By my definition, at least, this leaves out an exclusive or even leading role for computers and robots on the mission, because unless these machines have become completely autonomous in 20 years (in which case I don’t think we can count on them caring much about what interests mere humans) they will still be running on the tracks we lay for them.

So now you know why I want to talk about the extremely complex and controversial issue of human intelligence. Once you have a better idea what I mean by that term, I hope you will understand why I am suggesting that NASA should devote at least some time to examining its own views on this subject, and to the influence those views may be exerting on plans for future manned and/or unmanned missions to Mars.

I will begin not with my own work but with an extraordinary book written by the British psychologist Henry Plotkin, Darwin Machines and the Nature of Knowledge. Although this book is readily available in the United States B it was originally published in the UK in 1993 and issued as a Harvard University paperback in 1997 B it is not widely known and it is virtually never quoted in the contemporary American literature in cognitive science. The reason for this oversight, I have come to believe, is that Plotkin’s thought and writing are too clear and concise for American academic taste. Perhaps there are other reasons, but if so, they elude me.

To put his thesis into the briefest possible form, Plotkin proposes that we understand intelligence to be a special class of adaptive behavioral potentials found in a comparatively small number of species with somewhat unusual reproductive strategies. Most animals in this group exhibit what has been called a prolonged generational deadtime. What that means is that while these animals C like almost all other animals C receive their own unique, lifetime supply of genetic instructions at conception, unlike most other animals they do not make Ameal, mate, and morgue@ the short version of their personal biography. Animals with a prolonged generational deadtime delay paying off their genetic debt for a while C sometimes for several decades. And that’s long enough for the world to become a different place, perhaps a very different place, while they are otherwise occupying their time.

In order for us to understand the opportunities and the risks that this strategy creates not only for living individuals but for the future of their species, we have to arrange two adjacent generations on the same canvas. Our parents C yours and mine C effectively settled the details of our genetic inheritance at the moment they were conceived by their parents. Their genes predicted then what we would need by way of muscle mass to stand upright against the earth’s gravitational force, that we would want binocular vision, and sweat glands pre-set to react to a narrow range of ambient temperatures, and ears tuned to pick up the frequencies generated by the voices of other humans, and so on. Their genes not only arranged for us to receive all the standard equipment every new Homo sapiens needs, but prepared a schedule for bringing these systems on line at the right time and even gave us a curious and playful disposition so that we would find out for ourselves how it all works. All this stuff we would need and would need to become familiar with so that we, too, could become parents, constitutes what Plotkin has termed our primary heuristic C I’ll get back to that term a bit later, but you will recall from your classical Greek days that a heuristic is simply the generic term for a device which aids in learning or discovery. The point is that the human primary heuristic is so thorough, stable, and conservative that a gene package transported from an Egyptian who lived 5000 years ago would have served us just as well as the one we got from our parents.

But other aspects of our environment are far less stable than gravity or the amount of oxygen in the atmosphere. Indeed, looking specifically at the case of modern humans, who are the descendants of multiple generations of hominids that migrated over the entire face of the planet, we seem to have become not merely happy wanderers but compulsive agents for and architects of change of our own environments. And we reinvent ourselves wherever we go. Your parents and mine may have grown up in Africa, or Asia, or Europe, but you and I live in North America where the natives use spoken and written codes to communicate with one another that bear no surface relationship to the codes our parents used. We don’t use horses or carts to go to work, we use subways, cars, sometimes even airplanes. We play video games, buy food from all over the world in supermarkets or boutique grocery stores, watch in amazement as our wampum materializes and then turns to dust on our computer screens, and sometimes send people in small containers into outer space. None of these things that we do is specifically anticipated by anything in our our genes or in the genes of our parents.

Plotkin credits the British biologist C.H. Waddington with having noticed that human life exemplifies what he called the Auncertain futures problem.@ And we’ve just seen what that means. It means, if your Finnish parents decided what your genes would be in 1930 and you were born in Brooklyn in 1955, how come you not only speak English but sound as if your family has always lived in Brooklyn? How did you learn to get from Flatbush Avenue to Lincoln Center on a train without getting lost or killed? And how did you learn to bake a pizza and play the electric guitar? Plotkin’s sublime answer is, the human primary heuristic not only predicted that your world would be warm in the summer and cool in the winter, that you would have your first crush when you turned 14 and that you would like french fries the first time you ever tasted them, but that almost anything could happen in the world after your parents were conceived that it could not prepare you for. So it gave you a body with an unprecedented degree of operational flexibility and behavioral malleability so that you would not be undone by the avalanche of surprises waiting for you. It gave you the capacity to create your own personal kit of survival skills, which Plotkin calls secondary heuristics. These are the adaptive skills created, mostly through incidental learning or active instruction, as part of the lifelong process through which you calibrate yourself to the world not as it was in the good old days, but as you find it.

Your genes wed you to the present by giving you the capacity to discover at least some of the novel features of your world C not all of them C to analyze them, to relate them to more familiar features of the world, and to manipulate and rearrange them, and to craft a unique set of skills which would become such an intimate, fluent, and reliable part of your repertoire that they would look and behave as if they were genetic. Your so-called intelligence is not some number descriptive of a neuron or synapse count or a score on an IQ test, but the behavioral tool kit you built for yourself through experience. And it must be said that no matter how we labor to transform or reinvent ourselves, we have no way to Asave@ these discoveries or transformations genetically. So we devise extragenetic storage systems and experiential protocols for passing that information on to our young. We send them to school, and to libraries, to music lessons and summer camps, and to the Internet.

Let me briefly summarize: At conception every organism receives genetic instructions for the adaptations it will need to survive to sexual maturity. These instructions, which Plotkin calls primary heuristics, dictate how the organism will calibrate its own body to the physical details of its own immediate environment. The primary heuristic not only stages the acquisition of basic sensorimotor skills (when and how to swim, crawl, run, or fly) but sets critical perceptual filters for the recognition of sights, sounds, and smells that must be recognized and memorized immediately, initiates exploratory experiences in touching, vocalizing, walking, grasping, lifting, swinging, and so on, and directs the physiological unfolding of sexual maturation and the elaboration of mating behavior.

For some animals B and this is the case for all mammals, but especially for primates, anthropoids, and hominids (of whom we humans are the only surviving species) B survival has come to depend upon the acquisition of skills and behavior in an object world and a social world whose attributes and requirements change so rapidly that a high-level, specific preparation to respond to cannot be transmitted genetically. These animals have evolved organ systems that provide for discovery and learning in ways that can overcome the situational blindness of their primary heuristic. The adaptations and/or knowledge created by these special systems are called secondary heuristics, and the evolution of what we call intelligence is governed by the rules of the secondary heuristics.

Before moving on, let me make more concrete the distinction between a primary and a secondary heuristic. The human genome specifies that about 50% of the muscle mass of humans is located in the legs; for chimps, it’s a little less than 40%, which makes sense because humans rely almost entirely on the legs for locomotion, whereas chimps use their arms both for walking and moving through the trees. But the human genetic code says very little about how we will use our legs to walk B we can go barefoot, or we can put on shoes, or roller blades, or even skis, or travel by bicycle, if we are willing to take the trouble to learn by imitating others who already know how to do those things, or by having them teach us. It is the primary heuristic that codes for muscle mass, for the cadences of upright walking, and for a variety of common gait patterns because the force of gravity is very stable over enormously long spans of time. Through experience and training we gradually acquire higher level motor and cognitive skills that allow us to improve on simple walking, or to operate whatever special devices we may have created for that purpose. These are the secondary heuristics. Happily for us, probably because life was almost nothing but uncertain futures for the earliest hominids, our own brains have room to appropriate information about present context and the capacity to calibrate our perceptual-motor abilities as we grow.

I would like now to turn from the thesis of Plotkin’s book to the thesis developed in my book on the hand. In a phrase, for humans, evolution seems to have made the hand a sort of express lane for the definition and elaboration of secondary heuristics, a species-specific, high-bandwidth channel through which the human brain tunes itself to and communicates with the particular world in which the child finds herself or himself from the moment of birth. How did that happen?

When our earliest ancestors began life on the ground, neither the hand nor the brain were what they are today. Lucy, an australopithecine female who lived 3.25 million years ago, had several un-apelike anatomic features in the hand. As minor as these alterations to ancestral anatomy were B an increase in the length of the thumb compared to the fingers and the ability to rotate the index and middle fingers on their long axis B they gave Lucy’s clan the biomechanics needed for some new grips and hand movements. Most important of these was what is called the A3-jaw chuck@ B a grip that permits an irregularly shaped stone to be held tightly between the thumb, index and middle fingers. This grip is identical to that used by a baseball pitcher for an overarm pitch. Such a grip, of course, would have been extremely useful if the skill of throwing could be mastered for purposes of hunting or defense. No one knows how much aggressive or defensive overarm throwing the australopithecines actually did, nor do we know when know exactly when subsequent changes in the anatomy of the hand occurred or how they may have been exploited by Lucy’s descendants, but we do know that over the span of several million years, those of Lucy’s descendants who learned to take advantage of the hand came to dominate the bipedal world of the hominids and eventually outlasted all their competitors. When our ancestors came down from the trees, in other words, an upright walking posture had not merely relieved the forelimbs of their primary role in locomotion but had opened the door to a completely novel domain of perception, action, and interaction based in the hands, and our ancestors were the ones who walked through that door.

It is now widely acknowledged by anthropologists, archeologists, and cognitive scientists alike that the biologic success of humans has largely been due to an increasing dependence on tools. This success story is centered in the hand, beginning with Lucy’s 3-jaw chuck. Subsequent structural changes, mainly on the side of the hand opposite the thumb, allowed improved finger-to-finger contact and a greatly expanded range of both power and precision handling movements in manipulation of objects. There is still no accepted timetable for these later changes in hand structure, but the archeological record makes it clear that the exploitation of this potential to improve tools and tool use came very slowly. Stone knapping B that’s spelled with a k B is the deliberate chipping of rocks to make cutting tools that became a widespread practice and marked the appearance of Homo habilis about 2 million years ago. Somewhat more elaborate stone tools first appeared at about the time the worldwide migration of Homo erectus began a half a million years later. But the same tools that these pioneers carried with them across the entire face of the globe over a span of a full million years remained essentially unchanged for that entire period. Our earliest ancestors had restless legs, it seems, but were thoroughly contented with the modest state of their technology.

However difficult it may be for us to imagine this degree of indifference of human ancestors to tools and gadgets, we should not really be surprised. After all, it is highly unlikely that the brain would have been capable of controlling the novel and increasingly complex movements of the evolving hominid hand before the hand itself was physically capable of varying the hand grips and individual finger movements which are now part of our repertoire. From a neurologic and evolutionary perspective, in other words, the conservative position must be that the brain developed these control capabilities very gradually and modified them over time as experience with the hand defined its long-term role in hominid survival. And because of the extremely slow pace of genetic change at the species level, a very long period of time would have been required to assure that each new member of our species would arrive with an inborn potential for skilled hand use, plus the urge to take things apart and put them back together again.

In the first half of this century, anthropologists had not yet accepted that brain, body, and behavior could evolve in concert, or as a corollary that the structure and function of brain and hand can only have co-evolved. On the contrary, it was taught that the human hand was simply a primitive structure that had been elevated to importance in life by the gift of a larger and more advanced brain. But we now know better, and indeed it could be argued just as forcefully (and narrowly) that the truth verges on the opposite: that it was the newly configured hand B and the uses to which it could be put B that leveraged an ancient and highly successful simian brain into what it is now. Co-evolutionary theory leads us to see the process as reciprocal and complementary, without telling us anything about the specifics. But this conceptualization also leads us to a profoundly important inference about the modern brain: namely, that it diverges from other primate brains mainly in its having acquired a host of specialized neural operations to support sensorimotor and cognitive activities involving the hand.

Modern humans are so deeply and unconsciously immersed in a kaleidoscopic developmental and experiential world that we are for all practical purposes oblivious of the operation of this system. I was utterly unaware of this reality myself until I began working with injured musicians, and it was not until after I had been treating musicians with hand injuries that I began to wonder why they were always so upset when they were unable to work. Why should that be? For musicians, playing is not only real work but hard work, and yet for them, time off (even if it is paid time off) is uniformly unwelcome. Obviously, if musicians are simply emotional about everything B they are professional emoters, one might say B one could ignore their distress over interruptions in the daily routine. But there is more to it than that: musicians devote their lives to self-expression and are unusually liable to upset when physically unable to exercise their hard-won skills. When they are instrumental musicians rather than vocalists, obviously, their achievements depend upon the hands.

What about other examples of people whose professional work has an expressive component and depends upon skilled hand use? What about sculptors and painters, goldsmiths, machinists and blacksmiths, and other performing artists like jugglers, magicians, and puppeteers? These people, too, are passionate about their work. In the course of interviewing a number of such individuals, inquiring not only about their careers but about childhood interests, teachers, family support, and so on, I learned that for some people, at least, the hand is not merely a practical tool for making a living, but a master teacher and interpreter of the world, and the primary means of expressing ideas and feelings about it and of effecting change in it. There are certainly rudiments of this hand-based relation to the world in all of us, but these were people who, often when they were quite young, had staked their future on what they wanted to do with their hands, seeking inclusion in a professional community that defines itself on the basis of a specialized skill and sphere of knowledge.

The more stories of this kind I heard, the more powerfully I was struck by the parallels in the developmental path of all of these young people. It did not seem to matter whether it was music or fixing cars or quilting or juggling or building things out of wood or clay B a remarkable force was set loose when these young people began creating something personal with their own hands. Of course these were not the stories of the average child in our society, who grows up in a world that has largely reduced the hand to an incidental role in educational and working life. Given the way most of us live our lives, these stories B even when they were told by successful surgeons B made me feel as if I had stumbled across something akin to a Big Foot colony in the wilds of the Pacific Northwest. Whatever they were, they looked suspiciously like a behavioral or cognitive missing link in the human evolutionary story, a hidden subspecies of humans whose intelligence was geared to a time when we were tool users B which is to say, before we became device managers and computer users.

I finally convinced myself that these were not throwbacks at all, but rather unusually transparent expressions of the inborn tendency of most if not all humans to have a brain that expects the hand to be the prime bodily mediator of experience and communication. And the issue of communication brings me to a final point about the evolution of human intelligence. It wasn’t simply a matter of solitary hominids crouched over a pile of stones getting ready to go hunting. The anthropologist Peter Reynolds has emphasized that hominid tool manufacture and use cannot have been a solitary activity, except perhaps at the very beginning, two million or so years ago. As tools themselves became more complex constructions and as they came to be used for the manufacture of more complicated objects (weapons, clothing, shelters and the like) they required the cooperation of individuals working in groups, which is to say, the exploitation of diverse skills and knowledge within in a stable social context. And by the time Homo erectus had settled large parts of the planet, cooperative tool use in hunting and sheltering had become an essential underpinning of their survival strategy. This extraordinary revolution had taken place many thousands of years B hundreds of thousands of years, actually B before speech as we know it can have become part of the hominid behavioral repertoire. As a consequence of this experience, even language itself contains and builds on this ancient hominid dependence on skilled hand use and the increasing dependence of our ancestors of the tools they devised to help meet their needs.

It is not and never was the job of the human primary heuristic to predestine any child to be a carpenter or a mechanic or a computer programmer or a juggler. Its job was to allow for all of those possibilities, and many more besides, in the form of an entire suite of cognitive, inventive, and communicative potentials associated with an essentially open range of complex uses of the hand.

This brings me full circle back to my opening remarks, and in particular to my somewhat bold assertion that there is really only one reason to send humans on a mission to Mars, namely, because you don’t know what you were looking for and don’t know how to look for it. That is a mission for which humans are uniquely qualified, and a mission for which I can suggest no other satisfactory candidate. The absolute certainty of an uncertain future was a major driving force behind several million years of hominid evolution which produced the particular model of the primate brain which modern humans happen to possess, and that same ancient history is hardly irrelevant to your planning for our distant future.

For extra points, I would like to say a word not merely in favor of astronauts, but in favor of particular astronauts. I would side emphatically with Dr. Charles on the question of what sorts of skills belong on this mission, but for a stronger reason than the one he gives. It is not simply that you will need a handyman to fix broken pipes and broken bones, but that the handyman brings exactly the developmental pedigree you need to maximize the chances of finding what you are not looking for.

I notice that Dava Sobel will be talking on this series later this year. I am a great admirer of her books on John Harrison and Gallileo, and was intrigued to note that she points out that both men were accomplished musicians. Those of you who are fans of Patrick O’Brian’s novels know that both Jack Aubrey and Stephen Maturin were string players and that their teamwork over the years was greatly influenced by their experiences playing together. So I suppose I can safely recommend to NASA that the Mars crew be made up of handymen and handywomen who are qualified to play chamber music (or even perhaps jazz) together. And if they also happen to be the sort who make good dinner companions, you will probably have done all you can or even need to do to give human history a crack at achieving its second voyage of the Beagle.