To the editors:

Michael Fumento asks whether astronauts can contribute anything of value to the exploration of space, or whether we can, and should, rely on robotic space probes to do our exploring for us. He concludes that human spaceflight is unnecessary, dangerous, and a waste of money. There is much in Fumento’s analysis with which I disagree, but my main reason for responding is to correct the false impression given by his quotation from Lawrence Krauss that “the real science done by NASA has not involved humans.”¹ By any objective standard, this is simply untrue. There are numerous examples of scientific knowledge gained as a direct result of having humans in space. These include the monumental legacy of the Apollo missions, the repair and refurbishment of the Hubble Space Telescope, and knowledge gained in the life sciences.

Some twenty years ago, I found myself sitting next to a senior astronomer at a dinner party. When the conversation turned to space exploration, he made the bald assertion that “science got nothing out of Apollo.” For a long time, I was puzzled by this because, although it is not an uncommon view among astronomers, it is demonstrably mistaken. Eventually I understood. My senior colleague, an eminent ultra-violet (UV) astronomer, had fallen into the all-too-common trap of assuming that science was coterminous with his own sub-discipline. Because he felt that UV astronomy had got nothing out of Apollo—which is not strictly true, given that Apollo 16 carried a UV telescope to the Moon—he felt able to generalize this to the whole of science.² All too often, astronomers and physicists, such as Krauss, make similar assertions, seemingly without realizing that whole areas of science with which they are unfamiliar have benefited greatly from human space exploration.³

The science which benefited most from Apollo was not astronomy but geology. The study of lunar geology laid the foundations for the wider discipline of planetary geology, which now underpins our understanding of the origin and evolution of all the terrestrial planets. I have discussed the scientific legacy of Apollo in more detail elsewhere.⁴ It mostly arises from the analysis of the 382 kilograms of rock and soil samples returned to Earth from the six landing sites and the results of geophysics experiments, such as seismometers and heat-flow probes, deployed on the lunar surface.⁵ And, of course, the scientific legacy of Apollo keeps growing. Many scientists around the world, including my own research group at Birkbeck College, work with Apollo samples.

Even though science has, in fact, gained substantially from Apollo, it is still legitimate to ask whether the same knowledge could have been gathered without sending humans to the Moon. This is not a straightforward question to answer. It is true that much of Apollo science could in principle have been achieved robotically, since it involved collecting samples from, and placing instruments on, the lunar surface. Yet the diversity of samples collected and the range of geophysical instruments deployed would have required dozens, and perhaps scores, of individual robotic missions to achieve the same results. The three unmanned Soviet space probes Luna 16, 20, and 24 successfully collected and brought back 321 grams of lunar material during the 1970s. This was less than 0.1% of the amount returned by the Apollo missions. The Apollo material consists of more than 2,000 individual samples, collected from multiple locations around each landing site. The Luna material, in contrast, consisted of a single shallow core sample from each site. No practical or affordable robotic program would have returned the quantity or diversity of the Apollo lunar samples.

To my mind, the most important lesson from the Apollo missions is for the future. What Apollo really demonstrated is the efficiency of astronauts as field geologists on other planetary surfaces. Fumento makes light of Apollo astronauts playing golf on the Moon, but this unfairly trivializes the achievements of the astronauts working under extremely trying and dangerous conditions. Given these conditions, and the fact that each excursion on the lunar surface was limited to about seven hours by the oxygen reserves in their backpacks, I find the efficiency of the Apollo astronauts to have been little less than miraculous.⁶ Consider the final mission, Apollo 17, launched in 1972. During the course of three days on the lunar surface and only 22 hours outside the lunar module, the Apollo 17 astronauts traversed a total distance of approximately 35 kilometers, collected and returned to Earth 110 kilograms of rock and soil samples, drilled the deepest borehole ever made on another planetary body, and deployed a range of geophysical instruments. The latter included two heat-flow probes implanted at a depth of two meters and the only active seismic profiling experiment yet performed on another planetary body.⁷

It is interesting to compare this with the efficiency of robotic rovers, such as those being used to explore Mars. The fantastically engineered and long-lived Opportunity rover took eight and a half years to traverse the same distance that the Apollo 17 astronauts traveled in three days.⁸ Opportunity did not return samples to Earth, drill boreholes, or deploy geophysical instruments. Of course, these robotic missions are cheaper than Apollo—albeit not by as much as one might think. They have also revealed important knowledge about Mars, but in terms of exploration efficiency, there is simply no comparison. This fact is well recognized by scientists engaged in exploring Mars robotically, such as Steve Squyres, the principal investigator for the Spirit and Opportunity rovers. “The unfortunate truth,” Squyres remarked, “is that most things our rovers can do in a perfect sol [i.e., a Martian day] a human explorer could do in less than a minute.”⁹

And, again:

What Spirit and Opportunity have done in 5½ years on Mars, you and I could have done in a good week. Humans have a way to deal with surprises, to improvise, to change their plans on the spot. All you’ve got to do is look at the latest Hubble mission to see that.¹⁰

It is this level of efficiency that leads many of us to anticipate tangible significant scientific benefits from returning human explorers to the surface of the Moon, and eventually to Mars and perhaps elsewhere in the Solar System.¹¹

Fumento estimates the cost of Apollo science by dividing the overall cost of the program by the mass of samples collected, and then asserts that the resulting cost—about US$50 million per kilogram in 1960s money—was too high to be justified. Yet Apollo was not primarily a science project; it was a geopolitical project with its ultimate rationale firmly embedded in Cold War geopolitics.¹² For geopolitical reasons, the vast majority of the US$19.4 billion (about US$150 billion today) spent on Apollo would have been spent anyway, regardless of whether any science was performed or not.¹³ Looked at this way, the transformative science that was achieved by Apollo came essentially for free.¹⁴ Apollo provided logistical infrastructure on which science was successfully piggybacked, but the bulk of the infrastructure was not developed with science primarily in mind and was not paid for primarily from scientific budgets.

Consider a more recent example of science benefiting from the availability of a human spaceflight infrastructure. The Hubble Space Telescope (HST) was launched by the Space Shuttle Discovery in 1990. It is arguably one of the most productive scientific instruments ever built. The HST has only lasted so long and been so productive owing to no less than five human-tended servicing missions by the Space Shuttle—the vehicle that Krauss, quoted by Fumento, described as “a multi-billion-dollar failure.”¹⁵ Critics will point out that the HST did not need to be deployed from a Space Shuttle and that the primary goal of the first servicing mission was to correct a manufacturing flaw in the telescope’s main mirror. Yet the first servicing mission did far more than correct faulty optics. It also replaced failing solar arrays, solar array drive electronics, two gyroscope control units, two original magnetometers, and a processor in the flight computer. A new imaging system was also installed.¹⁶ Even if the HST had been launched with a perfect mirror by an unmanned rocket, without the human servicing missions it likely would have failed within its first decade of operation. Twenty years of extraordinary astronomical discoveries would never have been made. Four subsequent servicing missions replaced the solar panels, more failing gyroscopes, and the telescope’s computer system. Outdated cameras and spectrometers were also replaced with more capable versions. The entire scientific legacy of the HST has been due to the availability of a human spaceflight infrastructure in Earth orbit.¹⁷

After the Columbia accident in 2003, serious thought was given to the possibility of designing a robotic servicing mission for the HST to avoid the perceived risks of a human mission. This approach was found to be infeasible, and a decision was made to fly the fifth and final human servicing mission using the Space Shuttle Atlantis in 2009. The Final Report of the US National Research Council’s study that informed this decision has important implications for other aspects of the human spaceflight debate:

It is true that robotic systems can be stronger and faster than humans, can go places too dangerous for a human to venture, and can operate without fatigue while performing highly repetitive and precise tasks. However, it is very difficult to build a mechanical device (e.g., a robotic arm) that has dexterity comparable to a human’s limbs. It is even more difficult to build a computer system that can perceive its environment, reason about the environment and the task at hand, and control a robotic arm with anything remotely approaching the capabilities of a human being.¹⁸

The HST is another example of science benefiting from the availability of a human spaceflight infrastructure, in this case through the Space Shuttle, which was developed for a multitude of other reasons. That infrastructure has now been dismantled for the most part. As a result, the future potential of space astronomy is likely to be constrained. Consider the HST’s successor, the very capable but enormously expensive James Webb Space Telescope (JWST), due to be launched in 2021 at a cost of about US$10 billion.¹⁹ The JWST is not designed to be tended by astronauts and cannot be repaired or upgraded once launched. It only has a five-year design lifetime, and if it fails prematurely, that will be the end of it. Given the trials and tribulations of its development, it is unlikely to be quickly replaced. The time may therefore come when astronomers again wish for the existence of a human spaceflight infrastructure capable of maintaining astronomical facilities in space. Looking further ahead, we can imagine other benefits, such as the ability to build and maintain large lunar and space-based telescopes to study the early universe and planets orbiting distant stars.²⁰

From the earliest years of human spaceflight, life scientists have been studying the effects of the space environment on astronauts, and more recently this research has been extended to other living things.²¹ Much of this research is dependent upon a human presence in space. This work is often ignored by physical scientists, such as astronomers and geologists, possibly because they do not understand it. Nevertheless, biology is a science, and it does benefit from human space missions.²² Critics such as Steven Weinberg, as quoted by Fumento, argue that this research only has value if we want to learn how to keep humans alive and functioning in space. Since these critics deny the scientific benefits of human spaceflight, life sciences research in space appears little more than a “senseless and circular process.”²³ This ignores numerous fundamental biological insights that have resulted from this work, together with the potential for medical applications arising from an improved understanding of the effects of extreme environments on human physiology.²⁴

I have no doubt that diverse fields of science will continue to benefit from human spaceflight infrastructure in the future, but it is clear that such an infrastructure will be, as it has been, developed for a multitude of social, economic, and political reasons, of which science is only one. Fortunately, a number of wider societal benefits arising from an ambitious human space program can easily be identified.

1. The Apollo program was driven by essentially negative geopolitical motives, rooted in the Cold War competition between super powers. Space exploration also lends itself to international cooperation, as manifest in the fifteen nation-states peaceably cooperating on the International Space Station and the more recent International Space Exploration Coordination Group coordinating global space exploration.²⁵ Governments may come to perceive investment in high-profile international human space projects as a means of strengthening links between nations.²⁶
2. In the short-term, developing space infrastructure will employ many people in high-tech industries and stimulate technical innovation. Many of these innovations will have benefits and applications outside the space sector. Of course, space is not unique in this respect. Any sufficiently high-tech public works program would be expected to yield similar economic benefits. Nevertheless, economic motives for investing in space exploration do exist.²⁷
3. In the longer-term, there is the possibility that a developing space economy, based on utilizing space resources, could eventually lead to space infrastructure that will pay for itself. A number of commercial companies, and even some governments, are already thinking seriously about mining the Moon and asteroids for raw materials that could support new industries in space.²⁸ It is difficult to foresee how this trend will develop in the coming decades, but in such a case, science is likely to be a beneficiary.²⁹
4. For many people, especially young people, human space exploration is inherently exciting; it therefore acts to stimulate interest in science and technology. Only a tiny fraction of those inspired to study science at school and university will go on to work in the space sector, but society will still benefit. “Exploratory spaceflight,” Carl Sagan observed, “puts scientific ideas, scientific thinking, and scientific vocabulary in the public eye. It elevates the general level of intellectual inquiry.”³⁰
5. Fumento is right to criticize any argument that we might one day need to move to other planets if the Earth is no longer able to sustain our growing population. The problems associated with human pressures on the biosphere must be addressed here on Earth. Yet, civilization is much less robust than human life. It is subject to a wide range of existential threats, both natural and man-made. It follows that the long-term future of our civilization, if not our species, would be more secure if a future technological civilization occupied multiple locations in the Solar System, rather than being confined to a single planet.

Taken together, these elements form a compelling case for investing in ambitious human space projects. Whether or not such initiatives will ever prove sufficient to drive the development of a space-based civilization is impossible to predict. In this respect, I am happy to agree with Krauss: “I believe the future of the human species will eventually be in space, and that we will one day colonize other planets. But we have to be honest about this goal.”³¹

In truth, there are multiple reasons to pursue a human future in space. From a purely scientific perspective, it seems clear that if humanity does become a spacefaring civilization, then our understanding of the universe around us will be greater than if we do not.

Ian Crawford

Michael Fumento replies:

For all its length, Ian Crawford’s letter appears to come down to the expressions of someone who has watched the glory days come and go, a sailing ship captain in the days of steam.

As I wrote in my essay, the formula is simple. Machines become faster, smarter, and cheaper by the day. In space, they will always be safer. Human beings, far from evolving, are becoming ever more dependent on machines.

In 2012, a writer for Science cited a study by Crawford, which he summarized:

The Apollo 17 astronauts covered more than 22 miles in three days, a distance that has taken the Mars Opportunity rover eight years to match. Humans can drill for samples deep underground and deploy large-scale geologic instruments, something that no rover has achieved on another body.³²

Crawford was quoted: “In what was really only a few days on the lunar surface, the Apollo astronauts produced a tremendous scientific legacy. Robotic exploration of the moon and Mars pales in comparison.” Seven years later, Crawford repeats this argument in his letter.

In fact, Crawford seems trapped in a block of space amber dating back to sometime in the first decade of the century. Also in 2012, he penned a detailed article comparing human to robotic space exploration, explicitly based on a table showing robotic and human capabilities way back in 2004.³³ We can grant that in most ways humans were superior space explorers back then. But fifteen years have since passed. And hopefully there will be more ahead.

Crawford does not respond to my point that the Mars Curiosity rover carries advanced tools for sample and weather analysis that would be impossible for humans to carry for the seven years that the rover has operational.³⁴ Furthermore, compare the payload of a ship that carries only instruments to one that must carry humans and their massive life support systems. Each human passenger equates to a reduction in the payload.

Next year NASA plans to launch a far more advanced rover, the Mars 2020.³⁵ It is cheaper than the Curiosity: about US$2.1 billion versus US$2.5 billion.³⁶Yet the vehicle will be more rugged; part of the mission is to improve the odds of a successful landing, a problem that has plagued the non-US rover missions. This rover too will collect samples, but without the ability to send them back to Earth.³⁷ NASA has contracted Airbus to produce a “fetch rover” concept to bring the samples back.³⁸ But it is probably better to just study them in place. With virtual reality and haptic sensors, those so inclined can interact with the samples in numerous ways. That said, one may be surprised to hear we already have almost 200 samples of Martian rocks from meteorites flung off the red planet.³⁹ Rock collection does not get much cheaper than that.

Instruments aboard the Mars 2020 will include

• An X-ray fluorescence spectrometer that can identify the chemical composition of substances as small as a grain of salt.⁴⁰
• A radar that can penetrate up to 10 meters underground to detect structural layers, densities, buried rocks, meteorites, water, ice, and salty brine.⁴¹
• Sensors for temperature, wind speed and direction, pressure, relative humidity, radiation, and dust size and shape.⁴²
• An experimental device that will convert carbon dioxide into oxygen. It is projected that this device, about the size of a car battery, can eventually be scaled up to generate liquid oxygen propellant for return voyages to Earth.⁴³
• A SuperCam and an instrument named SHERLOC (Scanning Habitable Environments with Raman and Luminescence for Organics and Chemicals), both similar to the Curiosity’s ChemCam. Their cameras, lasers, and spectrometers will search for signs of past microbial life.⁴⁴ The SuperCam “can identify the chemical and mineral makeup of targets as small as a pencil point from a distance of more than 20 feet (7 meters).”⁴⁵
• A three-dimensional video and imaging system that can zoom in to examine distant objects.⁴⁶
• Microphones to be used during landing, driving, and sample collecting.
• The Mars Helicopter Scout, an experimental 1.8 kilogram solar-powered unmanned helicopter that will identify the best driving routes for the rover.⁴⁷

Transporting such an array of instruments around Mars would require an army of porters. If this seems unlikely, it should be kept in mind that even getting these devices to Mars in the first place would be compromised by the presence of humans. Rover missions with humans along for the ride would be forced to ditch important equipment to accommodate passengers, despite the fact that human expeditions are likely be considerably shorter in duration than robotic missions. Meanwhile, work has surely begun on a Mars 2030 project, or something of that nature. One can also imagine sending payloads of solar-powered drones tasked with mapping out the entire planet rather than selected landings sights, as is currently the case.

The early days of human space travel were truly glorious. Decades later, space travel has become little more than a commuter service for astronauts with no real jobs. It is much cheaper, faster, and safer, to send robots in our stead. Anyone feeling nostalgic can watch and re-watch highlights of the glory days from the comfort of their living rooms. Meanwhile, there is solace that all of this incredible equipment is being designed by humans.

At least for now.