Ted Taylor and the 135-foot-diameter library/cafeteria at General Atomics, November 1999.
In what Kurd Lasswitz described (in 1901) as "The Universal Library," elaborated by Jorge Luis Borges as "The Library of Babel" (1941) and revisited by Kevin Kelly (1994) as "The Library of Form," all possible books, all possible creatures, and all possible technologies have a place somewhere on the shelf. "When it was proclaimed that the Library contained all books, the first impression was one of extravagant happiness," wrote Borges. "All men felt themselves to be the masters of an intact and secret treasure. There was no personal or world problem whose solution did not exist."[128] It is the business of both evolution and invention to sift through the possibilities, cataloguing those combinations, however improbable, that make sense.
Nature and technology usually advance by increments, with sudden innovations appearing as the result of novel combinations or appropriations of features that already exist. The space of possibilities is infinite, but the library is not. It is finite but unbounded: if you do not find what you are looking for, you can always add one more book to the shelf. Ted Taylor sought to open a whole new wing at once.
"The variety of conceivable space engines is huge; we have so far worked hard on only a very small fraction of the possible ones," he explained in 1966. "I have made a morphological outline of possible space propulsion systems, classifying them according to whether the energy release is pulsed or continuous, the types of energy sources that are used, the numbers and types of energy conversion stages in the engine, and so on. If one randomly permutes the elements of this outline, one generates more than 1022 different space propulsion concepts, each of which makes logical sense! If each of these concepts were studied by one person for an hour, it would take a billion people a billion years to study them all!"[129] Ted posted the outline on his refrigerator at home. "Random generation of propulsion concepts from Table III is practically guaranteed to produce a concept that no one has ever thought of before," he reported. "I have found it impossible to reject, as clearly nonsensical, any of the dozen or so concepts which I have seen derived that way, mostly by my children. But every one of them has been a strange idea indeed."[130]
A universal library, whether of books, genotypes, or technologies, forms an expanding cloud of possibilities in a multidimensional space. The laws of nature form an outermost bound. A smaller cloud, condensed out of this atmosphere of possibilities, represents the organisms or technologies that can be assembled from available parts. Finally, a small central core—where we live—represents the books, organisms, or technologies that exist at the present time. Instead of building outward by small increments, Ted sought to develop Orion the other way around: start with the laws of nature; delineate the bounds first of possibility and then of practicality; finally, trace a path backwards to existing technology so as to advance not by increments but by leaps and bounds.
We have an unusually detailed record, for a secret project, of how the requisite conceptual leaps were performed. De Hoffmann insisted, says Ted, that "anytime anybody had an idea, write it down. And put the GA logo on it." When an idea was written down, it was reviewed by General Atomic's technical document center, and, if approved for distribution, was released internally as a GAMD (General Atomic Manuscript Document) report. De Hoffmann applied academic standards to General Atomic's technical literature, both in deference to academic tradition and because clear thinking was aided by written reports. "Just in the process of writing it up, you refine your thoughts and often find that something that is not quite wrong was not quite right," says Burt Freeman. It was also a good way to build an institutional memory and benefit collectively from individual work. "Any informal report would come over with a distribution list," explains Bill Simpson. "Fred was hell on cross-fertilization of ideas—and the distribution lists were supposed to be constructed with that in mind."
F. W. (Bill) Simpson, director of the library and document center, had been a librarian at Furman University in Greenville, South Carolina, until hiring on in 1946 with the Manhattan District at Oak Ridge, Tennessee. "I thought we were getting in on the ground floor of atomic energy," he says, "but after being there a while, I decided we entered the basement!" He was assigned the job of organizing the documents belonging to the research division of the Manhattan Project, then in complete disarray—"an unindexed, uncatalogued, unabstracted mass of information"—with no coordination between the different installations being transferred to the newly established AEC. The huge number of classified documents presented a challenge, but what to do with the first unclassified document was even more of a problem for the Manhattan District. "The guys from the declassification branch brought this manuscript around to me one day and said 'What can we do with it?' And it threw everybody into a tizzy. Heretofore, everything had been restricted, confidential, secret, and so on. What did we do with an unclassified report?"
Simpson helped organize the Manhattan District's (later the AEC's) collections, abstracting, indexing, and circulating them according to standard distribution lists. Freddy de Hoffmann noticed. Simpson was hired by General Atomic in April of 1956 and began ordering books and journals out of the General Dynamics office in Washington, D.C. "Fred wanted very much to have a library, which he regarded as an inducement to the people from the universities and national labs," Simpson explains. "Everybody that came in, the new hires, he would ask to make recommendations for the library. And Fred spent a lot of time on that, himself." In addition to assembling a research library, Simpson established a team of editors and typists—consisting largely of English majors, not scientists—to produce General Atomics own reports.
"I have one of the most unmathematical minds you have ever encountered," Simpson admits. "I enjoyed my associations with the scientists and always asked what they were working on. And they were always happy to talk about their work!" Simpson has no patience for scientists who say they cannot explain their work to someone who does not understand the math. "In my entire tenure at General Atomic, talking with aeronautical engineers, metallurgists, reactor physicists, fusion physicists, astrophysicists, I never had one instance of that. They all talked to me like I was an intelligent being." Project Orion produced an exceptionally lucid series of technical reports. No matter how mathematical in nature, the meaning is clear and the language understandable independent of the mathematical results. The title pages identify who did the work as well as who wrote the report.
The AEC's classification rules differentiated Orion's literature into two separate phyla—one branch that circulated freely (if internally) and one branch that was confined to a vault. Reports that mentioned "bombs," provided specific design details or dimensions of the ship, discussed or allowed inference as to either the yield of or fissionable material consumed by the individual pulse units, or discussed specific missions or military applications were classified—usually "Secret—Restricted Data," or S-RD. Any release of information was controlled not by the Air Force sponsors but directly by the AEC.
In this category were all the progress reports issued under the terms of the original ARPA contract and its Air Force successors. This includes a long series of preliminary, annual, and final reports, beginning with the Feasibility Study of a Nuclear Bomb-Propelled Space Vehicle, Interim Annual Report, 1 July 1958-1 June 1959, written by Brian Dunne, Freeman Dyson, and Michael Treshow, and edited by Ted Taylor, who remembers this as the document that "turned on the Air Force," including Mike May, later director of Livermore, who said "it was the best progress report he had ever seen." It remains classified S-RD. "It only covers the first year, but if I had to choose one document for declassification, that would be the one," says Ted. By the end of the project, Orion's final reports were being issued in four volumes, totaling more than 600 pages and constituting a cumulative handbook on the state of the art.
Ted
Taylor and the 135-foot-diameter library/cafeteria
at General Atomics, November 1999.
Hundreds of GAMD reports were issued covering every aspect of the feasibility, design, and possible operation of the ship. Some reports were two or three pages and some were two inches thick. A sampling of classified reports: A Survey of the Shock Absorber Problem; Trips to Satellites of the Outer Planets; Random Walk of Trajectory Due to Bomb Misplacement; Flight Characteristics During Takeoff Through the Atmosphere; Radioactive Fall-Out from Bomb-Propelled Spaceships; Multi-ICBM Weapon System; ORION Charge-Propellant Fire Control; ORION Parameter and Payload Study Based on 200- and 4,000-Ton Reference Design; Orion Fuel Requirements.
De Hoffmann encouraged the publication of as much work as possible in unclassified form, distilling the underlying science and carefully removing any reference to specific dimensions, military applications, or bombs. The unclassified literature reveals who was thinking what during Project Orion's seven years of work (see the appendix for a more complete list): The Absorption of X-Rays by Cold Materials; Flexural Vibrations and Stresses in a Flat Pusher; Shock Structure in a Medium of Finite Radiation Opacity, Optimal Programming for Vertical Ascent in Atmosphere; Minimum Energy Round Trips to Mars and Venus, Deformation Analysis of a Plate-like Structure Represented as a Grid of Beams, The Scientific Uses of Large Space Ships; Diffraction of a Shock Wave Around a Corner; Diffraction of Radiation Around an Opaque Disc; Minimum Energy Loss in a Two-Mass Spring System; Application of the Single-Scattering Approximation for Atmospheric Side-Scattering of Gamma Radiation into a Nuclear Space Vehicle; Viscous Flow of Ablating Grease Films; Hydrodynamics in the Interaction of X-Rays and Cold Iron; Preliminary Data on a Complete Life Support System for a Manned Space Vehicle; Preliminary Analysis of Meteoroid Protection for the 10-M Diameter Orion Engine; Study of the Effects of Using Lunar or Planetary Material for Propellant.
These reports develop Orion from first principles, defining the bounds of feasibility that the more classified reports, by adding engineering details, attempted to fill in. "It was all slide rules, and sometimes you used log tables," says Brian Dunne. "All the reports were typed up by hand." A large number of them end with the lowercase initials "br"—the mark of Betty Risberg, who "could type up equations without even stopping," according to Dunne. All this technical literature, establishing the theoretical framework for the ship, was both produced and collected in the document center that occupied the ground floor of the central building at General Atomic—which also formed a model of Orion, full scale.
This building was a trademark achievement for its architects, Pereira & Luckman, appearing as futuristic today as it did in 1959. A fluted, toroidal monument, with plate-glass windows, it is supported above a central courtyard on tapered, angled steel buttresses that give the impression it has either just landed or is ready to take off. As you ascend one of the curved steel staircases attached to the periphery of the central core, you do not feel you are entering a building; you feel you are climbing aboard. As a six-year-old, I watched my father head in through the classified security gate to work at General Atomic and thought the round building was the beginning of the spaceship. I was not completely wrong.
"To me, the library always was Orion, ready to take off," Ted recalls. "I saw it take off! I had repeated dreams about it. Imagine sitting up there, eating, on something rotating at about one-quarter g." The upper level is still the General Atomic cafeteria, and sitting near one of the peripheral windows you can imagine the ship rotating to produce enough artificial gravity to keep meals in place while passing through Saturn s rings. "We always imagined the ship with a big recreation area in the nose, and windows looking out forward and sideways, so we could see the rings of Saturn sweeping overhead as we passed through," says Freeman. "There would be heat shields covering the windows at takeoff and landing and during thrust maneuvers, but most of the time we would be cruising in space with the windows uncovered."
No one was thinking about Orion when the library was designed. And no one was thinking about the library when the Orion group, still at the Barnard Street School, established the basic dimensions of the ship. "Something with a gross weight of 4,000 tons was the way all of us were thinking about Orion at the very beginning," Ted explains. "A 4,000-ton vehicle, how big is that? Well, about 100 to 150 feet across. How much does 150 feet mean to somebody—me in particular? Nothing without comparing it to something else. It turned out, when I went out and paced the distance, to be the diameter of the General Atomic library. This was after the general scale of the thing was determined, while we were still at the schoolhouse. How did the transfer of attention happen? The library was obviously something to point at if one wanted to say roughly how big the thing was—or could be. Once having done that it was easy to visualize shrinking down this dimension closer to a delivery shaft for warheads. The shock absorber distances were about right if you just went from floor to floor. It was easier to me than pointing at a blackboard to say here's this building and just imagine that there are nitrogen-filled shock absorbers directly attached to pneumatic tirelike things at one end. You get a sense of how the shock absorbers looked. Were they spindly? No, they weren't. The columns that are out there are roughly half the diameter they should be. It was a very helpful way to think about the project as a whole.
"I remember leaning on a rail and pondering the thing after going down to Torrey Pines beach and eating my lunch down by the water, among the birds," Ted continues. "It was an object for reflection about the project, in a lot of different ways. The setting was perfect. And there was a model! It was a way to deal with questions like what is the flight of something three feet in diameter compared with the flight of something like this, how can it be relevant? Is it just a waste of effort? It was easy to focus attention on different components, to separate, for instance, the shock absorbers from the rest of it. But then you could glance at the building and see it back where it was supposed to be, and not get carried away with things that did not make sense, for reasons that would occur to you if you looked at this thing that was the right size.
Before anyone was thinking about a building as Orion, Freeman was thinking about Orion as a building. According to Harris Mayer, early in the project Freeman sought to establish the dimensional and structural bounds of Orion-type vehicles, given as a foundation the acceleration of a bomb-driven plate in space. "We have a building, the Orion ship, and it's being accelerated," says Mayer. "Well, OK, we have buildings on earth and they're accelerated, the acceleration of gravity. So he asked the question, 'What's the biggest building we can make?' And he carved out the whole field."
"He derived certain engineering parameters from physical first principles in an amazingly clear way," recalls Lew Allen. "He ended up showing that really the only number that mattered was the strength of materials. Once the strength of materials was set—as it was for, say, steel—all the other parameters of the vehicle naturally fell out. It was such a beautiful and simple way to look at that."
"Soon after we moved out to the new laboratory," remembers Ted, "I set for myself the task of doing some parameter studies, and I did them all graphically, and they were painfully slow. I showed them to Freeman and he said, 'That's a good idea,' and within a week he wrote down about a dozen simultaneous equations, solved, for the whole thing. He did analytically what I had been struggling to do graphically, with all the important parameters of the ship: total momentum transfer, shock absorbers one-stage, two-stage, and so on. That led him into his unfettered, no-limits study and within a couple weeks he was designing starships."
To begin with Freeman limited his parametric studies to orbital and interplanetary ships capable of taking off from the ground—described in a report titled Dimensional Study of Orion-type Spaceships, issued on April 23, 1959. "The dimensional study was less serious," Freeman explains, "but it answers the question, 'Did you explore the outer limits of this technology?' The answer is yes."
"What range of variation can be allowed each of the design parameters without violating general principles of physics and engineering?" Freeman asked. "The general conclusion of the analysis is that ships able to take off from the ground and escape from the Earth's gravitational field are feasible with total masses ranging from a few hundred to a few million tons. The payloads also range from zero to a few million tons. The number of bombs to be carried is independent of the size. The total cost of each trip in fissionable material and in atmospheric contamination is also roughly independent of the size of the payload."[131]
After defining the limiting parameters and establishing the algebraic inequalities that have to be mutually satisfied by any given design, Freeman presented the results as a series of graphs defining the boundaries of feasibility for three classes of ships:
SATELLITE means a ship with propellant velocity 30 km/sec and propellant mass 100 kg intercepted by the ship. This is the smallest ship which can go into orbit around the earth with reasonably economical use of fissionable material.
ORION means a ship with propellant velocity 60 km/sec and diameter 40 meters. It has a smaller mass than the nominal Orion M2 design, since the M2 was not optimized for minimum number of bombs. ORION is the smallest ship which appears economic for interplanetary missions.
SUPER-ORION means a ship with propellant velocity 60 km/sec and diameter 400 meters. It is the largest interplanetary ship which can take off from the Earth's surface. It can be economically propelled by H-bombs.[132]
The ship masses range from 300 to 8,000,000 tons. "It is clear that the larger sizes of the Orion system have immense promise for the future," Freeman explained. "A ship with a million-ton payload could escape from Earth with the expenditure of about a thousand H-bombs with yields of a few megatons. The fuel cost of such a mission would be about 5 cents per pound of payload at present prices. Each bomb would be surrounded by a thousand tons of inert propellant material, and it would be easy to load this material with boron to such an extent that practically no neutrons escape into the atmosphere. The atmospheric contamination would arise only from tritium and from fission products. Preliminary studies indicate that the tritium contamination from such a series of high-yield explosions would not approach biologically significant levels."[133]
Once you start to imagine launching a million tons into orbit, it is no longer an impossible leap to start thinking about building really large vehicles that could operate in deep space but would never get off the ground. "Freeman gave a talk about what's the largest thing you can do, never mind the engineering details," remembers Harris Mayer. "So this was a spaceship propelled by megaton hydrogen bombs. The pusher was made of uranium, and the neutrons on the uranium would make plutonium, so when you got to Alpha Centauri's planet, if there was one, you would just take the pusher off and build a nuclear reactor so you could have a colony. We thought this was absolutely marvelous, even though we weren't going to do anything about it. But the atmosphere encouraged us to do things like this. And Freeman did not work in a vacuum, he was interacting with all the people, including Ted Taylor, who was saying, 'Calm down, calm down.' "
A dense, five-page handwritten General Atomic Calculation Sheet, titled "High-Velocity Ships," survives from 1959. It is filled with brief statements such as: "1,000 or 10,000 km/sec in principle obtainable in nuclear explosions. Such velocities necessary to cross solar system in a month. For 1,000 km/sec exhaust we can think of masses of 104 tons and A ~ 1/2 g. Cannot take off from surface. Only 40% of mass used as propellant. So max ship velocity only 1/5 V. To get 100% need only scrape surface. Also possibly reuse fuel and even breed it. Ship in shape of hollow sphere with 1/4 solid angle window. Propellant is shit and fission products. The ship velocity is 1/2 V. Type II reaches 500 and Type III 5,000. Right for energy sources. Type III with mass-ratio 10 could reach 10,000. Take colony of several thousand with amenities of civilization to Alpha Centauri in 150 years."[134]
"This was just an informal talk I gave to the group rather early in the game, about wild extrapolations, just to give people a feeling for what the ultimate limits would be," Freeman remembers. The power source, however, was not science fiction, but hydrogen bombs such as those already sitting on U.S. and U.S.S.R. shelves. "Hydrogen bombs are the only way we know to burn the cheapest fuel we have, deuterium," Freeman explained. "I do not know exactly how efficient hydrogen bombs are, and if I did know I would not tell you. So I will put upper and lower limits on the numbers that we are not supposed to know exactly."[135] Taking a conservative guess at the efficiency of one-megaton bombs, Freeman estimated what it would take to reach velocities of 1,000 and 10,000 km/sec, in each case examining two different kinds of ship: one optimized for maximum acceleration, the other for minimum size.
The outer limits are constrained by the velocity of the bomb debris, the strength of materials, and the maximum temperature—1,000 degrees Kelvin—the ships surface can withstand. "That's just so the pusher won't evaporate," Freeman says. "With hydrogen bombs you have much higher temperatures of the gas coming in but much lower density so the gas remains transparent. As soon as it hits the surface it radiates away the heat, as long as the kinetic energy of the stuff coming in is not greater than the heat capacity of the pusher. It just says 'thou shan't melt the pusher.' "
The 1,000 km/sec ship has a total mass of 24,000 tons. "The difficulty with space ships in the 1,000-km/sec class is not the high cost per pound but the large size of the smallest feasible ship," explained Freeman in 1968.[136] The small-size 10,000 km/sec ship, with a pusher 150 km in diameter and a mass of 240 million tons, would take 30 years to accelerate to full speed, and 150 years to cover the four light years to Alpha and Proxima Centauri, our nearest neighboring stars. To reach 10,000 km/sec, 90 percent of the original mass has to be used as propellant, requiring either an extremely light structure, unfolded in space like a spinnaker or a parachute, or the jettisoning or consumption of part of the ship during the voyage, like a steamship burning its furniture as it nears the end of a trip. The fourth page of Freeman's notes is a table showing how the numbers work out for six different ships. The parameters are V, velocity; M, propellant mass per bomb; μ, mass of ship; N, number of bombs; A, acceleration; L, pusher diameter; b, shell thickness; τ, bomb period. The interval between explosions varies from 0.4 seconds to 50 seconds across the different designs. "For the small-size ship the times are long enough," says Freeman, "but for the high-thrust version it is too short. Four-tenths of a second is certainly not easy. Thirty seconds, then you can imagine opening and shutting a door, pushing the bomb out by hand."
However implausible—to reach 10,000 km/sec requires 25 million bombs—nothing beyond existing materials and technology is assumed. Cost is no constraint, since the project is envisioned as at least 200 years out, when, if the economy keeps growing at 4 percent per year, "the building of a ship for $100 billion will seem like building a ship for $100 million today."[137] The ultimate question raised by super-Orion is not how or when but why. "These numbers represent the absolute lower limit of what could be done with our present resources and technology if we were forced by some astronomical catastrophe to send a Noah's ark out of the wreckage of the solar system," Freeman explained in 1968.[138]
"By the time the first interstellar colonists go out they will know a great deal that we do not know about the places to which they are going, about their own biological makeup, about the art of living in strange environments," he says. Echoing his 1958 manifesto, he listed two goals for such a voyage: "assurance of the survival of the human species against even the worst imaginable of natural or manmade catastrophes" and "total independence from any possible interference by the home government."[139]
Freeman emphasizes that Orion was never intended for travel beyond the outer planets. "It's a very poor system for this kind of interstellar trip. 10,000 km/sec is only 1/30 of light speed, it's just far too slow to be interesting." Interstellar travel, if and when we get to it, will be found on a different shelf in die universal library, where, as Ted sees it, there are 1022-1 other propulsion concepts left to explore.