Alternate View Column AV-84
Keywords: killing superconducting super collider SSC high-energy physics acceleration
Department of Energy Congress bureaucracy
Published in the May-1997 issue of Analog Science Fiction & Fact Magazine;
This column was written and submitted 12/04/96 and is copyrighted ©1996 by John G. Cramer.
All rights reserved. No part may be reproduced in any form without
the explicit permission of the author.
This page now has an access count of:
My new hard science fiction novel, Einstein's Bridge (Avon, June-1997), is set in Waxahachie, Texas a few years after the Superconducting Super Collider (SSC) comes into operation. I started writing it in 1988, soon after my first novel, Twistor, was completed. My original plan was that the novel would take advantage of the publicity from the construction and operation of the world's largest particle accelerator, the SSC. My wife and I traveled to Waxahachie before the 1989 Boston Worldcon, visited the SSC Office, interviewed SSC physicists, and explored the region. The following year we visited CERN to sample the flavor of a working particle physics facility approximating the SSC in scale. (I was very taken with CERN, and since 1991 I have done my physics research there.)
Unfortunately, when my new novel was almost complete, it was brutally projected into an alternate universe. On October 27, 1993, when the SSC project was 20% finished, two billion dollars had already been spent, and 14.7 miles of the 54 mile oval tunnel to house the machine was already dug, the Congress of the United States voted to terminate the project. Fifteen thousand physicists, engineers, technicians, contractors, and support workers, the vanguard of the most important project in particle physics, found themselves unemployed, in many cases unemployable in their chosen professions.
Einstein's Bridge (so named because until the 1960s wormholes were called "Einstein-Rosen bridges") is a complete rewrite of the aborted SSC novel I had almost finished in 1993. It's about particle physics, wormholes, alien contact, and time travel, and it uses the SSC killing and associated politics as key plot elements. During the background research for this rewrite I developed a view of the root causes of this greatest of scientific disasters. In this Alternate View column I want to give my perspective on what happened to the SSC and what lessons can be learned.
The SSC Project [see in my Alternate View column in the March-1988 Analog] was to have been the world's largest and most powerful particle accelerator, the home of experiments that would discover the elusive Higgs boson and other wonders of the micro- universe, the facility that would provide the key data to break the code of Nature's innermost secrets, the machine that would restore the preeminence of the United States of America as the center for high energy physics. The SSC was to be a great technological leap forward, bringing liquid-helium-temperature superconductivity and ultra-high field magnetism to new levels of engineering accomplishment, seeding the advance of technology in such areas as nuclear magnetic resonance imaging, electric power transmission, high-speed computing, and efficient energy storage.
Hundreds of young particle physicists with fresh PhDs gambled their futures on the success of the mega-project and gathered in the new SSC buildings springing up in the black-dirt cotton fields south of Dallas to contribute to the heroic effort that would bring high energy physics to a new pinnacle of technical accomplishment and theoretical understanding. Thousands of other physicists looked on these chosen ones with envy, an envy that evaporated in an instant when the project was canceled.
The SSC project rose from the ashes of another canceled accelerator. In 1984 the ISABELLE collider was being built by the Department of Energy (DOE) at Brookhaven National Laboratory and was in deep trouble from two separate directions. First, its giant superconducting magnets didn't work. Second, the very reason for its existence, the anticipated discovery of the Z and W bosons of the weak interaction, had already been accomplished a year earlier by Carlo Rubbia and his UA1 group at CERN.
The leaders of the American high energy physics community huddled with Reagan Administration officials and devised a radical plan: ISABELLE would be canceled and its funds diverted to the design of a new and larger accelerator. The new machine was nicknamed the "desertron" because it would be so powerful and so large that it might have to be built in a desert area with cheap land and low population density.
The DOE created the Central Design Group, based at Lawrence Berkeley Laboratory in California, to design for the new accelerator. By 1986, the group had produced a "reference design" for the desertron, now renamed the Superconducting Super Collider or SSC. It would have a ring 53 miles in circumference filled with 8,600 superconducting magnets, each having a "bore" of 4 centimeters through which a beam of 20 TeV protons would pass. The cost of the machine was estimated to be around two to three billion dollars.
On January 29, 1987, at the beginning of the semi-final year of the Reagan Administration, the DOE presented its plan for the new SSC project to the President and his cabinet. The price tag was $4.4 billion, quoted in 1988 dollars omitting inflation and the cost of detectors that the new accelerator would need. After the DOE experts had made their pitch, Reagan recalled his days as a sports reporter interviewing a star quarterback whose watchword had been "Throw deep!" This was the advice he gave the DOE regarding the SSC project.
"Mr. President," remarked OMB Director William Miller, "you're going to make a lot of physicists ecstatic."
"That's probably fair," Reagan replied, "because I made two physics teachers in high school very miserable." With that the SSC received its start. Later in 1987 the DOE raised the SSC cost to $5.3 billion after including a small allowance for inflation and $500 million for detectors. The SSC then went into the DOE Budget as an approved project, and its initial phases were funded by an enthusiastic Congress.
When George Bush became President in 1989, the SSC cost estimate, with a more realistic inflation rate, rose to $5.9 billion. This was already 34% above the cost Reagan had announced. Bush chose as his Energy Secretary a spit-and-polish Rickover protege from the Navy's nuclear submarine program, Admiral James Watkins. Watkins' appointment was a disaster for the SSC. His management style, developed with large DOD nuclear submarine construction projects, was exactly wrong, as applied to a cutting-edge scientific project like the SSC. He treated the dedicated physicists and seasoned accelerator engineers of the SSC like crooked shipyard contractors trying to sell the government $5,000 toilet seats. Seeking personal control, he grossly increased the bureaucracy and paperwork associated with SSC oversight. He dismantled the existing SSC management structure, installing people that he trusted in key positions while displacing physicists and experienced accelerator builders from the decision chains. The SSC Program Office in Dallas swelled to one hundred bureaucrats, sixty permanent staffers plus forty more on temporary assignment from elsewhere in the DOE. Roy Schwitters, SSC Director, later characterized the DOE's massive oversight overkill as "the revenge of the C students."
When the SSC became a funded project, the new design team began a careful reexamination of the 1986 reference design of the Central Design Group. This reexamination was done under the close scrutiny of the nit-picking justify-every-step-with-a-paper- trail bureaucracy installed by Watkins. After much agonizing, the design team decided that the SSC design could only be guaranteed to succeed if the magnet bore was increased from 4 centimeters to 5 centimeters. This modification reduced the field of each magnet, and thus required more of them, a longer tunnel to house them, and more construction money.
This design change was carefully reviewed by a group of outside accelerator experts. While a minority report argued for keeping the 4 cm bore, the majority consensus supported the decision to change to a 5 cm bore as the conservative choice which would ensure successful operation of the machine.
A separate group of theoretical physicists, the Drell Panel, considered the option of holding to the old budget by reducing the SSC's operating energy and the physics goals of the project. They rejected this option as unacceptable. The 5 cm magnet design was adopted, and the change raised the price of the SSC to a hotly debated number that stabilized in 1991 at $8.6 billion, a 46% cost increase.
With these rising cost estimates, opposition to the SSC in Congress was surprisingly slow to develop. Even after the official DOE cost rose to $8.6 billion, an attempt in 1991 to kill the SSC project in the House failed by a vote of 165 to 251. However, the climate for expensive projects worsened as the US economy moved into recession in the early 1990s.
It is common wisdom in Washington, D.C., that it is dangerous for a large project to span more than one Administration. Bush was defeated by Clinton in 1992, and the SSC project came to violate this rule and suffer the consequences. In 1993 the incoming Clinton Administration made a budget-tightening decision to stretch out the SSC project, moving its date of completion from 1999 to 2003, increasing the overall cost of the project while reducing its yearly cost. The SSC cost rose to over $10 billion, a 16% cost It is common wisdom in Washington, D.C., that it is dangerous for a large project to span more than one Administration. Bush was defeated by Clinton in 1992, and the SSC project came to violate this rule and suffer the consequences. In 1993 the incoming Clinton Administration made a budget-tightening decision to stretch out the SSC project, moving its date of completion from 1999 to 2003, increasing the overall cost of the project while reducing its yearly cost. The SSC cost rose to over $10 billion, a 16% cost increase. The budget-conscious freshman Congressmen swept in with with Clinton in November of 1992 felt no responsibility for the decisions of their predecessors, and the SSC project became a tempting target of opportunity.
Clinton's new Science Advisor John Gibbons did not give active support to the SSC project, as had his predecessor, Alan Bromley, and Clinton's new Energy Secretary Hazel O'Leary, now famous for her million dollar travel excursions, proclaimed during her confirmation hearings that she was "not passionate" about the SSC. In September, 1993 when her passions were finally aroused, she took the counter-productive steps of re-shuffling major SSC contractors and increasing the already bloated oversight team to 140 bureaucrats in the Dallas DOE Office. Before the two critical votes in June and October, neither Clinton nor Gore was willing to make personal appeals to House Members on behalf of the SSC, as Bush had in 1992.
The final blow to the SSC came late in 1993 when the DOE's Baseline Validation Report was released. The validation group surveyed the sorry history of SSC cost escalations and concluded that extreme conservatism was needed. Their report advocated much larger safety and contingency margins and moved the completion date back to 2004, increasing the project cost to $11.5 billion or another 15% increase.
With this, rank-and-file members of Congress had had enough. They were fed up with the ever-rising SSC price tag, the evidence of poor management and DOE indecisiveness, and the heavy-handed attempts by Congressional Leadership to save the project. On October 27, 1993, by a vote of 283 to 143 the House rejected the Conference Committee report that would have continued SSC funding. The project was officially dead.
There was, of course, much more to the decline and fall of the SSC than this brief summary, focused mainly on the cost escalation problem, could include. Einstein's Bridge gives a fictionalized but more detailed account.
We must learn from our disasters. Here's my list of lessons from the SSC about future large scientific projects:
(1) For projects of the scale of the SSC, the DOE's "lowball" cost shaving which omitted inflation and detector costs is inappropriate. Cost estimates must be realistic from the start and must be treated as "sacred". If that price is too high, so be it. It would have been better for all concerned if the SSC had never been approved rather than canceled in midstream.
(2) Bureaucratic oversight of scientific projects should be applied with a light hand. Oversight overkill can destroy morale and break the budget of a project, as the SSC case demonstrates.
(3) If expensive major design changes become necessary, as in the case of the 4 cm to 5 cm magnet change for the SSC, the project should go back to Congress for re-authorization and renewed support. Similarly, stretch-outs to reduce budget impact but increase overall cost should only be done with the concurrence of Congress.
(4) Projects of the scale of the SSC should be designed, if possible, to span a single 8 year presidential administration. Incoming presidents feel little responsibility for the expensive initiatives and commitments of their predecessors, particularly if there has been a change of party.
(5) We desperately need some way of funding long-term scientific projects like the SSC with stable funding that is better insulated from the 2-year Congressional election cycle.
Doing good science is difficult enough under the best of circumstances. When politics is added to the mix, it may become impossible.
AV Columns On-line: Electronic versions of more than eighty past "The Alternate View" columns by John G. Cramer are available on-line on WorldWideWeb at the URL: http://www.npl.washington.edu/AV.
This page was created by John G. Cramer on 12/12/96.