This project was completed in August 1998 and resulted in the book Atomic Audit: The Costs and Consequences of U.S. Nuclear Weapons Since 1940 edited by Stephen I. Schwartz. These project pages are only occasionally updated and should be considered historical.
The Hidden Costs of Our Nuclear Arsenal
“Overview of Project Findings” by Stephen I. Schwartz —June 30, 1998
Stephen I. Schwartz is a Guest Scholar in the Foreign Policy Studies Program at the Brookings Institution, where he has directed the U.S. Nuclear Weapons Cost Study Project since May 1994. He is an expert on nuclear weapons, including the history and costs of the U.S. nuclear weapons program, weapons research, testing, production, and deployment, nuclear command and control and nuclear-related intelligence collection and analysis, environmental remediation and nuclear waste management, nuclear arms control agreements, and congressional oversight of nuclear weapons programs.
Prior to coming to Brookings, Schwartz was the Washington Representative for the Military Production Network (now the Alliance for Nuclear Accountability), a national network of more than 40 organizations addressing nuclear weapons production and environmental issues at the Department of Energy’s (DOE) nuclear weapons complex. From 1988 until early 1992, he was Legislative Director for Nuclear Campaigns with Greenpeace, where he lobbied and conducted research on issues pertaining to the DOE’s nuclear weapons and naval nuclear propulsion programs. Schwartz has also served as Associate Director of the Council on Nuclear Affairs (1987-88) and as Senior Research Assistant for the Adlai E. Stevenson Program on Nuclear Policy at the University of California at Santa Cruz (1985-87).
Schwartz writes frequently on nuclear policy matters. Atomic Audit: The Costs and Consequences of U.S. Nuclear Weapons Since 1940 (Brookings, 1998), which he edited and co-authored, is his first book. His articles and letters have appeared in the Atlanta Journal/ Constitution, the Bulletin of the Atomic Scientists, the Christian Science Monitor, Commentary, Disarmament Diplomacy, the Los Angeles Times, the New York Times, the San Jose Mercury News, the Wall Street Journal, the Washington Post, and the Washington Times, among others. Schwartz?s monographs include A Nuclear Weapons Primer: A Supplement to the UCSC Nuclear Information Handbook (Adlai E. Stevenson Program on Nuclear Policy, 1986), The ABM Treaty: Problems in Compliance (Adlai E. Stevenson Program on Nuclear Policy, 1987) and Rhetoric vs. Reality: Admiral James D. Watkins at the Helm?The Department of Energy, 1989-1992 and Beyond(Military Production Network, 1992). Schwartz earned a B.A. in Sociology (summa cum laude and college honors) from the University of California at Santa Cruz.
It is a privilege to be here today to share with all of you the fruits of four years of intensive research into what should have been a relatively simple question: what did the United States spend on nuclear weapons? Atomic Audit is truly the first book anywhere to ask and answer the question of what creating and maintaining a nuclear arsenal has cost the United States. Starting with the very first government-funded research into the military potential of nuclear energy and proceeding to the present day, we have attempted to document all significant nuclear weapons costs, from the well known to the obscure.
In the limited time we have today, I cannot possibly delve into every fascinating aspect of U.S. nuclear weapons—from the enormously costly but ultimately futile attempt to develop a nuclear-powered strategic bomber, to plans to deploy nuclear missiles underneath the Greenland icecap, to an early and highly secret effort to detect Soviet nuclear tests which inadvertently helped spawn the myth that aliens crash-landed in Roswell, New Mexico, to the numerous bunkers and command posts built in the 1950s and 1960s to allow military and civilian leaders to run the country during and after a nuclear war. For those stories and more you will have to read the book. But what I would like to do is touch on some of the more important findings of our work.
First, what did nuclear weapons cost the United States? From 1940 through 1996, we spent nearly $5.5 trillion on nuclear weapons and weapons-related programs, in constant 1996 dollars. This chart shows what the total is when estimated future-year costs for dismantling nuclear weapons and disposing of surplus nuclear materials, and for managing and cleaning up some 50 years of nuclear waste resulting from nuclear weapons production are added in. If we could represent $5.8 trillion as a stack of dollar bills, it would reach from the Earth to the Moon and nearly back again, a distance of more than 459,000 miles.
You can see that the majority of funds were spent not on building the nuclear explosives themselves?that proved to be relatively inexpensive given the scale of the program?but on the myriad delivery vehicles used to carry them to their targets. These included not only the well known strategic bombers and ballistic missiles, but also artillery shells, depth charges, and nuclear land mines. In fact, when we add the cost of deploying offensive delivery systems to those of defensive weapons, along with the costs associated with targeting and controlling the arsenal, we find that 86 percent of what was spent was spent on building a variety of launch systems and ensuring that not only could they be fired when ordered to do so but, more important, that they would not go off unless valid launch orders were issued.
As large as this total sum is, I want to stress that despite our best efforts it remains only a conservative estimate, a floor rather than a ceiling. The Government has never tried to track all nuclear weapons costs either annually or over time and as a result records in this regard are extremely spotty and in numerous instances non-existent. We know, for example, that in the early-to-mid-1950s, nuclear weapons were thoroughly integrated into what became known as conventional forces. This was done for two key reasons. One was that nuclear weapons were perceived to offer a “bigger bang for a buck.” Seeking to do more with less, this reputed benefit appealed to the Eisenhower administration, which authorized a massive expansion of the nuclear stockpile; when Eisenhower entered office in 1953, the stockpile consisted of some 1,400 bombs; by the time he left in 1961, the total had risen to more than 24,000, with the majority of these being for battlefield use.
The other reason was that the Army, seeing that the Air Force, and particularly the Strategic Air Command, was receiving a larger and larger share of the military budget, sought to capture some of these funds for itself. The Army therefore proposed and implemented plans for an array of nuclear weapons and only after these were deployed began to concern itself with how they might actually be used to defend Europe or the United States from the Soviet threat. What they discovered was that nuclear weapons, far from being cost-efficient killing machines, were in fact immensely expensive. Early nuclear weapons were quite complex and required large numbers of personnel to operate them, personnel unavailable for other duties. Wargames found that in head-to-head combat with a nuclear-armed adversary (a possibility surprisingly not considered when the weapons were first proposed) nuclear weapons would actually cost more than conventional ones. Because one nuclear weapon could kill or injure hundreds or thousands of troops at a time, large numbers of reserve forces would be necessary to maintain the advantage. Also, the large numbers of wounded would required an expanded medical corps, again further diverting soldiers from the field of battle. Despite these seemingly intractable problems, the Army was slow to shed its nuclear role, disarming completely in 1991 only after President Bush ordered the removal of thousands of tactical weapons from active duty forces.
We conservatively estimate that between 1946 and 1996, some 15 percent of the cost of what are known as “general purpose forces”was for nuclear weapons. That equals $1.2 trillion dollars. Changing the percentage to 20 would add another $400 billion to our total. Even though we suspect that the true figure for the 1950s and 1960s is significantly higher, perhaps approaching 50 percent, we have chosen to stay with 15 percent. I cannot stress enough that this is the single biggest unknown in our study.
$5.8 trillion is clearly a lot of money by any measure, but how does it compare when stacked against other government expenditures?
Here we see nuclear weapons spending ranked against all other federal government spending from 1940-1996, as documented by the Office of Management and Budget. The bar at a the extreme left is for national defense and totals $13.2 trillion. We have deducted our estimate of nuclear weapons-related spending from this bar. Next comes Social Security, at a nearly $7.9 trillion. Bear in mind however that much of this is not spending per se but funds collected from payroll taxes and redistributed to older Americans or placed in the trust fund. Coming in at a number three is nuclear weapons at almost $5.5 trillion. Welfare is a close fourth followed by interest on the national debt. I realize that the type on this is rather hard to read but the chart is in your packet and may also be found on page 5 of the book. What this chart demonstrates, among other things, is that nuclear weapons spending over this 56-year period exceeded the combined total federal spending for education; training, employment, and social services; agriculture; natural resources and the environment; general science, space, and technology; community and regional development (including disaster relief); law enforcement; and energy production and regulation. On average, the United States has spent $98 billion a year on nuclear weapons.
Where did all this money go?
Here we see annual spending totals for producing nuclear weapons materials. From 1948 through 1996, the United States spent $165.5 billion manufacturing plutonium, highly-enriched uranium, tritium, and other materials necessary to make nuclear explosives. That spike in 1953 is not a mistake. It represents the result of rapid expansion programs instituted after the start of the Korean War. In 1953, the government obligated almost as much money as was spent on the entire Manhattan Project less than a decade earlier. Those expansion programs were authorized and supported, by the way, even though officials knew that they would have no immediate impact on the Korean War because the materials could not be produced for some years hence. As you can see, the majority of production took place in the 1950s and early 1960s. So much highly-enriched uranium was produced as a consequence of these programs, that the United States unilaterally halted production in 1964 having achieved a huge surplus, a surplus that remains with us to this day.
This graph shows the other part of nuclear weapons production, the research, development, and testing budget. The Manhattan Project is over here on the left. The dip in 1976 is caused by the government’s changing the start of the fiscal year from July 1 to October 1. The increase in the 1980s is the result of the Reagan administration’s strategic modernization program which, ironically, did not lead to an increase in the size of the nuclear stockpile, as more weapons were retired than were built.
Today, the Department of Energy is proposing to spend at a least $4.5 billion a year on “stockpile stewardship” activities to maintain the nuclear stockpile for the indefinite future without nuclear testing and without large-scale production of new weapons. As you can see, this will be more than what we have spent on average over the entire Cold War (1948-1991), $3.6 billion, when hundreds to thousands of new warheads were being built annually and nuclear testing was a common occurrence. The Department has been insisting for more than a year that stockpile stewardship will in fact cost less than what it spent in the past. But when materials production activities are set aside and what used to be called research, development, and testing is compared with its replacement, the truth becomes clear. We have spoken with the DOE official who maintains the department’s historical budget database and he has told us that our figures are without a doubt more reliable than his because the DOE has no record of how many of its annual historical figures were derived.
Here we see the sum total of all known nuclear tests from 1945 through last month. The United States conducted more nuclear tests than all the other nuclear powers combined. The absence of U.S. and Soviet testing in 1959 and 1960 is the result of a mutually observed testing moratorium. The peak in 1962 occurred just prior to the enactment of the Partial Test Ban Treaty. That year, the United States detonated 96 nuclear weapons, 39 in the atmosphere. The Soviet Union detonated 79, all but one in the atmosphere. Fifty-seven percent of all the megatonnage exploded in the atmosphere–244 megatons?occurred between September 1961 and December 1962. That is the equivalent of 16,250 Hiroshima-sized bombs.
As I mentioned previously, the United States began amassing a sizable arsenal in the 1950s. Here we can see how a large portion of the stockpile–more than 36 percent–was for tactical weapons and how these weapons dominated the stockpile until the mid-1970s. The orange line represents an estimate about the size of the Soviet tactical stockpile, but a great deal of uncertainty remains about how many weapons the USSR actually produced. The arsenals of Great Britain, France and China are also represented here, but they are almost invisible at a the bottom of the graph, amounting to only a fraction of the superpower arsenals.
Here we see the evolution in acquisition costs for strategic nuclear delivery vehicles. Acquisition costs include research, development, testing, and procurement costs. Associated support costs–air bases and refueling tankers for bombers, missile silos for ICBMs, and submarines for SLBMs, are excluded from this accounting. As you can see, the most recent variant in each class of delivery vehicle is quite expensive. I have omitted the B-2 bomber from this chart because if I included it
this is what would happen. The B-2, by the way, is currently worth more than five times its weight in gold.
Of course, nuclear weapons are only useful if you can deliver them to their targets. Here we see the development of U.S. and Soviet bombers, ICBMs, and SLBMs. Here is the bomber gap. You will notice that it is sharply in our favor, despite the then-prevailing wisdom that just the opposite was the case. In fact, the rhetoric of the day actually served to increase the gap in our favor. The Soviet Union never fielded a substantial bomber force and for the most part did not engage in the sort of provocative posturing that we did in the 1950s and 1960s, such as routine incursions into Soviet airspace. Here is the missile gap. Again, we got it wrong, although the Soviets eventually built a large ICBM force and came to rely on it more than any other leg of their strategic triad. We hear talk today from India that in order to have a truly effective deterrent, one must have a triad of forces, with each leg countering the vulnerabilities of the other and complicating an opponent’s attack plans. But the triad as we know it was not the result of any sort of systematic plan. It simply evolved as the Air Force and the Navy (the Army was effectively prevented from competing in the strategic arena) built weapons in no small measure to deny the budgetary advantage to each other. As former Secretary of Defense James Schlesinger said almost a quarter century ago, the rationale for the triad is “just a rationalization.”
The rapid growth in the number of weapons led to a perverse process whereby as early as the 1960s targets were assigned to nuclear weapons rather than vice-versa. In fact, from 1961 until the late 1980s, there were always more targets identified than we had weapons, even though the stockpile was on the increase. This was due in large part to the secrecy surrounding the target planning process and the fact that for many years massive duplication existed in the targeting database, with dozens of warheads sometimes aimed at a installations that would be vulnerable to just one weapon. One particularly telling example relates to a new targeting guidance issued by Secretary Schlesinger in April 1974. In it, he called for U.S. strategic forces to “be able to destroy 70 percent of the Soviet industry that would be needed to achieve economic recovery in the event of a large-scale strategic nuclear exchange.” At the staff level, this was erroneously interpreted to mean “70 percent of every factory or industrial installation rather than 70 percent of Soviet economic-industrial capacity as a whole.” This error created a large phantom requirement for perhaps several thousand additional nuclear weapons. This mistake was only discovered years later during a thorough review of the targeting process. The bar for U.S. strategic warheads, by the way, is only for operational alert forces and does not represent the entire strategic stockpile.
Here we see the previous graph with the addition of a new bar showing the size of the stockpile in terms of Hiroshima-equivalent warheads. The bomb that destroyed Hiroshima nearly 53 years ago exploded with a force of 15,000 tons of TNT or 15 kilotons. In 1960 when the explosive power of the U.S. stockpile peaked, we had the equivalent of nearly 1.4 million Hiroshima-sized bombs. Today, even though the nuclear arsenal is substantially smaller, we still have the equivalent of 120,000 to 130,000 Hiroshima-sized bombs.
Defending against ballistic missile attack is something with which we’ve been preoccupied ever since Sputnik was launched in October 1957. Yet despite spending $100 billion (or about $107 billion if the last couple of years are added in) we are not much closer to creating a viable missile shield. And more than half of that has been spent just in the last 15 years. In the spring of 1958, after the Army proposed deploying what would have been the very costly and marginally effective nuclear-armed Nike Zeus missile to shoot down incoming Soviet warheads, then-Secretary of Defense Neil McElroy put the project on hold, saying, “We should not spend hundreds of millions on production of this weapon pending general confirmatory indications that we know what we are doing.” As with so much of our nuclear history, this warning still resonates today, as advocates of the speedy deployment of missile defenses press forward even as the proposed defensive systems fail test after test.
One reason that we face such a large and daunting “cleanup” program today is that so little money was spent on such activities in the past. This, coupled with an approach by the government that placed production of nuclear weapons and weapons materials ahead of everything else, mean that we face a bill of as much as several hundred billion dollars for a program stretching to 2070 and beyond to rectify past wrongs. And even today, as this graph shows, most of the money in the “cleanup” budget goes toward managing existing wastes. In all likelihood, the cost of cleaning up our nuclear weapons facilities will come close to or equal the cost of producing the weapons in the first place. These are lessons that India and Pakistan should bear in mind as they increase the size of their nuclear stockpiles. As in the United States and the former Soviet Union, the strict secrecy surrounding these programs increases the potential for officials to place production first, to cut corners, to look only at the perceived short-term gains of building nuclear weapons and ignore the very real and very dangerous long-term costs to the environment, and to public health.
Atomic Audit is not merely a historical exercise. Nuclear weapons are still very much a pillar of U.S. national security, even if the President insists that we don’t target anyone anymore. The de-targeting agreement with China announced with great fanfare by the White House over the weekend will have exactly zero impact on U.S. operational nuclear posture. We stopped actively targeting China in 1982. As for the 13 Chinese ICBMs that the CIA claims are targeted on the United States, we have no way of knowing if or how this agreement will be implemented. If China’s ICBM targeting systems are anything like ours or Russia’s, changing from one pre-programmed target set to another is about as easy as switching channels on a television with a remote control. The agreement may have genuine symbolic importance, but it has not done anything to ameliorate the underlying problems which have us and Russia still pointing thousands of warheads at a each other, warheads that can be launched in a matter of minutes once the order is given.
We are currently spending some $35 billion to operate and maintain our nuclear force, address the legacies of the Cold War–nuclear waste “cleanup” and victims of radiation exposure from nuclear weapons–and enact and enforce arms control agreements and try to develop missile defenses. That amounts to 14 percent of all appropriated funds for national defense, even excluding things like the unknown portion of the Navy’s costs for strategic antisubmarine warfare. These costs are significantly lower–by an estimated $22 billion or 39 percent–than they were in 1990. Yet going lower still is not a foregone conclusion. If the START II Treaty is not ratified by the Russian Duma, the United States may elect to keep thousands of warheads that would otherwise be retired on alert, increasing annual costs by perhaps $1 billion or more. If the nation chooses to deploy missile defenses, costs could rise further still.
Here is what our current nuclear spending looks like arrayed against all other federal government spending (this chart is in your packets). Slightly less than veterans benefits and services, somewhat more than natural resources and the environment.
We often get asked, how much would have been enough? How much was wasted?
Exactly how much of this country’s $5.5 trillion investment to date in nuclear weapons was “wasted” will remain a matter of debate, both because there has never been a fixed numerical goal or endpoint for deterrence and because “waste” is in the eye of the beholder.
When Admiral Arleigh Burke, Chief of Naval Operations, stated in 1957 his belief that the equivalent of 720 warheads on invulnerable Polaris submarines would be enough to deter the Soviet Union, the United States already had almost six times as many deployed. When retired Army Chief of Staff General Maxwell Taylor wrote in 1960 that “a few hundred missiles” (presumably armed with a “few hundred” warheads) would satisfy deterrence, the United States already had some 7,000 strategic nuclear weapons. And when Secretary of Defense Robert McNamara argued in 1964, that within a few years the equivalent of 400 megatons would be enough to achieve mutual assured destruction and hence deterrence, the U.S. stockpile had almost 17,000 megatons, 17 billion tons of TNT equivalent. In short, there has always been a tremendous gap between what informed military and civilian leaders thought necessary for deterrence and what was actually deployed, a state of affairs that has not changed with the end of the Cold War.
Many observers would classify the nuclear-powered aircraft program as a waste of more than $7 billion because it never produced anything approaching a workable concept and diverted critical resources away from more urgent programs such as the development of ICBMs. But what about the more than $1.6 billion dollars expended on Safeguard C, the 30-year effort to maintain the capability to resume atmospheric nuclear testing on short notice, a byproduct of the 1963 Partial Test Ban Treaty? Or the more than $400 billion expended on air defenses, much of it after it was ascertained that the Soviet bomber program posed no large-scale threat to the United States? What we can say is that at a minimum hundreds of billions of dollars were expended on programs which contributed little or nothing to deterrence, diverted critical resources and effort away from those that did or created long-term costs that exceeded their benefits (as in the case of the overproduction of fissile materials).
Moreover, the desire to quantify precisely what was “required” for deterrence diverts attention from the fundamental point that management of nuclear policy matters violated the core principles and practices of fiscally sound democratic governance. The 1950s notion that nuclear weapons provided “a bigger bang for a buck” was simply accepted despite contemporary evidence that this assumption would not hold up to careful scrutiny. The appropriate question is not how much or how little should have been spent (to which there will never be a single, unambiguous answer), but why numerous government officials over more than 50 years failed consistently to ensure that what was spent on nuclear weapons was spent wisely and in the most efficient manner.
While it can be argued that excessive or wasteful spending is a perennial problem in the United States, and while it may be tempting to compare the nuclear weapons program to welfare or agricultural subsidies or other entitlement programs in this regard, it is important to recognize one critical difference with respect to nuclear weapons: The costs of entitlement programs are well known. They are frequently debated in Congress and are readily available in Government documents to anyone who cares to look. The costs of nuclear weapons, by contrast, have never been fully understood nor compiled by the Government. Whatever problems were encountered in managing and disbursing entitlements, they were at a least well understood; indeed, the well known failures and abuses of the system have led to frequent and sustained calls to either pare back or eliminate particular programs, most recently in 1996. In more than half a century Congress has taken action to terminate nuclear weapons programs only a handful of times and has never held a hearing, debate or vote on the cost, scale, pace or implications of the overall program even though the potential for waste, fraud and abuse is at a least equal to that for entitlement programs, as indicated by the approximately equal share of spending for each.
I would like to conclude with our recommendations for how to improve the current state of affairs. Atomic Audit is a major contribution to our understanding of the costs of the Cold War, but it should be viewed as a first step, not the last. We therefore believe that Congress should pass legislation mandating an annual government-wide atomic audit. Such an audit should be done under the auspices of the Office of Management and Budget and should be submitted annually to the Congress with each fiscal year’s budget submission. It should account for all nuclear weapons-related costs in the previous fiscal year, the current fiscal year, and the fiscal year for which funding is requested. Such an effort would obviously require the cooperation of the Department of Defense, still the largest source of nuclear weapons-related spending, including intelligence concerning nuclear targeting, nuclear proliferation, and the like. The DOD has endeavored in the past several years to do some nuclear-related accounting by providing charts in the secretary’s annual report showing spending on strategic nuclear forces. Unfortunately, these have been misinterpreted by many to represent all DOD nuclear weapons spending, leading to a distorted view of the financial burden nuclear weapons continue to impose on the department.
Second, we believe the President should play a more active role in formulating nuclear weapons policy and requirements. The president currently signs off on the annual Nuclear Weapons Stockpile Plan which determines the overall size of the stockpile and the number of weapons to be built and dismantled, but presidents typically have little input or guidance to this plan before it reaches the White House. With nuclear weapons still consuming a sizable percentage of the military budget, it is vital to understand how the figures in the plan were derived. Moreover, these costs (except for those associated with environmental remediation and waste management) are driven by the Single Integrated Operational Plan, historically and today the most important factor behind the maintenance and deployment of the nuclear stockpile. We therefore call for an independent review of the current SIOP and the process by which it was developed.
Third, we encourage the Department of Energy and Secretary-designate Richardson to continue the openness initiative begun by Secretary Hazel O’Leary in late 1993. A complete history of the nuclear age cannot be compiled?nor the costs and consequences fully understood?without better and broader access to the original documentation held by the DOE and the DOD. In this regard, we are especially pleased that the DOD cooperated with this project by providing unprecedented access to the previously classified Future Years Defense Program Historical Database. This database, established in 1962, tracks the annual costs of every major defense program, and was of immeasurable help in conducting our audit. To facilitate its use by other scholars and the public, we are making available the aggregate spending totals for some 700 nuclear weapons and weapons-related programs on our World Wide Web page.
I would like to recognize three people without whom we would not have this data. First, my colleague John Pike, who recommended four years ago that we seek the declassification of the FYDP and for steering me in the right direction. Second, Jason Appelbaum, who spent more time than I know he wants to recall carefully entering the definitions for the programs and creating all the internal links. And finally, Tibor Purger, for designing the database for us and for making it a model of how the World Wide Web can be used to improve public access to and understanding of important policy matters.
Finally, we urge Congress to improve its oversight of nuclear weapons programs by focusing not just on the most expensive or most controversial items in the budget in any given year but rather on the larger strategic picture of how nuclear weapons would be used, how the various elements of the program contribute to deterrence and, not least, what constitutes deterrence in the post-Cold War era. Congress has begun emphasizing the broader picture to some extent in congressional scrutiny of the DOE’s environmental remediation and waste management program (especially from 1990 through 1994), where the overall goals and cost of the effort were examined in great detail while the individual programs at facilities throughout the country generally received less attention. Clearly, a mixture of the two approaches is needed to ensure a full and fair treatment of the budget. Yet with strikingly few exceptions, the annual congressional debate usually focuses on the minute details of a few programs at a the expense of the overall effort those programs are supposed to support. This approach can be likened to building a house by carefully examining the cost of only a few of the obvious elements, largely ignoring the rest, and rarely pausing to consider what the house will actually cost or look like, or if it will even meet one’s needs.
Given the structure of our representative democracy–particularly the myriad constituent groups and their ability to pull policy in many conflicting directions simultaneously–decisions about nuclear weapons budgets or any other item of public spending will never be entirely coherent. Nevertheless, the record to date demonstrates that Congress has been less than diligent in exercising its oversight responsibilities with regard to the nuclear weapons budget. Armed with the data we have provided and which we hope the executive branch (with Congressional prodding) will continue to provide, Congress will finally be able to make truly informed decisions not just on the basis of individual weapons but about the arsenal as a whole and about the sometimes hidden or overlooked costs (such as waste management, intelligence support, command and control) which inevitably accompany decisions about which missile, aircraft or submarine to acquire. That much of the current arsenal was purchased on the basis of arbitrary or strategically irrelevant decisions and justified by post hoc rationales should serve as an important reminder that programs, policies and weapons levels frequently cited as sacrosanct did not necessarily originate from an objective, clearly defined military purpose.
The time has come to consider carefully the costs and consequences to the United States, and the world, of producing tens of thousands of nuclear weapons and basing national security on the threat of nuclear annihilation. We have provided what we hope will be the starting point for such an assessment, focusing on the one aspect of the endeavor that until now has been largely ignored. We cannot rectify our mistakes or build on our achievements if such a crucial part of our nuclear history remains incomplete. Neither can we hope to prevent other countries from acquiring nuclear weapons if we do not fully comprehend the forces that have driven our own program and affect it still. Given the enormous sums expended and the substantial risks incurred, we owe it to ourselves and future generations to seek answers to these questions, to fill the gaps in the atomic ledger.
Note: The views expressed in this piece are those of the author and should not be attributed to the staff, officers or trustees of the Brookings Institution.
Copyright © 1998 The Brookings Institution