In this decade, terrorism has grown from an esoteric aspect of aggression and violence to a predominate means for international and intranational conflict resolution.[1]  It appears likely that as the smaller nations and weaker specialized interest groups of the world acquire the technology of modern war, both conventional and nuclear, they will increasingly turn to terrorism – just as the Palestine Liberation Army has done in the Middle East, as the Irgun and Stern Gang previously did against the British Empire, as guerilla groups in various Latin American countries do, and as the nineteenth-century eastern European revolutionaries did in order to bring down autocratic governments.

It is ironic that terrorist groups have been among the first to recognize that we live together in a planetary community, rather than a conglomerate of national communities.  The nearly threefold increase in the number of nations that have come into being since World War II makes this appear to be the most nationalistic of times.  However, most of the newer nations are caricatures, the products of historical accidents rooted in the imperialism of the last three centuries.  These new nations often have little in common, and many have no viable resources.

Nationalism was perhaps the main social, political, and historical focus of the nineteenth century, but in the twentieth century we have become one planet.  The consumption of irreplaceable elements by a relative handful of the human race affects the lives of all members of the race, as it always has; however, now the relationship is known and often felt strongly.  Couple this realization with the knowledge that the great power the ability to destroy both opponents of the moment and probably all of human society and one comes quickly to the few alternatives that exist to resolve conflict.

A main alternative is terrorism.  Formerly, terrorism was the prerogative of the powerless; it was a technique whereby a few determined men and women could affect the destinies of large empires.  This has changed.  Now we see examples of the technique being used by virtually all elements of society.

Terrorism, in this global setting, has become a predominant form of confrontation between differing subcategories of societies which seek to overcome each other, regardless of size.

This essay concentrates on nuclear terrorism throughout the world – a form of terrorism potentially so devastating that it must be considered meticulously.  Even though many might treat is as a very  low probability event, the margin for error is thin.

Borrowing from Willrich (1975; 12), terrorism is defined as “threats or acts of violence planned, attempted, or carried out by an individual or group with a specific political intent in mind.”  In the case of international terrorism, the definition is modified; such acts must fall “outside the accepted norms of international diplomacy and rules of war.” (Jenkins, 1975: 11)

This definition is broadly construed, and it includes the covert orchestration of surrogate warfare by nations.  In other words, terrorism may be conducted by individuals or groups acting solely on their own accord to achieve self-determined political objectives; terrorism may be secretly sponsored by other groups, organizations, or nations; or terrorism may be jointly undertaken by several parties pursuing overlapping objectives. Later implications of nation-state-inspired and – financed terrorism, an important but neglected dimension of the terrorist threat, are outlined.

Of the various forms that nuclear terrorism could take in the future, governmental policy and research have focused almost exclusively on the problem of terrorists manufacturing nuclear weapons from stolen fissionable materials.  There are understandable reasons why this definition of the terrorist problem has evolved.

So stated, however, the terrorist problem is narrowly misspecified.  The central objective of a safeguards policy should not be solely to prevent the illicit manufacture of nuclear bombs.  Rather, it should aim at preventing terrorists from acquiring a real or apparent nuclear weapon capability.

This improved specification enables us to identify several sorely deficient aspects of present safeguards policy and research.  In particular, only modest policy attention and no publicly available research address the prospect of terrorists stealing assembled nuclear weapons from military regimes.  Also, the public record suggests that virtually no consideration has been given to the possibility of unauthorized individuals acquiring an apparent or real capability to detonate nuclear weapons at their storage location, to arm and launch tactical or strategic nuclear weapons, or to direct armed forces personnel to execute nuclear strikes against other nations.  (Our concern extends to all nuclear powers, not just the United States.)

The possibility of any of these events has seldom been raised.  Only a handful of skeptics has questioned he adequacy of military nuclear safeguards, and seldom has any hard evidence been advanced in their support.[2]  Krieger (1975: 28) expounds the prevailing view:

“stealing an assembled nuclear weapon from a nuclear weapon nation would be the most difficult and the least likely route for terrorists to achieve a nuclear weapon capability … Additionally, there are reportedly sophisticated lock systems on the weapons themselves to prevent military weapons from being utilized by other than authorized personnel.”

The view that military nuclear weapons are today immune from theft or misuse contrasts sharply with views held, at least in the United States, in the 19650s and early 1960s.  The security and safety of nuclear weapons were controversial issues during that period, although earlier debate centered on the question whether safeguards were adequate to prevent the accidental or unauthorized use of nuclear weapons by armed forces personnel rather than nongovernmental terrorists. Such concern all but disappeared by the mid-1960s, and attention shifted to the development of safeguards for commercial nuclear power plants in the United States.  Several factors contributed to the reorientation.

First, policy makers and the public received repeated assurances from military quarters that highly reliable weapons safeguards had been implemented – measures that in fact were so reliable and effective as to be “fail-safe.”[3]  Defense Department officials maintained that the chances of a U.S. nuclear weapon exploding were “so remote as to be negligible.” (Phelps, 1961)  The development of supposedly “fail-safe” designs doubtless bolstered the public credibility of the military’s risk assessment, but confirmation came from other observers as well.  After completing a comprehensive study of military safeguards, Larus (1967: 42) was able to report:

“Today there is only a very slim possibility that a faulty communications signal, a mentally deranged airman, or any other mishap could trigger a Russo-American nuclear exchange.” 

Simultaneously, the issue lost much public visibility as Soviet allegations that U.S. military practices risked major nuclear accident or provocation abated.

Second, a series of dramatic shaped the reorientation.  An incident that attracted great attention, and which resulted in an abrupt awakening among policy makers to the threat of nuclear theft, was the loss in 1965 of over 200 pounds of weapons-grade materials from a Pennsylvania Atomic Energy Commission (AEC) fuel fabrication plant.[4]  As a direct result of this discovery, a new AEC Office of Safeguards and Materials Management was established in 1967 and was charged with responsibility for formulating nuclear industry safeguard policy.  In addition, the event led to the creation of an independent study group to assess the vulnerability of power facilities and marked the beginning of systematic research devoted to investigating terrorist threats to nonmilitary nuclear programs.

The literature since 1967 is not voluminous, but several excellent studies exist.  Unfortunately, the important but rather narrow range of issues on which these studies focus has helped foster the belief that theft of raw fissionable material is the nuclear threat posed by terrorism.[5]

A third influence contributing to the shift in U.S. safeguards policy was, and is, the rapid proliferation of nuclear technology.  In 1965, the nuclear power industry was nascent.[6]  Today there are scores of nuclear plants in operation in the United States alone, and several hundred are planned or under construction (Gapay, 1975).  As of December 1976, 229 reactors were operational or under construction outside the United States and 240 more were planned or on order (Walske, 1977).  The diffusion of nuclear technology throughout the world is accelerating, with the result that industrial safeguards are absorbing most of the public and official attention and resources.

The changing nature of nuclear technology is arousing even greater fears.  If the new generation of fast breeder reactors becomes operational, the number of plutonium shipments between processing plants and reactors will increase dramatically, vastly complicating the task of transport security, which is already considered the weakest link in the present safeguards systems.  According to one projection, assuming fast breeder reactors become a commercial reality, the amount of nuclear material in international transit each year will be enough to make 20,000 nuclear bombs (Gapay, 1975).

Finally, the upsurge in worldwide terrorism during the past decade affects perceptions of the threat of nuclear terror.  More than the absolute level of terrorism incidents, the historical trend and the realization that terrorism is unchecked are alarming.  A recent study made for the Energy Research and Development Administration (ERDA) found a sevenfold increase in terrorist incidents between 1969-1973 over the previous 1964-1968 period (HERO, 1974).  A recent downturn in this trend is heartening, but one cannot feel comfortable about a situation in which the ebb and flow of terrorist activity have less to do with governmental control than with the self-restraint exercised by terrorists themselves.

Contrary to popular belief, terrorism is not confined mainly to Middle Eastern and Latin American settings.  Indeed, according to ongoing work by Mickolus (1976), the Atlantic community has experienced more terrorism in terms of both incident location and nationality of victims than any other region of the world.  Latin America is second, followed by the Middle East, Asia, Africa, and Eastern Europe.  Although differing operational definitions of terrorism produce different results, the message is clear.  Terrorism is an international phenomenon that has eluded effective governmental control.

Security problems posed by a burgeoning nuclear power industry are cause for genuine societal concern.  However, as policy makers intensify the search for solutions with a view to shaping safeguards policy around a single prospect – terrorist theft of fissionable materials and construction of home-made nuclear devices – concern for safeguarding a large and ever-growing stockpile of the world’s nuclear weapons continues to fade.[7]  In our view, a thorough reexamination of the security of those weapons is long overdue.[8]

This recommendation stems in part from an evaluation we made of the technical and pr0cedural safeguards for what may be the Western world’s most well-protected nuclear weapons program – the Minuteman Intercontinental Ballistic Missile (ICBM) force in the United States.  After presenting a brief overview of the Minuteman security program, major findings are summarized, and several recommendations are advanced.  These findings are based on conditions which existed in the very recent past and which have recently undergone review by the U.S. Air Force and the Department of Defense.  We have been apprised that appropriate action has been taken to correct any safeguard weaknesses.  Although it is our position that general assertions of safeguards effectiveness are not sufficient, and that the burden of proof must be on those who assert the adequacy of remedial steps, the reader is cautioned against drawing specific inferences about the present situation from the evidence presented.  In developing a case which underscores the need for a policy of continuous and thorough safeguards evaluation, we make no claims concerning the workings of weaknesses of current Minuteman safeguards.  However, we do share the grave concern evidenced in the following remarks made in March 1976 by Congressman Ottinger:

“From classified material I have seen, as well as from unclassified briefings I have received from former high-ranking Defense Department personnel, all of which I hope this committee will take the time and trouble to explore thoroughly, I have every reason to believe that the protections are inadequate against catastrophe by way of theft, sabotage, unclear and overextensive delegation of authority, incompetence or incapacity of authorized personnel, unauthorized use, weakness of communications and command and control. [U.S. Congress, House, 1976; 13].” 

From what was publicly known as late as March 1976, it appeared quite possible that strategic nuclear weapons could be compromised in various ways. 

The Minuteman is a three-stage, solid-fuel missile with an intercontinental range, capable of carrying a one-megaton payload, and housed in a protected, concrete and steel underground silo.[9]  A silo is also protected by an elaborate security system, consisting primarily of a fence surrounding the perimeter of the silo, combination locks to gain access to the underground silo, and sensitive electronic instruments.  Under normal conditions, when all security systems are functioning properly, physical intrusion that “breaks” the silo perimeter wither above or below ground is registered by instruments and transmitted via cable to a computer visual display and readout located in an underground Launch Control Center (LCC) several miles away.[10]  In this manner, two-man LCC crews monitor the security status of a ten-missile flight simultaneously.  Security violations reported by the crew result in the dispatch of armed security police who inspect the site and silo for intruders and evidence of unlawful entry.

There are five flights, hence five two-man LCCs, in a 50-missile squadron.  Since all missiles and LCCs are electronically interconnected, the “normal” launch of any or all missiles in a squadron requires the cooperation of only two crews – no more, no less.  One LCC crew can launch any or all of a squadron, but in this “abnormal” situation the launch is delayed substantially, i.e. the missile lift-of reaction time is increased from about eight seconds to a matter of hours.[11]  The primary reason for this built-in delay is to allow adequate time to “inhibit” an illicit launch command generated by a single aberrant LCC.  Indications of such an attempt are automatically relayed via cable and computer to each LCC in the squadron, and each LCC crew is trained and responsible for instructing the squadron’s missiles, through a computer command, to disregard the unauthorized command.

Located in each LCC are two launch keys, one for each member of the crew, and the codes needed to authenticate presidential launch directives.  Only the launch keys, not the codes, are physical prerequisites for generating valid launch commands, the purpose of the codes being exclusively that of authenticating an execution directive.[12]  Contrary to popular belief, there are no mechanical “Permissive Action Links” (PAL) installed at LCCs to prevent cooperating crews from launching missiles without presidential authority.  The implementation of a version of PAL (called the “Permissive Enable System”) is underway, but there is only a meager store of evidence on which to define the program’s scope and purpose and to base an appraisal of its quality.

Other sensitive information found in each LCC includes lock combinations to missile sites which are passed to authorized personnel on site for silo maintenance, targeting information for missile sorties in the squadron, and code and message formats used to validate directives of various kinds, including war termination orders.

Carefully selected and highly trained crew members are the only personnel authorized to perform various ICBM system operations, and the security of the operation depends heavily on their professional integrity.  Technically, crew members can launch a nuclear attack with or without approval from higher authority.  Unless PAL or its equivalent forecloses this option, as many as 50 missiles could be illicitly fired.  Moreover, unless adequate precautions were instituted, an even more drastic option would be available.  Crew members could conspire in the formatting and transmittal of strategic strike directives, deceiving the full contingent of Strategic Air Command (SAC) LCCs, as well as higher authorities, into reacting to a spurious launch directive as if it were valid and authentic.  Or they could render the U.S. strategic force virtually impotent by formatting and transmitting messages invalidating the active inventory of presidential execution codes.  Finally, crew members could aid accomplices in stealing thermonuclear warheads from missiles on active alert.  Such weapons are many times more destructive than any atomic bomb that might be constructed from stolen fissionable materials.

The public’s abiding confidence in the probity of the professional officer corps assigned LCC responsibilities is well deserved; however, the margin for error for the proper functioning of the launch network is not as great as one might believe.  Without stringent safeguards, a single aberrant individual could “unsafe” the arming mechanisms of an entire Minuteman squadron, or facilitate the theft or sabotage of nuclear warheads.  Other acts of terror, except for the physical launch of ICBMs, would require the collaboration of only two individuals (one person in each of two separate SACLCCs).  Four individuals (two persons in each of two separate LCCs in the same squadron) acting in concert could succeed in mechanically launching one or more missiles.[13]

If these nontrivial risks were ignored or inadequately addressed, the terrorist problem would be compounded in the case where access to the launch network is not stringently controlled.  Given the enormous discretionary power held by whoever has LCC control, effective measures for denying LCC access to individuals or groups bent on carrying out an act of nuclear terror are self-evident security requirements.

In the recent past, such safeguards were poor or nonexistent.  Military personnel, e.g. maintenance airmen, and civilian contractors who possessed minimal security credentials were granted LCC access, and annually thousands of visitors holding no clearance whatsoever were permitted access to operational LCCs.  In the interest of public relations, the Air Force permitted ready access to the Minuteman launch network by practically anyone desiring it.

Requests for visitor access were routinely processed and approved.  The requesting party had only to provide a name and social security number, and authentication checks were not usually made.  As a matter of course, checks of individual backgrounds or motives for requesting LCC access were not made either.  Furthermore, within wide bounds, the number of individuals in a party was limited only by the capacity of an LCC – about eight persons.

Once military personnel and civilians are allowed inside an LCC, responsibility for them falls squarely on the shoulders of the on-duty crew members.  The present situation parallels that which existed several years ago in the area of airline security.  Aircraft flight personnel are manifestly incapable of curbing hijacking incidents, and not until major changes in airport security were implemented was the incident rate reduced to tolerable levels.  LCC crews are no more capable of thwarting launch network seizures than unassisted flight personnel are capable of foiling hijack attempts.  Added to all this are the facts that no acknowledged procedures or rules exist to prevent or prohibit groups of military personnel and/or civilians from gaining simultaneous access to LCCs in the same squadron, and procedures, i.e. technical orders for arming and launching missiles are unclassified and can be readily performed, especially if rehearsed in advance.

One must also recite the obvious point that silos and launch control centers are loated in desolate reaches of the heartland.  Reaction times to mount a counterterror offensive pinpointed at one or a few of these facilities would be measured in hours, not minutes or seconds.

Although the unfolding scenario contains all the ingredients of a nuclear disaster, the seizure of one or more LCCs would not necessarily lead to nuclear violence.  Even if the most serious loopholes were not closed by newly implemented changes in security, terrorists might be unable or unwilling to consummate the nuclear options potentially at their disposal.  But the mere seizure of control and the acquisition of a possible nuclear weapon capability would greatly enhance the credibility of any threats they might make.  In responding to threats of terrorists whose nuclear capability is even remotely plausible, authorities may feel compelled to accede to their demands as if the alleged capability were real.

Elaborate lock systems, personnel screening, e.g. crew security clearance and human reliability programs, and “no-lone-zones,” not-withstanding, there is little reason to have confidence that Minuteman safeguards are inviolable.  If this component of America’s nuclear force, so often hailed as epitomizing reliable command and control, has been or continues to be far less than “fail-safe,” then America’s nuclear force as a whole is implicated.  As noted earlier, it was just March 1976 when Congressman Ottinger was able to voice serious concern about the safeguards for all nuclear weapons programs – both strategic and tactical:

“Let us now turn to tactical weapons.  The situation with respect to safeguards against theft, sabotage, seizure, dangerous delegation of authority, unauthorized use, incapacity or incompetence of those authorized and ineffective communications, command and control is many times worse with tactical than with strategic weapons – and we should bear in mind that the differentiation between tactical and strategic weapons today is mostly a matter of mission – many weapons classified as tactical have destructive power many times that of the bombs we dropped on Hiroshima and Nagasaki.” [U.S. Congress, House, 1976: 13]

Corrective action is required to shore up commercial and military nuclear safeguards.  But the question remains, what programs should have priority and how much improvement is needed? The answers lie not only in identifying security defects, but in understanding terrorist objectives and capabilities.

Not much is known about these subjects, and even less is known about them in relation to the commission of nuclear terror, since the terrorist manufacture or detonation of nuclear weapons has never occurred.  Given the present state of knowledge and the potential consequences of failure, once could assume that some fraction of the terrorist population, however minute, will seek a nuclear capability.  On purely assumptive grounds, upgrading ill-protected or ineffectual security programs is justifiable.

The type of safeguards systems most appropriate for deterring or subduing terrorists should be designed with the following factors in mind:

• Likely tactical objective, e.g., achieving a credible nuclear launch capability, weapons theft, force degradation.
   
• Likely strategic objectives, e.g., punishment, concessions, fear and alarm, publicity.
   
• Terrorist capabilities.
   
• Pervasiveness of groups disposed to committing nuclear terrorism.

With respect to historical precedents relevant to the latter two categories, researchers are finding that: terrorists and terrorist groups are growing in number and widening the scope of their activities, they are well-financed and well-educated, there is an unprecedented extent of international cooperation, attacks against targets of wider variety and complexity are being mounted, more ingenious means of gaining access to and escape from these targets are being devised and used, and more sophisticated conventional weapons than ever before are being relied on.[14]

The deficiencies identified indicate that basic changes in Minuteman safeguards are required, changes that may also pertain to other strategic and tactical weapons systems.  The following recommendations do not exhaust the creative possibilities, but they appear to be promising and deserving of further consideration and evaluation.  They are listed in what we believe to be an order of comprehensiveness and probable effectiveness.  Similar assessments are also needed of all nuclear weapons systems, and these assessments would doubtless evince similar lists of recommendations for improved security.

Secrecy has probably engendered more public ignorance of the risk of illicit nuclear weapons detonation than of security problems arising in connection with the growth of commercial nuclear power.  In either case, public input in the development of nuclear safeguards policy is effectively limited by classification’s ubiquitous stamp.

Because nuclear terrorism is rapidly becoming more practical and legitimate, we need to go beyond examining the efficacy of safeguards in the context of terrorist capabilities and objectives.  We also need to look closely at the organizational, political, and economic factors that determine the shape of terrorist prevention policy more generally.