IRDISP-RESEARCH INSTITUTE FOR DISARMAMENT, DEVELOPMENT AND PEACE

ABSTRACT: Fine that there is the arms race, but what has Italy to do with it? Aren't the U.S. and the Soviet Union the promoters of such race? It is true that the two superpowers are the chief responsibles for the arms race. The chief ones but not the only ones. Italy has its share of responsibility as well. A smaller share, but not a negligible one. In absolute numbers, Italy's military expenditure in 1985 was the eighth of the world. As regards the number of men at arms, it is among the first fifteen countries. And the Italians are in the first six positions among the world exporters of armaments. The weight of the military sector on the whole of the Italian economy is still rather limited: The expenditure accounts for 2.7% of the gross domestic product; arms represent 2.7% of the wealth produced by the industry, and 2.3% of the exports. Moreover, the military threats to the security of Italy are less serious than those which many other international actors have to face - including many of our allies. Ther

efore, Italy is in a situation that offers many opportunities to contain the expenditure, experiment conversions to civilian uses of the military productions, and promote a realistic security policy aimed to achieve detente.

Unfortunately, these opportunities are dropped. In fact, there has been a tendency to expansion over the last decade which must be urgently stopped. As of the mid-seventies, Italy has become one of the major exporters of weapons systems, and its military expenditure exceeds the annual growth rates decided at the NATO level. That same period marks the rise of the supporters of a "new military role" for Italy in the Mediterranean. The "White Paper" presented by the Minister of Defense Spadolini in winter 84-85 summarizes and pinpoints these developments, obviously from the point of view of a person who supports them and hopes that they will continue. This book on the contrary highlights the doubts, the questions, the alternative proposals compared to what has been to this moment a monologue carried out by the establishment.

("ITALY AND THE ARMS RACE" - A counter-White Paper of defense - edited by Marco De Andreis and Paolo Miggiano - Preface by Roberto Cicciomessere - Franco Angeli Libri, 1987, Milan)

1. C3I AND UNITED STATES STRATEGIC DOCTRINE

by Marco De Andreis

1. Introduction

C3I stands for Command, Control, Communications and Intelligence. For the purposes of this paper, all refer to strategic nuclear forces.

In a discipline so inclined to acronyms and formulas as that of strategic studies, "C3I" was bound to be a hit. Only one of the four terms taken by itself, however, is enough to give an idea of the underlying problem. Thus, one can speak of control of nuclear forces: control of anything requires information (Intelligence), that has to be received and transmitted (Communications), sometimes in the form of orders (Command).

The study of control of nuclear forces is becoming more and more fashionable in contemporary strategy. Works of any scope scientifically dedicated to the subject are all, at least as far as I know, post-1980. In recent writings, and not only in specialised literature, it is almost impossible not to come across some mention of C3I. What's the secret behind this success?

In part it is due to cultural trends in the sector. In a sector in which "Much of what is offered today as a profound and new insight was said yesterday; and usually in a more concise and literate manner" (1), finding a new angle to the nuclear dilemma - whose horns were laid bare forty years ago by Bernard Brodie - can mean academic success for whoever discovers it and those who first climb on the bandwagon.

Going beyond fashion, however, there seems to be some logic to studying control of nuclear forces. This is often expressed as a metaphor of human physiology in which nuclear forces are the muscles and C3I the nervous system which moves the muscles when necessary. One could take it even further, identifying political and/or military authorities as the brain which takes the final decision on use of nuclear weapons (2). In any case, the metaphor is useful in clarifying that a break-down of the nervous system is capable of paralysing the entire organism, even if the muscles are in excellent condition and the mind lively. All who have written about C3I repeat that strategic studies have put too much emphasis on the calculation of forces, à la SALT, or on the polemics about doctrine, respectively the muscles and the value systems of the nuclear mind. Instead, the way in which strategic doctrines interact with weapons systems - or the question of control - has been largely overlooked.

All this apparently amounts to C3I's revenge on other fields of study of contemporary strategy. I say apparently, because, actually the opposite is true. The control system, per se, seems to be a jumble of strange, mainly electronic, devices. The only reason to talk about it is in order to judge doctrines and nuclear forces. Which means that, in the end, after having battled one's way through a multitude of technical details - all of importance - one ends up with the same old dilemma: do nuclear weapons have any political use in the contemporary world? What if deterrence fails? and so on.

In the first part of this paper an attempt will be made to describe the American C3I system and then evaluate it in the light of the existing strategic doctrine and the modernisation programmes brought in under the Reagan administration. As will be seen, strategic doctrine is the real centre of balance of this paper.

2. The Pentagon budget and C3I

Among the reasons for the current interest in nuclear control, the high priority is given it by the Reagan administration has its importance. This priority was recently reaffirmed. On 3 June 1986, the White House issued a "Summary of those Strategic Modernization Programs the President has asked Congress to exempt from defense cuts" (3): command, control and communications progammes have priority over all others, including the SDI and the modernisation of strategic forces.

In terms of budget, C3 (strategic and tactical) has risen from 5.9 per cent in 1982 to 6.5 per cent in 1985 of total Pentagon expenditure. In current value, this is 12.3 and 18 billion dollars respectively (4). On the average, more than half of the these funds are earmarked for strategic C3I, with 73 per cent going to acquisition of new hardware.

Bruce Blair, THE American expert in command and control (5), believes that this is really only the face value, or rather, that the Pentagon is not about to put its declared intentions concerning C3I into practice. In Blair's opinion, defence budget projections show that strategic C3I will vary between 1.3 and 1.3 per cent of the total in the financial years 1983-87. In addition, approximately one-third of those sums will be absorbed by a single programme, the communications satellite, MILSTAR. "The figures", he wrote, "raise some doubts as to whether the Reagan Administration's pledge to overhaul the nuclear command system is genuine" (6).

It is also doubtful whether the President's appeals can overcome the inclination of individual armed forces to continue acquisition of their own weapons system to the detriment of all other budget items: "Indeed, we are likely to witness greater pressure, not less, on C3I expenditures, and a continuing preference on the part of the services for weapons " (7). All the more so in the context of the budget freeze imposed by Gramm-Rudmann. Now let's take a brief look at how the American strategic C3I network works.

3. The Chain of Command

Control over the armed forces is in the hands of the National Command Authorities (NCA), that is, the President and the Secretary of Defense, or their representatives or successors. Control of nuclear weapons is the responsibility of the President, who is the only one who can authorise their use. However, "such authorization is unlikely without the consent of the Secretary (of Defense)" (8). Not only that, the chain of command calls for an intermediate step between the President's decision and actual use of nuclear weapons by a field commander: the Chairman of the Joint Chiefs of Staff. Authoritative statements confirms that, practically, control is shared by three people. Lyndon Johnson once joked about what would happen if he were to give a sudden order for an all-out attack on the USSR. "General, go get'em! You know what the general would say? "Screw you, Mr. President" (9).

What if a President is "incapacitated"? According to the 25th Amendment of 1976 and the Presidential Succession Act of twenty years earlier, the line of succession goes from the President to the Vice-President, then to the Speaker of the House of Representatives, the President pro tempore of the Senate and finally to the various secretaries, in order of seniority. Another key figure in the NCA is the Secretary of Defense. Here the line of succession is equally complicated, first the Deputy, then the Secretaries of the Army, the Navy and the Air Force, the Director of Defense Research and Engineering, the various Assistant Secretaries and the General Advisor to the Department of Defense (by seniority), the Under-secretaries of the three armed forces (as above) and the Assistant Secretaries of the three armed forces (as above).

What seems likely is that under extremely stressful circumstances, such as after a nuclear attack on the United States, the result would be a good deal of chaos. In practice, the first person available of all those listed above would become the legitimate successor. A perfect example was the confusion after the wounding of Reagan by a psychopath on 30 March 1981, in a moment of absolute calm, as far as international tension is concerned. With Vice-President Bush away, then Secretary of State Haig proclaimed himself the successor: "I'm in control here, in the White House, pending the return of the Vice-President" (10). As we have seen, at least two other people should have had precedence over Haig.

The black briefcase (which the Americans call the Football) that follows the President everywhere, clutched tightly in the hands of a member of his escort, has caught the public's fantasy. It contains the codes for authorisation of the use of nuclear weapons and the Black Book with the options of the Single Integrated Operational Plan (SIOP). Even if the President can be temporarily defined "incapacitated", it has not to be taken for granted that his successors are aware of the briefcase's contents. When, after Kennedy's assassination in November 1963, the Football was passed to Johnson, he had not the slightest idea of what it contained (11). Thereafter, the situation remained unchanged, at least up until Carter, who, after taking office "was astounded to find that the next commander in line, the Vice President, had never been involved, in any administration, in any of the top secret briefings about the SIOP and its attack options" (12). A series of briefings on the subject followed, attended by Carter, Vic

e-President Mondale and Secretary of Defense Harold Brown.

Except for Carter, and to a lesser degree, Kennedy, not even the presidents seem to have much familiarity with the unsettling contents of the Football. General William Odom, military assistant to Carter's National Security Advisor, Brzezinski, made the following comment about the last Democratic president's efforts to master control of nuclear forces: "I don't think that has ever been done before", in general the attitude of the other president "...toward command and control, particularly of the strategic forces, has typically been one of benign neglect" (13). We don't have much to go on concerning Reagan's knowledge of the subject, but the few indications we do have are both very reassuring. He has said that missiles launched by submarines are recallable; that "land-based missiles have nuclear warheads, while bombers and submarines don't"; and after learning that the Soviet SS-18 missile is larger than the SS-19, he commented, "they've even switched the numbers on their missiles in order to confuse us!" (14

).

4. Command Posts

In order to be in control of nuclear forces, the President has to make use of communications. And these have to be received somewhere. Thus, the NCA has to have command posts. The first is in the Pentagon, occupying the second and third floors. It is called the National Military Command Centre (NMCC), houses the Moscow-Washington Hotline and is ot hardened in any way. Concern for the vulnerability of the NMCC first emerged in 1954 and therefore, another protected command post called Alternate National Military Command Centre (ANMCC) was established at Fort Richie in Maryland, seventy-five miles from Washington. With the development of Soviet offensive forces, the ANMCC also became vulnerable. At that point, the only logical way to have a survivable command post was to fit out a B-747 for the purpose.

There are four of these aircraft, denominated E-4. The Pentagon originally asked for seven, but the number was reduced because, it seems, of the great expense of making the aeroplane and its equipment resistant to electromagnetic pulse (EMP) (15). (We will return to EMP later). The E-4s also have an acronym, pronounced "Kneecap", which stands for National Emergency Airborne Command Post.

Theoretically, an aircraft is certainly less vulnerable than a fixed installation. Yet, it too has its problems. It can, for example, be refuelled in the air, but its engines can't be lubricated. Thus, autonomy is limited to seventy-two hours, without taking into consideration the crew's fatigue. During a nuclear war, who knows how many airports would be available on which to land a B-474, refuel it, change flight crews and so on. Furthermore, debris and dust flung into the atmosphere by possible grounds bursts could seriously damage the engines. This may be of little importance to those who feel that an exchange of intercontinental weapons can't last for long anyway. If, however, like the current administration, one feels that it may continue for months, then the longevity of the Kneecap becomes a serious issue. Finally, there is the problem of crisis management: to have the NCA board the NEACP in a moment of extreme tension might be taken by the Russians as a declaration of war, which in turn might dissuad

e the President from taking a decision.

It would, however, be a prudent decision: it would take a Soviet SLBM less than ten minutes to reach--and destroy--the runway on which the E-4 is waiting for the President, or--and the result would be the same--to reach and destroy the White House. In the case of an ah hoc attack, then, would the NCA manage to board the Kneecap?

In 1977, Brzesinksi simulated an emergency. It ended in disaster. Not only did emergency procedure go beyond the time limit, but security almost shot down the helicopter that was to carry the President and Brzensinski to Andrews Air Force Base, where the E-4 was waiting (16).

In 1983, the Pentagon decided to move the aeroplane from Andrews, which is very close to Washington, to a secret base further inland. At the new airport, the aircraft is in constant ground alert, so as to be able to take off before an SLBM arrives, and perhaps land elsewhere to rendezvous with the President's helicopter. However, a senator from Indiana, Daniel Quayle, could not resist the temptation of boasting to his electorate that he had succeeded in having a new military base transferred to his state. The new base, at Grissom Airport, was that of the Kneecap (17).

Another important command post is that of the Strategic Air Command (SAC), located underground in Omaha, Nebraska. In time, it too was considered vulnerable and airborne duplicates were made. The United States Air Force has two flight groups of EC-135s (B-707s), called Post-Attack Command and Control Systems (PACCS) or, more commonly, Looking Glass. Since 1961, one or two of these aircraft are always in flight. On board there is, among others, an air force general, who is, theoretically, to carry out the orders of the SAC commander of the NCA.

Theoretically, because in the case of a successful decapitation attack, the NCA would, in fact, become the Looking Glass. The practical consequences od this is that in order to deal with situations of this kind, whoever is on board the PACCS can decide/authorise the use of strategic nuclear forces. This can take place in a number of ways: the aircraft can transmit the launch order, called Emergency Action Message (EAM) to land-based intercontinental missile (ICBM) Launch Control Centres (LCC); should the LCCs not be in operating condition, it can directly launch the ICBMs; it can do the same by means of other Looking Glass aircraft, if the latter have not been destroyed on the ground; it can order the launch of emergency satellites - that is, the Emergency Rocket Communication System (ERCS), a dozen Minuteman missiles based at Whiteman in Missouri which, instead of warheads, carry transmitters capable of sending a launch order to the other ICBM bases (18). Furthermore, the SAC aircraft can order bombers to h

ead for their targets, although this calls for the presence of other Looking Glass aircraft in the air space to create a link to the formation (rows) of B-52s. Lastly, it can communicate with the Navy's C-130s, which can in turn pass on the order to missile-launching submarines (SSBNs), to launch their SLBMs. The aircraft that communicate with the SSBNs are called Tacamo, from Take Charge and Move Out. Two are always in the air, one over the Atlantic and one over the Pacific and the rest of the two groups are in constant ground alert. The Tacamo cannot be refuelled in flight and thus has an autonomy of only ten to eleven hours, in comparison to Looking Glass' twenty-four hours (19).

5. Authority over Use of Nuclear Weapons

The problem raised by Looking Glass' capability to decide on the use of strategic forces, brings us back to the "legal" question of the authority over the use of nuclear weapons. At this point, it seems quite evident that the president's real powers represent a constitutional monarchy, at least. The authority of the head of executive is subject to numerous conditions and/or exceptions. To have an idea of the scope of the matter, let us go back to the concept of control: we can call positive control the guarantee that nuclear weapons will be used when the NCA so decides; whereas measures taken to avoid improper use can collectively be called negative control. The two forms of control are inversely proportional to each other, in the sense that maximising the one minimises the other and vice versa. Bottlenecks of this kind occur often. Perhaps the classical logical form is that of the beta error (accepting a false hypothesis) against the alpha error (rejecting a true hypothesis).

There are a multitude of negative control measures, some of which even involve the President who, in splendid isolation, would hardly be listened to, if not sent straight to the devil, as Johnson said. For all the other military who, in one way or another, participate in control, essentially two types of safety measures have been adopted: the first consists of having more than one person participate simultaneously in the procedure for use of a warhead; the second is the so-called Permissive Action Link (PAL), a numeric code, without which it is impossible yo activate a warhead. In some cases, such as the ICBM control centres, there is both two-man rule and PAL, that is, two individuals must convalidate and simultaneously insert the launching codes plus a series of other safety measures, one of which is a kind of veto right by the crew of another LCC.

With reference to non-strategic nuclear weapons, PAL is the main guarantee against unauthorised use. And lastly, for SSBNs, Looking Glass aircraft, and probably bombers and ships equipped with nuclear weapons (some attack submarines and certain ships from frigates on up), authorisation depends on more than one person - it seems that on SSBNs, almost the entire crew is involved (20). In other words, those serving on the craft either already have the codes or at least are familiar with the procedure for deployment of the various devices.

The reasoning behind this kind of choice is clear from the fact that in all cases mentioned here, there is a good probability that communications with NCA may be cut off. Or, as in the case of looking Glass, that the officer aboard may have to take the place of the NCA. Thus, the only way to make up for decapitation is by pre-delegating control. This means that in order to guarantee positive control, a quantum of negative control has to be given up. Although some people are keen on introducing PAL on naval vessels, in particular SSBNs, in general, the system of safety measures against unauthorised use is considered more than satisfactory. The probability of a scenario like that of Doctor Strangelove, in which a SAC general decides, independently, to send his B-52s to bomb the Soviet Union, is considered to be very low. It should be pointed out that this problem is conceptually quite different from that of a "false alarm" which will looked at later while considering early warning systems.

Finally, it should be remembered that the less tense the international situation, the more effective negative control. That is to say, the more nuclear forces are generated, which is jargon for increasing the state of alert, the more probable it is that a commander, or an entire crew, decides to not wait for orders from the NCA. As a consequence, alert becomes a problem in itself. This will also be dealt with later on. Let is suffice to recall here that Defense Condition (DEFCON) 1, which is the maximum on a scale from 5 to 1, of generation of US forces is, significantly, called cocked pistol.

6. Early Warning

Although nothing is stopping anyone from being the first to use nuclear weapons, the concept of deterrence suggests a reactive role. The reaction is set off by an outside attack. Logic, therefore, demands that one must be able to establish, first of all, if an attack is taking place and secondly, where it is coming from. Then it may be of help to know the size and target of the attack. All this is warning. And warning is none other than the I in C3I, since no warning can be given without information. An important decision of warning is time. "Strategic warning is warning of impeding attack prior to its actual execution. Tactical warning in contrast, refers to warning of an executed attack" (21). In general, the organisation and means described in this section are needed for tactical warning; strategic warning makes use of more or less all the other resources provided by information services: from spies to reconnaissance satellites.

The first means for determining a ballistic missile attack against the United States is constituted by the three satellites of the Defense Support Program (DSP). They are in geostationary orbit (at an altitude of around 36,000 Km): one over the eastern hemisphere and two over the western. They are equipped with infrared sensors that can detect the emissions of a missile in about a minute after launch. Its point of departure can be established with an approximation of from 3-5 Km. DSP satellites also carry systems for detection of nuclear particles, X-and gamma-ray sensors, and EMP sensors. For data transmission, DSP East relies on an unhardened reader station in Australia. From there, data are sent to the United States (California) by underwater cable. The two DSP West satellites, on the other hand, being in line of sight, can transmit information directly to Sunnyvale California, where the Satellite Test Centre is located. Besides sorting and transmitting DSP data to the North American Aerospace Defense Com

mand (NORAD), the centre at Sunnyvale also has the task of keeping the American satellites on their assigned orbits, with their sensors and antennae pointed in the right direction. It may also be defined as a "large, soft target". Besides NORAD, NMCC, ANMCC and SAC also receive data (apologies for the maze of acronyms). The first, however, plays the key role, being responsible for attack assessment--the number and direction of attacking warheads--and its transmission to other command posts.

Early warning satellites are integrated by a series of other sensors, mainly radar. They are:

a) the Ballistic Missile Early Warning System (BMEWS), three radar situated respectively in Clear in Alaska, Thule in Greenland and Flyingdale Moor in Great Britain;

b) two PAVE PAWS radar - one at Otis air base in Massachusetts and the other at Beale air base in California, which primarily warn against launch of SLBMs;

c) the Perimeter Acquisition Radar Characterisation System (PARCS), a kind of leftover from the dismantled ABM Safeguard System, which serves mainly for attack assessment;

d) FPS-85 and FSS-7 radar, both in Florida and designed to detect SLBM launches from the south;

e) the Cobra Dane radar in the Aleutian Islands in Alaska, the main role of which is "intelligence collection and verification".

In the words of Desmond Ball, "It is possible that some early-warning and attack assessment would be provided by signals intelligence (SIGINT) sources. These sources include more than two thousand signals intercept stations located around the world, electronic intelligence (ELINT) and communications intelligence (COMINT) satellites in relatively low earth orbit, and the geostationary SIGINT satellites...which monitor centimetre (i.e. microwave) and millimetre signals" (23).

Now considered of secondary importance--but modernised, nevertheless, as we shall see--are the radar set up to detect a bomber attack. The thirty-three installations of the Distant Early Warning (DEW) Line, extending from Alaska through Canada to Greenland and including two radar in Iceland, date back to the 1950s. Besides the DEW line, mention should also be made of a system of thirteen radar in Alaska, another line of twenty-four radar operated by the Canadians (PINETREE), air defence radar in Hawaii and two aerostat-borne radar in Florida.

The redundancy of early warning sensors has led to an important safety measure called dual phenomenology. In other words, more than one system has to signal an attack before countermeasures, of which the extreme is retaliation, can be taken. Complexity is welcome here: numerous checks and balances act as an insurance policy against false alarm caused by the always possible failure or malfunction of equipment. Nevertheless, there have been a number of cases in the post-war period of dangerous false alarms set off by flocks of geese, the break-down of a microchip, or a war-game programme forgotten in the NORAD computer. In some cases, bombers already had their engines started and were made ready for take-off. It doesn't take much to realise, though, that nothing serious has ever happened.

In this respect, the considerations made for negative control hold true here as well. That is to say that in a relaxed international climate, the entire early warning system works wonderfully in avoiding the kind of event we usually call "war by mistake". A serious crisis in which nuclear forces would already be in advanced state of alert, is another story altogether. In such a situation, there is the danger that a single phenomenology would be enough to raise the state of readiness of nuclear forces, and that a false alarm could well provoke a catastrophe.

7. Communications

The aggregate of means of communication which ensure the so-called connectivity between the NCA and nuclear forces is called the World Wide Military Command and Control System (WWMCCS). Pronounced Wimex, it's little more than a name, in the sense that it only partially integrates various independently developed and operated C3 systems. Communications systems include telephone lines, underwater cables, radio systems, satellites, etc.

The Pentagon makes us (and pays for it) of ATT's and GTE's commercial telephone network. Only a part of telephone communications are coded.

"Users of SATCOM (Satellite Communications) fall into three categories: high-data-rate peacetime users, including intelligence and diplomatic terminals; tactical forces needing moderate data rates but worldwide coverage, small mobile terminals and resilience to disruption; and nuclear forces and their commanders, needing low data rates, but performance under severe stress. Today's US Defense Satellite Communications System (DSCS), pronounced discus), Fleet Satellite Communications (FleetSatCom), and Air Force Satellite Communications (AFSATCOM) systems correspond roughly to this tripartite division" (24).

Discus and FleetSatCom make use of four to six satellites each, all in geostationary orbits. AFSATCOM, on the other hand, has a good number (around twenty-five) of transponders onboard other satellites, including those already mentioned, and others with a wide range of orbits. According to Ashton Carter, "The array of COMSATs (Communication Satellites) available to the US military is completed by a few Allied Systems, various experimental communications spacecraft, and out-of-service but still partially usable satellites. Services of COMSATs owned by civilian companies and other nations might, in some circumstances, be made available to the military" (25). Most communications satellites transmit in UHF (ultra-high frequencies) and SHF (super-high frequencies), while EHF (extremely-high frequency) systems are being developed. In general, the higher the frequency, the greater the following advantages: more data transmission capacity per unit time; shorter transmitting antenna length; greater resistance to jamm

ing; less distortion in passing through the ionosphere disturbed by nuclear explosions. Instead, the Survivable Low Frequency Communications System (SLFCS) operates at low frequencies. It is a worldwide VLF/LF (very low frequencies/low frequencies) transmitter network with two high power sites. Receivers are installed at headquarters, and command posts, on submarines, LCCs and bombers (via land sites). Low frequencies are practically the only means of communicating with submerged submarines.

The SLFCS, together with the airborne command posts (Kneecap, Looking Glass, Tacamo, etc), the ERCS, AFSATCOM and FleetSatCom makes up the Minimum Essential Emergency Communication Network (MEECN). The latter, "according to the Department of Defense, 'comprises a system that is intended to survive and to provide those links essential for sending the Emergency Action Message' in order to 'exercise deliberate and precise control of strategic nuclear options for the SIOP execution and termination'" (26). AFSATCOM alone has approximately nine hundred terminals at almost all nuclear force control centres and main command posts.

8. The Moscow-Washington Hotline

A means of communicating with the adversary somehow seems pertinent to C3I, be it in case of crisis, or in case of war termination. Aside from ordinary diplomatic channels, all that the two superpowers have is the Hotline.

The Direct Communication Link (DCL), the official name for the Hotline, was established by a Memorandum of Understanding signed by the US and the USSR in Geneva on 20 June 1963. Today, it is neither a line, nor hot, but in those days it consisted of a teletype machine at both ends (Washington and Moscow) of a telegraph cable spanning the distance via London-Copenhagen-Stockhoklm-Helsinki, plus a radio circuit via Tangiers.

On 30 September 1971, the USA and the USSR signed an agreement in Washington concerning improvement of the DCL. Two circuits were set up for message transmission, both via satellite (one Russian, the other American); the radio circuit via Tangiers was eliminated and the cable kept as a back-up. Furthermore, the agreement provided for an increase in the number of terminals in each country. This, it seems, at least as far as the Americans are concerned, has not yet been done.

The last DCL upgrade is quite recent: it dates back to 17 July 1984. Despite the month, relations between the two country were icy and the Geneva negotiations interrupted. This was reflected in the diplomatic form--an exchange of notes--which, at the Soviets' insistence, was hardly mentioned in the press. The Soviet notes in some way emblematic of Moscow's reluctance to talk about arms control at the time: "The Embassy of the Union of Soviet Socialist Republics acknowledges the receipt of the Department of State's Note of 17 July, 1984, which reads as follows: (quotation of the entire text of the US note). The Embassy of the USSR states that the Soviet side agrees to the proposal contained in the note of the Department of State. Therefore, that note, together with this reply, shall constitute an agreement, effective on the date of the Embassy's reply" (27).

The American note establishes what the latest improvements are. Via the INTELSTAT and STATSIONAR satellites, Moscow and Washington should be able to exchange facsimile communications. The agreement also provides for a series of devices for coding transmissions - devices supplied by the Americans to the Soviets in a classical transfer of technology. The facsimile system does away with the filter of the typist, who is a potential source of additional error. Furthermore, it allows for transmission of drawings, graphs, maps, etc. It does not eliminate the problem of translation, however. A report by US Defense Secretary Weinberger states that: "Translation is the slowest step in the direct communication process. A highly proficient language specialist can produce a full translation at the rate of one thousand words per hour and do a cursory review at the rate of six thousand words per hour" (28). Direct communication, verbally or by video, are excluded. The reasons, it seems, are as follows: emotional factors, s

uch as the tone of voice or facial expression, are to be eliminated in moments potentially laden with consequences. Both leaders must be able to consult with their respective staffs before making proposals or replying, which is incompatible with direct communication times.

9. The Evaluation of American Nuclear Strategy

It is now time to pause and try to understand what the American government expects of its nuclear C3I system. Referring specifically to this, the Defense Secretary of the Carter administration, Harold Brown, wrote: "The degree of flexibility required depends on how many different strategic options must be available and for how long. This in turn depends on the doctrine adopted for deterring and, if necessary, carrying out a nuclear war. In the simplest case, where the only response contemplated is all-out thermonuclear retaliation on military and urban industrial targets, no great sophistication is needed" (29).

The simplest case mentioned by Brown, given that it ever existed, was rejected in January 1961, when Kennedy, whose campaign platform included criticism of the MC-14/2's slight or total lack of flexibility gave it little credibility. This marked the beginning of a long search for flexibility, and therefore, of the proliferation of options and targets that characterise the evolution of American nuclear strategy in the last quarter of a century. Later, McNamara took quite a different stand, inventing, if not the term itself, at least the idea of Mutual Assured Destruction (MAD), a much less demanding conception of deterrence with regard to flexibility. It is quite probable that the original thrust of the first months of the Kennedy administration had much more lasting effects on SIOP than MAD.

All of McNamara's successors contributed to channelling American nuclear doctrine away from the simplest case referred to by Brown. The only exceptions are remarks made by two presidents, Nixon and Carter, to quite different ideas, respectively, that of "sufficiency" and "minimum deterrent". Developments remained on paper, however, and were soon belied by the course of the administrations themselves.

In a nutshell, the basic concepts underlying US nuclear strategy are:

a) credibility, that is, credible dissuasion must be able to respond to the level of violence and the politico-military targets chosen by the adversary. If these are limited, the response must be limited, too, since the threat of obliteration is not credible unless it is in response to an equally all-out threat;

b) denying the adversary's moves, not just threatening punishment, is, therefore, credible (denial vs punishment);

c) the capability of inflicting punishment in the form of assured destruction of a certain percentage of the population and the industrial capacity, must be kept in reserve;

d) if deterrence fails, the United States must try to limit damage and terminate hostilities as quickly as possible on favourable conditions;

e) deterrence does not give way suddenly, but degrades gracefully: during the course of the war, a margin of superiority is represented by forces kept in reserve, the use of which can be abstained from to suggest restraint to the adversary and invite him to do the same (intra-war deterrence);

f) the same margin of superiority makes it possible to increase violence, if necessary, or shift that responsibility onto the adversary (escalation dominance);

g) on the basis of all the foregoing principles, the main criterion is that of flexibility (flexible response), made concrete by the availability of a range of limited or selected nuclear options.

I believe it can be affirmed that these principles are behind the last three official American documents establishing guidelines for strategic doctrine: the National Security Decision Memorandum (NSDM) 242, signed by Nixon in January 1974; the Presidential Directive (PD) 59, signed by Carter in July, 1980; the National Security Decision Directive (NSDD) 13, signed by Reagan in October 1981. These are not public documents and are mostly referred to in other terms. For example, the first is known as the Schlesinger doctrine (after the Secretary of Defense of that time) or Limited Nuclear Options; the second is called the "countervailing strategy", the third has no particular label, but a series of press leaks published in the New York Times in May-June 1982 implied that it dealt with a doctrine concerning "how to prevail in a protracted nuclear war" (30). Each of these documents is, with minor retouches, a logical outgrowth of its predecessor. Therefore, we may as well take a look at the last.

Directives and memoranda are implemented in a document, the SIOP already mentioned, which offers the President a number of options. In the 1980 SIOP-5D, the options are divided into four categories: Major Attack Options, Selected Attack Options, Limited Attack Options and Regional Nuclear Options. "Most important, the SIOP always left the President with two special attack categories, one for pre-emptive attacks on the Soviet Union and another for the so-called Launch on Warning" (31). Potential targets in the Soviet Union can also be divided into four groups: nuclear forces, conventional military forces, military and political leadership, economic and industrial targets. None of these groups contain urban centres per se. The proliferation of options seems to imply a proliferation of potential targets. In SIOP-5D there are forty thousand. These are spread practically all over the globe: "There are, for example, thousands of targets in the Warsaw Pact nations, in China, in Cuba, in Vietnam, and even some targe

ts in unspecified 'allied and neutral territory'" (32). The number of targets and three categories of options short of Major Attack Options, attest to the Americans' desire to prepare for a rather protracted nuclear conflict. This brings us back to C3I, of which the degree of flexibility and sophistication depends, in Brown's words, on "how many different strategic options must be available and for how long".

Awareness that a flexible response is very demanding in terms of C3I dates back to the first criticism of massive retaliation. In December 1962, John McNaughton, then Assistant Secretary of Defense, claimed that American efforts to improve command and control "...are designed to assure that the United States forces can be used or not used, as the case may be, in a controlled and deliberate way, subject at all times to the direction of the highest civilian authority" (33).

While still at RAND, before becoming secretary, Schlesinger wrote that a new, credible doctrine "...requires a hyper-protected force for intra-war deterrence, with long endurance and excellent communications and control. It requires, in the counterforce stage, the ability to assess damage--and to reassign vehicles--thereby compounding the requirements for the command, control and communications system" (34).

As far as the Carter and Reagan administrations are concerned, it should be stated that both PD-59 and NSDD 13 were preceded by other documents--PD-58, PD-53 and NSDD-12--which all have to do with C3I. According to the New York Times, the current American administration is aiming at strategic forces and C3 systems capable of sustaining "...controlled nuclear counterattacks over a protracted period while maintaining a reserve of nuclear forces sufficient for trans-and post-attacks protection and coercion"; communication systems "...must provide the capability to execute ad hoc plans, even subsequent to repeated attacks...these systems should support the reconstitution and execution of strategic reserve forces, specifically full communications with our strategic submarines" (35). In other words, "the President (Reagan) considers a modern command system as one that can survive a nuclear attack and continue to function for another six months" (36).

10. Scenarios of Limited Nuclear War

It should be clear from the brief description in the preceding paragraphs that the American nuclear forces control system is both complex and sophisticated. Whether it is ad flexible as a flexible response requires is another matter. In the following pages we try to evaluate all the factors, including C3I, that can affect the implementation of American strategic doctrine.

In order to do that, let's imagine an attack, make up a scenario. It has to be a limited attack. The need for limited options to respond to limited attacks has been a guiding light in the evolution of American nuclear strategy, and has, in turn, called for a long-lasting and resistant C3I, as we have seen.

We could, therefore, hypothesise a Soviet attack by a handful of warheads on some of the less densely populated areas of American territory, avoiding, either by chance or on purpose, the essential ganglia of the control system. The American president could respond to such an attack with the most limited of the SIOP options, presumably a response similar to the offense received. In this manner, perhaps, an exchange could continue for the six months the Reagan administration plans to prepare for. This scenario is so absurd that we should be ashamed of even discussing it. What advantage could the Soviets possibly gain from a "war" like this? And what would be the Americans' advantage in limiting retaliations. The author does not believe that these questions can be answered.

Taking into consideration the fact that available information about Soviet doctrine would seem to rule out any "opening moves" of the kind just mentioned, a scenario of this kind has little weight on literature. But little does not mean none.

"If the Soviet Union were for some reason to launch an attack that included fifty nuclear weapons and killed one million Americans," wrote Harold Brown, "would adherents of such a doctrine (in the context, Brown is referring to MAD) advise the President to launch an all-out attack that would kill one hundred million Soviets and bring a Soviet response that would kill an equal number of Americans?" (37). The key here seems to be those words "for some reason". For some reason, the Soviets, or somebody else on their behalf, could do almost anything, no matter how absurd. But the rationality of the counterpart has to be assumed to some extent, since a "rational" nuclear doctrine cannot be expected to dissuade irrational actions.

Another equally strange scenario, but very successful until a few years ago, is that of a counterforce attack against ICBM sites. Should response to such a limited attack be equally limited? "In the case of a Soviet attack against the 1,054 ICBM silos, estimates of collateral fatalities range from eight hundred thousand to fifty million" (38). The lower figure is the result of a Pentagon study presented to Congress in September 1974. Assumptions were very optimistic and criticism was so overwhelming that a year later the Defense Department went over its calculation and concluded that the deaths would total 18.3 million. The higher figure stems from a study carried out by the Arms Control and Disarmament Agency (ACDA).

In any case, a comprehensive Soviet counterforce attack would not be limited to ICBM sites. It would include bomber and submarine bases, command and control centres and nuclear weapons deposits. The aforementioned 1975 Department of Defense study estimated the number of deaths to be 21.7 million, whereas a more recent study, carried out by three Princeton University researchers set the number of victims due to blast and fallout between thirteen and thirty-four million, without counting medium and long-term effects (39).

It is true that each person is free to establish his/her own idea of "limited", but it seems reasonable to assume that in a country which has lost a total of one million, two hundred thousand men in all the wars fought in the last two centuries, with no civilian victims except in the Civil War, the figures indicated above can hardly be interpreted as having that meaning. On the other hand, in his annual report to Congress in 1979, Harold Brown wrote: "I myself continue to doubt that a Soviet attack on our strategic forces whose collateral damage could appropriately be responded to without including some urban-industrial targets would immediately do away with any claim to maintaining a limited nuclear war.

These considerations hold true even more if parts are reversed: "The collateral effects of US counterforce attack against the Soviet Union are likely to be somewhat greater than those of the Soviet counterforce attack...About half of the twenty-six Soviet ICBM fields are located West of the Ural Mountains and several of them are near some of the most densely populated areas of the USSR. The impact of this basing is exacerbated by Soviet wind patterns" (41).

Going back to the hypothesis of a Soviet attack, it should be pointed out that that country's doctrine encourages the most comprehensive of possible comprehensive counterforce attacks. Albeit on the basis of different assumptions and with a different terminology (42), the Soviets also consider deterrence the primary objective of their strategic policy. Nevertheless, should deterrence fail, the USSR intends to use its nuclear forces immediately, massively and in a pre-emption mode, in order to degrade those of the adversary and limits its own losses. In their own way, then, the Soviets believe in damage limitation and counterforce attacks, too. But that is where similarities with the United States' concept end, since massive use is the opposite of the selective attacks so dear to Americans. In fact, "There is no serious consideration in the Soviet literature of such concepts as controlled escalation and limited nuclear war", and even more important with respect to the topic of this paper. "The US notion that

attacks should be withheld from command-and-control assets, at least to the extent that such assets are needed to control escalation and conduct negotiations, is not reflected in Soviet military doctrine" (43).

11. C3I and Limited Nuclear War

The first C3I problem in carrying out a limited and strictly centrally-controlled nuclear war would be caused by the remarkable vulnerability of main command posts, both fixed and airborne, to SLBM and/or ICBM attacks. Very probably, the airports where tanker planes are based would also be among the targets, thus greatly reducing the flight autonomy of aircraft like the Looking Glass.

We have already discussed other questions relative to NCA vulnerability and succession. An important corollary, however, concerns negotiations for termination of hostilities. It was mentioned that American doctrine aims at war termination on favourable conditions, if possible. Who would the attackers negotiate it? Who would be in control? An air force general or a rear-admiral? How could these negotiations take place, lacking the DCL? This problem is reciprocal: the four categories of SIOP-5D targets also include political and military leadership. If the United States were to implement one of these options, it is not quite clear who would be left to negotiate a cease fire with.

Even early warning functions can easily be jeopardised. Besides their intrinsic vulnerability--ADAT will be discussed separately--the three DSP satellites all depend on ground stations (see page 50) which are not and cannot be protected against the effects of nuclear or conventional explosions or simple acts of sabotage. Land-based radar systems are equally vulnerable. Not only that, apart from the three in BMEWS, all the others are located at bomber, ICBM or tanker bases, thus simplifying Soviet counterforce attack calculations.

It might be said that an attack on early warning systems is the same as, or a form of, early warning itself. Yet, it would cancel the already poor capability of scaling a response to the size and target of the adversary's attack, which is a cornerstone of controlled escalation.

The capability of scaling the response is considered poor because attack assessment seems rather uncertain. It is the responsibility of NORAD, whose headquarters, although buried under Mount Cheyenne in Colorado, are considered vulnerable. If NORAD stops functioning or is no longer able to communicate with the other command posts, the latter will continue to receive data from the early warning sensors but won't be able to process it. Even under ideal conditions, however, two important drawbacks prevent accurate characterisation of an attack. The first is the limitation of the sensors: should an attack involve more than a handful of missiles, neither DPS satellites nor radar are able to distinguish the warheads from the delivery systems, determine their number and predict the points of impact (i.e. the targets). Only PARCS has a limited warhead tracking capability, but it cannot establish targets with certainty either. Furthermore, PARCS has a limited range of action: attacking ICBMs would already have covere

d two-thirds of their trajectory and the time left in which to make use of the limited predictions of radar would be around ten minutes. Finally, it seems that NORAD is poorly equipped for data processing, no matter what the quantity. The cause can be traced back to a decision taken by the Pentagon in the seventies--despite NORAD's protests--to acquire a Honeywell series 6000 computer which operates by means of batch processing, an extremely slow process. Attempts have been made to improve the situation, but "the net result of NORAD's computer development effort has been to create for the 1980s a clumsy computer system that, when it works, just about matches the capabilities of the 1960s equipment that was supposed to be replaced" (44).

As can be seen, an ineluctable problem of C3I is time. Even assuming that in the near future NORAD will acquire a more sophisticated computer and that the various early warning sensors are able to provide all data needed to characterise an attack; even assuming that, there is no way to reduce politico-military decision-making times. Les than ten minutes can elapse from the time an SLBM breaks the water to the moment of target impact; the corresponding time interval from when an ICBM leaves its silo is around half an hour. In this time, the NCA must meet, assimilate the psychologically devastating information coming in, "rationally" choose the most suitable of SIOP options and order its execution. It seems probable that under those conditions, the choice would boil down to the alternative between responding massively and not responding at all, realising that in the latter case, retaliation would be decentralised to military men further down the chain of command.

The time factor affects the practicability of one SIOP option: ICBM launch on warning. In other words, "Too many things have to happen within twenty-five minutes or less: tactical warning would have to be processed through the chain of command; the President would have to hear it, believe it and authorise firing; the authorisation would have to be transmitted back and confirmed to missile crews, who would have to believe it and then turn their keys. On paper, this can all be accomplished within the time available; in the heat of crisis, with terrified people facing apocalypse, such a smooth-flowing, rapid sequence is scarcely conceivable" (45).

All told, the vulnerability of early warning systems, their low attack assessment capabilities and the short time for politico-military decision-making are all factors that cast serious doubts on the feasibility of a controlled response. As a consequence, the assumptions behind today's American strategic doctrine seem to be rather unrealistic from the outset.

The communications systems--forgetting about satellites for the moment--have in common that kind of vulnerability to the thermal and dynamic effects of nuclear explosions, to the effects of conventional explosions and to acts of sabotage, that is typical of large fixed installations, often equipped with large antennae or equally exposed apparatus.

In some cases, there's no need even to physically destroy the receiving or transmitting facility. All that has to be done is make the contents of the communications unintelligible through jamming. Some frequencies and some equipment are more resistant to electronic countermeasures than others. In general, however, "Soviet military forces have a formidable array of jamming devices, and electronic warfare units with radio Direction-Finding (DF) and jamming equipment are involved in most Soviet military operations" (46).

Of the many effects of nuclear explosions on communications, electromagnetic pulse (EMP) deserves separate mention. The intensity of EMP is measured in volts per meter (V/M) and, in general, it is directly proportional to the power of the warhead and the height of the explosion. Many of the elements of control systems are the best conductors of EMP: antennae, electric wires, telephone cables, support towers, rails, aluminium aircraft cockpits. Pulse effects vary from temporary failure of equipment (jumps in the power absorbed by an electronic unit can, for example, stop it from operating, because of the safety measures that are usually incorporated in it) to permanent failure of a component such as a fuse or a transistor. The real problem, however, seems to be represented by the fact that "A single explosion at two hundred miles above the centre of continental United States would generate EMP over almost the whole country as well as parts of Canada and Mexico. Very few weapons--certainly less than five--coul

d apparently blanket the entire United States with as much as 50,000-100,000 V/M" (47). Various measures for hardening against EMP have been taken and they refer to almost all elements of America C3I, but it is rather doubtful whether the entire system can somehow be made invulnerable to pulse.

12. Strategic Submarines (SSBNs)

To any observer of American strategic policy it is clear that ballistic missile submarines are the key to it. That is, they exemplify the inconsistencies that many are attempting to point out in Washington's strategic choices.

It is known that submarines on patrol--and half the American SSBNs are in that state at any given time, as compared to twenty percent of Soviet craft--are practically invulnerable, due also--again in the case of the Americans--to the scarse capabilities of Soviet anti-submarine warfare (ASW). On the other hand, for reasons which we are about to see, SLBMs are the kind of nuclear force least suitable for implementation of the current doctrine; a doctrine based on the ability to control nuclear war, on placing limitations on its own use of force, on non-spasmodic attacks limited in intensity, purpose and dimensions, on the ability to communicate rationally and on a willingness to settle, rather than to resort to brutal revenge.

Nevertheless, faced with a trade-off between vulnerability and the nuclear weapon system characteristics best suited to doctrinal requirements (accuracy, retargeting capability, communications ease, attack modulation, etc), those in charge of American strategic planning have always opted for the former. In fact, more than half of all American warheads carried by strategic vehicles are on SLBMs. It should be pointed out, however, that debate on this subject is often distorted by an overestimation of the vulnerability of ICBMs and bombers.

As for the ICBMs, the accuracy of SLBMs is increasing with each successive missile generation. The Trident D-5, soon to enter into service in the American arsenal, will be the first SLBM officially acknowledged to have counterforce capability. The lowering of CEP is, however, only partially due to progress made in the missile's guidance system. In the case of SLBMs, a key role is played by systems having nothing to do with either the delivery system or the submarine, namely, navigation satellites. In order to hit a target which can't be seen, one has to know one's own position, and know it precisely, because an error is reflected onto the target. That precision is ensured by satellites.

In peacetime, obviously, there is no problem. In war, things are slightly different. In the first place, the submarine may not always be in a position covered by the satellite's orbit. Secondly, communications, be they direct or via ground station, are always at high frequencies (VHF and UHF) and in order to receive them, the unit has to raise its antenna for a few minutes, with the risk of revealing its position. Thirdly, both the satellite and the ground station are vulnerable. If, as seems probable in a protracted nuclear exchange, the umbilical cord with navigation satellites is cut off, the possibility of carrying out surgical strikes with SLBMs is almost eliminated.

The last generation of American SSBNs, the Ohio class, has twenty-four launching tubes for as many missiles. Each missile carried a number of warheads, eight in the case of the Trident C-4 and ten on the new Trident D-5. Thus, with the C-4, the total number of warheads per submarine amounts to 192, for the D-5, 240. It is presumed that a submarine's missiles would all be fired together--compatibly with the operation's technical times--because once one missile has been launched, the platform can be located and therefore becomes vulnerable. But it is hard to believe that the Soviets would consider an attack involving 192-240 warheads as limited.

The surest communications with submarines are at low frequencies. The lower the frequency, the better the signal penetrates the water, therefore, the submarine does not have to expose itself. It is superfluous to add that land-based stations transmitting in VLF are all extremely vulnerable.

Airborne command posts like Looking Glass and the Navy's tacamo can also transmit in VLF, but it is not an easy operation. Besides the already mentioned vulnerability of these aircraft and their short flight autonomy, the antenna needed to transmit in VLF constitutes a special problem. It is miles long and weighs a ton. "Unfortunately, it is hard to fly the plane with such a heavy antenna trailing behind it and to keep the wire from oscillating. Because of the considerable drag the antenna creates, even when it is only partially unreeled, the plane must keep to straight and level flight, or to very gradual turns, in order to keep the antenna stable. If it does start to whiplash, it can pull itself loose or force the crew to cut it (the co-pilot and the antenna operator have devices for doing this). The plane does not carry a spare" (48).

The last form of communication with SSBNs is via the ERCS. Mounted on and deployed with Minuteman missiles, the system is likely to be one of the victims of a Soviet counterforce attack. Should launching be achieved, however, the ERCS transmits in VHF, thus bringing back to the problems already seen. Furthermore, the kind of message delivered by this kind of system, the EAM, is a simple go code, falling somewhat short of the requisites of controlled escalation.

Recapitulating, the SSBNs have a very limited set of targets and little retargeting capability. If they receive a message, it's a simple launch order, which they cannot even confirm having received or question without exposing themselves irreparably. If no message arrives for a prolonged period of time, this could be the same as a launch order because it is probable that the crew would deduce that there has been an attack on the United States. As there is no PAL, as we have seen, if the crew decides to launch, it can do so. If one missile is launched, all are launched.

13. Satellites and Anti-satellites

Satellites important for C3I can be broken down into four groups on the basis of the missions they are designed for: communication, reconnaissance and surveillance, navigation and meteorological. Only the fourth category has not yet been discussed. All that needs to be known, though, is that specific atmospheric conditions can affect the performance of certain weapons systems - from bombers to ballistic missiles. Therefore, it is useful for the military to have up-to-date data available.

A satellite which, although intact, cannot carry out its mission, is useless. Some means of disturbing or preventing staellites' operation without destroying them, have already been discussed: attack on ground stations, jamming. Another way is deception: an example is the coding of telemetric emissions during ballistic missile flight tests, deceiving intelligence satellites and enraging the Standing Consultive Commission (SCC), the body in which Soviets and Americans discuss cases of apparent SALT violations. Another way is by "blinding" an early warning satellite, that is, saturating its infrared sensors by means of a massive heat source. Finally, nuclear explosions above or at the limits of the atmosphere can produce, besides EMP, direct damage to satellites situated in lower orbits.

There are also numerous means, both existing and possible, for attacking and destroying satellites. Anti-ballistic missiles (ABMs) can also be used for anti-satellite (ASAT) purposes. The USSR has an operative one, the Galosh, while the Americans should have the Spartan in their arsenal. Moreover, the Americans have experimented with a non-nuclear ballistic missile interceptor called Homing Overlay Experiment (HOE). ICBMs and SLBMs can be programmed to make their warheads explode at the apogee of their trajectory at an altitude of approximately 1400 Km, with results similar to those obtained using an ABM nuclear missile.

Land-based lasers can, perhaps, now already damage low orbiting satellites, while the possibility of space mines also exists - exploding satellites already in orbit near quarries at the right time. It's rather improbable that there are already mines in space, since the parallel movement of an unidentified satellite to another would hardly go unnoticed by the owner of the latter.

The last kind of threat is ASAT, that is, weapons systems developed specifically for anti-satellite warfare. The USSR and the US have one ASAT each.

The Soviet system operates on a principle similar to that of mines: it must first be put into orbit and then maneuvered into proximity to the quarry. At that point, ASAT explodes and its fragments will supposedly hit and incapacitate the target. Between 1968 and 1982, the USSR carried out twenty tests of its ASAT: the most generous estimates indicate that thirteen were a success. Of the seven failures, in six cases the interceptor had an infrared guidance system, as compared to the radar system in the other cases. The system was tested at altitudes ranging from 200 to 1000 km. Potential victims include American photo-reconnaissance satellites and some navigation and meteorological satellites. They do not include communication and early warning satellites nor the eighteen Navstar navigation and four GOES meteorological satellites.

The American system, on the other hand, consists of a two-stage missile. The first stage is the AGM-69 missile and the second, the Altair rocket called Thiokol. The latter carries a warhead called the Miniature Homing Vehicle, weighing 1200 kg and equipped with an infrared guidance system. The warhead does not explode, but destroys the quarry on impact. The entire system is five and a half meters long and 50 cm wide. It's mounted on a F-15 fighter plan and launched at an altitude of 10 to 15 km. Adding the range of the two stages to this, it is presumed that the American ASAT can hit a target at an altitude of up to 475 km, thus posing a threat to Soviet photo-reconnaissance, ocean surveillance and, perhaps, some communications satellites. I say perhaps, because the latter are on very elliptical orbits called Molnyia orbits, from the name of the satellites themselves. While the apogee is 40,000 km away and clearly out of reach of any present day ASAT, the perigee is at only 440 km. However, at this altitude,

the satellite is also travelling at maximum velocity, making the ASAT's mission a good deal more complicated. Theoretically, an American ASAT should be able to hit a satellite at the perigee of a Molnyia orbit, since it works on the basis of crossing the orbit. This is not true of the Soviet weapon. As it is co-orbital, it may well reach the perigee, but not be able to develop the velocity needed to approach the quarry. Soviet early warning satellites are also on a Molnyia orbit with the perigee at 688 km. To make them vulnerable, the Americans should increase the range of their ASAT - something they are already working on.

Another advantage of the American system is that the launching platform, the F-15, is very flexible. Going back to the example of Soviet satellites in Molnyia orbit, their perigee is over the southern hemisphere, where it's always possible to find an air base for an ASAT-carrying F-15. The Soviet system, on the other hand, is much less flexible. To illustrate this point, Ashton Carter imagined an attack on American TRANSIT navigation satellites that are on a polar orbit at an altitude of 1100 km. "The Soviet ASAT", writes Carter, "has been tested at this altitude but never in polar orbit; we will assume, however, that polar intercepts entail no extra difficulties. The give TRANSIT satellites are arranged in five orbital planes about 36 degrees apart. The Soviet ASAT must wait until the earth's rotation brings the Tyuratam launch site underneath the orbit of one of the target satellites. This happens about every two and a half hours. Even if Soviet ASATs could be launched at this rate, it would take twelve ho

urs for Tyuratam to pass beneath all five TRANSIT planes. During this twelve-hour attack on its navigation satellites, the US would surely become aware that attack was in progress. How useful is such capability to the Soviets?" (49).

Talks about banning anti-satellite weapons are at a standstill. The American administration seems to have little or no interest in the matter. Congress, instead, insists on a moratorium on ASAT testing (a moratorium that the Soviets have maintained since 1982) but has not been successful in convincing the executive. The last test of the American system dates back to August 1985.

American C3I depends more on the functioning of a limited number of satellites than does Soviet C3I. "Thus, where does the net US advantage lie - in allowing satellites of both sides to operate under a rule of law, or in fostering an era of vulnerability for all satellites?" (5=. The answer seems obvious, but evidently it's not. In the first place, probably observing the adversary's present capabilities, the Americans feel able to win the ASAT race, or more in general, the race for militarization of space. Furthermore, the US military has specific concerns. It seems that "The official rationale for restarting a US ASAT programme was the need to disable Soviet ocean surveillance satellites" (51), in that the latter can serve to direct Soviet fire against US Navy carrier battle groups. Here, too, then, other considerations have priority over the need to ensure the means for implementation of American strategic doctrine.

14. New US C3I Programs and Limited Nuclear War

What are the practical implications of the vulnerability of the American strategic C3I system?

"It would require only about 50-100 warheads", wrote Desmond Ball, "to destroy the fixed facilities of the national command system or to effectively impair the communications links between the National Command Authorities and the strategic forces" (52).

On the other hand, according to Bruce Blair, "As of 1985 Soviet strategic forces could quickly overwhelm virtually all ground-based C3I...There are no more than 400 primary and secondary US targets, including the 100 Minuteman launch control centers. With 7,000 deliverable weapons in their strategic arsenal, Soviet planners can easily commit two warheads to every US C3I target" (53).

Many of the current modernization programs will not substantially change the situation just described. For example, even if they will soon be able to make use of phased array technology, the three BMEWs radar--in Greenland, Alaska, and Great Britain--will always remain large, fixed, vulnerable installations. The same is true of the other two PAVE PAWS under construction - one at the Goodfellow air base in Texas and the other at Robins air base in Georgia.

The Navy's Tacamo are to be transferred to E-6As (B-707s) which will give them a theoretical flight autonomy similar to that of Looking Glass, but certainly won't eliminate all the other problems of communications with submarines that we have seen. Furthermore, the E-6As will be fitted out with the equipment taken from the C-130s, no more and no less.

The Distant Early Warning System for warning against bombers will be up-dated with thirteen new AN/FPS-117 long-range radar, plus 39 short-range radar in Canada. Many people wonder about the expediency of such an effort: the Soviets have only a small number of bombers and the Pentagon has various other means for detecting an attack - from reconnaissance to ELNIT and COMINT satellites.

Nevertheless, yet another radar program, of the Over the Horizon/Backscatter (OTH/B) type is being started. The system operates at long ranges--over the horizon, as the name indicates--with the radar signal bouncing off the ionosphere. This system can also detect low-flying objects. The OTH/B is ideal for bomber warning, but it can't be installed in the north--where the shortest air route between the United States and the USSR lies--because the Arctic region is full of electromagnetic anomalies and the ionosphere is unstable. When the various OTH/B radar sites will be completed within the next decade, they will cover all possible attack routes against continental United States, "...except...the direction from which Soviet bombers are most likely to come" (54). At worst, they'll just offer the Soviets a handful more soft targets.

The real threat that the OTH/B are supposed to deal with, however, is cruise missile attack. It's not an immediate, but certainly foreseeable threat: as usual, the Soviets are making every effort to catch up. Many comments have minimized the threat posed by cruise missiles, claiming that their relative slowness makes it possible for air defense to detect and neutralize them. In reality, their small dimensions and the low altitudes at which they fly make that rather difficult, if not impossible. The military on both sides know this and have taken necessary precautions. Some of them are fuelling the controversy on respect of agreements.

Construction of a large phased array radar in central Siberia, at Krasnojarsk, is the most incontrovertible case of Soviet violation take up by the Americans. In Art. VI of the ABM Treaty, both parties "agreed not to deploy in the future, radars for early warning of strategic ballistic missile attack except at locations along the periphery of their national territory and oriented outward". Few have noted, however, that the United States is committing the same violation, presumably for the same reasons, with its two new PAVE PAWS in Georgia and Texas. "The former will be 260 km (160 miles) from the Atlantic Coast, the latter will be 220 km (140 miles) from the Mexican border and twice that far from the Gulf of Mexico. If these radars has been built closer to the coast, they too would have been vulnerable to a sea-launched cruise missile precursor attack of the kind that may have motivated construction of the radar near Krasnojarsk. Both new American radar installations could be considered in contravention of

the ABM Treaty not only because of their 'inland' locations but also because their combined field of view could cover as much as two-thirds of the continental US. On these grounds the USSR has officially charged the US with a violation of the treaty" (55).

This administration is setting up many other new strategic C3I programs, DSCS satellites are continually being up-dated: we are now at third generation, version B, "with improved anti-jamming capabilities, more redundancy and solid-state amplifiers replacing some traveling-wave tubes", according to the description in Aviation Week and Space Technology (56). Also for communications, the MILSTAR, a satellite destined to operate at orbits higher than geo-synchronous orbit--at an altitude of 100-120 km--is being developed. The MILSTAR will use very high frequencies - the lower portion of EHF and higher part of SHF. The purpose of some of these measures becomes clear when one takes into consideration that the higher the satellite, the safer it is from possible ASATs, and that the Soviets do not have EHF jamming devices. The network will be composed of three satellites. Critics of this decision maintain that it would have been better to have aimed at redundancy--for example with a large number of simpler satellite

s, or by proliferating UHF transponders on all kinds of satellites, as in AFSATCOM--rather than relying on a few ultra-sophisticated and expensive systems. At the moment, the US also has a problem of non-availability of delivery systems for putting all these satellites into orbit: on 28 January, the Challenger disaster stopped the Shuttle program, probably until 1988. On 18 April, the Titan rocket exploded, causing another delay of several months in launches. Air Force Secretary Edward Aldridge recently announced that when the Shuttle takes to the air again, "we'll be more than twenty Department of Defense payloads behind" (57). In the meantime, it's probable that a program will be approved for a new rocket, similar in size to the European Ariane, and ready by the end of the decade.

Two other programs still have to be mentioned: the Ground Wave Emergency Network (GWEN) and the Mobile Ground Terminals. The first will be an addition to the telephone network composed of a large number of receiving and transmitting stations called nodes. Each node will have many others within its range and automatic systems will take care of sending messages in the shortest and/or most practical way.

The second program consists of approximately four hundred vans fitted out as Discus terminals. As can be seen, in both programs, attempts are being made to decrease the vulnerability of certain systems by multiplying their numbers.

None of the present American C3I modernization efforts can, however, be interpreted as squaring the accounts of the country's strategic doctrine. Too many links will always be weak--the NCA, the early warning sensors, the satellite ground stations, etc.--to think that controlled exchange of nuclear warheads can go on for months. Without taking into account that the countermeasures race is on. As has always happened in the past, the Soviets will catch up, in this case with ASATs for higher orbits or with EHF jamming devices or with other, still unforeseeable systems.

"The notion of controlled nuclear war-fighting is essentially astrategic in that it tends to ignore a number of the realities that would necessarily attend any nuclear exchange", wrote Desmond Ball, adding that "there can really be no possibility of controlling a nuclear war... . The allocation of further resources to improving the survivability and endurance of the strategic command and control capabilities cannot substantially alter this situation" (58).

Bruce Blair's opinion is equally forthright: "Once deterrence fails, it fails completely; the rudimentary design and short endurance of our nuclear C3I system nullifies the whole conception of multiple, time-phased counterforce exchanges. The pursuit of a bargaining advantage by means of limited attack is purely intellectual construction that has little or no relevance to present circumstances" (59).

It's unusual to come across such a severe stance in specialized literature, especially in that bearing the imprimatur of the IISS of the Brookings Institute. On the other hand, there are good reasons for common sense to rebel here, such as the forty thousand SIOP-5D targets, or the idea of fighting a nuclear war for six months.

15. Crises, War and Nuclear Forces

There's a grey area that separates peace from war: it's called crisis. The ambiguity of the crisis situation is reflected in the impossibility of interpreting the events that determine it in an univocal way. A simple precaution may be interpreted by the adversary as a prelude to an attack; an action aimed at demonstrating firmness may be taken as a provocation; a conciliatory move may be seen as a sign of indecision.

The recurring nightmare of observers of contemporary security problems is that of a repetition of the specular mobilizations which in 1914 made the crisis following the assassination in Sarajevo precipitate into the First World War. The situation today, however, presents some distinct characteristics. The first and most important is the presence of nuclear weapons. This has two aspects: the extent of the destruction that these weapons can provoke is terrifying; the time needed to cause it is very small. Whereas it once took years to bring another state to its knees, "Nuclear weapons can do it quickly" (60). The second characteristic is the product of the gigantic C3I systems set up by the US and the USSR. Theoretically, they are able to take note in real time of the slightest perturbation in the balance of opposing military forces and take timely corresponding measures. "In certain respects, American and Soviet strategic forces have combined into a single gigantic nuclear system" (61).

A crisis between the two superpowers is bound to be experienced by the protagonists as a terrible threat, coupled with an equally terrible sense of urgency. The probable result is that any precautionary move involving nuclear forces--such as raising their alert status--would be interpreted as a provocation and imitated. The generation of nuclear forces would be accompanied by a slackening of negative controls and soon, the conviction would spread that it is only a matter of minutes before the other opens hostilities. At that point, one may as well limit damage as much as possible by aiming at the other's nuclear forces and C3I centers. All this, it will be said, corresponds to Soviet doctrine. But it's also one of the SIOP options--preemption--an option that the Americans have never denied possessing and one that the military probably take as being the most realistic in case of war.

The American and Soviet C3I systems are painfully inadequate for a long and controlled nuclear war. On the other hand, they are fine for preemption. For example, communicating a launch order to a submarine before an attack does not present particular difficulties. In the context of a preemptive attack option, which would have to be massive and counterforce, an SLBM like the Trident D-5 makes sense. After an attack, as SSBN, no matter what missile it carries, is not flexible enough for existing US doctrine.

This does not mean that the Americans or the Soviets are eager to attack first. It only means that, if a crisis between the two of them were to spiral, both would have incentives, means and plans for shooting first.

What has been described must be kept quite separate from a cold-blooded attack - which is much less probable. Strategic jargon distinguishes between the two with the words preemption and prevention.

There are a lot of anecdotes about the American military's propensity for preemption, some bordering on the macabre. In September 1957, Curtis LeMay, the SAC commander who supposedly inspired Stanley Kubrick - made the following declaration to the members of a government commission: "If I see that the Russians are amassing their planes for an attack, I'm going to knock the shit out of them before they take off the ground". The commissioners objected: "That's not national policy". And LeMay replied: "I don't care. It's my policy. That's what I'm going to do" (62).

On the other hand, with regard to prevention, in 1961, the Pentagon submitted to Kennedy a study according to which an American attack on the Soviet Union could have resulted in the neutralization of the adversary's nuclear arsenal - at the time the Soviets had four operative ICBMs. With a little bit of luck, American losses could have been limited to two or three million people. No one in the administration considered implementing the plan (63).

These two episodes illustrate the practical differences between prevention and preemption, and the different attitudes of the military and civilians. Among other things, one of the effects of crises on decision-making processes is also the progressive transfer of authority from politicians to the military.

16. American Nuclear Strategy: Is it a Bluff?

Why so much emphasis on a crisis scenario? Because it is the context in which realistic questions about nuclear weapons arise. If the same realism is applied to possible responses, then it may well be concluded that American ones rather resemble the Soviet ones. In both countries it is understood that the only mental state suited to the use of nuclear weapons is one of desperation: convinced of the other's imminent attack, I try to limit my damages by attacking first. Here damage limitation can be considered rational only if one overlooks the fact that there's no other choice (talk about options!) once the imminence of nuclear war has been ascertained. But if there's no other choice, how can it possibly be rational? In fact, no one can actually believe in limiting damage in an absolute sense; only fractions of either country's strategic arsenal--in the order of thousandths--can cause millions of deaths (64).

Why, then, does American strategic policy--or perhaps it would be better to call it declaratory policy--continue not to acknowledge the facts and pursue what, in the light of what we have discussed, seems like wild fancy?

One answer is that the search for options--what the US doctrine is all about--has a strong persuasive effect. It's understandable that any authority imagining him/herself called upon to use nuclear weapons, wants to illude him/herself that the outcome need not be catastrophic. Comforting illusions always do well.

But the answer that this author finds most convincing is another, and that is, that it's a kind of bluff. It's often said that a doctrine aimed at rationalizing and controlling nuclear war is dangerous because it can instill too much self-confidence in the party adopting it, thus lowering the nuclear threshold. It's probable that this is the impression the American government wants to give to its adversary.

This can be illustrated by the chicken game: two planes race towards each other at dizzying speeds. The pilot who turns away--chickens out--first, loses. It's obvious that no one would ever play this game without someone with known suicidal instincts. Ergo, to be a champion of the chicken game, one has to give the impression of having an extra reason for wanting to win the game. With reference to the nuclear balance, declaring belief in the controllability of the war is like having an extra reason for winning if necessary. Better yet, if backed up by some past examples of daring: from 1946 to 1973, the United States sent out political signals using nuclear weapons (alerting them) nineteen times. The Soviet Union only once (66). Furthermore, "The Soviet level of peacetime readiness is also significantly lower than that of the United States" (67).

In other words, the Americans are trying to convince their adversaries and themselves that they are still able to gain some political advantage by threatening use of nuclear weapons (coercion). Nevertheless, the fact that the last alert dates back to 1973 is significant: the margin is becoming narrower and narrower. We can only hope that no American president will ever take the doctrine very seriously, and that no Soviet leader will ever call the bluff.

17. Arms Control and Foreign Policy

The situation exemplified by the chicken game is often more elegantly called competition in risk taking. That it is still thought that some political advantage can be gained from a competition in taking risks with nuclear weapons is certainly unsettling. This is the reason behind certain appeals for moderation, such as that of John D. Steinbrunner, director of foreign studies of Brookings Institute: "...a full two-sided alert in a crisis must prudently be considered tantamount to war and should not be undertaken for reasons less powerful than those required to justify war itself. This principle requires strict discipline over impulses to use strategic forces to send signals of resolve" (68).

Others, instead, believe that the competition in risk taking should be regulated by bilateral agreements, with measures of arms control. In the United States, the first person to make proposals of the kind was Senator Henry Jackson in 1982. Two years after his death, the idea was taken up later by two of his colleagues, Sam Nunn and John Warner (69). In the two senators' opinion, the US and the USSR should establish "nuclear risk reduction centers" in respective capitals, operating twenty-four hours a day and in contact with each other and the highest political and military authorities. Alternatively, the proposal calls for a single center in neutral territory, manned with military and civilian personnel from the two countries.

The Reagan administration has taken up the proposal and passed it on, without much conviction, to the Soviets (70). For once, the President seems to be right; the protagonists of a really serious crisis would be the superpowers' highest authorities. One or more center with relatively secondary figures would inevitably be bypassed, if not looked upon with suspicion, by respective leaderships. Finally, bodies of the kind could even be counterproductive: too much confidence in their effectiveness as an emergency brake could easily lead to careless behaviour.

As can be seen, quite different arms control agreements are needed to stabilize the situation. Yet, no important actions in this direction are in sight, after almost six years of Reagan leadership. Quite to the contrary, SALT II--never ratified by the US--has been officially scrapped, and we are now, paradoxically, witnessing a Soviet "peace offensive" of some substance - just to think of the unilateral moratorium on nuclear tests, respected by the USSR for a year (August 1985-August 1986) and then extended another six months. In fact, no one can figure out how much--or worse, whether--Reagan is interested in an agreement with the Soviet Union. In the same way, it's still uncertain whether Reagan is convinced that the US and the USSR must co-exist. On the other hand, the only effective way of dispelling the spectre of serious crisis is by starting out from that conviction and building upon it a relationship of cooperation between the US and the USSR, between the East and the West.

So, in the end, it's fundamentally a foreign policy problem, a problem that cannot be solved by any discussion, no matter how sophisticated, on nuclear deterrence.

----

Glossary

ABM: antiballistic missile

ACDA: arms control and disarmament agency

AFSATCOM: air force satellite communications

ANMCC: alternate national military command center

ASAT: antisatellite

BMEWS: ballistic missile early warning system

C3I: command, control, communications and intelligence

COMINT: communications intelligence

DCL: direct communication link

DEFCON: defense condition

DEW: distant early warning (line)

DSCS: defense satellite communications system

DSP: defense support program

EAM: emergency action message

EHF: extremely high frequencies

ELINT: electronic intelligence

EMP: electromagnetic pulse

ERCS: emergency rocket communications system

FLTSATCOM: fleet satellite communications

GWEN: ground wave emergency network

HOE: homing overlay experiment

ICBM: intercontinental ballistic missile

LCC: launch control center

MAD: mutual assured destruction

MEECN: minimum essential emergency communications network

NCA: national command authorities

NEACP: national emergency airborne command post

NMCC: national military command center

NORAD: north american air defense command

NSDD: national security decision directive

NSDM: national security decision memorandum

OTH/B: over the horizon/backscatter

PACCS: post attack command control system

PAL: permissive action link

PARCS: perimeter acquisition radar control system

PD: presidential directive

SAC: strategic air command

SALT: strategic arms limitation talks

SCC: standing consultative commission

SHF: super high frequencies

SIGINT: signal intelligence

SIOP: single integrated operational plan

SLBM: submarine-launched ballistic missile

SLFCS: survivable low frequency communications system

SSBN: ballistic missile nuclear submarine

Tacamo: take charge and move out

UHF: ultra high frequencies

VHF: very high frequencies

VLF: very low frequencies

WWMCCS: world wide military command and control system

Notes

1. Lawrence Freedman, 'The Evolution of Nuclear Strategy', St. Martin's Press, New York, 1981, p. XV.

2. It's no coincidence that the term decapitation is used to indicate an attack aiming at eliminating a country's leadership.

3. Cf. United States Information Service, 'Daily Wireless File', 4 June 1986.

4. Cf. "Growth in Funding Yields Strategic Tactical Benefits", 'Aviation Week and Space Technology', 9 December 1985.

5. Former researcher at the Brookings Institute and author of the volume mentioned in the following footnote. Blair was also consultant to the Pentagon for a brief period of time. He was dismissed after having written a report on C3I for the Office of Technology Assessment of Congress that the Administration withdrew from circulation, Labelling it SIOP-ESI. Only the President, the Secretary of Defense, the Chairman of the Joint Chiefs of Staff and the Deputy Secretary of Defense can read such documents. Cf. "The Ultimate Secret: A Pentagon Report its Author Can't See", 'The Wall Street Journal', 19 February 1986.

6. Bruce G. Blair, 'Strategic Command and Control - Redefining the Nuclear Threat', The Brookings Institute, Washington, DC, 1985, p. 247.

7. Ibidem.

8. Desmond Ball, "Can Nuclear War be Controlled?", 'Adelphi Paper', No. 169, IISS, London, Autumn 1981, p. 38.

9. Richard K. Betts, "Surprise Attack and Preemption", in Graham T. Allison, Albert Carnesale, Jospeh d. Nye, Jr. (editors), 'Hawks, Doves and Owls - An Agenda for Avoiding Nuclear War', W.W.Norton&Company, New York and London, 1983, p. 205.

10. Quoted in Peter Pringle, William Arkin, SIOP - The Secret US Plan for Nuclear War, W.W.Norton&Company, New York and London, 1983, p. 205.

11. Paul Bracken, 'The Command and Control of Nuclear Forces', Yale University Press, New Haven and London, 1983, p. 202.

12. Pringle and Arkin, op. cit., p. 39.

13. Ibidem, p. 216.

14. Strobe Talbott, 'Deadly Gambits', Alfred A. Knopf, New York, 1984, pp. 273-4. SLBMs cannot be recalled; bombers and missile-carrying submarines have nuclear warheads; the number of Soviet missiles are assigned by the CIA.

15. Blair, op. cit., pp. 262-3.

16. Pringle and Arkin, op. cit., p. 217.

17 Daniel Ford, 'The Button - The Pentagon's Strategic Command and Control System', Simon and Schuster, New York, 1985, p. 137.

18. "It is also possible that a parallel, invulnerable ERCS system exists on the ballistic-missile submarines". Jonathan B. Tucker, "Strategic Command-and-Control vulnerabilities: Dangers and Remedies", 'Orbis', Winter 1983, p. 946.

19. There are three other airborne (on KC-135s) command posts for as many regional commands with nuclear responsibility: Silk Purse for the European Command, Blue Eagle for the Pacific Command and Scope Light for the Atlantic Command.

20. William Arkin and Richard Fieldhouse in 'SIPRI Yearbook 1984', p. 469, indicate that it takes four members of an SSBN crew "to validate launch orders and execute the attack". Furthermore, they maintain that bombers have PAL.

21. Bracken, op. cit., p. 5.

22. 'SIPRI Yearbook 1984', p. 474.

23. Ball, op. cit., p. 40.

24. Ashton B. Carter, "Satellite and Anti-Satellite: The Limits of the Possible", 'International Security', Spring 1986, p. 55.

25. Ibidem, p. 56.

26. Quoted in 'SIPRI Yearbook 1984', p. 478.

27. Text printed in 'Bollettino USPID', Year II, No. 1/2, March 1986, p. 214.

28. Ibidem, p. 207.

29. Harold Brown, "Strategic Forces and Deterrence", 'ACIS Working Paper', No. 42, Center for International and Strategic Affairs, University of California, Los Angeles, August 1983, p. 32.

30. The articles in the 'New York Times' are all written by Richard Halloran and appeared on 30 May, and 4 and 21 June of 1982. On the "Schlesinger Doctrine", see Lynn Etheridge Davis, "Limited Nuclear Options - Deterrence and the New American Doctrine", 'Adelphi Paper', No. 171, IISS, London, Winter 1975/6. "The Countervailing Strategy" is the title of an article published by Walter Slocombe (Deputy Undersecretary of Defense for Policy Planning during the Carter Administration) in 'International Security', Spring 1981.

31. Pringle and Arkin, op. cit., pp. 187-88.

32. Ibidem, p. 188.

33. Quoted in Davis, op. cit., p. 2.

34. Quoted in Freedman, op. cit., p. 378.

35. Quoted in Blair, op. cit., p. 28.

36. Ibidem, pp. 7-8.

37. Brown, op. cit., p. 34.

38. Ball, op. cit., p. 27.

39. Cf. William Daugherty, Barbara Levi and Frank von Hippel, "The Consequences of 'Limited' Nuclear Attacks on the United States", 'International Security', Spring 1986.

40. Quoted in Ball, op. cit., p. 29.

41. Ibidem, p. 28.

42. Cf. David Holloway, 'The Soviet Union and the Arms Race', Yale University Press, New Haven and London, 1983, pp. 31-35.

43. Ball, op. cit., p. 32.

44. Ford, op. cit., p. 82.

45. Richard Betts, 'Surprise Attack, Lessons for Defense Planning', quoted in Blair, op. cit., p. 235-6.

46. Ball, op. cit., p. 13.

47. Ibidem, p. 11.

48. Ford, op. cit., pp. 155-6.

49. Carter, op. cit., p. 80.

50. Ford, op. cit., p. 206. Emphasis in the original.

51. Deborah Shapley, "Strategic doctrine, the militarization and the 'semi-militarization' of space" in Bhupendra Yasani (ed.) "Space Weapons - The Arms Control Dilemma", SIPRI, Taylor & Francis, London and Philadelphia, 1984, p. 64. Also see, Carter, op. cit., p. 89.

52. Ball, op. cit., p. 35.

53. Blair, op. cit., p. 182.

54. Ford, op. cit., p. 212.

55. Miroslav Nincie, "Can the US Trust the USSR?", 'Scientific American', April 1986, p. 40.

56. 9 December 1985, p. 49.

57. Quoted in D. Sanger, "US Plans to Shift Military Satellites from Shuttle to New Midsized Rocket", 'International Herald Tribune', 35 June 1986.

58. Ball, op. cit., pp. 36-7.

59. Blair, op. cit., p. 5.

60. Thomas Shelling, 'Arms and Influence', quoted in Michael Mandelbaum, 'The Nuclear Question - The United States and Nuclear Weapons' 1946-1976, Cambridge University Press, Cambridge (Mass.), 1979, p. 3.

61. Bracken, op. cit., p. 59.

62. The episode is recounted in Fred Kaplan, 'The Wizards of Armageddon', Simon and Schuster, New York, 1983, p. 134.

63. Ibidem, pp. 298-301.

64. This is well known to anyone having had responsibility in the sector and constitutes a restraint, which should not be underestimated, against not only the use of nuclear weapons as such, but against reaching a crisis so serious as to bring their use into consideration. McGeorge Bundy, national security advisor to Kennedy wrote: "In the real world of real political leaders, whether here or in the Soviet Union, a decision that would bring even one hydrogen bomb on a city of one's own country would be recognized in advance as a catastrophic blunder; ten bombs on ten cities would be a disaster beyond history; and a hundred bombs on a hundred cities are unthinkable". "To Cap the Volcano", 'Foreign Affairs', October 1969, p. 10.

65. "American nuclear policy consists of four different components: employment policy, acquisition policy, declaratory policy and deployment policy...Declaratory policy gives guidance to American officials on what they say publicly about the employment and acquisition policies". Davis, op. cit., Footnote 2, p. 1.

66. Cf. Holloway, op. cit., p. 51.

67. Jonathan B. Tucker, op. cit., p. 951.

68. "Nuclear Decapitation", 'Foreign Policy', Winter 1981-82.

69. About the Nunn-Warner proposal, see 'Bollettino USPID', No. 1/2, March 1986, pp. 199-201. In general on the same question, see also William Langer Ury and Richard Smoke, 'Beyond the Hotline: Controlling a Nuclear Crisis', A Report to the United States Arms Control and Disarmament Agency, Washington, DC, 1984.

70. Cf. Michael Gordon, "US, Soviet Discuss Accidental Nuclear War", 'International Herald Tribune', 7 May 1986.