INTRODUCTION

No less than in other fields, fashion also plays a role in strategic studies. With regard to nuclear matters in particular, Command and Control or, said more elegantly, the management of nuclear operationsManaging Nuclear Operations (Washington D.C.: The Brookings Institution, 1987) is the title of the most comprehensive study on Command and Control published so far. The book is edited by Ashton B. Carter, John D. Steinbruner and Charles A. Zracket. Reviewing the book, Richard Ned Lebow heavily criticized its title's wording: "The infelicitous inclusion of the word 'managing' in the title seems to indicate a consensus among the authors that nuclear operations can be managed...Equally disturbing is the use of the word 'operations'. It is an euphemism for nuclear blackmail and nuclear war." See his "Operations Manual," Arms Control Today, December 1987. In this paper, however, I will also use the acronym C3 (Command, Control and Communications). Although it may be a far cry as elegance goes, concision is on it

s side.

is definitively à la mode. In a discipline 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,"Lawrence Freedman, The Evolution of Nuclear Strategy, (New York: St.Martin's Press, 1981), p. XV.

it is increasingly difficult to impress the reader with a fresh, new analysis. Yet the handful of studies on nuclear Command and Control published in the '80sDesmond Ball, "Can Nuclear War Be Controlled?," Adelphi Papers 169, (London: International Institute for Strategic Studies, Autumn 1981); Paul Bracken, The Command and Control of Nuclear Forces, (New Haven and London: Yale University Press, 1983); Bruce G. Blair, Strategic Command and Control - Redifining the Nuclear Threat, (Washington D.C.: The Brookings Institution, 1985); Daniel Ford, The Button - The Pentagon's Strategic Command and Control System, (New York: Simon and Schuster, 1985).

are doubtless among those which in recent years have succeeded in doing so.

Beyond fashion there is, however, substance. As many authors of C3 studies claim, for example, too much attention has been paid to SALT-like calculations of forces, and too much polemical effort has been spent in endless debates on nuclear strategy; whereas the ways in which a given strategy interacts with a given set of forces has been largely overlooked. Such interaction is clearly of crucial importance in determining if and how nuclear forces can perform the missions assigned to them in accordance with strategic doctrine; i.e. if and how authority can be maintained, orders executed and control exercised, following prearranged plans or ad hoc wartime contingencies.

Calling attention to nuclear Command and Control has however had the effect of reinvigorating the debate on nuclear strategy and forces - which is somewhat paradoxical given the aforementioned claim of those who pioneered C3 studies. On the whole, in fact, these studies have cast serious doubts on the tenets of the current U.S. nuclear strategy. In other words, the idea of controlling a nuclear war by means of limited options, intra-war deterrence, escalation dominance and flexible response would be highly unrealistic in the light of the existing and foreseeable Command and Control systems.

From a European point of view there are good reasons to be interested in nuclear C3 issues. First, NATO relies on the threat of general nuclear war as its ultimate deterrent and therefore any European member of the alliance has a clear stake in the credibility of the U.S. nuclear strategy. Second, while scenarios of U.S.-Soviet inter-continental nuclear exchanges are often discussed as if erupting from nothing, were they to occur they would most probably be preceded by nuclear exchanges precisely on the European continent.

Third and most important, however, a European may be tempted by the idea of following the example of what has been done at the strategic level and of using the C3 yardstick to gauge the credibility of MC 14/3, the NATO doctrine in force since 1967, better known as Flexible Response. This paper is, in many respects, the product of such a temptation. Its author doubts, though, that a final word on NATO's Flexible Response can ever be said, whatever the yardstick used to help in the judgement. Nonetheless, looking at the issue from a C3 perspective can highlight a number of choices and trade-offs, if not dilemmas, that are inherent both in the doctrine and in the theatre nuclear force posture of the alliance. And in their interactions as well.

French and British forces are treated here as outright independent - despite the differences between them as far as their relationship with the rest of the Alliance is concerned.French nuclear forces are under exclusive French control. British nuclear forces are assigned to SACEUR, "Except where Her Majesty's government may decide that supreme national interests are at stake." Joint U.S.-U.K. Communiqué, following meeting in the Bahamas, 18-21 December 1962, quoted in Duncan Campbell, The Unsinkable Aircraft Carrier, (London: Paladin, 1986), p. 110.

They are therefore excluded by this study. Quite apart from the fact that information about those forces' Command and Control systems is very limited,See, however: Shaun Gregory, The Command and Control of British Nuclear Weapons, (Peace Research Report No. 13, School of Peace Research, University of Bradford, December 1986); Lawrence Freedman, "British Nuclear Targeting" and David S. Yost, "French Nuclear Targeting," both in Desmond Ball and Jeffrey Richelson (eds), Strategic Nuclear Targeting, (Ithaca and London: Cornell University Press, 1986).

such exclusion should simplify an already overly complicated matter.

"Connectivity" is a term often used as a synonym of C3. It implies something that is in between, namely between the political authority and the forces. Political authority, in turn, is expected to be exercised in accordance with a given doctrine about the employment of the forces. This paper thus begins with a brief look, in the first two chapters, at both the NATO doctrine and the theatre nuclear armoury.

I. THE DOCTRINE: PLANNING NATO'S NUCLEAR USE

The concept of "forward defence" - i.e. the defence of allied territory must take place as far to the east as possible - has gone unchanged since the inception of NATO. With the notable exception of some proponents of alternative ideas, like 'territorial defence' or 'defensive defence,'Debate on these issues has blossomed in recent times. See, for example, the Bibliography edited by Bjorn Moller for the series NOD - Non-Offensive Defence, published in 1987 by the Centre of Peace and Conflict Research at the University of Copenhagen.

forward defence has also enjoyed a remarkable degree of consensus throughout the Alliance in its almost forty years of history. However, when NATO in 1956 adopted the doctrine of massive retaliation, or MC 14/2, the role of forward defence was mainly that of a so-called "trip-wire" for U.S. nuclear forces. Consequently, the chief means of forward defence - conventional forces - were not given particular prominence.

Things changed with MC 14/3, adopted by NATO in December 1967. The new doctrine called for a more balanced mix of conventional, theatre nuclear and strategic nuclear weapons. According to the Alliance's official description of its own doctrine: "The purpose of this balance of forces, while retaining the principle of forward defence, is to permit a flexible range of response combining two main capabilities: to meet any aggression at the level judged to be appropriate to defeat the attack, and to be prepared to escalate the level deliberately, maintaining firm political control, if defence at the level first selected is not effective."The North Atlantic Treaty Organization: Facts and Figures, (Brussels: NATO Information Service, 1984), p. 139.

This is not the proper place for a history of NATO's painful and rather long process of doctrinal change in the sixties.For an excellent treatment of this issue see David N. Schwartz, NATO's Nuclear Dilemmas (Washington D.C.: The Brookings Institution, 1983), chapter 6.

It is worth recalling, however, that the main reason that induced the Alliance to revise its strategy must be sought in the growth of Soviet nuclear retaliatory capability against both Europe and the U.S.

Despite well known European resistances, this led, first, to a shift in emphasis toward the notion of forward, conventional defence. In other words, NATO decided to be less hopeless about its prospects of countering the Warsaw Pact conventional threat on its terms.

The second major consequence was to give theatre nuclear weapons (TNW)In this study Theater Nuclear Weapons are those U.S. nuclear weapons deployed on the European territory for U.S. and NATO use. As can be seen in chapter II, naval nuclear weapons and strategic weapons are excluded, even though they may gravitate around Europe or be assigned to the NATO commander. Non-strategic weapons (range below 5500 km) deployed elsewhere (in the U.S., in South Korea etc.) are similarly excluded.

a central role in the allied deterrent. Being no longer risk-free to escalate to the point of massive nuclear retaliation, and still feeling insecure about its conventional capabilities, NATO sought in TNW those margins of superiority needed for the credibility of its deterrent posture. The prospect of escalation raised in MC 14/3 directly regards TNW. When NATO would consider its conventional defences "ineffective" and would resort to nuclear weapons is deliberately left unclear. This ambiguity is widely thought of as a deterrence 'multiplier'.

When Flexible Response was adopted, though, "a degree of ambiguity was also necessary in order to allow the American and European allies sufficient scope to interpret the strategy in accordance with their preoccupations and perspectives."J. Michael Legge, Theater Nuclear Weapons and the NATO Strategy of Flexible Response (Santa Monica CA: The Rand Corporation, 1983), p. 9.

Again such 'internal' ambiguity revolves around the role of TNW: "[for] the Europeans...the threat to use TNW represented the best way of 'coupling' the U.S. strategic deterrent to the defence of Europe, and [for] the Americans...it offered the best hope of preventing a major land battle in Europe from escalating to an all-out strategic exchange."Ibid. p. 10. David N. Schwartz pointed to the same ambiguities when he wrote of two versions of Flexible Response: "The version presented in Athens [at NATO Ministerial Meeting, 5 May 1962] by McNamara foresaw grave dangers in escalation, particularly in escalation over the nuclear threshold...The other version, explicitly supported by von Hassel [F.R.G.'s Defense minister in 1964] and others held that a stronger deterrent lay in the credibility of NATO's threat to escalate." Schwartz, (op. cit. in note 3), p. 177.

It is in this context that the role of theatre nuclear weapons in NATO doctrine should be assessed. It is indeed a pivotal role as testified, among other things, by the threefold growth of TNW deployed in Europe in the years between the first discussions on (1961) and the final adoption of Flexible Response. An arsenal, however, subsequently matched by a no less impressive Soviet growth in the same category of weapons. It is a fact of life that NATO lost whatever margin of superiority may have supported its capability, if not its willingness, to escalate. And it is not by chance, perhaps, that criticism of Flexible Response became stronger and stronger as time went by.

Adoption of Flexible Response did not mean in itself planning how to use TNW in support of the escalation option. This task was taken up by the Nuclear Planning Group (NPG),The NPG was founded in 1965 as one of the working groups of an ad hoc Special Committee on Nuclear Consultation. Since the establishment of the Nuclear Defense Affairs Committee (NDAC) in 1967, the NPG meets twice a year at the level of Defense Ministers. See Paul Buteux, The Politics of Nuclear Consultation in NATO 1965-80 (Cambridge, GB: Cambridge University Press, 1983), chapter 2. Full membership was first limited to the U.S., U.K., the F.R.G. and Italy. In 1979 membership was enlarged to the inclusion of any interested NATO country, that is all the allies but France and Iceland.

which first concentrated on the so-called "initial use". Not surprisingly, options for NATO's first use"'Initial use' and 'first use' are essentially synonymous...It would normally take the form of a discrete package of weapons covered by a single political release authorization and be limited to a specific time period; all subsequent authorization would be regarded as follow-on use." Legge, (op. cit. in note 5), p. 18, note 1.

of TNW ranged between the two extremes of "demonstrative use" and large-scale, massive use, the former being, in its simplest form, a single nuclear detonation over an inhabited area and with no direct bearing on the ongoing hostilities. Either extreme was, in the end, rejected. Demonstrative use apparently on the grounds that it may be perceived as a sign of indecisiveness and reluctance instead of resolve, thereby making at best no difference as to the outcome of the war.A very critical approach to demonstrative use is taken by William W. Kaufmann. See his "Nuclear Deterrence in Central Europe," in John D. Steinbruner, Leon V. Sigal (eds), Alliance Security: NATO and the No-First-Use Question (Washington D.C.: The Brookings Institution, 1983), pp. 36-7.

Rejection of demonstrative use, however, was never as thorough as was the case with large-scale use. This point was confirmed by Secretary of State Alexander Haig in November 1981, when he "revealed that NATO contingency plans still included the option 'to fire a nuclear weapon for demonstrative purposes.'"Desmond Ball, "Targeting for Strategic Deterrence," Adelphi Papers 185, (London: International Institute for Strategic Studies, Summer 1983), p. 17.

Work on initial use was completed as early as December 1969, when the NPG endorsed a document called "Provisional Political Guidelines for the Initial Defensive Tactical Use of Nuclear Weapons by NATO". Details of the document are not of public domain. According to one of the best accounts of the Alliance's nuclear planning, however, it seems that "NATO's objective would be essentially political and that initial use would therefore be very selective."Legge, (op. cit. in note 5), p. 20, (emphasis in the original).

Since in the same study "large scale" use is defined as "100 to 200 weapons rather than 10 to 20,"Ibid., p. 20, note 1.

the latter figure should be correctly taken as appropriate for a "very selective", initial use.

Apart from specific numbers, it seems that what the NPG had in mind was an initial use of limited (but of some) military significance and of great (but not exclusive) political meaning - the much debated signal of resolve. In other words it is a compromise, which neither solves, nor escapes the problems raised by the two extremes of demonstrative use and large scale use.

Subsequent studies in NATO concentrated on the no less thorny issue of follow-on use. Again, massive attacks were considered and rejected as incapable of granting the Alliance a military victory at an acceptable cost - first of all in terms of collateral damage on NATO territory. In the end it was decided that "Follow-on use should have the same purpose as initial use (to persuade the enemy to cease its aggression and withdraw), and the nature of the use should therefore still be selective and be designed to meet this political requirement."Ibid., p. 27.

To reach this conclusion took the NPG almost six years, that is from 1969 to 1975. In the following years the Group set out to study other problem areas, notably the composition of the TNW force. It can well be said, therefore, that the question of follow-on use was never seriously answered by the NPG. However, one wonders whether such a question could have any answer at all, apart from wishing the issue away by equating it to the situation NATO would face when considering first use. Too many variables, in fact, would have to be taken into account to plan any meaningful follow-on use: above all, the largely unknowable Soviet reaction to an unprecedented NATO decision to cross the nuclear threshold. There are indeed some clues as to how the U.S.S.R. thinks its TNW should be employed; they will be mentioned later on in this paper. Suffice here to say that, according to what it is known of Soviet doctrine in this regard, the prospects of protracted, selective nuclear exchanges on the European theatre are n

ot very bright.

As recently as October 1986, General Political Guidelines governing nuclear use were approved at the NPG meeting at Gleanagles, in Scotland. Not much is known about their actual content. It seems, though, that they cover follow-on use, the use of nuclear weapons at sea and the use of NATO Intermediate Nuclear Forces (INF) to strike Soviet territory.See Geoffrey Manners, "Major NATO nuclear review under way," Jane's Defence Weekly, September 27 1986. According to another source, "Elles tracent clairement le chemin depuis 'l'emploi initial' jusqu'à 'l'emploi succesif' et à la 'reponse nucléaire généralisée'." Moreover, "le territoire del l'U.R.S.S. n'est plus considéré comme un 'sanctuaire'." Nouvelles Atlantiques, October 24, 1986.

After the December 1987 U.S.-Soviet Treaty on the elimination of intermediate and shorter-range missiles, however, these guidelines may already be obsolete. Only a massive shift of emphasis from missiles to airplanes could partially make up for the elimination of Ground Launched Cruise Missiles (GLCM) and Pershing 2s, but clearly at the expenses of conventional capabilities that are arguably already overburdened by nuclear missions.

The new guidelines have been discussed in conjunction with the new theatre nuclear force structure emerging from various NATO decisions since 1979; and it is likely that this is what they reflect, more than any substantive, conceptual innovation. An innovation, moreover, that it is hard to conceive of, since the problems raised by NATO's nuclear use were basically the same in 1986 as in 1975.

Looking at how the NPG worked over the years it is worth noting, in passing, that the bulk of the studies were carried out by ad hoc committees or working groups formed almost invariably by British, German and American representatives. The only notable exception is a study on the employment of Atomic Demolition Munitions (ADM), which was carried out mainly by the Turks and the ItaliansSee Buteux, (op. cit. in note 7), p. 120 and Legge, (op. cit. in note 5), pp. 23-4.

- the ADMs are now due for withdrawal.

This is perfectly understandable if one considers: the relative importance of Germany and the Central Front in NATO defence; that more than half of the U.S. TNW stockpile in Europe is deployed in the F.R.G.; that Britain is a nuclear power; the particular structure of the NPG, with full members and rotational members, up until 1979.

Nonetheless, on a factual basis, one should not forget that nuclear consultation in NATO has been mainly a tripartite business. The role played by Italy, the Netherlands and Belgium in the INF deployment, as well as the opening of the NPG, are signs of change toward larger allied involvement. In March 1987, however, Bonn and London, confronted with Gorbachev's version of the zero-zero INF formula, rushed to consult Paris, before any other European capital.See James M. Markham, "Europe's Triangular Initiative," International Herald Tribune, April 3, 1987.

Political guidelines set forth by the NPG are subsequently translated into operational plans by the military. According to a U.S. Army Field Manual, "Advanced planning for nuclear strikes or counterstrokes is essential to timely employment."U.S. Department of the Army, FM 100-5 (Washington D.C.: GPO, July 1976), p. 10/7. This manual was superseded in August 1982 by a new edition, wherein nuclear operations are treated in a much shorter way.

The highest level for such planning is the Nuclear Activities Branch of the Supreme Headquarters, Allied Powers Europe (SHAPE). There the Nuclear Operations Plan (NOP) is developed under the responsibility of the Supreme Allied Commander Europe (SACEUR).

The basic partition of the NOP is between selective use and general nuclear response. Planning for the former category is done in so-called packages. "A package is a group of nuclear weapons of specific yields for employment in a specified area, within a limited timeframe to support a tactical contingency."Ibid. (emphasis in the original).

A "Hypothetical package", for example, could involve the use of 42 nuclear charges (2 ADMs, 10 nuclear missiles, 30 cannon delivered, 5 aircraft delivered) within a time frame of 10 hours and a time span of 180 minutes.Ibid., p. 10/8.

However, targeting sections at various levels, down to the Army division,See Daniel Charles, Nuclear Planning in NATO - Pitfalls of First Use (Cambridge MA: Ballinger Publishing Co., 1987), p. 95.

contribute with "subpackages" of targets to the plans for selective use. It is also important to bear in mind that, according to several sources, packages larger than the above mentioned "hypothetical" one are considered. "For example a 'corps package' is envisioned to consist of between 100 and 200 nuclear weapons 'depending upon threat, mission, terrain and population characteristics' in the area of employment."Ball (op. cit. in note 10), p. 16.

According to Catherine McArdle Kelleher, "The packages, as planned and as released, are of various sizes, with 50-100 weapons being a widely-circulated definition of the range of 'usual' packages.""Managing NATO's tactical nuclear operations," Survival, January/February 1988.

Although still included in the category of selective use, even larger use should be involved in the so-called "theatre nuclear warfare" option: "Theatre nuclear warfare is the ultimate level of nuclear warfare which SACEUR would be authorised to conduct, subject of course to Presidential authority."U.S. Security Issues in Europe: Burden Sharing and Offset, MBFR and Nuclear Weapons, September 1973. A Staff Report. Committee on Foreign Relations, U.S. Senate, 93rd Congress, (2 December 1973), p. 21.

On the other hand, the NATO general nuclear response, as the term indicates, involves the use of large numbers of nuclear weapons, especially of longer-range systems. It is likely to occur only in conjunction with the execution of the Single Integrated Operational Plan (SIOP), i.e. the plan for the employment of U.S. strategic systems. To this end targets in the general nuclear response are deconflicted with those of the SIOP's target list, thanks to the work of a group of NATO officers detached at the U.S. Joint Strategic Target Planning Staff (JSTPS) in Omaha, Nebraska.

Targets in the NOP include practically all that have some military significance: air fields, rail lines and rail yards, Soviet nuclear forces, Command and Control facilities, military headquarters, air defences, tank and troop concentrations, supply routes, bridges, dams, harbours, choke points etc. Some of these targets are in NATO or neutral/non-aligned countries.

II. THE FORCES

In 1960, at the end of the Eisenhower Administration, there were approximately 2500 U.S. TNW deployed in Europe. In 1968 they peaked at about 7200, to remain at that level till 1979. This number did not include warheads onboard U.S. vessels and U.S. strategic systems assigned to NATO command (SACLANT at first, then SACEUR) - Polaris nuclear ballistic missile submarines (SSBN) were committed to the Alliance at the very beginning of the Kennedy Administration.M. Leitenberg, "Background materials in tactical nuclear weapons (primarily in the European context)" in SIPRI, Tactical Nuclear Weapons: European Perspectives (London: Taylor Francis, 1978), pp. 16-7, Table 1A.7. p. 112 and pp. 115-6.

In 1979 a process of phased reductions started, when in conjunction with the INF modernisation, NATO decided to withdraw 1000 nuclear warheads (mostly from Honest John missiles), plus 572 others on a one-by-one basis with the deployment of Pershing 2 GLCMs.

Then in October 1983, at a meeting of the NPG in Montebello, Canada, the Alliance decided to withdraw a further 1400 warheads from Europe by the end of the decade. This step also included less publicised, and therefore less knowable, plans to modernise the TNW stockpile. It has been understood ever since, however, that such modernisation would not imply substitution of any of the 1400 warheads slated for withdrawal.

As of 1985, however, the situation should have resembled what is listed in Table 1.

A number of things have changed since then.

- The "Treaty between the United States of America and the Union of Soviet Socialist Republics on the elimination of their intermediate-range and shorter-range missiles" (the treaty on the so-called double zero) was signed in Washington, December 8, 1987. If ratified and implemented it should lead to the destruction in three years of 120 Pershing 2s and 309 GLCMs deployed in Europe at the time of the signing.

- The government of the F.R.G. has pledged to phase out, and seek no replacement of, its 72 Pershing 1A missiles, once the two superpowers have implemented the double zero. With respect to the situation depicted in Table 1, that implies a further reduction of 220 warheads.Under the terms of the Treaty, the U.S. will also destroy 430 more non deployed missiles; 170 of them are Pershing 1As in storage in the U.S. See "INF Treaty Specifies Methods of Destruction for Banned Missiles," Aviation Week and Space Technology, December 14, 1987.

- According to several sources,See William M. Arkin, "Fewer warheads in Europe," The Bulletin of the Atomic Scientists, August/September 1986; Geoffrey Manners, "SACEUR's plans for nuclear stockpile," Jane's Defence Weekly, 25 October 1986.

ADMs have been completely withdrawn. In June 1987, however, there still were 300 ADMs in the U.S. stockpile;See "U.S. nuclear weapons stockpile (June 1987)," The Bulletin of the Atomic Scientists, June 1987.

perhaps approximately 100 of them were deployed in Europe.See the chart "NATO without INF" in Arms Control Today, May 1987, p.7.

- The Dutch at first attached a condition to their acceptance of GLCM in November 1985, namely that nuclear roles for Orion P-3 Anti Submarine Warfare (ASW) Aircraft and F-16s be discontinued, but then Prime Minister Ruud Lubbers announced their willingness to retake those roles the very day the treaty on double zero was signed.See "Dutch Cancel Plan to Cut NATO Role," International Herald Tribune, December 9, 1987.

Despite much talk and public debate, therefore, as far as the Netherlands are concerned there should be no change after 1985.

- Scrapping of Nike Hercules Surface to Air Missiles (SAM) and Honest John should also be well underway. For the former, which will not be replaced by a nuclear system, there were only 75 nuclear warheads in the U.S. stockpile as of June 1987.See "U.S. nuclear weapons stockpile (June 1987)" (loc. cit. in note 4).

- Nuclear artillery shells represent another likely area of the current reductions. Both the W33 8-inch and the W48 155 mm atomic projectiles are ageing, poorly controllableBoth have mechanical combination lock as Permissive Action Link (PAL) and no command disable feature. See Thomas B. Cochran, William M. Arkin, Milton M. Hoenig, Nuclear Weapons Databook - Volume I - U.S. Nuclear Forces and Capability (Cambridge MA: Ballinger, 1984), pp. 47 and 54.

and limited in range. Their successors, the W79 and W82 respectively, have an Enhanced Radiation (ER) capability and as such are not deployable in Europe, given the open and widespread opposition to the neutron bomb on that continent. For this and other (economic) reasons, the U.S. Congress "mandated that only 925 of them be produced and that no neutron warheads could be produced after October 1984. This constraint means that only about 500 of those projectiles would be available for Europe - excluding the neutron warheads produced before October 1984 and those slated for the Pacific and Marine Corps."Arkin (loc. cit. in note 3).

- In perspective, the future of the Lance follow-on nuclear system is also in jeopardy. The Pentagon is seeking congressional approval to equip the Army Tactical Missile System with a nuclear warhead. But such a move has been blocked for four years and appeared unlikelySee "Pentagon Wants New Missile To Be Nuclear," International Herald Tribune, March 16, 1987.

before the treaty on double zero was signed. It cannot be ruled out, though, that pressure will mount in the coming months, both in Europe and in the U.S., for a NATO nuclear modernisation program in the missile category left uncovered by the treaty (up to 500 km range).

- As far as aerial bombs are concerned, the older B28, B43 and B57 are being replaced by the new B61. As of June 1987, there were 1000 B28s (both strategic and tactical), 975 B43s (both strategic and tactical), 1195 B57s and 2150 B61s (tactical version) of these bombs in the American nuclear arsenal.See "U.S. nuclear weapons stockpile (June 1987)" (loc. cit. in note 4).

Thus, it seems that in this category of weapons no substantial reduction will accompany modernisation. After the treaty on double zero, moreover, aerial bombs are also likely to be particularly dear to those NATO military planners who place high value on the capability to strike Soviet territory with nuclear systems based in Europe. In H%3d %4d %4d %2d %5d %5d %5d %3d %1d %5d %5d %5d %5d %5d %5d %5dHthe light of the most recent NPG work (see chapter 1), this school of thought seems to have gained the upper hand. Reductions are therefore unlikely in the near future.

In sum, in a few years, the U.S. nuclear weapons stockpile in Europe might well look like the situation depicted in Table 2. Although one must not forget that that only gives a highly hypothetical composition of the stockpile, two considerations can be made. First, a reduction of more than 50% would have taken place in about ten years, thanks to both negotiated and unilateral steps. Second the sheer numbers seem typical of a transitional phase; they are such, in other words, as to satisfy neither the 'doves' (who can rightly claim that there are still too many of these weapons, as they are still in the thousands), nor the 'hawks' (who fear that the U.S. is now well on the slippery slope of 'denuclearisation' of Europe). Clearly a NATO review of its military doctrine and arms control options is in order. It is in such a context that the role, if any, of TNW in the allied defence urgently needs rethinking.

To complete the picture, however, one must not forget that the U.S. can bring a number of other nuclear systems to bear on the European theatre. They range from the 400 Poseidon Submarine Launched Ballistic Missile (SLBM) warheads assigned to SACEUR,Five Poseidon submarines, with a total of 800 nuclear warheads, are assigned to SACEUR. Each sub, though, carries 160 warheads, so that at any given time there can be on station only a multiple of 160. The widely quoted figure of 400 is not such a multiple. It is simply half of 800. I am indebted to Gianluca Devoto, who called my attention to this point.

to artillery shells stored in non-European locations and comprise bombs aboard carriers, aerial and depth bombs, as well as warheads for the Lance missiles. For example, "Sixth Fleet Ships carry approximately 300 nuclear warheads for land attack, anti-air, and ASW operations;"William M. Arkin and Richard W. Fieldhouse, Nuclear Battlefields (Cambridge MA: Ballinger, 1985), p. 113.

A6 and A7 (the latter being replaced by F-18s) carrier bombers "are dedicated to SACEUR's Selective Strike Plan."Cochran et al. (op. cit. in note 8), pp. 207 and 210.

As of 1985, these additional U.S. nuclear systems committed to Europe amounted to some 2000 warheads overall,See Catherine McArdle Kelleher, "NATO Nuclear Operations," in Carter et al. (op. cit. in Introduction, note 1), Table 14-1, p. 448.

not including Sea Launched Cruise Missiles (SLCM). As of mid-87, 328 nuclear, land-attack SLCM were deployed on U.S. vessels,See IISS, The Military Balance 1987-1988, p. 202.

out of a program of 758 such missiles.

There are other interesting features of the U.S. TNW stockpile in Europe that are worth mentioning. The first regards the breakdown between warheads intended for U.S. use and for allied use. As shown in Table 1, 'only' one-third of the 1985 armoury was assigned to allied delivery systems. True, that still meant 2000 nuclear charges: certainly more than a token capability. Looking at these figures, however, from the standpoint of the ratio between nuclear warheads and delivery systems of U.S. forces as opposed to allied forces (see Table 3), one cannot but conclude that the American contribution to the defence of Europe is heavily nuclearised. It can be argued that this situation is the result of long standing European pressure to see its defence coupled to the U.S. nuclear guarantee, more than any American penchant for the nuclear battlefield. This implies, in turn, that the Europeans, while constantly complaining about the NATO conventional imbalance vis-à-vis the Warsaw Pact, have in effect undercut the fu

ll impact American forces might have in redressing such a perceived imbalance.

A second noteworthy aspect of the NATO theatre nuclear armoury is its regional distribution: more than half, about 3400 warheads, are in West Germany; 20%, about 1200, are in Britain, with another 20% in three countries of the Southern Flank (Greece, Italy and Turkey); only slightly more than a hundred are deployed in Belgium and the Netherlands; none are on the Northern Flank, or in Spain, Portugal, Iceland, Luxemburg and CanadaFor these figures see Arkin and Fieldhouse (op. cit. in note 13), p. 147. There has also been a report about SACEUR's "plans to move a number of nuclear artillery shells and bombs to three countries on the southern flank." Jane's Defence Weekly, 18 October 1986.

- in most cases this is the result of an explicit policy of the country concerned not to deploy nuclear weapons, at least in peacetime. To the extent that one can single out a NATO military rationale to deploy nuclear weapons at all, however, this must perforce be one and only one, i.e. valid for any of the allied members. Thus, regional distribution should simply stem from the relative weight of the various sectors of NATO defence: West Germany, in fact, has more than half of the stockpile. But this is evidently not the whole story. Some countries (Norway, Denmark) take exception on purely political grounds, although given the proximity of the Soviet threat it would not be surprising to know of some deployments of TNW on the Northern Flank. On the other hand, for a country like Italy one can only think of a rather far-fetched military rationale to such deployments, especially short-range, battlefield systems.On this point see Marco De Andreis, "The Nuclear Debate in Italy," Survival, May/June 1986.

Arguably, in this case the reason is again political (though opposite that of, say, Spain), namely that the nuclear burden-sharing can give the country some leverage, if not prestige, within the Alliance.

III. NATO NUCLEAR COMMAND AND CONTROL

Authority over the TNW use

Authority over the use of U.S. nuclear weapons, no matter what their range, yield, delivery means or place of deployment, cannot but rest in the hands of the U.S. President or his "duly deputised" alternates or successors. For all practical purposes, however, the President shares his control of the U.S. armed forces, nuclear weapons included, with the Secretary of Defence (together they form the National Command Authorities, NCA) and the Chairman of the Joint Chiefs of Staff.

Political authority is exercised through the military chain of command: the World Wide Military Command and Control System (WWMCCS) links to the NCA five major commands. These are: the North America Aerospace Defence Command (NORAD), the Strategic Air Command (SAC), the Pacific Command (PACOM), the Atlantic Command (LANTCOM) and the European Command (EUCOM). All but NORAD have nuclear weapons assigned. In particular, the bulk of NATO TNW is assigned to EUCOM.

In order to make sure that the U.S. has the capability to retaliate even in case of a NCA decapitation, it is likely that those in charge of the major commands know the codes to arm the nuclear warheads under their jurisdiction.To avoid unauthorized uses, a substantial portion, which includes TNW, of U.S. nuclear weapons are equipped with codes. Without the knowledge of the relevant code either the nuclear warhead cannot be armed, or (the older generations of artillery shells and the ADMs) it cannot be taken out of its container. In addition to the rather obsolete mechanical combination locks, there are four PAL categories - category F being the most recent and secure, with which, for example, GLCMs and Pershing 2 are equipped. Another safety catch is the so-called two-man rule: the various procedures needed to arm a warhead must be accomplished simultaneously by at least two individuals with the same authority. For the scope of this paper, the effectiveness of all these measures as guarantees of peacet

ime negative control is taken for granted. For a detailed treatment of the matter, however, see Donald R. Cotter, "Peacetime Operations: Safety and Security," in Carter et al. (op. cit. in Introduction, note 1), pp. 17-74 and Dan Caldwell, "Permissive action links," Survival, May/June 1987.

This is true, for example, for the SAC Commander and for the lower ranking officers onboard the Post Attack Command and Control System (PACCS), the airborne SAC command post also called 'Looking Glass'. It is reasonable to assume that the same thing holds true for the regional commanders. Generally speaking, in fact, "Distributing codes too widely could compromise control. Holding the codes at too few locations could compromise survivability under enemy attack. Force survivability was given high priority."Cotter (op. cit. in note 1), p. 51.

It should also be noted that it is the regional commanders who are responsible for TNW operational plans. Plans that, although co-ordinated and deconflicted with the Single Integrated Operational Plan (SIOP) are distinct from it. The President, for example, holds the SIOP options and could in principle, assisted by his own advisors, personally direct a strategic nuclear exchange. On the contrary, the NOP options (packages or otherwise) are held by the U.S. Commander in Chief Europe (CINCEUR) - who is also NATO Supreme Allied Commander Europe (SACEUR). The President can consent to the release request of a given TNW package; but he cannot chose on his own initiative from the various NOP options, or direct a theatre nuclear exchange by-passing CINCEUR-SACEUR. "In a technical sense, the President cannot order SACEUR...to fire a nuclear weapon; he can only release the weapon to him."U.S. Security Issues in Europe (op. cit. in Ch I, note 23), p. 20.

The problem of the authority over the use of TNW is complicated by the fact that those in Europe should be employed in the name of NATO collective defence. Thus, U.S. political authority has to somehow act as the collective political authority of the Alliance. The simplest solution would be a predelegation on the part of the non-nuclear allies to the U.S.In this paper the author keeps arguing as if only the U.S., among the 16 members of NATO, had nuclear weapons.

Another solution, though rather nightmarish, might be a sort of in-NATO horizontal nuclear proliferation. Despite the fact that the latter solution has, at times, found more supporters than the former, neither has ever been seriously considered. After the failure of the more ambitious and complex schemes, like the Multi Lateral Force (MLF), the puzzle of nuclear burden sharing has continued to absorb many of the Alliance's energies. What has finally been agreed upon tries hard to reconcile the fact of life that only one country has the bulk of NATO TNW with the ideal notion of a collective allied authority on their use.

Nuclear burden sharing between the U.S. and a number of NATO allies is governed in peacetime by bilateral agreements. Such agreements are of three types. First, an "Agreement for Co-operation on the Uses of Atomic Energy for Mutual Defence Purposes", which establishes the general framework for the exchange of information on nuclear weapon systems. Second, a "stockpile agreement which covers the location of nuclear weapon storage sites, their funding and their security."Buteux (op. cit. in Ch I, note 7), p. 212.

As far as security is concerned, this is jointly provided by the U.S. Custodial Units - which have the physical control over the warheads - and the host country's troops. The latter normally deal with the maintenance of security around the storage sites and during the weapons' movements in host territory. A third type of agreements is called 'Program of Co-operation': it covers the numbers and types of nuclear weapons that the U.S. assigns to those allied delivery vehicles certified as nuclear capable. Homologous services of the countries concerned are responsible for the Programs of Co-operation. Since 1967 the matter is the object of an "annual report from the Secretary of Defence to the allied Defence Minister"; also "in the event that 'significant changes' occur in the stockpile, a briefer report is made, but there is no requirement that consultation with the host country should take place."Ibid.

The various Programs of Co-operation "are subject to Congressional scrutiny, if not specific approval."Kelleher (loc. cit. in Ch I, note 22).

The same cannot be said of the allies' parliaments to which, as is the case in Italy, for example, the executive often denies knowledge of the contents of the agreements.

Despite this network of bilateral agreements and all the efforts made, in bodies like the NPG, to foster nuclear consultation, the question of what authority an ally can have on the use of an American nuclear weapon arises almost spontaneously. For those TNW earmarked for allied use - the so-called 'dual-key' systems - authority is clearly shared between the U.S. and the relevant ally. In particular, the latter has veto power, or negative control, over the nuclear use, since prudence dictates exercising at least residual national authority over the launch crew, even in the most cohesive of alliances. Apart from a last minute change of mind or about face, an ally can also refuse to commit its forces to the integrated military structure, thereby avoiding any undesired nuclear use of its delivery vehicles. The opposite circumstance - a joint U.S.-ally decision to use a 'dual key' system after proper consultation - would produce an interesting situation. Namely that, once the warhead were armed by the Ameri

can Custodial Unit and passed along to the ally's launch crew for use, all control would be in the hands of the ally - for precisely the same reason cited above with respect to the ally's negative control. If one, rightly or wrongly, assumes that dispersal of warheads from storage sites is simultaneous to the release of codes, then "dispersal is a step in the direction of suddenly creating several small new nuclear powers."Charles (op. cit. in Ch I, note 20), p.55.

In sum, if the U.S. shared control of all its TNW stockpile with its allies by means of controlling only the nuclear warheads, it would face the very unpleasant situation of having its positive nuclear control in the theatre limited by the negative control of the ally and of relinquishing all control once the warheads were released. Thus, it is no surprise that 2/3 of NATO TNW are not intended for allied use.

As for the portion of TNW for U.S. delivery vehicles, there is no question: the U.S. has both positive and negative control at any stage of the sequence, from authorisation to delivery to the target. The allies have no physical means of exercising authority - apart from the very extreme one of a violent take-over of the hosted U.S. base where nuclear warheads are stored.However extreme, this is a possibility that the American government takes into account. See, for example, the testimony of the then Assistant to the Secretary of Defense for Atomic Energy, Donald R. Cotter, in U.S. House of Representatives, Military Construction Appropriations for 1979, Hearings before a subcommittee of the Committee on Appropriations, February 23 and 24, 1978 (Washington D.C.: U.S. Government Printing Office, 1978) p. 210. On her part, Catherine McArdle Kelleher, has written of a "de facto barrier" built by the host countries around U.S. nuclear facilities, due to the fact that the former are involved in the warheads di

spersal. See "Managing NATO's tactical nuclear operations," (loc. cit. in Ch I, note 22). An unpleasent example of what all this may mean in practical terms was given by the confontation between Italian soldiers and U.S. special forces which took place at the Sigonella air base in Sicily, in October 1985, over who would take custody of the hijackers of the cruiser Achille Lauro. See Bill Keller, "Reagan's Order Ended U.S.-Italian Confontation in Sicily," International Herald Tribune, October 21, 1985.

They have, however, political means to exercise influence, if not authority.

The often quoted 1962 Athens Guidelines provide for consultation among allies about any NATO nuclear use, "time and circumstances permitting". In special consideration, moreover, are to be held the views of those countries from which nuclear weapons are to be launched or are to be used over.

Here a no less quoted question comes into play: does all this amount to a veto power on the part of the allies? The answer, for all the above mentioned reasons, is as follows: yes, for the third of the TNW stockpile for allied use; no, for the other two thirds for U.S. use. A veto power for this latter and more numerous portion has been consistently denied on reasonable grounds: any deterrent role TNW have would be greatly reduced in the eyes of the Soviets, were it to be conditional on an accord among up to 16 allies. If this is so, however, one must admit that the deterrent value of the TNW for allied use is greatly reduced, being dependent on the agreement of at least two powers.

Any other consideration is pure conjecture. How to interpret, for example, the well-known caveat "time and circumstances permitting"? Presumably, NATO would be hard pressed to retaliate in case of a Soviet first use. But the nature and scale of such use would make much difference as for the time and circumstances of the consultation.

Also, the special weight accorded to the countries on whose soil the weapons are launched and/or used can be read in many contradictory ways. Is the nearby third party, an ally nevertheless, going to be ignored?As for 'dual-key' systems, Catherine Kelleher is rather forthcoming on this point: "National decisions are required; Allied consultation is only desirable." Ibid.

does the special weight coincide with a power of veto? could consultation ever go beyond what, in the first chapter of this paper, has been called "mainly a tripartite business" between the U.S., the F.R.G. and the U.K.?

These are clearly unanswerable questions. Or, to put it differently, the answers vary widely according to the observer. "As a matter of fact the United States should have little rational interest in starting a nuclear exchange in Europe, and certainly not over the objections of the allies" - writes an American.Charles (op. cit. in Ch I, note 20), p. 129.

"Unless the technical means to implement a veto are there, even the most tightly drafted document is no avail" - writes a Briton.Campbell (op. cit. in Introduction, note 4), p. 314.

As noted in the first chapter, in its public version of Flexible Response, NATO stresses that escalation must take place under "firm political control". This clearly translates into an important commitment for the allied political authorities. Of course, one wonders whether and to what extent these same authorities have knowledge and familiarity with the Alliance's nuclear planning and consultation machinery.

First comes the President of the United States. Even though there are practically no direct signs of a U.S. President's interest in NATO nuclear matters, testimonies do exist as to his knowledge of the strategic, non-NATO ones.

According to Richard De Lauer, former Under Secretary of Defence for Research and Engineering, "President Carter was the first President ever to visit the National Command Post and sit down where he was supposed to sit and at least be briefed on what it all means. President Nixon never did, Johnson never did, and some of the security advisers, like Kissinger, never went down there."Ford (op. cit.in Introduction, note 3), p. 89.

A former Director of the White House Military Office, Bill Gulley, declared: "No new President in my time ever had more than one briefing on the contents of the Football, and that was before each one took office, when it was one briefing among dozens."Ibid. p. 90.

General William Odom, military assistant to former National Security Advisor Zbigniew Brzezinski also said that, with the exception of Jimmy Carter, "the president's attitude toward command and control, particularly of the strategic forces, has typically been one of benign neglect."Peter Pringle and William M. Arkin, SIOP (New York and London: W.W. Norton Company, 1983), p. 216.

What about Ronald Reagan? Apparently, "Carter volunteered to sit down with his successor to discuss the subject (the SIOP options), but this offer was rebuffed by President Reagan."Ford (op. cit. in Introduction, note 3), p. 91.

Furthermore, it is unlikely that a President who has said that SLBM are recallable; that "land-based missiles have nuclear warheads, while bombers and submarines don't"; that the Soviets "switched the numbers on their missiles in order to confuse us!"Strobe Talbott, Deadly Gambits (New York: Alfred A. Knopf, 1984), pp. 273-4.

(he had learned that the SS-18 is bigger than the SS-19) - it is unlikely that such a President be very familiar with the intricacies of strategic nuclear options, let alone NATO nuclear options.

As for the European allies, it does not seem that nuclear crisis management is very well practised by the relevant political authorities. Simulations are held, like the WINTEX exercises. Their main benefit, though, "according to a number of participants, is making sure that the technical equipment for nuclear release is in place. The players in WINTEX, particularly in the case of the United States, are not senior officials who would actually occupy those positions in a crisis; their places are taken by underlings, and serious debate on the political issues raised by nuclear use does not take place in this forum."Charles (op. cit. in Ch I, note 20), note 5, p. 73.

Also, as already noted in chapter 1 with regard to the work of the NPG, nuclear knowledge is likely to vary substantially depending on the country, with the U.K. and the F.R.G. probably playing a leading role. It remains to be seen, however, whether a deeper involvement by national military and diplomatic personnel in the day-to-day work of nuclear burden sharing, can translate easily into the attitudes of ministers or chiefs of governments. In some cases, like Italy for example, where the political turnover in cabinet posts is high and the political debate is dominated by domestic issues, it seems very unlikely that nuclear knowledge be widespread at the top.

Finally, the consultation machinery, both formal and informal, does not seem to be adequate to cope with the demands of a decision on nuclear use. First of all, there is the Defence Planning Committee (DPC). There, each ally, with the exception of France, is represented on an equal footing and has the opportunity to clarify its stance to its partners. But precisely such broad participation may very well make the DPC not the best place to agree on a decision so laden with dire consequences as nuclear use, especially in light of the NATO tradition of taking decisions by consensus, rather than by majority.

Furthermore, the relevant governments in the various capitals would hardly limit themselves to the DPC as the sole means of communication. After all, in that NATO Committee, officials of relatively second rank (permanent representatives) would simply rephrase their governments' instructions. It is reasonable to predict that almost all governments would at least try to get in direct touch with the nuclear powers, that is the U.S. and, perhaps, Britain. The likely outcome would be a good deal of chaos - or, in any case, a situation certainly not very conducive to rapid consensus. Additional problems, which would further slow down the whole process, may arise from the need to protect the communications from Soviet Signal Intelligence (SIGINT). Or from the relocation of the governments concerned to alternate command posts, perhaps to bunkers or the like. Nor should the use of a dozen different languages among NATO countries be underestimated in its potential as an additive to clumsiness in communications.

Again the Italo-American crisis of SigonellaSee note 9.

can help to grasp the magnitude of the latter problem. According to one report: "Mr. Reagan telephoned Rome shortly before midnight Thursday [October 10, 1985] to ask for permission for the Egyptian jet and its U.S. escort to land at Sigonella. Mr. Craxi gave the authorisation and the planes landed a half hour later. But the sources [Italian officials] said that Mr. Craxi and Mr. Reagan differed over which country should have jurisdiction over the hijackers: They said that the two leaders argued in a series of telephone calls over three hours before Mr. Craxi prevailed.""Reagan Says Plane Interception Is Intended to Warn Terrorists," International Herald Tribune, October 12-13, 1985 (emphasis added).

As a matter of fact the two leaders could use no common language and were forced to talk through an interpreter. It goes without saying that, although very serious issues were at stake on that occasion, they pale with respect to a decision on nuclear use.

It is only realistic, then, to expect that all the consultation permitted by time and circumstances would take place only between Washington, London and Bonn, although one may add Paris, for the obvious and mutual interest of co-ordinating the use of the relevant nuclear arsenals.

Command Posts and Chain of Command

NATO command posts and chain of command are an illustration of how the existing organisation is the result of compromises between military requirements, on the one hand, and political and economic constraints, on the other.

Avoiding the collocation of several command centres together, or more generally, with other vital military installations (like communications nodes, for example) is clearly desirable from the standpoint of vulnerability. But it is also costly, not only in financial, but also social terms: it is doubtful, for example, whether the European public would look favourably upon a further proliferation and geographic spread of military installations. Such proliferation is also likely to produce a number of organisational problems, as will be shown in the following discussion.

With regard to the Alliance's supreme commander, it seems clear that an understandable concern with survivability led to the multiplication of command posts: SACEUR-CINCEUR has his headquarters in Casteau (Belgium), Vaihinigen (F.R.G.), High Wycombe (U.K.), Mildenhall (U.K., more properly the base of a fleet of 4 'Silk Purse' KC-135 airborne command posts), plus a semi-hardened facility of unknown location."Possibly Belgium." William M. Arkin and Richard W. Fieldhouse, "Nuclear weapon command, control and communications," in World Armaments and Disarmament - SIPRI Yearbook 1984 (London: Taylor Francis, 1984), Table 13.2, pp. 464-5.

According to another source, the same commander, in both his capacities, has eight command centres available - but it is unclear how the four 'Silk Purse' planes are accounted for in this figure.See Ashton B. Carter, "Assessing Command System Vulnerability," in Carter et al. (op. cit. in Introduction, note 1), Table 17-1, p. 561.

Once the national forces of the Alliance were placed under the command of SACEUR, "EUCOM in Stuttgart [that is Vaihinigen] would split into two groups: one with NATO assignments and one with U.S. national responsibilities. Staff from the first group would take refuge along with the supreme allied commander, Europe, in the new wartime bunker being constructed in Belgium. The second group would retreat to the U.S. EUCOM wartime headquarters recently built in England [i.e. High Wycombe]."Bruce G. Blair, "Alerting in Crisis and Conventional War," in Carter et al. (op. cit. in Introduction, note 1), p. 98.

Another split would occur in the U.S. Army Europe (USAEUR) staff in Heidelberg, part of which would go to the headquarters of Allied Command Europe (ACE), Central Army Group (CENTAG) at Seckenheim (F.R.G.), while the rest would transfer to the wartime USAEUR headquarters in Massweiler, also in F.R.G. Presumably, similar plans have been drawn with regard to the staff of the U.S. admiral who is at the same time Commander in Chief U.S. Navy Europe (CINCUSNAVEUR, headquarters in London) and Commander in Chief Allied Forces Southern Europe (CINCSOUTH, headquarters in Naples).

On the whole, however, "It is doubtful whether such organisational rearrangements and command and staff movements would be conducive to orderly planning and timely operations - yet they would probably occur at precisely that point in a conflict when orderliness and timeliness would be at premium."Desmond Ball, "Controlling theater nuclear war," Paper prepared for an International Conference on Technology, the Arms Race and Arms Control, Union of Scientists for Disarmament/Unione Scienziati per il Disarmo, Castiglioncello, Italy, 15-30 September 1987, p. 10.

Moreover, the potential ubiquity of the highest military authority is likely to create some serious problems of communications with the other decision-making fora of the Alliance. The DPC and the NATO Military Committee are both in Brussels. The nearest command centre of SACEUR is 60 km away, at Casteau. The solution of "dispatch[ing] an emissary to Brussels to brief the Defence Planning Committee on the latest intelligence and on NATO military options,"Blair (op. cit. in note 23), p. 83.

would certainly not streamline an already cumbersome decision-making process in the heat of a crisis.

The major command centres, together with the NATO command chain are depicted in charts 1, 2 and 3. As can be seen, there are several instances of collocation of command centres and other C3 facilities. A case in point is Naples where, in addition to CINCSOUTH, the following commands are located: Allied Air Forces Southern Europe (AIRSOUTH), Allied Naval Forces Southern Europe (NAVSOUTH), Western Mediterranean Area (MEDOC), South East Mediterranean Area (MEDSOUEAST), Submarines Mediterranean (SUBMED), Maritime Air Force Mediterranean (MARAIRMED). Also in Naples are the Headquarters of Maritime Surveillance and Reconnaissance Forces Sixth Fleet, the Headquarters Area ASW Forces Sixth Fleet, a Control Station for the NATO Satellite Communications (SATCOM) System and a Control Station for the U.S. Navy Fleet Satellite Communications (FLTSATCOM) System, which should be more than enough to make Naples a so-called lucrative target.

The chain of command of an alliance with 16 members and such a vast geographical scope has to be complex. Sometimes to the point of causing paradoxes: the British Admiral who is Commander in Chief Channel (CINCHAN), one of the three principal commands in NATO, together with SACEUR and the Supreme Allied Commander Atlantic (SACLANT), in wartime would also be Commander in Chief Eastern Atlantic Area (CINCEASTLANT) and as such a subordinate to SACLANT.See Ball (op. cit. in note 24), p. 9.

Dual hatting - the same person holding both a national and an allied command - is also widespread from top to bottom. If one assumes that the national military, no matter what their roles in the NATO integrated structure, would strive to remain in touch with their own national political authorities,"Some experts speculate that high military commanders of all NATO countries are instructed to check with their national leaders before passing on [a nuclear] order." Charles (op. cit. in Ch I, note 20), p. 128.

then "An obvious although quite impolitic question concerns the ultimate loyalties of these two-hatted commanders...How would CINCHAN act if directed to one action by NATO and another by the British political authority? Although British forces assigned to SACEUR require a specific order from him to use British nuclear weapons, would the commander 2 ATAF [Allied Tactical Air Force] go against a negative SACEUR decision if he was instructed by the British Prime Minister to use a Jaguar or Tornado or Harrier with nuclear weapons to save an element of the British Army on the Rhine (BAOR)? Alternatively, would the CINCEASTLANT refuse to pass on an order from SACLANT to use a British nuclear depth bomb to defend a U.S./NATO Poseidon SSBN if the British Government remained opposed to any use of nuclear weapons?"Ball (op. cit. in note 24), p. 11.

Actually, the allied division of labour among military of various nationalities gives rise to some resentment even in peacetime."In an interview with Die Welt newspaper, Gen. [Hans Joachim] Mack, who retired from his NATO post [of Deputy SACEUR] last September, said that during his appointment certain staff talks had been held by SACEUR - then Gen. Bernard Rogers - of which he had not been told." Heinz Schulte, "Former Deputy SACEUR attacks NATO command," Jane's Defence Weekly, 23 January 1988.

On the whole, the observer is led to share the following assessment: "Because of its political composition and traditions, NATO's military chain of command is more like a web than a chain, especially in its nuclear aspects."Desmond Ball et al., Crisis Stability and Nuclear War, "A report prepared under the auspices of the American Academy of Arts and Sciences and the Cornell University Peace Study Program, January 1987" (Ithaca, NY: Cornell University Press, 1987), p. 30.

Early Warning

"NATO itself possesses no independent intelligence-gathering machinery."Blair (op. cit. in note 23), note 9, pp. 81-2.

There are, however, some exceptions to this rule.

Set up in January 1980, the NATO Airborne Early Warning (NAEW) Force is a truly multinational unit with 3000 personnel from eleven nations. Its fleet is composed of 18 Airborne Early Warning and Control System (AWACS) E-3A, to which 6-7 more aircraft of the same type, operated by the British but assigned to NATO, will be added in the future.Italy will soon order four AWACS aircraft, but they will not be assigned to NATO, as far as it is known.

NAEW is under the command of SACEUR, has its main operating base (MOB) at Geilenkirchen (F.R.G.) and three forward operating bases (FOB) at Trapani (Italy), Preveza (Greece) and Kanya (Turkey), plus a forward operating location (FOL) at Oerland (Norway). The task of the NAEW Force is to provide radar coverage against aerial threats flying at low altitudes.See Mario Lazzaretti, "In volo dalla FOB di Trapani," Rivista Aeronautica, n. 1, 1988 and NATO Facts and Figures (op. cit. in Ch I, note 2), p. 183.

Since 1973, the NATO Air Defence Ground Environment (NADGE) is also in operation. NADGE is an integrated air defence command and control system comprising 84 radar sites stretching from the North Cape to Eastern Turkey. In recent years, several improvements have been made to the network, especially on the southern flank.Ibid., pp. 182-3.

Apart from NAEW and NADGE, though, the allied Early Warning is the result of the combination of the member states' intelligence activities. Against a ballistic missile attack, for example, NATO relies entirely on U.S. systems, like the Ballistic Missile Early Warning System (BMEWS, three radar sites in Alaska, Greenland and Great Britain) and the Defence Support Program (DSP, three early warning satellites in geostationary orbits). From the standpoint of an attack against Western Europe, these systems leave a number of gaps. The BMEWS radar site in Fylingdales (U.K.), for example, is pointed in such a way as to leave uncovered a substantial part of the Mediterranean Sea and, more important, the Atlantic Ocean from which, theoretically, a Soviet SLBM launch could take place. The DSP coverage also has blind spots that, depending on the satellites' position, can even coincide with those of the Fylingdales radar (see charts 4 and 5). It should also be taken into account that the flight time of a Soviet ball

istic missile headed for Europe is so short (10-15 minutes) as to severely test the attack detection and confirmation capabilities of the BMEWS (up to ten minutes).A detailed discussion of BMEWS and DSP in the context of a ballistic missile attack against Britain is in Gregory (op. cit. in Introduction, note 5), pp. 32-43.

Warning information from the whole array of U.S. sources is received at EUCOM headquarters in Vaihinigen, where the European Defence Analysis Centre is also based. The centre handles electronic intelligence from all theatre sources. EUCOM is linked with its three major components: U.S. Army Europe (Heidelberg), U.S. Air Force Europe (Frankfurt) and U.S. Navy Europe (London). All these three components, in turn, provide further intelligence, particularly the Navy command, which gathers data received from two Fleet Ocean Surveillance Information Centres (FOSIC), one in London and the other at Rota, Spain.See Blair (op. cit. in note 23), note 7, pp. 79-80.

In case of a major crisis, "the Situation Centre at NATO headquarters in Brussels would receive warning from the various national governments. The centre is permanently staffed by military and civilian personnel from the member countries and depends on the exchange of intelligence among the permanent delegations."Ibid., note 9, pp. 82-3.

The magnitude of the non-U.S. contribution is difficult to gauge, but it should be fair to assume that Washington provides the lion's share of NATO intelligence.

On the whole the adequacy of NATO early warning is also difficult to measure. Strategic warning against a major Soviet conventional attack is generally expected to be timely and effective, whereas tactical warning against some Soviet nuclear uses could present problems.

Communications

Drawing a clear picture of NATO communications is no simple task. In this respect it is fair to say that nuclear command and control is a much more complex issue at the theatre, than at the strategic level.

At the beginning of the seventies the Alliance launched a program in this field called NATO Integrated Communications System (NICS). According to its official description, "The NICS concept is based on the development of a common user automatically switched grid network, employing sophisticated computer-driven switches for all forms of voice, telegraph and data traffic."NATO Facts and Figures (op. cit. in Ch I, note 2), p. 180.

The system was to be implemented in two stages: Stage I should have been completed in 1987. It "involves the implementation of a series of stand-alone sub-systems to automate NATO's voice and telegraph communications". The real challenge, therefore, seems to be Stage II, which more ambitiously "will involve the geographical and technical expansion of these sub-systems and their melding together into a single fully integrated system."Ibid.

In practice NICS (see charts 6 and 7) is based on three elements:

- "A terrestrial transmission grid stretching from Northern Norway through Europe into Turkey, with sub-systems like the tropospheric/line-of-sight Ace High and the Central European line-of-sight CIP67."Joachim M. Sochaczewski, "The role of communications in NATO," Military Technology, 6/1984. The author was Assistant Chief of Staff Communications and Electronics of SHAPE.

CIP67 is an analog microwave network consisting of about 60 nodes.See Gope D. Hingorani and Rupert Brand, "Architectural Framework for the Evolution of NATO Integrated Communications System," Signal, October 1985.

Ace High network was built in the early sixties and, according to one source, serves mainly for crisis communications among NATO political authorities.See Campbell (op. cit. in Introduction, note 4), p. 188.

The system has 49 tropospheric scatter links and 40 line-of-sight microwave links, 570 voice, 260 telegraph and 60 data circuits - but no more than a dozen very vulnerable nodes.See Ball (op. cit. in note 24), pp. 18-9.

- A Satellite Communication (SATCOM) system, with 21 fixed ground terminals but only 2-4 master control stations.Ibid.

- The Initial Voice Switched Network (IVSN) and Telegraph Automatic Relay Equipment (TARE) voice and message switching facilities. With 24 operational switches the former is intended to serve some 3000 direct and 7000 indirect subscribers. The latter has 18 operational switches, serving some 700 low and medium speed subscribers. Within IVSN the establishment of a NATO Secure Voice Network (NSVN) is envisaged.See Hingorani and Brand (loc. cit. in note 41).

NICS can be considered a system for strategic communications within NATO. It links SACEUR with his major subordinate commands and to army groups. But below Corps level national systems are the rule.

"NATO battlefield communications, for example, involve at least six systems: Ptarmigian (U.K.), RITA (Belgium and U.S.A.), Autokonetz (West Germany), Zodiac (Netherlands), Tritac (U.S.A.) and Mobile Subscriber Equipment (U.S.A.). Of these only Ptarmigian and Zodiac are directly interoperable with one another and none are directly interoperable with...NICS."Gregory (op. cit. in Introduction, note 5), p. 99. On her part, Italy is developing yet another system , called Catrin.

The more one descends towards battlefield level, the more the situation worsens as far as interoperability is concerned. Even the four U.S. services cannot communicate directly with each other. During the 1983 invasion of Grenada a U.S. Army officer "had to use his telephone credit card to call Ft Bragg in order to communicate with the Navy's fire support system only a few miles away"; worse yet, in the course of the Achille Lauro hijacking, "President Reagan aboard Air Force One had to speak with Secretary of Defence Caspar Weinberger on open channels because their two scrambling devices were not interoperable."Rick Marshall, "Battlefield Communications: Technology's Challenge," Defense Foreign Affairs, October 1986. On October 7, 1985, four Palestinian terrorists hijacked the Italian cruiser Achille Lauro, with 454 passengers on board. One of them, an American citizen named Leon Klinghoffer, was killed by the hijackers. After two days the ship was released, following tense negotiations mediated by th

e PLO, the Italian and the Egyptian governments. The assassination caused the American interception of the Egyptian plan that on October 10 was carrying the terrorists to Tunis and the subsequent attempt to take custody of them recalled in note 9.

In case of a critical function, like Identification Friend and Foe (IFF), "U.S. aircraft are equipped with systems which operate in the D-band (formerly known as the L-band) while West German systems operate in the E/F-band (formerly known as the S-band)."Ball (op. cit. in note 24), p. 12.

But IFF is only part of the larger problem of NATO Air Command and Control systems, be they for offensive, for air transport or for close air support operations. "All these [national] systems operate independently from each other with their own, dedicated sensors. Exchange of information is often possible only by telex or telephone, if at all. The data links used are slow and are limited in their capacity. In addition, most of the systems are inadequately protected, or are unprotected, against jamming and ELINT (electronic intelligence) and are vulnerable to enemy attack."K.G. Benz, "ACCS, a C2 system for NATO's air forces in Europe," International Defence Review, 11/1984.

To improve this rather poor state of affairs, NATO launched a new program called Air Command and Control System (ACCS), with the aim of integrating the national systems. Quite apart from the fact that, very much like NICS, it will not be working before the year 2000, in the end ACCS "could well correspond in practice to an improved NADGE system working together with NAEWS."Ibid.

Major NATO communications networks are duplicated by some analogous national systems: the U.S. Automatic Voice Network (AUTOVON) and Automatic Digital Network (AUTODIN) systems correspond grosso modo to NATO's IVSN and TARE; parallel to NATO's SATCOM, are the British Skynet and the American Defence Satellite Communications System (DSCS), with about two dozen ground terminals in the European theatre.See Ball (op. cit. in note 24), p. 18.

According to one source, however, DSCS, Skynet and NATO SATCOM are interchangeable and British military communications have been relying on DSCS satellites since 1977, when the last Skynet satellite went out of operation.See Campbell (op. cit. in Introduction, note 4), p. 197.

Similar to NATO's Ace High is the more modern Digital European Backbone (DEB) for U.S. Communications, whose more than 100 sites (with a dozen nodes) have been emplaced between 1979 and 1983 and stretch from Italy to Britain. DEB communications are encrypted.Ibid., pp. 187-8.

About half of the NATO telecommunications circuitry is carried by the national PTT networks.See Ball (op. cit. in note 24), p. 20.

Their security, hardness and redundancy vary widely from country to country - the German system is highly reputed, for example, while whoever has used an Italian telephone is unlikely to appreciate the efficiency of that system. The national networks are all unhardened against the electromagnetic pulse (EMP) and are permeable to Soviet SIGINT. As for cross-border connectivity of PTT systems, there are a limited number of very soft choke points.Ibid.

Communications for nuclear management can count on a number of dedicated systems: the European Command and Control Console System (ECCCS), which is installed at nuclear weapons storage sites and is used mainly for peacetime management;See Campbell (op. cit. in Introduction, note 4), p. 184 and Arkin and Fieldhouse (op. cit. in note 21), Table 13.5, pp. 480-1.

the U.S. Cemetery Net and NATO Last Talk radio High Frequency (HF) networks,Cemetery Net is being replaced by a similar system called Regency Net, according to Arkin and Fieldhouse, Ibid.

whose black-humour denominations easily convey what they are intended for, namely the sending of nuclear release and Emergency Action Message (EAM); SACEUR Status Control, Alerting and Reporting System (SCARS) II, defined by one author as a "special, highly secure communications link", to be operational, however, "in the future"Kelleher (loc. cit. in Ch I, note 22).

and whose bearer systems will be both IVSN and TARE;See Hingorani and Brand (loc. cit. in note 41).

the Air Force Satellite Communications (AFSATCOM) system, based on a number of transponders aboard various spacecraft, transmitting in Ultra High Frequency (UHF) to ground terminals at bomber and missile bases, nuclear storage sites and some field units.

Vulnerabilities

C3 facilities are inherently more vulnerable than the weapons they are intended to manage. Moreover, while the number of the key nodes of a given C3 architecture is, as a rule, finite, the list of potential threats against them is virtually open-ended and not always the product of the hostile intentions of a determined enemy.

Command posts can be rendered mobile or airborne, for example, but not to the point of impeding centralised control. Thus, there are not many such instances in NATO. Hardening is another solution; but again it is not very widespread and there appear to be very few underground command centres in the European theatre. None of them "could withstand blast overpressures of more than 1000 psi, and hence they would be unable to survive deliberate nuclear strikes."Ball (op. cit. in note 24), p. 16.

Likewise, a majority of the nuclear storage sites appear to be protected against conventional attacks, but not against nuclear ones.Charles (op. cit. in Ch I, note 20), pp. 47 and 51.

Radars, antennae, satellite ground stations and, more generally, communications systems simply cannot be hardened in any significant way and are very vulnerable even to conventional weapons' attacks. As noted in the preceding paragraph, moreover, they are parts of larger networks that invariably have a limited number of nodes, or choke points, that, if destroyed, can incapacitate the whole network.

According to Desmond Ball, there are only 60 key C3 sites in Europe. "Indeed, it is likely - he wrote - that the destruction or incapacitation of less than 20 of these 60 sites would totally immobilise command and control across the whole European theatre."Ball (op. cit. in note 24), p. 15.

He also noted that "all but a dozen" key C3 facilities could be destroyed by the use of conventional ordnance and that the Soviet Union can count on some 380 Spetsnatz teams "specifically organised" to fulfil this purpose.Ibid., p. 18.

Apart from physical destruction, however, threats abound. Communications in HF, for example, are common because of the frequency's characteristics: large bandwidth, small and low-powered antennae, long-distance propagation. But these features also make it easier "for hostile forces to listen in on communications, to jam them from far away and to locate the transmitters", and draw "many users to the band, resulting in a danger of self-jamming as legions of friendly transmitters come on the air at once."Ashton B. Carter, "Communications Technologies," in Carter et al. (op. cit. in Introduction, note 1), p. 241.

The Soviets, who make large military use of HF, are also likely to know well how to exploit jamming against NATO forces. Generally speaking, therefore, "Crisis communications, communications supporting U.S. projection forces overseas, and communications supporting theatre forces have much more to fear from jamming than strategic forces."Ibid., p. 263.

Once the nuclear threshold is crossed, the disruptive effects of the EMP on electronic apparatus would in all likelihood be devastating, from friendly and enemy forces alike. For example, "A single [nuclear] explosion at 200 miles above the centre of the 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 - could apparently blanket the entire United States."Ball (op. cit. in Introduction, note 3), p. 11.

It should not be impossible to the Soviets, therefore, to do the same over Western Europe. Even more so, if one takes into account that in NATO Europe "We have EMP-protection for some sub-systems, but we have not yet an EMP-protected network."Sochaczewski (loc. cit. in note 40).

With regard to the various weaknesses in the NATO C3 system described so far, the Alliance seems to have considered a number of remedies, like increasing redundancy, hardness and rapid repair capabilities. "We have looked into all these options - writes a knowledgeable NATO officer - and a few more and found out that they are either unaffordable or difficult to design."Ibid.

Organisational problems loom large in C3 programs, even when their scope is not alliance-wide. The U.S. Mobile Subscriber Equipment (MSE) program involves five army corps and an area of 37,500 square kilometres to be covered with some 6200 stationary and 1900 mobile users; by 1992 32,000 telephones, 10,000 trucks, 10,000 radio terminals, 300 node switches and 5600 generators will have to be delivered. To learn how to use the system, seven training centres will be set up in the U.S., Germany and South Korea, with the prospect of receiving some 20,000 students.Marshall (loc. cit. in note 47).

Very much like with the Strategic Defence Initiative (SDI), software for C3 systems is often a very serious hurdle: "the whole software design process at least for military systems is unpredictable, uncontrollable and ineffective...Projects are ridiculously late, system quality is unacceptable, resource estimates are wildly inaccurate and development is out of control."Sochaczewski (loc. cit. in note 40).

Unsolvable dilemmas: alert, dispersal, release

Approaching nuclear command and control diachronically, that is from the standpoint of the transition from peace to war through a crisis, reveals a whole set of troublesome problems.

It has always been clear in NATO that nuclear weapons would quickly turn into a sort of 'hot potato' in the hands of the Alliance's decision-makers as soon as they had to face the decision of raising the normal peacetime status of the forces to higher alert levels.The NATO Alert Condition (LERTCON) ladder has five rungs: LERTCON 5 (peacetime), LERTCON 4 (military vigilance), LERTCON 3 (simple alert), LERTCON 2 (reinforced alert), LERTCON 1 (general alert). NATO has never gone beyond military vigilance. See Blair (op. cit. in note 23), note 4, p. 77.

Paul Bracken's work has convincingly shown the extent of the intermingling of nuclear and conventional forces in NATO defence,See his The Command and Control of Nuclear Forces (op. cit. in Introduction note 3), chapter 5.

a fact that, as far as the former are concerned, substantiates the well-known saying "use them or loose them". Since it's clear that as long as TNW are kept in a limited number (slightly more than a hundred) storage sites (whose location is arguably known to the Soviets), they are a strong incentive to pre-emption, the question would soon arise whether to disperse them, if not use them. Dispersal, in turn, could be misread by the Soviets as being a signal of impending use and therefore could well be as strong an incentive to pre-emption as its opposite, i.e. inaction. Finally, the relationship between alert and dispersal on the one hand, and release and codes' notification on the other is totally unclear.

Some authors have, in fact, argued that a number of factors (the presumably impervious task of matching weapons with delivery vehicles, the likely difficulty and/or disruption of communications) authorise the conclusion that TNW will leave their storage sites together with the release of the relevant codes.See Ibid. p. 168; Ball (op. cit. in note 24), p. 28 and Charles, (op. cit. in Ch I, note 20), pp. 54-5.

Others have hypothesised a more phased process, rightly pointing to the different characteristics of the various nuclear weapons involved. Aerial bombs are co-located with fighter-bombers - a portion of which are in quick reaction alert (QRA) even in peacetimeAllied QRA aircraft in peacetime raise thorny questions, namely how the allied political authorities manage their forces. With nuclear bombers ready to take off in a matter of minutes, allied heads of state should have some equivalent of the U.S. President's 'Football' always accompanying them - which does not seem to be the case.

- and do not require the highly visible step of dispersion. "These forces - wrote Bruce B. Blair - can be moved to alert in a relatively quiet, low-key fashion."Blair (op. cit. in note 23), p. 91.

On the other hand, increasing the number of fighter-bombers in QRA amounts in practice to a sharp reduction of conventional capabilities at precisely the moment when they are most needed. SACEUR, in fact, would hardly risk nuclear assets in conventional sorties. Furthermore, the vulnerability of FGAs, as well as their bases, is so acute as to make alert a relatively minor problem. One guesses that their use in a nuclear role is the question that decision-makers would soon have to face.

As long as battlefield TNW are there, however, they will continue to represent an intractable problem for NATO crisis management. A problem that, far from being solved, is only compounded by exercises in modelling like the widely quoted 'bottom-up' release process.See U.S. Department of the Army (op. cit. in Ch I, note 17), p. 10-9. For an excellent discussion of the 'bottom-up' and its brethren 'top-down' release models see Kelleher (loc. cit. in Ch I, note 22).

Are the allied political authorities ever likely to release nuclear weapons to field commanders in such a short time as to make a battlefield nuclear use militarily meaningful? The answer, in all probability, is no. It would be much more realistic to ask if they would at least grant dispersal, but "According to one former participant, the [NATO Wintex] exercise did not distinguish between the decision to distribute warheads to delivery systems and nuclear release, at least in the occasion he participated."Charles (op. cit. in Ch I, note 20), p. 50.

For longer-range systems in particular, the release request would probably be initiated at the top by SACEUR. This does not guarantee, per se, a quicker decision on the part of NATO's political authorities. SACEUR's option of sending them an 'early notification' messageSee Ball (op. cit. in note 24), p. 30.

could expedite the process, but it could also make things worse in terms of the more general political management of the conflict, since the early introduction of the nuclear variable is hardly conducive to coolness and thoughtfulness.

All along the path of this agonising decision-making process, however, the threat of Soviet pre-emption never disappears: "through SIGINT and other means, the Soviets authorities would be able to keep abreast of NATO alert, dispersal and release developments."Ibid., p. 31. For the impressive SIGINT efforts and capabilities on the Soviet part see Ibid., pp. 21-6.

Distribution of TNW to a potentially large number of field units would make nuclear use almost uncontrollable, especially if coupled with the vast array of destruction caused to C3 facilities by the preceding conventional phase of the war. "In short [TNW] are reserved for use in situations in which command and control might be at the point of maximum disarray."Blair (op. cit. in note 23), p. 108.

War Termination

The threat of escalation, including escalation to nuclear warfare, is central to NATO's doctrine of flexible response. It is clearly not so much a strategy to win the war, as a means of bringing deterrence to bear well within the war. In other words, it is at the core of the notion of intra-war deterrence. In order to exercise deterrence through the threat of escalation, however, one must also be confident in one's capacities to de-escalate and bring the war to a halt at all rungs of the ladder.

War termination is undoubtedly the missing tessera in NATO's mosaic. In the light of the preceding discussion, in fact, the observer finds it hard to believe that it would be feasible to recall hundreds of nuclear weapons back to their depots, once they had been distributed to dozens of field units. A number of isolated detonations are very likely, if one assumes the no less likely breakdown of a good deal of communications that would occur at an advanced stage of conventional/nuclear fighting. This is all the more true, if one bears in mind the standard scenario of NATO's nuclear first use, that is the dreadful event of a massive Soviet invasion of Allied territory. It is not unfair to predict that in such a scenario an isolated NATO field commander would have very few incentives not to fire whatever he finds at his disposal.

As far as bombers are concerned, de-escalation seems even more difficult. Recallability is possible in principle, although it would be hard to translate into practice, given the fact that the aircraft mission most probably requires the crew to stay incommunicado till delivery to the targets. Nor is it known of any 'fail-safe' procedure for tactical bombers, given their proximity, as opposed to strategic bombers, to enemy territory.'Fail-safe' means that the bombers will turn back to their bases, once they reach a certain point en route to their targets, unless they get the order to follow-on.

"The chief difference - wrote Paul Bracken - between the central and theatre war forces is that the weapons of the former are owned by command systems specially designed for the nuclear age...Theater nuclear forces, on the other hand, are given to ordinary Navy, Army and Air Force units...In Europe, or at sea, where both sides are put in this position, de-escalating a conventional or theatre nuclear war seems more difficult than de-escalating a central exchange."Paul Bracken, "War Termination," in Carter et al. (op. cit. in Introduction, note 1), pp. 206-7 (emphasis in the original).

Europe and the notion of limited nuclear war

Europeans have never been enthusiastic supporters of limited nuclear war theories. The reason is quite straightforward: above one's head any war is bound to look unlimited in nature. On the other hand, a substantial part of the European political establishment in this post-war period has not only accepted but also actively advocated the NATO policy of first use, inasmuch as this policy warrants, in their view, the coupling with the U.S. central strategic arsenal, thereby enhancing deterrence. One can indeed argue that in all probability a nuclear war would not remain confined to EuropeFor a very convincing argument in this direction see Ball (op. cit. in note 24), pp. 32-5.

- though the American side would have good reasons for at least trying. A European concerned about the robustness of the allied deterrence may well feel reassured by that.

Perceptions of nuclear issues, however, always have two sides: often the very feature that enhances deterrence also makes things worse should deterrence nevertheless fail. From this latter point of view, as things stand, a European has little reason for comfort: beyond the point of single-digit nuclear explosions - a point that it is easy to predict would be soon overcome - lies catastrophic destruction, no matters what happens, or does not happen, to Soviet and American citizens.

There is some evidence to illustrate this point.

Some of it is as old as the introduction of nuclear weapons in Europe. The June 1955 NATO exercise 'Carte Blanche' simulated the use of 335 tactical bombs in West Germany, Northern France and the Netherlands. It turned out, quite optimistically, that such use would have resulted in 1,500,000 dead and 3,500,000 injured (not counting those affected by radioactivity).

"Studies and war games done in the 1960s showed repeatedly that even under the most favourable assumptions about restraint and limitations in yields and targets, between 2 and 20 million Europeans would be killed in a limited tactical nuclear war, with widespread damage to the economy of the affected area and a high risk of 100 million deaths if the war escalated to attacks on cities."Alain C. Enthoven, "U.S. Forces in Europe: How Many? Doing What?," Foreign Affairs, April 1975.

More recently an Italian physicist, Andrea Ottolenghi, ran a computer simulation of a 'limited' nuclear exchange in Europe. In his model the target list is composed of: missile bases; command sites; communications sites; nuclear storage sites; air-bases and naval nuclear bases. Nuclear storage sites for artillery shells, ADMs and SAM, the majority of military installations in or near large cities and political headquarters were excluded. Thus, he hypothesised that 94 targets (3 in Eastern countries and 91 in Western countries) were attacked with one 150 kt air burst; 285 targets (94 in the East and 191 in the West) with one 150 kt ground burst; 91 targets (11 in the East and 80 in the West) with one cluster of three 150 kt ground bursts. In total, about 98 MT would be used to attack 470 military installations.

The number of fatalities due to the blast and thermal effects was found to be 7.4 million. The total casualties from these effects (deaths plus injuries) amounted to 15.6 million. Estimates of fatalities from fall-out vary, in the simulation's results, from 58.3 to 86 million, depending on the month (and therefore on the prevailing seasonal meteorological conditions), the degree of protection and the LD-50 (dose of radiation assumed as lethal for 50% of the population exposed).See World Health Organization, Fortieth World Health Assembly, Effects of Nuclear War on Health and Health Services, doc. n. A40/11, 18 March 1987, Annex 4, C.

These figures are not surprising if one takes into due account that the average population density in Europe is eight and four times respectively that of the U.S.S.R. and the U.S.A., and that in Europe (excluding the Soviet Union) there are almost 400 cities with a population of more than 100,000.

In table 4 the reader can find the yield of the nuclear warheads considered in this study. Yields vary from sub-kiloton to 1.45 megaton. It is intuitive, however, that any use of these weapons in packages of at least tens of warheads would be unlikely to produce very limited effects. Corresponding Soviet warheads are, on average, even more powerful.

Nuclear effects would also devastate Europe were nuclear reactors (about 300 of them are in operation, under construction or planned in Europe) struck. "One year after the explosions, the fall-out from the reactors and the other nuclear facilities might be equivalent to the detonation of 30,000 megatons of nuclear weapons. Over a period of 25 years, the accumulated doses would be 10 to 50 times higher than from the weapons alone."Joseph Rotblat, "Consequences in Europe of a nuclear conflict," in Michel de Perrot (ed), European Security: Nuclear or Conventional Defence?, Proceedings of the IVth International Symposium organized by the Groupe de Bellerive, Geneva 8-10 December 1983 (Oxford: Pergamon Press, 1984), p. 11.

It seems really difficult to grasp how the Alliance could manage problems of this order of magnitude if one recalls the stress caused to Western European democracies (in terms of information, emergency health measures and political decision-making) by the April 1986 accident to the Soviet nuclear reactor at Chernobyl - no more than a trifle if compared with the effects of even the most limited nuclear use.

Political leaders are increasingly aware of all this. Former F.R.G. Chancellor Helmut Schmidt, for example, recently declared that "once you explode the first two or three [tactical nuclear weapons] on German soil, as far as the Germans are concerned the war will be over...German soldiers would try to get home in order to look whether their wives and children or their parents are still alive or not and wouldn't really bother much about the rest."Quoted in Elizabeth Pond, "W. Germany's Schmidt touts a Europe independent of U.S.," Christian Science Monitor, November 20, 1986.

Schmidt's comments are the political translation of military concerns that are rarely aired. On the battlefield, "strike planning may call for delivery of nuclear weapons in close proximity to friendly forces directly engaged in combat against concentrations of enemy forces. Under those circumstances the safety of his own forces can present serious and potentially irreconcilable conflicts to the targeter."Theodore A. Postol, "Targeting," in Carter et al. (op. cit. in Introduction, note 1), p. 402 (emphasis added).

In practical terms: "Even if friendly troops are outside the range of threatening blast effects, light and heat from the detonation can set tents and clothing on fire, light weapons can warp and jam from differential heating, and retinal burns or flash blindness can occur. Although the use of very low yield weapons and high heights of burst can greatly reduce the levels of radioactivity in the target area and downwind, friendly forces could still be subjected to large amounts of radioactivity."Ibid. p. 403.

In this study, the Soviet view of theatre nuclear warfare has never been considered. Its importance is crucial, however, for it takes the commitment of both parties involved to keep a war limited and under control. In recent times the Soviet side has moved sharply away from that certain penchant for undifferentiated use of conventional and nuclear means that was detectable in the past. The use of conventional weapons to destroy NATO nuclear assets is currently emphasised, as was recalled above with regard to Spetsnatz special forces. Should NATO cross the nuclear threshold, though, it is still very doubtful that the Soviets would play their nuclear cards according to Western notions of escalatory steps, damage-limitation and intra-war deterrence.

"What must be emphasised is that Soviet military strategy and operational art are structured in such a way that any NATO nuclear use seems to call for extensive concentrated nuclear strike deep into NATO territory. Preparations for a NATO 'demonstrative' use of nuclear weapons intended to display resolve might therefore in fact serve to spark a full Soviet [Theatre Nuclear Forces] strike."Stephen M. Meyer, "Soviet Theatre Nuclear Forces - Part I: Development of Doctrine and Objectives," Adelphi Papers 187 (London: International Institute for Strategic Studies: Winter 1983/4), p. 30.

Again, it is not difficult to predict that such a posture would result in dire consequences for Europeans, troops and non-combatants alike.

CONCLUSIONS

Some considerations can probably be drawn from the preceding discussion.

- Almost twenty years of attempts at planning nuclear use in NATO, be it first or follow-on use, have raised more questions than they have been capable to solve. Unless a very limited, demonstrative use, were to prove successful in persuading the Soviets to stop and retreat, any subsequent nuclear use would be largely indifferent to planning, since there is no way to predict with a high degree of confidence what the Soviet reaction would be. Uncertainty may enhance deterrence, but here it clearly works both ways.

- As a consequence, nuclear force levels can be based on no convincing rationale. Arms control, weapons' obsolescence and a certain mood in the U.S. Congress against TNW have all concurred in recent times to almost halve the U.S. nuclear arsenal for NATO. But there are still thousands of TNW, i.e. far too many for any use that is not bound to destroy what it is supposed to defend. The present nuclear force level, moreover, weighs heavily against conventional capabilities, particularly as far as American air power is concerned.

- Attempts at sharing political authority over the use of U.S. TNW for NATO have produced a maze of contradictions and dilemmas. The allies cannot be sure there will be such time and circumstances as to permit consultations, whereas the U.S. cannot discount the physical means to exercise negative control on the part of the allies. Political leaders on both sides of the Atlantic do not seem to be fully aware of the decisions they would face once the nuclear cards were put on the table. Looking at the history of nuclear consultation in NATO, chances are that the majority of the allies would not, or could not, take part in any discussion over nuclear use and that such discussion would not take place in the formal fora. If the product of political consultations, it is unlikely that nuclear release would be so timely as to be militarily useful as well.

- The military chain of command reproduces the same tensions between national and collective, alliance-wide interests that can be detected on the political side: questions about the ultimate loyalty of dual-hatted commanders are a clear example. As a matter of fact nuclear weapons seem too powerful, even symbolically, to be managed by complex organisations, political as well as military.

- Vulnerabilities of C3 facilities are countless and appear to have a synergic effect, that is, the vulnerability of the nuclear C3 system is greater than the sum of the vulnerabilities of its components. By the same token, devising a working software for complex operations - and nothing seems more complex than keeping a nuclear war in Europe under control - is always more difficult than procuring the hardware.

- The dynamics of alert and dispersal of NATO nuclear forces are likely to make any major crisis extremely unstable. Well before being confronted with questions of nuclear use, political decision-makers would be hard pressed to find a way out of the dilemma between inaction on the one hand and higher alert levels and dispersal on the other. Worse yet, both decisions could trigger Soviet pre-emption. Here is another instance in which uncertainty works against NATO, insofar as it detracts from crisis stability.

- Beyond the most limited demonstrative use category, no simulation or exercise has ever produced results that can be taken as acceptable levels of damage for Europe. Nor is it clear how a nuclear war in Europe could be brought to a halt.

Do all these considerations also tell something about the NATO doctrine of Flexible Response? They probably do, with regard to the letter of what is known of MC 14/3. The key words in this respect are "under firm political control." Escalation, including escalation to nuclear warfare, must take place "under firm political control." Chances are that not even military control would be effective in a theatre nuclear war, let alone political control.

As for the spirit, i.e. the thousand interpretations of Flexible Response, clearly nothing conclusive can be said either way. Does, for example, the impossibility of keeping nuclear war under control undermine deterrence? Probably not, for the prospect of taking a path of spasmodic nuclear exchanges to utter catastrophe certainly does not invite any opponent in his right mind to bullish behaviour.

It is fair to say that the European reading of Flexible Response that has prevailed so far is wholly consistent with the idea that nuclear war is not, and should not be, controllable. That is, the quicker it is supposed to escalate into strategic exchanges the better, since nothing enhances deterrence more than putting the homeland of both superpowers at risk. In this sense, Europeans have achieved a strategic masterpiece: with practically no nuclear weapons of their own, they believe to have succeeded in deterring both the U.S.S.R. and the U.S.A. - the latter from thinking in terms of limiting nuclear warfare outside his territory.

With such a powerful constituency against the notion of controlling nuclear use, it is also very unlikely that NATO nuclear C3 will ever benefit from dramatic improvements. In addition to the structural problems that I have attempted to underline in this paper, there is, in fact, also a question of political will. Proliferation and hardening of military facilities is not welcomed in wealthy, industrialised democracies. Standardisation and interoperability - vital features indeed for C3 systems - come a distant second to the unbending practice of protecting national economic interests. As a result, programs are systematically behind schedule, and once implemented are much less far-reaching than the initial ambitions, as is the case with NICS and ACCS. Paradoxically, the observer can find this state of affairs quite heartening, since it means that no one in NATO seriously believes that war is likely. Or that controllability really matters for pre and intra-war deterrence.

Still broadly speaking, the American way of reading Flexible Response tends instead to emphasise notions of limitations, controllability and firebreaks. This author must say that, quite apart from their feasibility, such notions have the merit of signalling an awareness of a basic fact of contemporary strategic life, i.e. that NATO's opponent also possesses the whole array of nuclear weapons. Basically, European obsessions with coupling and escalation are symptoms of an unending nostalgia for the golden age of massive retaliation, when the American nuclear threat made sense to deter even conventional war, since the opponent could not retaliate in kind. That age long gone, most Europeans have attempted to re-create it, while Americans have at least tried to find some new answers to the situation confronting the Western alliance. That the answers found so far are probably wrong and impractical is another matter.

Massive vs. limited nuclear use: the pendulum keeps swinging to these two extremes, and the tables keep turning. For all they cherish escalation, when it comes to nuclear planning, Europeans insist that first use be limited and demonstrative, while Americans see it, rather, as massive or, in any case, "militarily significant". As for follow-on use, the Europeans prefer deep strikes into Soviet territory, while Americans would rather limit the geographical scope to NATO territory or the periphery of the Warsaw Pact.See Legge (op. cit. in Ch I, note 5). It seems that European views prevailed on both accounts. On this point, see Chapter I of this paper.

With nuclear war difficult to control and probably impossible to limit, and no longer with any meaningful margin of superiority to exploit, the Alliance's threat of first use looks tantamount to suicide. This leads to the question: is a suicidal threat credible?

Again the question is not new: Massive Retaliation was once found to be suicidal, and therefore an incredible answer to potential Soviet actions that were limited in scope. To say that Flexible Response is incredible because it is based on a suicidal threat has so far sparked no more than nostalgia for better days on the one hand, and endless intellectual exercises to discover residual margins of superiority along fictitious ladders, on the other. Suicidal threats are largely credible: sane people, as one must assume Soviet leaders to be, would not dare call the nuclear bluff on the grounds that NATO strategy is suicidal and as such will never be implemented.

The real trouble with suicidal threats is sustainability. In fact, they call for rather reckless behaviour in world affairs, in order to substantiate the claim of being able to go to extremes. And they ask a lot psychologically, since one must somehow always be ready to make good on the threat. Both attitudes are not acceptable to the Western public. A fact that is apparent for the former attitude and it is increasingly clear for the latter.

Whatever one may think of the peace movements that opposed the INF deployment, it is a fact that they sparked important changes in European thinking about security. These changes are now by and large reflected in the stances of major political forces, like the British Labour Party and the German Social Democrats. Insofar as nuclear weapons are concerned, it is perhaps symptomatic that in the March 1988 NATO summit of heads of state it was decided that even the word 'modernisation' could be poorly received by the European public. The concept was thus watered down to "keeping up to date when necessary."NATO Press Service, "Declaration of the Heads of State and Government Participating in the Meeting of the North Atlantic Council in Brussels (2nd-3rd March 1988)," Press Communique M-1 (88) 13.

Europeans will probably continue to subscribe to the following assertion: "What matters most is to concentrate not only on the prevention of nuclear war, but on how to prevent any war, conventional war as well."Karl Kaiser, Georg Leber, Alois Mertes and Franz-Josef Schulze, "Nuclear Weapons and the Preservation of Peace," Foreign Affairs, Summer 1982 (emphasis in the original).

What they are increasingly unlikely to buy are the most common implications of that assertion, namely that in order to deter any war one has to accept a low nuclear threshold, substantiated by the presence of thousands of uncontrollable and suicidal nuclear weapons.

In sum, the only way to keep the pendulum from swinging back and forth from massive retaliation to flexible response is to accept the notion that nuclear weapons' doctrine and force levels must be based on one simple requirement: deter an opponent's nuclear use. The minimum deterrent that would result from it does not require thousands of TNW and would probably fit much better the C3 systems that NATO is realistically able to field. Conventional war should be deterred through the planning of conventional operations, although nuclear weapons will continue to dissuade from any war by their mere existence, no matter how 'high' the nuclear threshold may be perceived in a world of minimum deterrence.

Abbreviations and Acronyms

AUTODIN Automatic Digital Network

AUTOVON Automatic Voice Network

ACCS Air Command and Control System

ACE Allied Command Europe

ADM Atomic Demolition Munition

AFSATCOM Air Force Satellite Communications

AIRSOUTH Allied Air Forces Southern Europe

ASW Anti Submarine Warfare

ATAF Allied Tactical Air Force

AWACS Airborne Early Warning and Control System

BAOR British Army on the Rhine

BMEWS Ballistic Missile Early Warning System

C3 Command Control and Communications

CENTAG Central Army Group

CINCEASTLAN Commander in Chief Eastern Atlantic Area

CINCEUR Commander in Chief Europe

CINCHAN Commander in Chief Channel

CINCSOUTH Commander in Chief Allied Forces Southern Europe

CINCUSNAVEUR Commander in Chief U.S. Navy Europe

DEB Digital European Backbone

DPC Defence Planning Committee

DSCS Defence Satellite Communications System

DSP Defence Support Programme

EAM Emergency Action Message

ECCCS European Command and Control Console System

ELINT Electronic Intelligence

EMP Electromagnetic Pulse

ER Enhanced Radiation

EUCOM European Command

FGA Fighter Ground Attack

FLTSATCOM Fleet Satellite Communications

FOB Forward Operating Base

FOL Forward Operating Location

FOSIC Fleet Ocean Surveillance Information Centers

GLCM Ground Launched Cruise Missile

HF High Frequency

IFF Identification Friend and Foe

INF Intermediate-range Nuclear Forces

IVSN Initial Voice Switched Network

JSTPS Joint Strategic Target Planning Staff

LANTCOM Atlantic Command

LERTCON NATO Alert Condition

MARAIRMED Maritime Air Force Mediterranean

MEDOC Western Mediterranean Area

MEDSOUEAST South East Mediterranean Area

MLF Multi Lateral Force

MOB Main Operating Base

MSE Mobile Subscriber Equipment

NADGE NATO Air Defence Ground Environment

NAEW NATO Airborne Early Warning

NAVSOUTH Allied Naval Forces Southern Europe

NCA National Command Authorities

NDAC Nuclear Defence Affairs Committee

NICS NATO Integrated Communications System

NOP Nuclear Operations Plan

NORAD North America Aerospace Defence Command

NPG Nuclear Planning Group

PACCS Post Attack Command and Control System

PACOM Pacific Command

PAL Permissive Action Link

QRA Quick Reaction Alert

SAC Strategic Air Command

SACEUR Supreme Allied Commander Europe

SACLANT Supreme Allied Commander Atlantic

SAM Surface to Air Missiles

SATCOM Satellite Communications

SCARS Status Control Alerting and Reporting System

SDI Strategic Defence Initiative

SHAPE Supreme Headquarters Allied Powers Europe

SIGINT Signal Intelligence

SIOP Single Integrated Operational Plan

SLBM Submarine Launched Ballistic Missile

SLCM Sea Launched Cruise Missile

SSBN Submersible Ship Ballistic Nuclear

SUBMED Submarines Mediterranean

TARE Telegraph Automatic Relay Equipment

TNW Theatre Nuclear Weapons

UHF Ultra High Frequency

USAEUR U.S. Army Europe

WWMCCS World Wide Military Command and Control System