Document Number (FOIA) /ESDN (CREST): 
Document Page Count: 
Document Release Date: 
July 30, 2014
Case Number: 
Publication Date: 
December 1, 1981
PDF icon DOC_0000617249.pdf946.16 KB
Approved for Release: 2014/07/29 000617249TITLE: CIA Versus DOD: Competing MisconceptionOf Strategic AssessmentAUTHOR:(b)(3)(c)VOLUME: 25 ISSUE: Winter YEAR: 1981(Approved for Release: 2014/07/29 000617249 A collection of articles on the historical, operational, doctrinal, and theoretical aspects of intelligence. ?All stfatements of fact, opinion or analysis expressed in Studies in Intelligence are those ofthe authors. They do not necessarily reflect official positions or views of the CentralIntelligence Agency or any other US Government entity, past or present. Nothing in thecontents should be construed as asserting Or implying US Government endorsement of anarticle's factual statements and interpretations.Approved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249The Psychological Assumptions areRarely ArticulatedCONPQENTIALCIA VERSUS DOD: COMPETING MISCONCEPTIONS OFSTRATEGIC ASSESSMENT*(b)(3)(c)Aping the adversary's strategic capability -suggests to observers, notdetermination, but rather the lack of assurance of the frontier tenderfootin the Western movie who carries two oversized guns to town todemonstrate his readiness for 'high noon'."Maxwell TaylorRecent changes in American strategic doctrine have prompted .shifts in theyardsticks used to gauge strategic sufficiency. The changes in doctrine have also led toa heightened perception of the importance of these measures. This paper examinesmeasures of the strategic balance now in vogue in the Central Intelligence Agency andthe Department of Defense. It argues that they are inaccurate, misleading, and evendetrimental to the success of deterrence.Strategic forces mean nothing in the abstract. They must be assessed in terms ofthe missions they are expected to perform. For the United States this mission hasalways been deterrence, although deterrence of what has never been altogether clear.From time to time officials have spoken of extended deterrence, supposing thatAmerican strategic forces could prevent or at least discourage military action by theSoviet Union against the United States and its allies. At other times, these forces havebeen described as having the more circumscribed task of deterring a nuclear attackupon the United States. Former Defense Secretary Harold Brown's statements in hislast Annual Report for 1981 illustrate the fuzziness that surrounds the officialconception of deterrence. Twice in this document he declares: "Despite some initialillusions, most of us have recognized for many years that strategic nuclear capabilitiesalone could only deter a narrow range of contingencies." These he defines "as nuclearattacks on the United States, our forces overseas, and on our friends and allies." On thesame page, however, where this statement appears the second time, he states that"Nuclear forces also contribute to some degree, through fear of escalation, todeterrence of nonnuclear attacks." Sandwiched in between two iterations of the firststatement is yet another pronouncement on deterrence that extends its scope evenfurther. The Secretary declares that it is capable not only of "discouraging recklessaction in a crisis" but of "minimizing aggressive behavior over the long term." [1]To be fair to Harold Brown, we must acknowledge that his lack of precision aboutthe political utility of strategic forces is but a reflection of the ambiguity on thissubject in the government, the media, and academia. In part this is due to the natureof the beast; it is very difficult to know just what impact a state's strategic arsenal hason its friends or adversaries. But such ambiguity also reflects our failure to developadequate theory about the psychological relationship between threat and response andforce and will. In the absence of good theory, efforts to determine the utility ofReprinted from Contra, September 1981.CONFID TIAL 9Approved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CONF NTIAL CIA versus DODdearrence often verge on the tautological. The official literature is full of statementsto the effect that "deterrence has worked because the Russians have not invadedWestern Europe." Even more sophisticated attempts to assess deterrence generally failto spell out either the psychopolitical mechanisms by which it is supposed to operate orthe levels at which it is assumed to be effective. [2]These studies fail to provide a logic for formulating strategic missions or designingforce structures. Successive administrations have accordingly succumbed to thetemptation of devising strategic doctrine in response to the capabilities of theirweapons. This is a complete reversal of the Clausewitzian dictum that militarycapability should reflect the political objectives it is designed to achieve. Nothingillustrates this point better than Presidential Directive 59. The new doctrine was littlemore than a recognition of the ways in which a series of incremental technologicalchanges transformed the capabilities of American strategic forces. Each increment ofchange was the result of an innovation that improved the performance characteristicsof some component of those forces. Collectively, these improvements give the UnitedStates the capability to carry out a sophisticated range of counterforce options into theSingle Integrated Operational Plan (SIOP). Doctrine was belatedly developed siALpromulgated by the Defense Department and the White House to take these changesinto account. It was made public at a time the Carter administration hyped to impressthe public with its toughness toward the Russians. [3]StrategicEven though successive administrations have not been clear about what they hopeto deter, they have not hesitated to redefine the force requirements thought necessaryto do the job. In the 1960s, when McNamara decreed that the ability "to destroy theSoviet Union as a 20th century nation" was sufficient to guarantee deterrence,strategic sufficiency was measured in terms of the capability to wipe out an arbitrarilydetermined number of Soviet cities. As the Russians possessed no capability duringmost of this period to attack the continental United States, little thought was devotedto the credibility of the US commitment to use nuclear weapons in retaliation.American strategic planners could also be confident about their ability to attain theirstipulated operational objectives as the Soviets lacked the means to defend themselvesagainst bombers and, later, nuclear submarines and intercontinental missiles. Theability of bombers to penetrate Soviet air defenses was the only question that arousedenduring concern. But this was partially allayed by. the redundancy of Americandelivery systems and for a while the relative invulnerability of ICBMs and SLBMs.The adoption of "essential equivalence," an outgrowth of the SALT process, ledto a corresponding shift in the standards used to determine strategic sufficiency. As adoctrine, essential equivalence divorced the meaning of strategic forces from thewartime missions these forces might be called upon to perform. Instead of usingmission as a yardstick as McNamara had, the Nixon and Ford administrations soughtto assess strategic sufficiency by comparing American forces to present and projectedSoviet force levels. Several static comparisons of forces were devised for this purpose,among them numbers of missiles, launchers, deliverable warheads, and equivalentmegatonnage. [4]Static indicators proved to be inescapably murky measures of strategic suffi-ciency. They ignored the reality that both the Soviet Union and the United States haddeveloped their forces in response to different political needs, exploited differentlevels of technology in their design, and developed different doctrines for theiremployment. Soviet and American strategic forces represent the expression of twodifferent societies, each with its own political-military tradition, distinctive world10 CONF ENTIALApproved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CIA versus DOD COt?-?'..fFtgatITIALvie*, and means of evaluating its security. Direct force comparisons, regardless of theparticular measures used, are largely meaningless. As the SALT debate revealed, theyalso have the disadvantage of being politically very charged.For these reasons the Intelligence Community, the Department of Defense, andthe Carter White House became disenchanted with both essential equivalence andstatic indicators. In January 1979 Harold Brown summed up this consensus in theintroduction to his Annual Report: "In designing our strategic nuclear forces, what weneed for deterrence and stability cannot be dictated by any simple comparison withthe forces of the Soviet Union, even though we must take those forces into account inour planning." [5] In the place of static indicators, defense planners and intelligenceanalysts developed more dynamic measures of the strategic balance, measures thatattempt to take into account the performance characteristics of the forces on bothsides. This dynamic analysis attempts to calculate the results of both sides allocatingstrategic forces against specific target sets. It uses the outcomes of these theoreticalexchanges to describe the existing state of the strategic balance and offers them asbenchmarks against which to formulate strategic needs.The appeal of dynamic measures of strategic sufficiency has been enhanced 153,'recent changes in doctrine. The "countervailing strategy" is based on the premise thatdeterrence requires the United States to be able to deny the Soviet Union any possiblerelative advantage that it might derive from a nuclear war, limited or all-out. Thedoctrine emphasizes the targeting of Soviet missile sites, command and control centers,and other military assets. Static indicators of the strategic balance are incapable ofshedding light on the extent to which the United States can destroy these targets.Qualitative asymmetries in the strategic forces of the two superpowers also mean thatstatic indicators are useless as a means of determining their relative capability todestroy given target sets. Dynamic analysis seeks answers to both these questionsthrough the mechanism of theoretical force exchanges. It attempts to assess absolutecapability in terms of the damage each side can inflict upon the strategic, military,and economic assets of the other. It measures relative capability as a function of thedestructive potential each side retains after absorbing a first strike.Dynamic measures have been made the cornerstone of the most recent efforts bythe Central Intelligence Agency, the Studies and Gaming Agency of the Joint Chiefs ofStaff, and the RAND Corporation to determine the relative strategic capabilities of theUnited States and the Soviet Union. The CIA, which pioneered the use of dynamicanalysis, has employed it in its National Intelligence Estimates of Soviet strategiccapabilities (NIE 11-3-78 and NIE 11-3-81). In these Estimates, Agency analystscalculate the potential of each side's ICBMs to attack the retaliatory forces of the otherand then model such attacks in order to determine residual destructive potential.Residual destructive potential is defined as the gross number of missile reentryvehicles (RVs) and bomber weapons that each side could be expected to retain afterhaving been attacked. These remaining weapons are then compared on the basis oftheir lethal area potential and hard-target potential, that is, in terms of their utilityagainst both area and point targets. These estimates are calculated under conditions ofsurprise and preemption. Surprise refers to a "bolt from the blue" attack, launchedwhen both sides were in a day-to-day attack, that arises out of a crisis situation inwhich both sides have generated forces. [6]The National Intelligence Estimates that incorporate this analysis make nopretence that it reflects the outcome of an actual nuclear exchange. The Estimatesacknowledge several ways in which it is artificial. First of all, there are the kinds offorces used. Only ICBMs and SLBMs?and bombers for the Americans?are em-ployed in the initial strike, and they are all targeted against the other's missile silos,CON FIIYENTIAL 11Approved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CONI-111SIAZAL CIA versus DODbombs& bases, and submarine pens in order to reduce as much as possible itsretaliatory capability. In a real attack both sides, but especially the United States,could make use of other kinds of forces. The United States has significant strategicassets in Europe and the Far East, on land and at sea, that could be used to augment astrike against the Soviet Union.Most strategists also expect that both superpowers would attack a wider range oftargets than missile pads and bomber and submarine bases. Soviet and Americandoctrines explicitly call for the destruction of each other's military assets, in particulartheir command-and-control facilities and conventional force capabilities. The esti-mates also use admittedly arbitrary targeting doctrine; two-on-one ICBM attacks onsilos and one-on-one attacks against the projected MX shelters. As with the choice oftargets this is a reflection of the damage-limiting objective assumed for the purposes ofthe analysis to motivate both sides. For the same reason the Estimates assume that bothsides ride out a first strike, whereas in a real exchange it seems more likely that theywould attempt to launch from underneath an attack. The Soviets actually espouse a"launch-on-warning doctrine- and have made considerable strides in the last fewyears toward developing the capability to do this. Finally, the estimates make no,..attempt to calculate the reliability and accuracy of systems or of command-and-control networks in wartime conditions. The figures used reflect peacetime perfor-mance and are based on the results of tests that both sides carry out under relativelyideal conditions.CIA Versus DoD: How they differFor some or all of the reasons noted above many strategists are dissatisfied withthe results of the CIA's strategic analysis. Liberal critics charge that the Agency'sadmittedly unrealistic assumption of a damage-limiting strategy for both sides has theeffect of exaggerating American vulnerabilities to a Soviet first strike. Conservativecritics allege that the Agency model minimizes the Soviet threat by failing to take intoaccount more realistic and advanced Soviet preparations to fight a nuclear war. TheDefense Intelligence Agency and the senior intelligence officers of the militaryservices have gone on record with this complaint.The Agency defends its dynamic analysis in terms of its contribution to assessingthe efficacy of the American, deterrent. The key requirement for deterrence, asdefined by successive Secretaries of Defense, is the capability of the United States toabsorb a first strike and still retain sufficient forces to destroy a broad mix of Soviettargets. The CIA argues that residual destructive potential, calculated in both absoluteand relative terms, is the most direct measure of this capability and that trends withrespect to it are an appropriate indicator of the overall strategic balance. Put anotherway, deterrence is above all a psychological phenomenon, a state of mind that resultsfrom the belief of Soviet and American leaders that they have everything to lose froma nuclear war. The assumption is that leaders' perceptions of the outcome of a nuclearexchange do not derive from a detailed understanding of what might really happen insuch a conflict but depend upon an abstract and relatively simple conception of thestrategic balance. To the charge that its analysis is artificial, the Agency could replythat its purpose is not to simulate an actual exchange but to derive numbers for anidealized measure of strategic balance?residual destructive potential?that itself hasmore symbolic than operational significance.The Defense Intelligence Agency (DIA) and the three services with strategicmissions insist that only the analysis of a comprehensive two-sided exchange, modeledon more realistic assumptions about doctrine and performance, can convey validimpressions about the relative strategic capabilities of the superpowers. This may be12Approved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CIA versus DOD CON? NTIALtruelbd such an approach is entirely scenario dependent. In addition to knowing wholaunches the first strike and whether one or both adversaries have their forcesgenerated or on day-to-day alert, it requires detailed knowledge of their attack-and-response strategies as well as accurate information about the wartime performance oftheir weapon systems and command-and-control networks. There is no consensus inthe strategic community as to how to estimate the outcome of such a conflict. Theproblem is compounded by the fact that a nuclear war fought in the late 1980s wouldinvolve weapon systems that have not yet been deployed, or even tested in someinstances, by either side. Nevertheless, the choice of attack strategy and the assignedperformance characteristics of the weapons used determine the results of anysimulation.The uncertainty associated with efforts to make a realistic model of a "compre-hensive" exchange of forces deprives such estimates of much of their analytical value.An analyst can come up with almost any outcome he wants if he chooses theappropriate doctrine, rules of engagement, or indices of operational performance. Thenature of this problem is readily apparent in the reaction within the strategiccommunity to efforts carried out under Defense Department auspices to model-enuclear conflict. Briefings on these exercises rapidly degenerate into heated confronta-tions about the methods and numbers used to build and test the model: If Americanstrategists cannot agree among themselves about how to simulate a nuclear war, it isreasonable to suppose that whatever consensus the Intelligence Community eventuallyreaches will bear only a chance resemblance to the conclusions of their Sovietcounterparts, who would approach the problem with their own doctrinal assumptionsand data bases. The entire purpose of assessment, however, is to analyze relativestrategic capabilities through Soviet eyes in order to assess the health of our deterrent.When confronted with this dilemma, the artificiality of the CIA analysis, singled outas its major drawback, might fairly be described as its greatest virtue.On one level the conflict between the CIA and the Department of Defense is astruggle over turf: the services contend that the Agency's dynamic analysis represents a"net assessment," a prerogative of the Defense Department. But the conflict alsorepresents the clash of two different intellectual approaches to the problem. The CIAapproach stresses the importance of the psychological roots of deterrence. Agencyanalysts have accordingly constructed stylized scenarios for force exchanges that arekeyed to this conception. DIA and the services conceive of deterrence in moreoperational terms. They accordingly want to calculate, as realistically as possible, thelikely outcome of a nuclear exchange. For them the most important indicator ofdeterrence is the military balance that would prevail after the mushroom cloudsdissipate, measured not just in terms of total forces but in the relative capability ofeach side to use those forces in a coordinated and intelligent manner.Proponents of both approaches justify their choice by asserting that it moreclosely approximates the way in which the Soviets assess relative strategic capabilities.The military and its supporters argue that Soviet adherence to a war-fighting doctrinecoupled with the preparations they have allegedly made actually to fight a nuclearconflict reveal the need to assess deterrence in terms of the outcome of such a war.CIA analysts who favor more abstract measures of strategic capability stress thedeterrent objective of this war-fighting doctrine. In point of fact, Soviet intentions andeven more so, their judgments about our capabilities and intentions are sufficientlyobscure to permit analysts to make a case for either point of view. This leeway tends tomake strategic assessments theological exercises carried out to advance the parochialinterest of the institutions the analysts represent. Even when this is not the case,assessments still tend to reflect ingrained institutional ways of looking at the world. Putcrudely, the sevices are socialized into judging everything in terms of militaryCONFI NTIAL 13Approved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CONPtQENTIAL CIA versus DODcapatfiliti'es, while analysts at CIA, depending upon their particular office, displayvarying degrees of sensitivity to the political side of things. Individual analysts in eachof these institutions on the whole are more sensitive to information that suggests thatthe Soviets see the world in the same way that their particular subunits of thebureaucracy do.CIA and DoD: What They Have in CommonSo far our investigation has focused on the ways in which the CIA and Defenseconceptions of assessment differ. But there are also important underlying affinitiesbetween them. Both assess deterrence as principally a military relationship. Both userelative residual military capability, albeit calculated in different ways, as theirmeasure of that capability. Both assume that this approach to deterrence reflects theSoviet way of thinking. The approaches differ only in their technical details, in thekinds of residual capability that are considered important, and the ways in which theyare to be measured.The psychological assumptions that the two approaches share are rarely articu-lated, let alone critically evaluated by those who use them. The reason for this is prob;=--ably the lack of controversy associated with them within the intelligence andpolicymaking community. By contrast, the kinds of residuals that are important andthe means of determining them have been the subject of extensive analysis because ofthe controversy that surrounds them. The nature and validity of their psychologicalassumptions are nevertheless the more important questions, for the choice ofassumptions has more bearing on the outcome than the choice of means used.The psychological assumptions common to both the CIA and Defense approachesto strategic analysis do not stand up under scrutiny. Take the use both make ofresidual military capability. There is nothing wrong with residuals per se; theyindicate something about the military balance that might prevail in the aftermath of astrategic exchange. Both approaches, however, go far beyond this by attempting toemploy residual capability as a measure of will.Nuclear deterrence depends upon an adversary's belief that his opponent, ifattacked, will carry out his threat to use nuclear weapons. Both the CIA and itsDefense Department critics assume that the best indicator of the probability of thatresponse is the relative strategic capability of the adversaries. The costlier the response,the less likely it becomes. Both sets of analysts believe that deterrence will no longer becredible if the relative cost of reprisal is very great. In such a situation they conceive ofthe deterrer as being deterred from retaliation by virtue of the superior strategicpower of his adversary. This is the principle of which Paul Nitze and others havebased their scenarios of a disarming Soviet first strike. They worry that if the Sovietsdestroy the Minutemen, and with them the American counterforce capability, thePresident would be compelled to accept the attack as a fait accompli becauseretaliation against Soviet cities, his only real option as they describe it, would elicitequal if not more destructive retaliation in return. [7]While cost is certainly one component of will, it is wrong to assume that it is theonly component or even the most important one. The measurement of will in terms ofcost reflects a narrow formulation of the decisionmaking process. It rests on theassumption that people, national leaders in this instance, maximize their "utility"regardless of the stress or uncertainty associated with a decision and its possibleoutcomes. It further assumes?and this is quintessentially American in approach?thatutility can be equated with material and physical well-being. This disregards a host ofother values?emotional, intangible, unquantifiable?that history reveals to be at leastas important for most people, including Americans. Would the South, for example,14 CON1Akt1TIALApproved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CIA versus DOD COt1DENTIALhaverrebelled if material well-being and physical comfort were the primary concernof its leaders and people? Why did the Confederacy continue the struggle attremendous human and economic cost long after leaders and soldiers alike recognizedit to be a lost cause? Any number of other instances can be cited where'a people wit-tingly fought a determined struggle against great or even impossible odds. FromMasada to the Irish Easter Rising, from Thermopylae to the resistance of thebeleaguered Finns in 1940, history records countless stories of peoples who wagedcostly struggles with little or no expectation of success. Honor, anger, or national self-respect provided more compelling motives for action than did pragmatic calculationsof material loss and gain. At the same time, the French experience in Algeria and thatof the American debacle in Vietnam reveal, if ever proof were needed, that superior,even vastly superior military capability, is no sure indicator of resolve. [8]National will is a social amalgam of many components. Among the mostimportant of these are probably the nature of the adversary, what people believe is atstake, the cohesiveness of the society, its historical traditions, and the character of itsleadership. Depending upon the circumstances, any of these considerations can bemore influential in determining national will than calculations of relative militacapability. Several examples will illustrate this point. In 1882 the British fleet subduedEgypt by firing a few cannon balls at the Khedive's palace in Alexandriann 1956, fol-lowing the massive and successful Suez invasion, the British and French withdrewfrom Egypt because they were forced to recognize that their armies were relativelyuseless weapons against aroused nationalist opinion. In 1940, France was defeatedrapidly because the rise to power of Hitler had aggravated the deep and bittercleavages within French society, cleavages that came close to paralyzing its army andgovernment. Britain, by contrast, became a more cohesive society in response to theNazi threat. Even if Germany had successfully invaded the British Isles, it is likely thatits people would have continued to resist domination. The Czechs and the Poles alsoprovide contrasting examples of national will, in this case perhaps attributable todifferent national traditions. Twice in recent memory, in 1938-39 and 1968,Czechoslovakia capitulated without a fight. In 1938, Prague had the military means toresist and possibly force France and Britain to come to its aid. The Poles waged fierceand futile struggles against foreign domination in 1848, 1939, and 1944-45. Theexpectation that they would do so again has probably prompted Soviet leaders toexercise more caution in the current crisis than might otherwise have been the case.These examples should make apparent that there is at best only a marginalassociation between military capability and national will. It is therefore quiteerroneous to use a measure of relative military capability to determine whether or notthe United States would respond to a Soviet attack. It is also farfetched to assume thatthe Soviets would calculate American intentions in this way; that in the aftermath of aSoviet nuclear strike they would expect an American President to tote up the residualmegatonnage or RVs of both sides and sue for peace or even surrender if his side cameup short. If the Soviets behaved this way, they would have rolled over and playeddead in 1941 following the initially devastating results of Hitler's onslaught againstthem.By dint of their own national experience, Russians, more than most people, mightbe expected to understand the role of leadership and morale?what they refer to intheir writings as -moral forces--in determining the resolve of an army or people.Soviet leaders are likely to ask themselves just how much would the American peoplebe prepared to sacrifice in order to preserve their influence in Europe, theindependence of their closest allies, or their own domestic integrity? What about theadministration's mettle? Is the President the kind of person to recoil in horror fromusing America's nuclear arsenal regardless of the circumstances, or would he have theCONPQENTJAL. 15Approved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CONEt&NTIAL C/A versus DODcourZge or insanity?take your pick?to give the launch order in response to a Sovietattack or even confirmed preparations for an attack? Given the incredibly stressfulcontext in which such a decision would be made, the character and personality of thePresident is almost certain to be a more decisive factor in influencing the decision tolaunch than arcane calculations of relative military advantage handed to him by amilitary adviser. The extent that Soviet leaders take such considerations into account,or at least recognize their importance, their analysis of resolve is more in accord withthe realities of human behavior than the narrowly technical approach of theirAmerican counterparts.The preceding discussion indicates that most of the really important componentsof resolve are scenario dependent. This brings us to our second important criticism ofthe use of residuals as a measure, of deterrence: they measure the results of war interms of the means by which it is waged instead of the ends it seeks to achieve. Thisdivorces military conflict from the political context in which it occurs and from whichit derives meaning.Wars rarely start because one side believes that it has a military advantage.occur when when leaders become convinced that force is necessary to achieve importantgoals. War, as Clausewitz observed?and the Soviets proclaim?is an'extension ofpolitics by other means. Its scope, strategy, and timing are determined, or ought to be,by the political objectives for which the war is fought. It follows from this that anyassessment of deterrence must be predicated upon some notion of the objectives thatthe Soviet Union might go to war to achieve. Only then can we attempt to determinewhat must be done to deny the Soviets these objectives and thereby deprive them ofwhatever incentives they might have for going to war.Soviet objectives are an extremely contentious question and perhaps the reasonwhy analysts have shied away from this approach. Paul Nitze, Richard Pipes, andother "Team B" types argue that the Soviets are out to conquer the world and are will-ing to pay the necessary price to do it. Most analysts take a more moderate view; theyargue that Moscow would only resort to war to protect its vital interests. Needless tosay, there is no agreement among even these analysts as to the nature of these interestsor the ways in which a war might come about.One way around this problem might be to recognize that the -bottom line" forthe Soviet Union, indeed for any country, in a nuclear war must remain thepreservation of its political system and territorial integrity. This would be the primaryobjective in a defensive war and the obvious precondition for the attainment ofobjectives in an offensive war. Recognition of this reality has important implicationsfor deterrence because the preservation of a state's political system and territorialintegrity is likely to be a much more difficult task in a nuclear war than in aconventional one.A nuclear war between the superpowers is almost certain to far exceed the mostferocious conventional war in its destructiveness. Even a "limited" nuclear warbetween the superpowers consisting only of strictly counterforce exchanges can beexpected to leave perhaps 20 million dead on each side. A more general exchangecould kill more than 100 million people in each country and deprive the survivors ofthe means of restoring their standard of living within their own lifetimes. To thephysical destruction of nuclear war, whether limited or general, must be added thepsychological incapacitation of the survivors. This is a phenomenon as yet littleunderstood, but nevertheless likely to have a serious retarding effect upon a society'sability to reconstitute itself.16 CON ENTIALApproved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249C/A versus DOD 6014RLittITIAL)?tWe have no data on the social consequences of a massive nuclear strike.Hiroshima and Nagasaki, mankind's only experience with nuclear warfare, were merepinpricks in comparison to the death and destruction that would result from even alimited nuclear exchange between the superpowers. Because of our lack of knowledgeabout the social and psychological repercussions of nuclear war?and perhaps for fearof what we would discover?the American strategic community has largely ignoredthis aspect of the problem. Instead strategists have based their calculations and theirdoctrine on technical phenomena about which they feel more confident. Residualweaponry, circular errors of probability, and blast overpressure have been made thebedrock of American strategic analysis. Such an engineering approach to the subjecthas blinded analysts to some important social truths about nuclear war.The first of these pertain to the disruptive destructive effects of nuclear war uponsociety. These are certain to be greater than is recognized by the official or quasi-offi-cial literature on the subject. As it is, these estimates of damage have been criticizedfor minimizing fire and radiation fatalities and for disregarding deaths arising fromsecondary causes such as the disruption or even breakdown of public order, andsanitary, medical, and food distribution facilities. Some recent studies suggest thatr..._postattack deaths may not peak until a week or more after an attack. Yet most officialcasualty projections calculate only first-week fatalities. Most, if nor all, of theseestimates also ignore the psychosocial cost of a nuclear holocaust, the price paid not bythe victims but the survivors. [9]Robert Jay Lifton, who studied the survivors of Hiroshima, identified what hecalled the "death in life" syndrome: "a grotesque, absurd, collective, unacceptable,and unabsorbable death" that survivors carry with them for the rest of their lives. [10]Even those who were able to pursue successful careers continue to be affected by "anunending lethal influence," a sense of being the victim of a force that threatens thespecies. Hiroshima, it must be emphasized, was the victim of a primitive low-yieldweapon and subsequently, the recipient of long-term medical, economic, and socialassistance from a society left untouched by nuclear devastation. The psychologicalaftereffects of a nuclear war between the superpowers are certain to be morepronounced and more widespread than those associated with Hiroshima. They maypose as great a challenge to the long-term survival of a society as the physicaldestruction caused by blast, fire, and radiation does in the short term.Nuclear war may leave survivors in such a profound state of shock that in its im-mediate aftermath they cannot respond to whatever political and military authority isable to reassert itself. This phenomenon will be all the more likely to occur if peoplehave not been adequately prepared beforehand for the horrors of the postattackenvironment. Survivors who are not psychologically incapacitated may be very hostileto authority if they see it as responsible for the catastrophe. In conventional warsgovernments usually attempt to rally support-by portraying themselves as the defenderof all that is sacred.Conventional wars retain a human flavor regardless of their scale or level ofdestruction. Combatants and civilians are aware that everything that happens, nomatter how unintended in consequence, is the result of human decisions and action.[II] They also "learn" something about the enemy in the course of the war, eitherfirsthand or through the media. They can form an "enemy image" upon which to dis-place their own frustrations, insecurities, and aggressive drives. Knowing this,governments have often gone to war to deflect such hostility from themselves. Theyalso count on the enemy's actions, for example, bombing raids or atrocities againstcivilians, to rebound to their domestic political advantage. As the German onslaughtagainst Russia in World War II revealed, even the most unpopular government can, inthe long run, benefit from the depredations of a barbarous invader.CONFTBENTIAL 17Approved for Release: 2014/07/29000617249 Approved for Release: 2014/07/29 000617249COt,ENTIA1 CIA versus DODXri intercontinental nuclear war will be different. It will be short, perhaps lastingonly hours or days. It will also be highly impersonal, as neither the adversary whopushed the button nor the missiles that deliver his warheads will be seen. There maybe no combat at all in the traditional sense, and certainly none that will impinge uponthe consciousness of the heartland populations. This will deprive soldiers and civiliansalike of a palpable image of the enemy, so necessary to make the devastation of warpsychologically comprehensible.Nuclear war will also differ from conventional war by virtue of the magnitude ofits destruction. The psychological effects of say 20 million dead in the course of anhour and a half are certain to be different and greater than the impact of the samenumber of fatalities sustained during the course of four years of war. This combinationof impersonality and instantaneity may make nuclear war more closely resemble anatural disaster than a war in the minds of the survivors. If this is true, it may havesome hitherto unanticipated but important political consequences. Studies of naturaldisasters reveal that those who warned of disaster are likely to become the focus ofacute hostility when their predictions turn out to be correct. [12] This is so, onepsychologist speculates, because people tend to interpret disasters as forms of persona..?punishment and warnings as threats of punishment. [13] Thus, a government thatattempts to prepare its people for the possibility of a nuclear war during an escalatingcrisis may be punished by the survivors as a result.The preceding argument is admittedly speculative. It highlights, however, asecond important truth about war in general and nuclear war in particular: thesurvival of any political system in war is at least as much a function of its legitimacy asit is of its military capability. To the extent that a government is seen as culpable forwar and by extension for the damage resulting from it, its chance of survivaldiminishes. This is one more reason why, for purposes of assessing deterrence, militarymeans cannot be divorced from the political context in which they might be used. Thiswill determine whether or not a government will be supported by its people, and suchsupport seems essential for postattack conventional military operations.Astute statesmen have always been sensitive to the need to develop and maintainpublic support in war. Bismarck's manipulation of the Hohenzollern candidacy andhis editing of the "Ems dispatch" are often cited as examples of how a clever politicalleader can manipulate an adversary in order to generate public enthusiasm for war.Our own national experience offers examples of how public opposition to war can actas an effective check upon leaders. Woodrow Wilson was prevented from enteringWorld War I until Germany's unrestricted use of submarines in 1917 broughtAmerican public opinion around. Franklin Roosevelt was even more constrained bypublic antipathy to intervention in World War II. His constitutionally dubious andintentionally provocative efforts at goading Hitler into declaring war on the UnitedStates got nowhere until the Japanese bombed Pearl Harbor. More recently, theAmerican failure in Indochina was at least in part attributable to the inability ofsuccessive presidents to maintain popular support for that war.While democracies are more vulnerable than authoritarian governments to thevagaries of public opinion, the latter must also demonstrate some degree of sensitivityto public sentiment in order to maintain support for military sacrifices. Stalin wasaware of this need and in the fall of 1944, for example, gave considerable pause beforeadvancing beyond the borders of the Soviet Union into Eastern Europe for fear of itseffect upon Soviet conscripts. [14] Afghanistan may yet illustrate the vulnerability ofthe Soviet system to public opinion if the Red Army continues to sustain a steady rateof casualties without bringing an end to that war.18 CO>1131TIALApproved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CIA versus DOD COrl fib??ITIAL*The) staggering loss of life and destruction that can be expected to result fromeven a limited nuclear war make it advisable, perhaps even essential, for leaders,regardless of what kind of political system they manage, to "prepare- their people forsuch an event. This would require educating the public about the nature and effects ofnuclear war. A carefully orchestrated public relations campaign designed to convincepeople that their government, if it resorts to nuclear weapons, will do so only as a lastdesperate measure in defense of the country might also be considered. Serious effortsto prepare people for nuclear war, however, are likely to be psychologically unsettlingto those being instructed and politically damaging to those responsible for theinstructing. Perhaps for these reasons neither the United States nor the Soviet Union,which has an extensive civil defense program, have chosen to enlighten theirpopulation about the realities of nuclear combat.The survival of a political system in wartime and its ability to reconstitute itself inthe aftermath of nuclear attacks are also influenced by conditions independent of thespecific political context of the war. The previous legitimacy of the regime, the natureof its economic base, the extent of its ethnic or racial tensions?what Stalin might havecalled -permanently operating factors"?will play a crucial role in this regard. Any as7,-sessment of deterrence must take them into account, not only objectively but more im-portantly in terms of how they are perceived by the adversaries. This is admittedly an-other difficult task. A useful beginning can nevertheless be made by describing someof the most important of these factors and the ways in which they might retard or faci-litate the respective political and economic recovery of the superpowers from anuclear war.Viewed from this perspective the Soviet Union has a number of vulnerabilitiesthat could pose serious difficulties for its leaders in the aftermath of a nuclear conflict.For a start, there is Eastern Europe. The Soviets are not popular in any of these coun-tries with the possible exception of Bulgaria. In Poland and East Germany they areloathed by the local population. Both countries are astride the main communicationroutes between the Soviet Union and Western Europe, the most likely theater ofconventional operations. Hostility and obstruction?perhaps even tolerated by localauthorities in Poland?could cause serious problems for the Soviet military. If theSoviet Union were greviously wounded by a series of nuclear strikes, some of thesatellite governments in Eastern Europe might seize the opportunity to establish theirindependence from Moscow. The prospects for this could be substantially enhanced bya Western targeting strategy designed to destroy the tentacles of Soviet power in thesecountries while leaving the national armies and industrial plants and populationcenters relatively intact.Beyond Eastern Europe, the Russians face considerable opposition within theirown multiethnic empire where their political primacy is resented in varying degreesby the other nationalities. In wartime the Soviet nationality problem could becomeacute if dissident nationalities perceived the central government's power to be waning.Resistance to Bolshevik domination in the Civil War and to Soviet rule in World WarH followed this pattern. In this regard, some of the dissident nationalities would profitfrom the fact that European Russia is certain to suffer more devastation than SovietCentral Asia because of its denser concentration of military and industrial targets. TheMuslim and Caucasian minorities in particular can be expected to fare much betterthan their Russian counterparts in terms of their physical survival, usable resources,and access to food. Dissident Muslims and other rebellious groups throughout theSoviet Union would also profit from the fact that Soviet conventional militarycapability, especially its command and control, will be targeted 'for destruction in anyconflict other than the most limited nuclear exchange.19Approved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CON TIAL C/A versus DODtestive nationalities would also benefit from their geographic location. Most ofthem reside along the peripheries of the Soviet Union and, in many cases, alongsidestates that harbor territorial irredenta against the Soviet Union. The number of suchstates has increased since the physical expansion of the Soviet Union during andimmediately after World War II. One must assume, as the Soviets probably do, thatsome of these states, perhaps China, would be tempted to take advantage of Sovietweakness to make good their claims.Economically, the Soviet Union is also susceptible to the disruptive effects of anuclear war because of its highly centralized economy. Planning, production, anddistribution decisions are made by ponderous bureaucracies at the several levels of theSoviet hierarcy. Seweryn Bialer comments: "The key characteristic of the Stalinistmodel of economic growth was its lack of economic self-generating, self-regulating,and adjusting features. To run at all, let alone to perform well, it required anenormous political edifice to provide the decisionmaking and the push, the regulation,supervision, and coordination." [15] The need for direction from above remains thehallmark of the Soviet economy. Its hierarchical structure permits political leaders arange of economic choices denied to leaders in a capitalist society, but it is also mar?,vulnerable to disruption.The economy of the United States might be likened to a worm. Cut off any part,even its head, and the rest continues on and even regenerates the missing section. TheSoviet economy in contrast more closely resembles an animal with a highly developedcentral nervous system. It is capable of complex behavior based on the coordinatedefforts of its many parts, but sever its spinal cord and the rest of the beast quickly dies.During World War II the Soviets performed prodigious feats in moving entireindustries away from battle zones and reconstructing them in safer rear areas, brick bybrick and factory by factory. They could do this because their administrativeapparatus was intact and able to coordinate and coerce the labor force. A nuclear warwould decimate or disrupt the functioning of these cadres so that surviving centers ofproduction would be without organized access to raw materials, energy, or markets.Economic reconstruction, especially in the initial postwar period, could be a slow andtortuous process.The Soviets seem aware of their economic vulnerabilities. Their civil defenseprogram aims at protecting not only the political leadership but economic managersand workers in key industries. Plans for protecting the economy and therebyenhancing Soviet capability for postattack recovery include the evacuation, dispersal,and sheltering of economic cadres and the emergency shutdown and relocation ofessential industrial installations to low-risk areas. The Soviets have also developed someprocedures for evacuating cities and resettling their populations in the countryside,although these measures have never been rehearsed. Evacuation, if successfullyimplemented, would significantly reduce immediate casualties in a war in which citieswere attacked.Critics of "Mutual Assured Destruction" have pointed to the Soviet civil defenseprogram as an indication of Soviet willingness to fight a nuclear war. [16] This is a veryquestionable assertion. A moderate civil defense program, a fair description of theSoviet effort, is no more a sign of intent to wage war than fastening a seat belt in a carindicates a willingness to have an accident. Both are prudent preparations forundesired, even feared, outcomes that the actors in question may believe is beyondtheir power to avert: It may be more revealing to inquire whyAmericans tend to rejectboth seatbelts and civil defense.More pertinent to our point is the question of just how much benefit the SovietUnion would derive from its civil defense effort in a nuclear war. This is clearly20 taf<ITIALApproved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CIA versus DOD CONFtEIA1beyond the scope of the article. For the sake of our argument, however, it is useful totouch upon some of the more important aspects of this question. The first of these isthe fact that all major Soviet civil defense plans require hours to days of advancewarning to implement. They would be useless against an American -bolt from theblue- or in preparations for a similar Soviet strike, as they would telegraph Sovietintentions well in advance of the attack. Large-scale American economic reconstruc-tion would also benefit from a redundant transportation system. The thick net ofhighways, a questionable luxury in peacetime, would prove indispensable in theaftermath of a nuclear war. The highway system is so developed that enough of it maysurvive to allow the minimum necessary movement of goods and people to beginreconstruction. In the Soviet Union where transportation relies principally uponrailroads, the destruction of major railheads and bridges would more effectively haltmovement. Communications with the Soviet Far East would become all but impos-sible as the only land route connecting it to European Russia is the Trans-SiberianRailway and the as yet unfinished Baikal-Amur line. The US highway system confersanother advantage: it has facilitated the more even geographical distribution ofmeans of production. This enhances the prospects of some industrial base survivingeven a second wave assault. Soviet industry by contrast is much more concentrated, asit has developed along major rail lines. Missile sites and major military installations aresimilarly situated, making Soviet industry more vulnerable to collateral damage in alimited nuclear war.ConclusionsThe preceding discussion has brought out three important attributes of deter-rence. All of them reveal the inadequacy of residual military capability, howeverderived, as a comprehensive measure of strategic assessment.First of all, military capabilities reveal nothing about resolve. Resolve is afunction of leadership, a people's traditions and expectations, and their perception ofwhat is at stake. Strategic analysis that ignores the political context of a war ismeaningless for any purpose other than testing and developing operational capabili-ties. Unfortunately most American analysts adopt this narrow approach. This isreflected in official assessments of the strategic balance which analyze decisions forwar or peace in terms of technical calculations of relative military advantage.Secondly, the ends of war?not its means?should be the yardstick of strategicanalysis. The minimum objective of any state in a nuclear war is almost certain to bethe survival of its political system and the maintenance of its territorial integrity. TheSoviet Union may be more vulnerable than the United States in this regard by reasonof its centralized economy, hostile neighbors, and restive minorities. For the UnitedStates, therefore, deterrence rests at least as much upon its relative ability to recoverpolitically and economically from a nuclear war as it does upon its relative militarycapability to inflict damage. One can fairly dispute the precise advantage that eithersuperpower enjoys with respect to recovery or the extent to which they accurately per-ceive the reality of their situation. Any assessment of deterrence that ignores suchconsiderations, however, is incomplete and misleading.Finally, nuclear war, even limited nuclear war, is certain to be more destructivethan most studies of-the subject acknowledge. For in addition to the immediate effectsof blast, fire, and radiation there are the longer term effects of radiation, thebreakdown of health care, food processing, and distribution systems and the psycho-logical impairment, temporary and permanent, of the survivors. Historians suggestthat it took almost a century for Europe to recover from the effects of the BlackDeath. The effects of a nuclear war could prove more enduring.C01.101156.QpAL 21Approved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CS-RMS1<TIAL CIA versus DODLeaders of nuclear nations are more likely to be deterred from using nuclearweapons if they recognize the staggering and perhaps incalculable costs of nuclearwar, regardless of its relative outcome. Awareness of these costs may?and should?constitute a more potent deterrent than any degree of relative nuclear advantage. Ifthis is so, the current fixation on military capability is not only misleading butdangerous. (Confidential)References1. Department of Defense Annual Report, Fiscal Year 1981 (Washington, D.C.:Government Printing Office, 1980), pp. 5, 65, 63.2. This criticism has also been made by Alexander L. George and Richard Smoke,Deterrence in American Foreign Policy: Theory and Practice (New York..-Columbia University Press, 1974), passim; Robert Jervis, "Deterrence TheoryRevisited," World Politics, 31 (January 1979), pp. 289-394.3. On Presidential Directive 59, see Richard Burt's articles in The New York Times,5, 11, and 12 August 1980.4. Concise treatments of the evolution of strategic doctrine from the Kennedy toCarter administrations are provided by McGeorge Bundy, "Maintaining StableDeterrence," International Security, 3 (Winter 1978), pp. 5-16; Francis P.Hoeber, "How Little Is Enough?," International Security, 3 (Winter 1978), pp.53-73; Walter Slocombe, "The Countervailing Strategy," International Security, 5(Spring 1981), pp. 18-27.5. Department of Defense Annual Report, Fiscal Year 1980 (Washington, D.C.:Government Printing Office, 1979), p. 12.6. The general approach and results of the CIA's analysis are described in theDepartment of Defense Annual Report, 1981, pp. 123-126. NIE 11-3-78, itsresults and methods, have been discussed by Michael Getler, "Soviet's PowerSparks Intelligence Rift," The Washington Post, 9 May 1980, pp. Al, A18;Rowland Evans and Robert Novak, "The Strategic Imbalance, The WashingtonPost, 3 December 1980, p. A19; Richard K. Betts, "Strategic IntelligenceEstimates: Let's Make Them Useful," Parameters, 10 (December 1980), pp. 20-26;also see the testimony of Richard Pipes and Douglas Garthoff before US Congress,Permanent Select Committee on Intelligence, Hearings, Soviet Strategic Forces,(Washington, D.C.: Government Printing Office, 1980), pp. 2-34.7. Paul Nitze, "Deterring Our Deterrent," Foreign Policy, 25 (Winter 1976-77), pp.195-210; "Assuring Strategic Stability in an Era of Detente," Foreign Affairs,(January 1976), pp.207-232; "A Plea for Action," The New York Times Magazine,7 May 1978. For a critique of Nitze's view, see Alan Tonelson, "Nitze's World,"Foreign Policy, 35 (Summer 1979), pp. 74-90 and John Steinbruner and ThomasM. Garwin, "The Balance Between Prudence and Paranoia," InternationalSecurity, 1 (Summer 1976), pp. 138-181.8. Robert Jervis, "Deterrence Theory Revisited," pp. 314-322, and "Why NuclearSuperiority Doesn't Matter," Political Science Quarterly, 94 (Winter 1979-80), pp.617-634, makes a similar point. He distinguishes between a -balance of resolve"and a "balance of military power," arguing that it is the former that counts.22 C....60FtFUZLNTIALApproved for Release: 2014/07/29 000617249 Approved for Release: 2014/07/29 000617249CIA versus DOD CONFID TIAL9. See, for example, US Congress, Senate Committee on Foreign Relations, Effects ofLimited Nuclear Warfare, and Briefing on Counterforce Attacks, (Washington,D.C.: Government Printing Office, 1975); US Congress, Office of TechnologyAssessment, The Effects of Nuclear Warfare, (Washington, D.C.: GovernmentPrinting Office, 1979); Sidney Drell and Frank von Hippel, "Limited NuclearWar," Scientific American, 235 (November 1976), pp. 27-37; Kevin N. Lewis,"The Prompt and Delayed' Effects of Nuclear War," Scientific American, 241(July 1979), pp. 35-47, and the forthcoming, Arthur M. Katz, Life After NuclearWar: The Economic and Social Consequences of Nuclear Attacks on the UnitedStates, (Cambridge: Ballinger, 1981).10. Robert Jay Lifton, Death in Life: Survivors of Hiroshima, (New York: Simon andSchuster, 1976).11. Eric J. Leeds, in his provocative study, No Man's Land: Combat and Identity inWorld War I, (Cambridge: Cambridge University Press, 1979) suggests that thebewildering and terrifying experience of trench warfare on the Western front wasbeyond the ability of combatants to understand in human terms. He explores theways in which they sought to impose meaning and order upon their situation by'-,invoking magic and ritual.12. See C. Fritz and E. Marks, "The NORC Studies of Human Behavior in Disaster,"Journal of Social Issues, 10 (January 1954), pp. 26-41; Martha Wolfensten,Disaster, (Glencoe, Ill.: Free Press, 1957), pp. 11-30; Irving L. Janis, "Psychologi-cal Effects of Warnings," in George Baker and Dwight Chapman, eds., Man andSociety in Disaster, (New York: Basic Books, 1962), pp. 55-92.13. Wolfensten, p. 20.14. Voitech Mastny, Russia's Road to the Cold War, (New York: Columbia UniversityPress, 1978), pp. 156-157.15. Seweryn Bailer, Stalin's Successors: Leadership, Stability, and Change in theSoviet Union, (New York: Cambridge University Press, 1980).16. For some of the debate on civil defense, see T. K. Jones and W. Scott Thompson,"Central War and Civil Defense," Orbis, (Fall 1978), pp. 681-712; Daniel Goureand Gordon H. McCormick, "Soviet Strategic Defense: The Neglected Dimensionof the US-Soviet Balance," Orbis, (Spring 1980), pp. 103-127; William H. Kincade,"Repeating History: The Civil Defense Debate Renewed," International Se-curity, 2 (Winter 1978), pp. 99-120.The above information is Unclassified.CONPtENTIAL 23Approved for Release: 2014/07/29 000617249