114 1 fJI: Nuclear Command, Control, Cooperation

the probability, for example, up to 0.8, then we should increase the strength of attack up to 15 warheads. And so on (the numbers are relative). Each side has approved no-mograms and schedules, which may be periodically reviewed and refined, but can be used in a concrete study as fixed initial data. This is true for all components and curves.

But all this applies to separately analyzed elements. "For the enemy," however, we have to distribute all resources for the entire graph. This is also not a new problem, and has been solved in many studies. We will pay attention only to two crucial aspects.

First of all, we have to determine which strategy is better for the enemy. Often it is thought that this criterion is the least number of our warheads launched. But we are already convinced that point estimates provide too little information. It is easy to dem-onstrate that we can arrive at the same minimal number Nwh under a whole series of very different offensive strategies. It would, therefore, seem more correct to use the entire curve of the distributive function F(N) (see 3.5.8) as the criterion. Here each strategy means a curve, so there is no ambiguity. Which curve is better? Let us leave this question to the consumer's taste. We will express some of our thoughts at the end of this chapter. For now, let us decide that we do have a fixed criterion.

Even that is not enough. To resolve the distribution problem, we have a criterion, an apparatus for each component, and data on the enemy's resources (approximated)! Now we have to go to the object of suppression -that is, our cohesion graph.

An attentive reader would ask: "But we do have our own cohesion graph; what's the problem?"

But the trick is that the graph in our possession is not sufficient. The enemy should have the above-mentioned data, too. Let us think - the solution for this distribution problem is a strictly bilateral process. It is as if we are thinking for the enemy, suggest-ing the strategies rational for him. How can he distribute his weapons rationally with-out any knowledge of the object to be suppressed? Thus, we have to consider the high-est possible degree of knowledge on his part.

Here, it seems, we have arrived at our first really serious difficulty. The enemy knows his resources; we can reach agreement with him on the final criteria and nomograms for each component, but how do we take into consideration his knowledge of us? What does he know about the components, structure, and properties of our cohesion graph? Realistically, he needs to know about our SNF groupings, command and control sys-tem, and the GIS. But no side will completely reveal such data "off hand" because of concerns about its national security interests. Without those data, we will not be able

to obtain results for the minimal deterrence levels. What are we to do? We've come to one of the most important questions in this whole discussion - the

question of availability of data, the question of secrecy. This is an obstacle on the way to our final goal, and as such, it should be resolved. .

A few years ago, even posing this question made no sense. Today, there is a hope that it may be positively resolved, based on the experience each side has acqwr in reducing nuclear weapons themselves.

Indeed I it has b .

The Methodology; The Estimate O/ Nucteap Deterl'ence Using ca I J J 5

1 come poss1bl t d 1 sue l formerly super s o c a. stfy th fil t part f delivery vehicles - Jcjr t areas as the number and ty fo our cohesion graph,

1 as we as th d 1 pes o nude 1 over reducing them W etr ep oyment ar as and t . . ar war leads and the graph the . hy shouJdn t we b able to d, h o e tablish mutual control

' command a d o t e same fi tl interests of botll sid ? n control system and the I . 1 or le second part of

es. , WH' lOut h . Unfortunately, our tb ' k' aunmg security

' m mg has b h prevads that even if dedas if' . een c anging too lowly E come possible, command dlcatton of information about the. ven now the opinion

This is certainly a mist: an_d information proper has be-access to command and contr . n t e ftrst place, without reso/vi u be. minimal deterr nc I _ol systems data, we cannot ev lk ng tlus question of command and conti:o1 na yzmg only the weapons themsel en ta . .the problem of

systems and th I . vesl WltJIOut 1 1 d' . econd, we ought at least to tt e ' . ' WllJ not allow this probl nc u mg their

the perspective of whi I a mpt to look at the command d em to be resolved. l part are "d 1 an control .c. nattonaJ security p l ec a sifiab/ " witll t h . syst m trom

. . et lap Oltr fc . . ou at:n111lg th . even tned to analyz tl . L eazs rue omewhat exaggerat d? U e Interests of

Let us break all th us.. et us try to do it now at least e 1. p to now, no one has e mam dat h ' on t le mo t

thre groups. a on t e command and con 1 s general level. TL f . ... tro system and the 1 . le 1r t group of d t . mto

cb a a mclude tl t -c . ea other, which make n la lw.ormatlon both side dude, for example th 1 o _ense to conceal. Tlus will b know very well about

' e ocat.10n d b .e[ e ea y to agr . Tl . mand posts and commu . . an c I . characteristics of th . e on. ley 111-and communjcation sy t DlcatJon_ facilities; the presence of gl tationary com-nications technolocr . h ems; maJor cable routes; the par . o. a command, control ecaJ princi.ple of tobhy, t nomenclature of the main com ameters of standard conunu-

. . e command d ponents ofth . 1 . rnents, tlus Ji will . I . ao control systems and I . e gtap ,I and gen-The s cond . camp .ment our cohesion graph . After additional agree-

. group cons1sts of th d cu:cumstances . e ata that shou ld b

Js group t Lf .c not e d 1 ' are of no importance fo . se . !ails into two parts. On s ec assi_ led under any

Potentia/ opponeotr :olvmg our problem. For xample th data that State lumbe ") . now the codes for blocki 1 Jete nev.er Js a need to 1 t

Th r , or the valu fi ng nuc ear wea ( 1 e other sub.gro . es or the encryption ke s fi t 1 so-caHed ll1auce tabilit up mdud the data that have. . y or commurucation charnlel

b declassified. .e over the graph's perf or- fteguencies mo J!e nuclear weapons carriers and, .or example, the schedule of

at determine ;h ormbati?n transmission times and tl command posts, working ra-are c . e pro ab11i ti h o 1ers. Tho . h

onsldered in th c aracter of th e t e very data the enem mentioned tandard initial components' performance, and

ln Ows th approx/ . ows t!1at such and such ty of no:ograms and schedules. For ow the frequ 11c;:; may b by post exists, he

posJtron change A . I ocatJOn but does n approp t ' not na e nomograms ill w an wer the

118 \ fJI: Nucleal' command, contl'ol, coopel'atlon

. . ere uired probability. Coordi-. means are needed to destroy 1t W1th th ~is a complicated business,

questwn of what . th every component of the grap. . . 1 it is technically nating these nomogram~ Wl d t '1 and a lot of time, but m pr~nop e

. 1 t f attentwn to e al 1 secunty. requmng a o o h' . not a direct threat to nat10na . h . d is sure they know feasible. In any case, t 1S lS . those about which nelt er Sl. e One ex-

Finally, the third g~:is:fa~:::~:vant for estimatin~ the coh:sl;:::~r~~ the infor-everything, and wh1c b f mobile elements of vanous type . t' tes We will call

d the num er o d ct our es 1ma ample lS ata on . . . 1 d ta and allow us to con u d "fully" with a

. ld fll ut the m1t1a a , d ing the wor mat10n wou 1 o f. 't'al data "fully revealed, en ow ' netration into our

1 ted set o llll 1 t the enemy s pe such a comp e . are fully protected agams . . . mation remains in-very specific meamng: wt'e keys etc. This area of classlhed mfothr degree of stability

b " ncryp wn ' h upon e "state num er, e th t this "full revelation" only touc es the probabilistic charac-tact. But let us note a l t 's performance. Even so,

d and contro sys ern in the comman . d t would allow

f much data has remame . ld be highly desirable smce 1 ter o . the third group wou . 'tin a hurry. The reason

Reveahng data .from lt but neither side is gomg to reveallh "looked into it" yet.

us to start estim~tmg r:~a:the final result wou~d be. No o:e s::uation, the unknown is clear: no one nows b k from that pomt. In sue a A d there would be no way ac f

n . . to anyone. . "insurance" stage or would be fnghtenmg h ld be a logical intermediate ste~, an. to the third group,

This is why there s .ou . conditional initial data be on~mg rantees that mutual theoretical studieS u~l:; goal of such studies is to obtam so~:{:e:s of further taken from a broad range. 1 tely convinced as to the purpose

t become comp e would allow us o. ual research. It is work along those hnes. b ht us to the idea of such mut

ts have roug All previous argumen absolutely necessary.

"' 7 2 Towards Confidence Through Jol~t ~esearchrch on the deterrence problem? First ilol ts in favor of JOlnt resea

What are the argurnen " . ugh: . nfidence and guarantees. "not wanting lS not eno

is the lssue.of co Earth wants a nuclear war. Yet, from the vagueness No senslble person on . . stability. We have to get away . above roaY

one would like to have c?nhdence lhnwith the use of the approach ofudthtnedence indica-. J01nt researc .. d r" 0 e err of today's sttuatwn. 1 theoretical picture, an or e bl everyone to

help establish, first of all, a g~e~:. 1 data. In other words, it ~~ul~ e~; e uld provide tors across the broad range ~ mt 1~ measuring fear . Thls m ltse wo

f h. d1menswn -

f 1 the scale o t lS new . d' tor to a ee f d d ce m lea some ground for c~n 1 e~~e. ld benefit from reducing the .eterren the risk of wal'

The fact that nelther sl e woub'l' the overall situation and mcrease h to desta 1 lZe

level dangerous enoug f uarantee. ee if it is aol'l would itself serve as a meas.u~e o g ept may also serve as a guarant ,

The stability of the sufftclency cone

Ths Mslhodotogy: The Estimate oti'Jucteal' DetBI'I'ence Using C3 1 117

firmed by joint research ( ee .5.6). By that we understand that in peacetime, no side would be able to achieve such economic superiority over another that the degree of risk would make aggression acceptable.

Thus we can expect a "double guarantee' situation, in which neither side wants or can afford nuclear aggr s ion. While everyone now accepts thi intuitively, we will be abl to prove this quantitatively.

There is reason to hope for confirmation of a stability area within the concept of ufficiency. Thi is betokened by the properties of the cohesion graph as a network.

When the network reaches a certain size, the expense of expanding it becomes lower than the expen e required to maintain its suppression at the previous leveL To make this clear, let us review a simplified example.

Let us talk about a group of mobil command posts of the highe t level. We are interested in the probability of at least one of them urviving. First variant - in the group there are four command posts. In order to eliminate one with a probability of 0.3, the enemy needs to allocate 10 warheads. The cost of one command post is one conditional unit. The cost of a strike by one warhead is 0.1 conditional unit. In this case, th probability of at least one command post surviving is 1-(1 -0.3)4 = 0.75, and the relation hip of the costs ' 'group of command posts: attack" is 1:1. econd variant -we increase the number of command posts to six. With the same relationship between costs, the probability of on post surviving wilJ increase to 1-(1 -0.3)6 = 0.87. There-fore, the enemy will need more attack assets to maintain th same probability 0.75. For the attacker, the probability of urvival of ach command post should not be higher than 0.2 . To achieve this re ult, the attacker need to aU ocate not 10 but 15 warheads per command post, o that the relationship of costs will get worse for the aggressor, 1:1.5 instead of the formerl: 1.

Obviously, to reduce the probability of survival of at least one po t to, for example, 0.5 or 0. , the aggressor will have to allocate substantially more offensive weapons for this task. At that, the de~ nding side is able to maintain this probability at higher levels with the help of lower, asymmetrical efforts.

This example illustrates only in principle a concept of sufficiency that is stable. If we consider not only a group of mobile command po ts but instead the entire cohesion graph, everything will certainly become much more complicated. But the general trend, of cost ratio changing in favor of the d fending side as the deterrence indicator de-creases, should remain the same. To completely exclude (or adliev very high prob-a~ility for excluding) an unacceptable retaliatory damage would probably be irnpos-~ble. It is hardly possible to overestimat the sigrllficance of such a conclusion if it can e proved quantitatively. As noted earlier, this could solve a lot of problems. It could demonstrate, for example, that no matter how great the potentia l aggressor's

efforts are dLlring peacetime, no matter how much he spends, and even with an optimal att~~k trategy, he will not be able to reduce the probability of receiving a retaliatory st(ike of at least 800 warheads to a value lower than 0.2 to 0 . . Is this a little or a lot?

118 1 f?: Nuclear command, conrrol, Cooperation

Right now we are not making that e,timate. The Unpodant point hcre is whethe< such

an area of 'tability exists, while locating and confinning it would be the pmpooc of joint '"'ea

600) _

:: :~:~ ~:::;~o~rect, _and ''"Y ~ :::0~:=~~~: ::;~e " It '~em' thi; ,houl~ o earners, C3 and GIS, as our alread r: '1' ~~amp e. Let us look at Y 1amr rar broom " (S p ee rg. 29).

ucud .... ! F'(N) I

I I

clearly unacceptable d a mage

pcud

1.0

0.8

0.6

0.4

0.2

0

Fig. 28: lndepende

I

- , I ~ ------- I a --------=+--=- ------- -====~-

600 800 1,000 1,200

nee of results under "a" and "b" i mpact strategies.

When using the " " ._.-kin a strategy th Su'J g the "h d , . ' . e n my allocates the . This mean an le, thus bnogmg the probabili . maJor part of h>S asset for And wl ' that out of 100 sunulation , the "b . ~. of rts performance down to 0. 5.

1at rs the srtuation with th ' loom remams undamaged in '" 0 centrated h " . e weeper here? th . ::J case . ..: on t e handle" th .. , . . mce e enemy's att k uons, then : e swe p r ' better off I ac wa con-high a d umber of delwery vehicles and tl . . n almost all computer simula-. n may b 'd 1.err warhead t k' . 10 the "broom"e consr ered ufficient for unacce tabled a >ng pad m the attack i

If w altogether is: p ~ 0. 5. p amage. That i , what we have ., e change 0 the "b" cud swe per," w u . trategy and concentrate th ' . even mme I e wr achieve the effect en in Fig 1 11 ~emy s mam attack on the in the "sw re La~le fashion :say, the probability f. t . ," handle" will perform in an

eeper the situation will wo. ho ~. s per ormance wiiJ go up to 0 7 o t rsen, sot atm o I b .. vu n y a out, for xampl '60 of100

120 I IJ: Nuclellt' Command, Conll'ol, Coopel'ation

aggressor's impact

Probability of performance of "handle" P=O.S

Flg. 2S: "Broom's'' condition under "a" Impact strategy.

0

0 0

Majority of nu~le~r carriers survrve,

N>CUD

. h ture will look like this: d 't of the final result distributiOn, t epic If we present the same picture through the ensl Y

f(N)

P=0.50

0 CUD N

Fig. 30: The density of distribution (to Fig. 29).

aggressor's impact

~------~-------/ Probability of performance of

"handle" P::0.7

The Methodology: The Estimate of Nucleal' Detei'Pence Using CS 1 121

0

Majority of nuclear carriers and

communications channels to them are destroyed

Fig. 31: "Broom's" condition under "b" Impact strategy.

simulations will the number of warheads taking part in the strike be more than 600. We are interested in the general result for the whole "broom." Multiplying the probabili-ties of successive events, we will obtain:

P(N~600) = 0.7 X 0.6 = 0.42 (Fig. 31)

So, even if the means of attack ace 'edistdbuted fwm the same pool"' stmtegies change, the final 'esults may be conside,ed to be independent of them. Results are determined by the structu,e and characte,isti" of the "handle" and the "sweepec" It ~Possible to have such a poo' "sweepe," that, using the "b" strategy, not one compute, simulation will 'esult in the numbe, of wacheads being mo,e than 600. Then, the general 'esult is P(N;>600) ~ 0, even though the "handle" may be an ideal one in this Situation. Each part of the "b,oom" plays its part and makes its own contdbution. lo l'he usef~lness_and possibility of joint 'eseacch is influenced by the following simple gtcal cons1deratJon :

First of all, the ve,y idea of mutual detenence implies a mandatmy exchange of in-

122 l C': Nucteal' command, Contl'ol, coopel'alion

formation. Each sid , after determining for itself a uf!icient level of defensive paten tial, is obligated to demonsttate this to the opposite side. It i further required to prove that with uch a potential, the pmbability of unacceptable damage for the aggressor would be of an appropriate value. Each side is most interested in seeing that the other side has this knowledge, or the deterrence factor simply will not work. uch a nece.sity is not felt very sttongly right now because the current strategic nuclear force are very large. Yet, when searching for minimally acceptable defensi v levels, such mutual pmofs

will become crucial. cond, this can be financial ly beneficial !or both sides. The pr ence of full rand more reliable initial data made available during joint research will allow the parties to come do e to truly sufficient minimuros and avoid unnecess"')l expenses. If the exist-ence of the above described areas of stability is confirmed, both sides may avoid con-

ducting expensive but pointless programs. Third, there is no risk involved for either side. Each of the parties has the right to

stop at the level that it deems sufficient for itself. And after conducting joint reeearch, it will be able to obtain more solid support for its decision. How much initial data will

be revealed depends, just as at the present time, on the specific situation. Such joint research could have as its main goal fuller accounting of all major factors

in the process of nuclear arms reductions, not only the initial levels of strategic forces such as composition, characteristics, deployment a< exchange. It is, after all, not the potential itself, but the expected result of its uee, that acts as a deterrent. And the latter, as we have ,een, is most fully characterized by the pwbabilistic distribution value. joint design and study of such distributions by both sides, and then using those studies to make recommendations to

relevant agencies, is a viable way to speed up nuclear disarmament. It seems that a logical fi

124 1 IJI: Nucleal' Command, ContJtol, CDopet'lltion

seems to be the most logical one. So far, at least, there has been no sensible alternative. If one is presented, it will not change the whole method but will only influence the initial data.

ow let us present an intermediate summary. A priori, we have already overcome some methodological challenges. We have de-

cided that our first priority i to establish theoretically a range of assumed val ues for the degree of risk, and to convince ourselves of the stability of the concept of suffi-ciency. This would be best done through joint research. If the safety margin of reli -ability is proved, the way will be open for the next practical steps. Then the two parties may cooperate on initial data that is lacking on a third group without any major risk to themselves. This process will be, it seems of iterative character, with a gradual ap-proach to the "truth. " The path goes from redundant reliability towards a minimum rea onable reliability.

The transition from theoretical re earch to practical applications of the work will be a sophisticated business requiring patience, mutual understanding, and initiative on both sides. The results obtained will not be "points" but will be located within a cer-tain range. At this stage, each side could justifiably rely on the following principle: any re ult obtained jointly, based on some accepted level of declassification for initial data, will be considered the "true" one . . As the initial data become clearer and influence the precision of the final result, this will approach the "real" truth.

All this, however, will come at a later date. Today's agenda is to organize joint re-search between Russia and the United tates, along with other partners, in this area. The purpose of this book i to establish the necessity of such research. All the follow-ing steps mu t be considered without losing sight of this final goal, or the motivation to achieve it. We also do not consider as closed the discussion of how to guarantee accept-able truthfulness in the final results, ince thi major question will be crucial for the two other problems in implementing our method: creating the mechanism for using it, and making choices.

3.7.3 The Mechanism tol' Applying 1/Je Mei/Jod Let us assume that a methodological apparatus for "measuring fear" has been de-

veloped through the efforts of both sides. A unified model has been created, and the existence of an area of deterrence indicator stability is proven theoretically. Now, the participants have started to discuss coordinating their initial data. This last step is, in reality, already a part of the negotiating process on strategic nuclear forces, since imple-menting it would get us closer to specific quantitative indicators for the command and control systems, GIS, and nuclear weapons delivery vehicles groups. Naturally, each side will check those results "at home" before coordinating any part of the initial data with its partner. Each side's behavior during negotiations will be based on those pre-paratory results.

Fig. 32 provides a schematic illustration of using the method during the negotiating:

The Methodology: The Esllmar, 'B Of Nucteal' DBiel'l'ence Using ca I 125

aggressor's Impact

r---- _---r----.--1 lr-----!._

II nuclear potential

(NP) I

command and control system

(C 3) I 1-:----_j I_ _!!clprocal ~tlons o!_Eefending side

---Minimization of NP and c3. balanced with regard to economic considerations ~nd reliability guarantee (Independently by defending side)

(P(W~Wreq.))

wroq, ---1

Predicted result .01 NP realization _

distribution function F'(W)

l r-----1 I I

deterrence factor _ degree of risk for

attacker (P(W~Wreq))

Minimization of (P(W> to level ot minimal d ~~eqJ) (b eterrence y agreement between sides)

Suggestions at negotiations

w- damage for attacker_a_s_a-re _________ J w with the defending side suit of reciprocal actions

req. - requested (agreed) value unacceptable for the attac~~~amage, which is

Fig. 32: The method for minimization of nuclear potential and C3f

actor. p

Ig. 32 demonstrates that th f comes from the realization _e mal result of retaliatory actions b . tees the unblo k' of Its nuclear potential (NP) th y the defendmg side

c 10g and us f 1 rough (C3) h ~y the aggressor's attacks ;ho nucleahr weapons. Since both NP and c'3 w I~hflguaran Is used as a ' e resu t as a random ch are 10 uenced

~ndicatordu~i:~t::~:i:~terrence i~dicator in the illus:::t~~:. ;~~ba~ility P(W~Wreq) Is a possibility of f1 g 'bl ons t~ review questions of risk det h Sid~ ~auld use this

We m exi y mampulating its p and W ' errence, stability, etc. There . ay assume that it is 'b components

tisk accord possi le to establish "th however t~:; to ~e .participants' mutual agreement r;shold" values for the degree of and mai~tai'n eacfi Side has the right to keep its "ow . .. ron: the chart it becomes clear

a sa ety m T . n estimate f th d ' Weapons, C3 and a~giO, his means that each art o e egree of risk, siderations 0 1 t~e GIS Independently, based on ec p ~er may reduce his nuclear balanced s~ n ~a ter many independent and joint ~?omic p~tential and other con-

No matt gghestwns at. the negotiations. es Imates wdl each side present its . er owtempt h' .

Q\Jite doubtful . 10~ t IS picture of joint pro 'ects . . . that it will not d~ t? Its highly unusual character~ wh : Its Implementation may seem dl!ring negotia:a e I~ own errors or be deceived? Ever~~~ar~ntee does any side have

ns. et us try to analyze a few pot t . 1 10g .ep.ends on the behavior en Ia negohatmg strategies.

126 1 f7: Nuc/eal' Command, ContJtot Coopel'ation

The idea that the participants are interested in preventing a nuclear conflict as well as minimizing the cost of guaranteeing deterrence should be at the core of this analysis. The goal of analyzing potential behavior strategies at the negotiations is to try to mini-mize the level of distrust, both toward the other participant(s) and toward the potential results; in other words, to obtain certain guarantees.

Let us consider a possible Russian approach to the assumed strategies of the United States during such negotiations. This is just an example. It is logical to assume that the U.S. side is entitled to similar evaluation of the Russian strategy, taking into account their own peculiarities and interests. Can such an analysis, based on simple logical conclusions, help us find, in a preliminary fashion, some guarantees?

The following considerations should have priority: The participants have worked out a joint methodological apparatus together, which

gives them a model for estimating results they would like to predict. They have also agreed on maximum (critical) levels of indicator P(W;:::W ) according to which they

req

evaluate the results obtained in the course of research. Russia evaluates the consequences of U.S. strategy according to three main criteria:

does it change the degree of risk of nuclear war; does it reduce or create obstacles to the process of strategic nuclear force reduction, assuming this to be one of the main goals of world politics; and what extra expenditures will Russia incur, when purposeful out-lays are separated from wasteful ones?

Our research indicates that those three factors are interrelated, so that, as a final result, improvements of SNF C3 increase the value of the indicator P(W;:::W ), while

req

reduction of the SNF themselves decreases it. It is possible to regulate the indicator through balanced changes in the components and characteristics of SNF and SNF C3

When estimating possible U.S. strategies in the course of research, we assume the following attitudes of the United States toward the risk of unacceptable damage as a result of Russia's retaliation in response to a U.S. attack. For the majority of the public, the very possibility of unacceptable damage, regardless of its magnitude expressed quantitatively, is enough of a deterrent. For a minor segment (aggressive military/ political circles, members of the military industrial complex, etc.- the so-called "hawks") just acknowledging such a possibility is not a sufficient deterrent. Additional quanti-tative estimates of the degree of risk P(W;:::W ) will be necessary. This factor ma:y req play a deterring role only if it is officially and openly declared.

Let us demonstrate, considering all of the above, one possible Russian assessment of projected U.S. negotiating strategies.

Three lines of U.S. behavior might be used during joint research with Russia to ob-tain final calculations for the value of P(W;:::W ). (See Table 4 ). req

Strategy A: the "honorable" one, with the most reliable initial data possible. Two results are possible here: one, when the value of calculated indicator P(W;:::Wreq) is high, i.e. higher than the mutually established critical level; and two, when it is lower than the critical level.

The Methodology. Th E: . 'II Stimate Of Nuc/eal' Detel'l'ence Using ca I 127

Strategy "A" ("h onorable") Strategy "8" c results of calculations are: Strategy "C"

aims to ~ 0 P(W>W )>P increase the value goal is to Risk of req ern. P(W?.W )

W,.~) tnq crlt. nuclear war small high P(W>W )

high req,

small Additional expenses of small Russia for C3 system high small

high

Continuation of STAAT accelerating

process (reducing nuclear decelerating arsenals of both sides)

accelerating decelerating

(but it increases

Attitude to the given U.S. strategy from

the risk of war)

U.S. side -most acceptable for both:

U.S.: partly Russian side - unacceptable acceptable

most acceptable unacceptable dangerous/

Russia: unaccep!able Conclusion and unacceptable due to extra expenses expedient measures favorable situation: necessary to

U.S.: refuse "8" possible to conside agree on joint U.S.: can use, maybe next steps towards additional

strategy; .Russia: to try to

nuclear arms Russia: for reduction.

measures for insurance, to

rncrease P due to not increasing

provide only own additional ., P(W>W,oq) redundance: +llP

expenses, but also with U.S. help.

Table 4 Evaluar h . mg t e possible strategies of the United St t . .

a es srde durrng joint research .

Strate B 1 . . . gy . h is one con ciousl a. 1 . . artrflaal changes in some initial rt: bso:~ UA1 ea .mg the value ofP(W? W ) through

trategy . same a b b , rnencan and Ru ian '"'' sa ove ut 'th L P(W~ W ) ' WI ~.ne onsciou goal of 1 . . 1

\M '"" oweung t 1e value of . e have to work under th as . .

inme jf tl sumpttOn that 1t w Jd b th . le trutial data was distorted by tl . . ou e lOlpo sib1e to fu JJy d ter~

. ~lr strategies and our responses we 1e ruted States. Th refore, while evaluatin. !;~~onnatio~. . ur ideas of how,the ~:t~~r~:~t to consider :he possibility of s ud~

. to posstble results should b f I e evaluates Its own strategies wit! P1esented in Tabl 4 va ue to us her ho . " ) A e . . e assumptions are also n1a s ~ result of estimating pos ible U . . .

y draw the following preLiminary co . . clt:I~tegJes m the course of a joint resear h .. , n us1ons: ' "

Strategy A a) If hi h

ioio.t g values for P(W~W ) are obt . . tel:lce ~a~~lations possible, we ~~y consid:~n:: m t~e course of the most "honorable" IDe nuc~ Ie~ed. In this case, we may consider the mam ;oal of effective nuclear deter-

ear lOrces. e ways lOr further reductions of strate-

128 1 fJl: Nuc/eal' Command, Contl'o~ Coopel'ation

b) If low values for the rune indicators P(W~ W ) are obtained, the situation should rcq

be perceived as dangerous, mainly because the "hawks" will be trongly tempted to declare that aggression would not result in retaliation. hould such results be declared officially, the ensuing situation would have to be regarded as unacceptable for both sides, especially because the goal of further reducing strategic nuclear force becomes more difficult. Under these conditions, both sides together wi ll have to look for ways to increase the value of P(W~W ) to a required level. There are several possible ways

t L'\1

to do this: improve 3 of Russia's F, or reduce the effectiveness of a U . . conven-tional and nuclear attack agains the RussianS F command and control system, as well a against the delivery vehicles themselves. This cou ld be done, for example, by redeploying S B s to the areas further removed from Russia, limiting theater and space armaments, etc. aturally, further Russian steps for reducing strategic nuclear forces wiJl have to take into account such considerations.

Strategy B If strategy B is pursued by the United tates, the calculated value of indicator

P(W?.W ) is higher than the "true" one obtained as a re ult of "strategy A. '' ur '""I

(Russian) side may not have this information available. This situation .is al o dangerous and unacceptable for both sides. The nited tates

would stiJl be tempted to commit aggression, because within the United tates, those involved with the issue would know the true degree of risk, while society in general wouJd not. On the surface, moreover, it would look as if it were possible to further reduce strategic nuclear forces. uch a measure, however, would actually increase the danger of war, since the value of the indicator P(W~W ""~)would become even lower. Also, Russia would not feel a need to increase expense for improving the 3 NF. For all these reasons, it is very improbable that the United States would use strategy B for joint research.

"'or Russia, such aU. . strategy is also dangerous because of potential di informa-tion. Because establishing for a fact that information provided by the United tates is or is not really disioformation is very difficult, we would hav to take out an insurance policy creating a safety margin in the value of indicator P(W~Wrcq) . The ways to do that are described under strategy A.

Strategy C Due to intentional distortions by the United tates, the value of P(W?.W rcq) is lower

than the 'true" one. It is clear that in reality the risk of war is small. This i all the rno~e so since as in the ca e of strategy A (b) measures to incl-eas the value of the matn indicate~ would be taken. The implications ofthi situation are different for the United States and Russia. . hi

It is not beneficial for Russia because we would have to incur additional (and JL1. t ~ case unnecessary) expense for improving 3 NF. r, due to the seeming weakness 0

The Methodology: The EsOmate of Nuc/eap DeiBI'I'ence Using ca I 128

the c3 SNF, We would have to slow the process of red . For the United States strateg C . "b . ucmg nuclear strategic forces

I ' Y Is poss1 le f t d al t Is tr~e that the risk of war would be small , 1 no. 1 e , under certain conditions.

would mcur additional expenses. Th I a~d mi~ht be further reduced. Russia be slowing down the process of st teo? y nelgative pomt (for the United States) would t t 1 ra egic nuc ear force red t. N

s ra egy eaves room for maneuver. uc IOns. evertheless, this

General Conclusion When arranging or a JOmt research usin th

the merican to be "honorable" . r 1 g. e su~gested approach, we may expect m wrmu atmg the '( 1 d .

toward lowering the calculated val f. d. Ir Im Ia ata, With a certain trend h . ues o m 1cator P(W>W ) I .

act Wit the maximum openness d - req . n this case, we should an accuracy of "( 1 d .

sures to create additional safety m . r h 1111 Ia ata, takmg subsequent mea-h ld b argm wr t e value of P(W>W )

s ou e taken not only by us but b t . h - req Those measures Th h ' e me Wit counterm b h . us t e conclusions obtained throu h anal . ea~ures ~ t e United States.

mahon as some kind of a guarantee Wgh"l h~Sls may be .VIewed m the first approxi-"bl 1 e t 1s exampl .11 possi e ways to choose and is b f" e Is JUst an 1 ustration of the F . ' y no means mal or co 1 t ous. Irst of all, the conclusion th t "t . . . mp e e, two advantages are obvi-

d r a 1 IS Impossible to t t ll 1 d amage wr the aggressor may serve b . r o a y exc u e unacceptable

I . . as a asis wr guara t r h . cone uswn Is confirmed by testing th . . . I d n ees wr eac side. If such a . e 111Itla ata across b d I" .

ranges, It may become a serious det r roa ' rea Ishcally achievable . errence 1actor S d h mg Jomt research by the participants . . I. econ '.sue a result, achieved dur-th . . , may 111Itse f serve to

e negotiatiOns process and contrib t t . mcrease mutual trust during useful to consider an open excha ub e o Its success. Based on this, it would seem

h h , nge etween the two sid f h eac ot er s possible strategies. After all the 1 e~ ~ . t eir own estimates of expenses while maintaining a reliable le~el ofoa of such activities-:- to avoid additional

mutual deterrence- Is very attractive.

'';: ~he Pl'obtem ot Choice: How Much Is Needed top DetePPence'~ ow, et us assume that the intermediate ste . . . .

pleted. Areas of stability have been th . lpl s m our J0111t work have been com-b d eoretica y pro t oun anes of these areas - that . 1 f P ven o exist. The quantitative

rat d Th Is, va ues o and N ( t bl ) e . e next point on the a d . st. st s sta e - have been elabo-

v . gen a Is an analysis f th f . . anants of the totality of nuclear weap d I" o . e cost-su ficiency of alternative

tern, and GIS components in order t ons e IVery vehicles, command and control sys-tee reliable deterrence wid~ minim 1 o select the mos~ rational one that would guaran-sake of simplicity. a expenses. We will call this totality "SNF" for the

What con d F. Sl erahons should we use for this h . ? Irst of all h c 01ce.

of st . , we ave to consider the possibilit of h . . rategy 111 his attack - from th " d , y t e aggressor employmg any type

VIctory (b) e a venturers strate " ( ) h in h. . The aggressor can make this ch . . ~y a to t e strategy of Pyrrhic

g Is choice of weapons to be u d . Olhce !~mediately before the attack by alter-se agamst t e given targets Th t . h

a Is w Your evalua-

130 l 17: Nuclear command, conwot, cooperation

1 'll have a dual character that logically follows from the dual nature of

tion of the resu t w~ d N) ht evaluate the deterrence indtcator (P. an .. d t' n how a potential aggressor mtg . t

Second, we should take tnt? con.st ehra to e of our calculations. Let us n~t or~~ d N obtamed m t e cour . . " that 1s at tne

the values of p st an ~t b daries for "coropressmg the s~ung, . gh to those are the maxtmum oun p and r are Lmpresstve nou

:ag~est possible peaks of arms race. Itfhtehseps~~~agt~eJ. liS~ a tiny bit more at the expldense bof h h "compress ? It wou pro -be a deterrent, t en w y . . . on't do any good anyway. .

huge financial outlays, espeCl;llyt~~c:e~s~essly intense competition;, dream~~g :~:~~ bl make sense not to plan or . but to follow the natura p .

a y natural and incredibly expenstve we.aponsW, should rather continue developmg super f the ttme. e 0 d GIS -development tha: is characte~ts~;ac~r against strategic nuclear weapons, .an of bilat-weapons, includmg means o ~ ns that during the subsequent senes

. This m turn mea t for the extreme but at an ordmary pace. N chosen as the reference pam s The eral estimates, the values of P andt , than the established values of P st and Nl st' f

f k hould be grea er d' ith the rea pace o strategies o attac ' s 'th less effort, correspon mg w " . gs" would be pressed down Wl spnn development. p 33

Ons and countermeasures . P nd N as seen m tg. .

weap . l nee potnts a pr' We will call these practtca re ere P'

Ppract.

F'(N} I I I

I I I I f-----+ -----

Pstable. (limit)

- --- 1 I 1 __ L -- - --

----~- I I I I I \ I I I

N stable. N pract. (limit)

.. gs" contraction.

N

F. 33' Limit and practicallevels of spnn 19 of

1' . t for the probabthty . e case of arms race, the lower tml .. 0 5 Accord

For example, m an _extrem . . "normal" pace of development tt ts . . late tne unacceptable damage ts 0:3, ':"htle m a in an "ideal'' situation, we ~an calcu aod ingly, applying the Pyrrhtc vt~to.ry strategy 300 warheads, while realistically toda:eads.

. . magnitude of retahatwn at, sa~, here at the level of 600 war mmtmum h' . . mum wtll be somew in the foreseeable future t ts mtm

The Methodology: The Estimate at Nuclear Deterrence Using IJI 1 131

r, consideri-ng th real situation, we have moved rea 3 of our picture up and to the right. We will do further work with this area.

What i our goal as a defending ide? We need to have an F comp ition (includ-ing 3 and the I ) that would be able to provide for th curve of the distributive function '( I) through Area. under any type of enemy attack. trategy, from trategy "a'' (adv nturer's strategy) to strategy "b' (Pyrrhic victory)_

rn principle, we have to consid r different variants of composition of F while modifying both maximum cm-ve ; that i , under nemy attack strategi s fTom "a" to "b." Let us, how ver note the foUowing circum tance. p to th.e pres nt moment as we have already mentioned everal times in this chap er, no one has addr ssed the prob-ability of damage (P), yet there .i much lebate about the magnitude of damage {W ), and the concept of unacceptable damage remain quite varrue. lt wiJI r main this way for a long time, and, quite possibly, over larg r ranges.

Therefore, to simplify our further argum nts, let us introduc the foUowing device: we will analyze the variants of the S fF campo itions while m difying only the "b" curv , leaving the 'a" curve at a given fixed level. hi device is j ustified because the "a" curved monstrates th probability f normous, known to b unacceptable dam-age (up to 1,200 Ol' 2,000 warheads for example), and fluctuation within the 0.3 to 0.5 range < re not very significant for this kind of probability. We have not yet learned to operate with such oncepts. Let us rem mb r al o that the expenses for thi curve go mostly for the I and the "brooms handl ." ' o, we will assum hat in all the variants of omposition f Tf we will be r vi wing, the e expenses will remain constant and their jze will be such that they will deter the enemy if h decides to u. the "a" strategy.

The "b' curv is a very di{feren story. ince the concept of unacceptabl damage is so vague, it makes sen e precisely h r to review the alternative variants whil taking i11to account th differeuc sin co ts forth other part of the cohesion graph - the nuclear deljvery vehi les themselves and thos lement of the command and control y tern that we have called ''the broom.' ' W will, naturally, consid r g neral expen es when using the "sufficiency-cost" criteria while taking into account both curves 'a" and "b."

ne other consideration ha to do with the time fa t r. 1 ot on ly money andre-sources bu also time are crucial to change any weapon development plan . For x-ample, if th potential aggr sor uddenly decid d to change the exp cted picture, catch us unawares and g below the deterrence thresholds in order to later create the condi-tions for an attack, he could not ucceed. Yi ars are needed for developing and intro-~ucing into production and ervice new types of weapons, artd to cone al them today is Lmpossible. In thi , cas the pot 1tial victim would have time to take countermeasures. Such measures cost less than the enemy' (as we have demonstrated earlier). will com-P~nsate for th aggressor's efforts, and the hresh ld values of Lhe deterrence indicator wl]~ remain at the a me level. With a unified model and joint research, this fact will be eastly demonstTable to the other side, and make i reconsid r any abrupt changes in si:J)ategic tability.

132 \ Cl: Nuclear command, Contl'ol, Cooperation

Based on the above considerations, let us now consider as examples a series of re-

sults that could be obtained by analyzing alternative variants of the SNF composition

for a specific period of time. All numbers are conditional. Let us assume that there are four such variants altogether. First is the existing com-

bination of weapon delivery vehicles, the GIS, and the command and control system. All other variants involve increasing certain components of this combination. We are not going into the details of what components are increased (according to their content and characteristics) right now. It is important, it should be done, and it should be done with expenses in mind. But this is "internal" work, while here we are limiting our-selves to an analysis of the general picture. Therefore, we will designate the expenses

for SNF (costs- C) as !).C. The charts for the distributive function F'(N) for the given variants are shown in

Fig. 34:

F'(N)

1.0

0.8

0.6

0.4

0.2

0 1,000 1,200 200 400 600

BOO

Fig. 34: Distribution function of final result for alternative variants.

These graphs provide the most detailed and complete characteristics of the given

variants, and thus should serve as a basis for detailed analysis. Wben the variants are reviewed by a broad circle of mers, however, it is more useful and convenient to""

tables correlated with these graphs, such as Table 5. "Clear! y unacceptable damage" (CUD)- that which is maximally possible to achi "

with the selected initial data. How should a lay user view this table? The most important thing to consider wh"'

The Methodology: The Estimate of Nucteal' De*""" . ..... Bnce Usmg ca I 133

P(Nc.Nud) under value of N ud

Expenses SNF 6C

variant (conditional under "b" impact

under"a" impact

- -units) clearly 600* 800 1,000 unacceptable

damage

1 -- 0.4 0.3 0.2 0.5*

2 2 0.7 0.55 0.4 0.6

3 3.5 0.8 0.7 0.6 0.6

4 6 0.9 0.8 0.7 0.6

li . a~le 5. Evaluatmg SNF alternatives from the point of view of the deterrence effectiveness.

flgunng this out is what da 1 . . mage va ue the a glven time (ud). Let ~s assume that th. h :gressor p~rceives as unacceptable at a heads.. . . lS as een estabhshed at the level of 500 war-

Besldes, lt ls useful to see practical . table noted by*): ' maxlmum reference points N d p (. pr an pr 1n our

PP,= 0.5 and N = 600 pr

No": .for the analysis of the table itself ( of stab1hty p = 0 3

d N Let us remember the the t. 1 b . F " an "~ 300) , ore rca oundanes lrst the user look t h . ("a") Th . . s a t e potential result of the ene , " . e probablllty of retaliation wh 1 . " my s adventurer's strategy"

ceptable dam h ose va ue ls guaranteed" b . th . age tot e aggressods, in this first h to nng clearly unac ca e practtcal boundary of the area of stability he:": att elevel of 0 . 5 . It is located on ~t re~uce it. But is this a lot or a little &:m ';h me:ns that as of now the aggressm

ofBadurp y'~'Law" is highly respected in the Uni e pomt of v>ew of deterrence/ on! Luck, meanmg that if something bad t: States - hece we call it the "Law

prJ, ~~mpt. In our case, "something bad" can aptn it will, and on the first and de a tty of 0.5 . This is no Mur h ' L ~eanst e end of the world, and with a thr~~ent? It seems to be. At leas;, ~:thi~w ~n ~~tty; this is a S0/50 chance. Is it a a Prov:~d ~p to 0. 6 as in the. table, or even hfgh: i~ ~ ~e potenbal victim to rai>< this

IJ . a1>~e counter questwn' how much e "':enough money. (Let us ask trern:~g fmished with the "adventurer's s~:~;d ~?~ hke for. "full" confidence?)

. ow tempted would the aggressor be to us!yh t, e use~ Wl.ll regard the other ex-t e Pyrrhlc vlctory" strategy? How

134 I 17: Nucteal' Command, Conrrol, CoopBI'alion

Would the damage be that unacceptable even if "Pyrrhic" is this victory in reality?

retaliation is highly probable? . d th t 600 warheads is close to the lower bound-Let us look at the table. If we consl ~~ had . the beginning that Nud = 500), then

ary of unacceptable damage (we esta ~ Oe 4 l~e can say that our "sweeper" is not variant 1 seems weak since P(N2':610d0d) -1 . ~hat these conditions are enough of a de-

. f h nemy shou ec are complete. Even l t e e . ht to chose variants 2 or 3. terrent for him, we neve~theless.~ave the :~:le" at the level of 800, not 500 warheads,

If the damage is constdered unaVcc~p t 3 r 4 seem most likely. If Nud = 1,000 . h ld we choose? anan s o . tc which vanant s ou . . . ll more expensive vanants, etc., e .

warheads, we may have ~o revtew potent:t :nd resent as concisely and graphically as What is left now is to mtegrate th.e ~e.s di~ons state the lower boundary of unac-

'bl What have we got? If our mtttal con posst e. h d we recommend: If ceptable damage as 500 war ea s, . d ntees P(N2':600) = 0.6-0.7. we

I . 1 t penstve an guara 6 0 8 Choose variant 2. t ts eas ex . . d ' t will be higher: P(N2':600) = 0. - . .

sive variant 3, thts m tea or d" select the more expen . I "0 roducts are guarantee . .

Dear users, make your chotce. ur p d l of the probability we are seekmg h f presente va ues 1

Let us note here that t e ran~e o . P(N>600). Therefore, for genera use, . . 1 " k"06tsP and0.8ts - .

is not JUSt a stmp e or . . . f cu\ f the deterrence indtcator: we may suggest the followmg ormu a or

p (oo --+ 600) = 0.6 --+ 0.8. . . 1 of com letely unacceptable damage.

The infinity sign here mdtcates thh~ lkevef ore fnteresting and elegant formula, but 'll be able tot m 0 am

Perhaps someone Wl h ld be able to understand tt . the important point is that everyone s ou d't ' 1 by way of example. But even

d h of course con 1 wna ' b ld ot All figures presente ere are, . . ' h commendations stills ou n

lt f mt researc ' our re d t if they were obtained as a resu o JO no ri bt to - recomm.end a specific \ ~vel o e er-_ be taken up because we cannot - have ~h h uld be solved democratJcally. Wh~t renee. This is a task for everyone, at1~ onhie 'ttat ~ oR commendations presented in thdlS

t 1 a method rot t as 1 accor -is presented here ts a o~ ' . T order of choices, no cone uswns chapter should be percetved as suggestmg an

ing to specific values.

C3 in Russia and the United States: A Comparative Analysis

4.1 Russian C3: History ol Establishment and Development

4.1.1 The Beginning Stage (195918841

The formation of the SNF C3 began and has improved together with the formation and development of the forces themselves. With the appearance of the first nuclear weapons delivery vehicles (long-range bombers, short-range and medium-range bal-listic missiles), they were taken under the strict and centralized control of the supreme level of the Soviet military-political leadership. The main functions of nuclear com-mand and control - blocking unauthorized use, development of operational use plans and perfecting procedures for direct management of nuclear strikes - were the exclu-sive prerogatives of the General Staff.

The lower level c8mmand and control organs of the Ground Forces, Air Force and the Navy had the tasks of routine command and control over the weapons' con-dition and maintenance.

Initially, engineering brigades for nuclear ballistic missiles were included in the com-position of the Supreme High Command's reserves. The 1959 appearance of a new military service- the SRF -was a huge step in regulating command and control of the rapidly developing nuclear weapons, and was used to create better conditions to de-velop the doctrinal and technical foundations for using this fundamentally new means - as they thought then - of warfare.

Today, everyone officially recognizes that nuclear armaments must not be used, and that they may be regarded only as a peculiar means of deterrence. However, this per-spective assumes the necessity of a more-or-less logical theory of their use.

The nuclear doctrine of the early 1950s freely accepted the possibility of nuclear war. On Sept. 14, 1954, in the Soviet Union's southern Urals village ofTotskoe, a 20-kiloton bomb exploded at an altitude of 350 meters. Thirty-seven minutes after the explosion, "attack" troops wearing protective clothing passed through the explosion's epicenter. The fact that nothing immediately happened to them gave the Soviet estab-lishment a basis for developing a broad campaign to disperse the fears associated with nuclear weapons. For some time this succeeded. In spite of the widely known long-term consequences associated with Hiroshima and Nagasaki, many in the Soviet Union, especially the military, began to think that "there was nothing to fear in it." Unfortu-nately, the inertia of this impulse proved quite strong and, for many years, defined the character of military-ideological goals.

136 1 Cl: Nuc/eal' Command, Contl'ol, Coopel'ation

During a revi w of the Totsko exercises, ikolai Bulganin, then minister of defen e, drew attention to the 11ecessity of developing troop operations in a nuclear contaminated zone. ikita Khrushchev further used these "results" in the political truggle with Malenkov, accusing the latter of ignoring r at experimental data which, according to the general secretary of the Communi t party, confirmed the admissibi lity of nuclear war.

fGcially, this position was embodied in the so-called " okolovsky's trategy. ' Its appearance is connected with the publication of Military t-rategy, edited by Marshal V. D. Sokolovsky. ln the book, basic elements of conducting a nuclear war were formu-lated, and they defined the main trend in the subsequent development of oviet military theory, incl uding the operational and teclu1ical aspects of nuclear command and control.

The main question then was the ab.i lity to maintain command and control of troops under nuclear warfare conditions. fficial assurance that this wa the case in the a-viet military were pure bluff, and wer actually refuted in the eneraJ taff and the scientific research institutes' classified materials. Everyone understood that at that time there was no technical basis for command and control of nuclear war.

At first, in the proces of establishing nuclear forces, command and control was ac-complished using the existing communications network ofthe military districts, other services and civilian administrations. The overall technical level of military and civil-ian communication wa quite low. Here, antiquated types oft lephone and tel graph equipment wer in use, as well as radio stations n1any of which w re pre-World War II design. ecure communications wer practically nonexistent, and therefore the sole means for transmitting important information was ciph red communications, with all of its complications and low efficiency. pecial difficulti arose in orgaruzmg com-mand and control of the new RF sites due to their isolated location.

Additionally, from the first day of the nuclear forces life, demands fo.r reliability, efficiency and ecrecy in th ir command and control became rigid. All existing meth-ods had to be inve tigated in order to mitigate the sharp contradiction betwe .n big wishes and very limited capabilities. For that, the scanty xesources of the local civil networks of the. 1inistry of omm.un.ications were exploited to their limit in the inter-ests of the new users. The best mainline channels of cabl communications with Mos-cow were relinqui bed to RF on an around-the-clock lease.

But even this wa not enough, especially for the proposed period of military opera-tions. Therefore, ba ed on a special government decree an operational system was~traduced, allowing nuclear forces in extraordinary circumstances a significant addt tiona! number of the Ministry of Communications' able and radio channels. ll .of these channels were pre-assigned to specific ites, and their transfer to the UUJtS was regularly practiced during exercises and training. Besides, urgent subscriber coP nection by any ofthe other civil ian channels, in case all available military channels went out of order, was often used. In this case, by transmitting down the line the set pa~s

. f th . . . l d district word Vozdukh! (Air attack!) or Samolet! (Au-era tl), e extstmg regwna an L bl d 'cation between commumca.tJ.ons centers, wtbtm one to two mmutes, ena e commuru

CS in Russia and the Unhe.d ,,.. . Quiles. A Compapatfve Analysis I 137

nuclear sites separated h The absence of indu~~:u:~~a~~;:c~~r:~o~::nds of kilometers.

would correspond to the hi'gh . ns of command and control which reqUirements fo d th

made substitutes. During the c . , rce. e creatiOn of temporary, home-. h. se lOrmatiVe years m t. . . .

wit m SRF, naval and airborne SNF . . ' nova IVe activities were prevalent d d 1 umts. Different ty f an Isp ays, crew training simulators d p~s ~ command post panels

crypto-stability level were created b an ~el~ure c~m.mumcatwns equipment with set Y spec1a Ists Withm th automated systems did not pla d . . . . ese umts. Of course these

I . . Y a ecisive role 10 Imp 1 ' contra . Still, Independent in t' . . . rovmg nuc ear command and . ven IOns were distmgUished b . . .

ways, such Initiatives by young officers hel ed y ongi?ahty and, in several of a new standard of work. p to speed the creatwn and introduction

. As a whole, the old technological base created lar . lmg nuclear forces. The Min. t f C . ~e problems m managing the fledg-

Is ry o ommumcatwn ' bl command and control system operated k d . . s ca e network, on which the wor e mtermittentl .c.t h m commumcatwns with sites Th f . y, Oil en avmg long breaks b . e use o radw comm . . ecause of secrecy considerations S t ll't . U~Icatwns was severely limited C . a e I e commumc t d 'd . ommand and control deficienci b . a ~ons I not exist in practice.

. . es ecame especially t bl . trammg exercises, where it was clear th t th I n~ Icea e m command and staff

1 a ey were s owmg d .c. h . operatwna command meth d Th . . own llirt er refmement of . o s. e transmissiOn of ord d . .

were earned out, as a rule by t I h . h ers an receivmg of reports . ' e ep one, Wit a large n b f .

catwns. The documentation of I' d d urn er o correctiOns and clarifi- genera 12e ata wa d rom the staff assigned to follow the t' f s carne on manually, with officers I d . opera IOns o each co d d' se ecte officers who took note fd t . . rps an IVIswn, and specially-

. o a a on missile launch t F o entire exercises was to try to 1 'f h . . es, e c. requently, the purpose

d h c an y ow many miss 1 I' . . an ow many were destroyed by the "enem " I I es accomp Ished their mission, new weaponry required a fundamentall d 'fJ' t became clear that the fundamentally

y I erent command and control system.

4.1.2 The li'ans/Uon to lh A The . . . e .'lltomatlon ot Command and Conll'ol rt9851970J

growmg sophisticatiOn of nuclear . '1 brought to light the question of t . . missi e weaponry and methods of its use

. a ransition to aut t d ex~enence of the first several years f oma e command and control. The latlOn of the basic requirements fo o ;~mbat duty allowed a sufficiently clear formu -trol (avtomatizirovannaja systema ~a ulur~ automated system of command and con-reliable, fast and secure transmis . pra~ enzdya) . It ha~ ~o support, first of all, highly

Of the thre . d . Sions o or ers for missile launch. c II e mentwne reqUirements s e d b . a. YaU scenarios "gamed" d . . ' p e . ecame the mam one, in that practi -

strik unng exercises envisag d th h es, after which the Soviets ste . e. e exc ange of massive nuclear

(he\vever, in th . . f ~ m was, JUstly, considered to be completely d t d th .t\ e opmwn o Soviet and Am . es roye e encan ex . tnencan system) Alth h h perts, the same outcome applied to

:~:~:~::atthe peopl~ insid:ut~e ~;1i~;;;_J:~~:t::~:~~;~:~:~bt:umt tahke givehn ~ir-e a command a d 1 e a c Oice

n contra system so survivable that it would be undam ~

138 1 fJl: Nucleal' Command, Conll'ol, Coopel'ation

aged in nuclear war, or to make one so fast that it would succeed in launching all mis-siles in the intervening time from the warning signal of an attack to the arrival of the enemy's first nuclear warheads.

The emphasis was placed on the so-called response-encounter strike (OVU - otvetno-vstrechnyy udar) - the American terminology, launch on warning (LOW). In reality, such a concept did not in the least disturb simultaneously gamed follow-on operations after exchange of strikes, i.e., to re-establish the command and control system, to re-target the remaining missiles and to conduct repeat launches for the "final'' victory. This was the concept of response strike (OU - otvetnyy udar), or launch under attack (LUA).

War gaming on military maps had to be supported by actual development of appro-priate command and control means by industry. It was decided to finance both con-cepts, although, notably, less financing was allotted for the second one. At the early 1970s successes in space technology predetermined the selection of LOW with the strengthening of the ground missile strike EWS by satellites. The SNF's command and control system now had "improved eyesight."

Besides the struggle to gain seconds, necessitated by essence of LOW, high reliabil -ity would be required from the future automated command and control system, both in peacetime and in limited conventional warfare. This requirement was not easy to accomplish, if one keeps in mind the deficiencies of the existing communications net-work on which the automated system had to rely. And, of course, besides the major function of bringing the launch orders, the new system had to support day -to-day management of troops and remote monitoring of the missiles' technical condition.

By the beginning of the 1960s, there were two main directions for developing such a command and control system. The first was characterized by the idea of using special integrated circuits in the automated system. This, according to the designers' intent, was supposed to provide relatively flexible logic for using nuclear weapons, i.e., to change, if necessary, the automated system's working algorithm by re-programming. This direction was promoted by the Moscow Research Institute of Automatic Equip-ment, led by Vladimir Semenikhin (subsequently, member of the Academy of Sciences).

The other proposed direction for an automated system used elements with rigidly-fixed ("hard-wired") logic, produced on ferrite-cores (Ferrit-ferritovye yacheyki) . Taras Sokolov is rightfully regarded as the father of this direction. At the end of the 19 50s, while working as an instructor at the Leningrad Polytechnic Institute - LPI, Sokolov had already worked out and successfully put into practice the ferrite-cores idea. In 1961 he established a research design bureau attached to the LPI (presently the Scien-tific Production Corporation - the Impuls Corp.) and invited his most capable stu-dents to come there. These were, basically, young graduates of LPI who challenged the recognized authorities from Moscow. By 1964, the LPI Design Bureau compl ted the development of a highly reliable ferrite logic board sealed in a special compound. Based on it, the first experimental pieces of SRF command and control eq uipment were built by the end of 1965. 115

cstn Russ/a and the Unned States: A Comparatille Analysis I 138

. In the history of the oviet military-industrial co mstances of real competition n . h . mplex, there have not been many .d d OJ a uge order More oft tl

Cl e on a monopoly basis ac d. : en lan not, matters wer de-uld . cor mg to an edtct from b Tl .

co b constdered an xc ption and th b . a ave. le story just discussed undoubted u efulness of uch , f elsu .sequent course of events onfinn d the

p 11 I a orm o se ectLOn ara. e r s ~ch and developm nt ended with. com . . . .

ducted m the first half of 1966 . If . pettt:Jve prototype testmg, on-. JQ 'LOscow and Leruna. d 1 'h

tammg the 'R " automated . d ,ra e truggle was for ob-f d comman and control sy t d un amental efforts in tllose s em or r, Slnce the country's

. . years were concentrated 0 . rnent of thts s rvrce. fficers of th ,r . . n upportmg raprd establish -tl . Itru try of Defense e h le troop (ma.mly from the SR ) . 1 searc mstltutes and from a1 . wer asstgned to th M d L .

tlon tnals. I was among tl1 m. oscow an enmgrad opera-. The comp titive trials, which lasted for ev .

viCtory."6 Tt turn d out that . b eral month ' wer capped by 'okolov's d . . equtpment ased on th r 'd r . . excee ed m r bability th anl d . tgr ~ernte-core -!ogre notably

. a ogous evJces based on s~;JJ . c . Cutts. t the same time a reJ t. e1 [J ' "' u:npenect mtegrated cir-. a 1 v Y rna set of c d d mt~ the automated system developed by the LPI omn~an ~1 r 1 ort ' "hard-wiud" fictent for RF command and co tr I f h . p ctal destgn bureau, was fully suf-

th n o o t at penod Forth k f b' .. note at later the Leningrad gro al d . . e sa eo o ~ectJvtty, we will 'bl up so rna the trm1Sltlon . d all 1 e computer logic, but with greater . 1. bil' gra u y, to more flex -I . . re ta tty.

n sptte of_ the d ftmte re istance of several h.i hi . . . . the tll n mmtster of defense 1 1 a) . g . y placed mdtvtduals, m particul

140 1 fJI: Nuc/eal' Command, Conti'O~ Coopet'ation

The functions of the highest level of nuclear command and control, the most impor-tant of which is making decision together with the political leadership on use of nuclear weapons, have not changed much since then . At the same time, the technical founda-tion of command and control - the system of command posts and corresponding means of automated command and control - have evolved quite substantially.

Until the early 1970s, work in the highest levels of SNF command and control had mostly not been automated. Information from the troop was collected by telephone and telegraph. The necessary calculations and generalized data input to the display in the General Staff's Central Command Post were petf ormed by hand. The same can be said about the work of the NCA. During a period of threat, they were supposed to gather in one of the heavily protected underground bunker , which could, in principle, survive nuclear strikes, but which were not equipped with effective means of auto-mated command and control.

The General Staff's Central Command Post transmitted nuclear employment or-ders to the troops primarily via the Monolit system. The Monolit simultaneously trans-mitted via cable and radio coded signals to be decoded at reception points by opening special packages, which were kept under strict control in the safes of the troops' com-mand posts. These packets had been prepared beforehand in the General Staff, then circulated to the troops, and were periodically updated.

The Monolit system had existed even before the establishment of the Soviet nuclear forces. One must give credit to the military signals specialists who ensured for many years that Monolit improved troops' command and control. Many crews were virtuosos when it came to receiving Monolit's radio transmissions under powerful jamming by the "enemy. ' Individuals, designated Champions of the U.S.S.R. and Europe in ham radio operation, frequently served on duty at the most crucial elements of the command posts.

However, the new weapons' stringent requirements for command and control pin-pointed the main shortcomings of traditional types of communications, including the Monolit system. After the order was received through that system, the missile launch operations could be carried out (or not) only by the crews at the lowest levels of the nuclear forces. Remote launches or their urgent cancellation, originating from Mos-cow, were impossible. The old system's speed was also no longer satisfactory.

Members of inspection teams, which frequently visited the troops during exercises, time and again witnessed how outdated Monolit had become even from the psycho-logical point of view. In a decisive moment, a duty officer of the combat crew might fail, sometimes for several minutes, to cut open the packets with scissors because his hands were shaking so badly. One does not know what frightened the officers more -an imaginary nuclear war or the real inspector from Moscow, calmly observing this unique procedure. In any case, minutes - crucial for evaluating the combat readiness of a regiment or a division - were wasted.

The problem of using scissors was recognized as so serious that troop inventors were entrusted to find the solution. The packet was constructed with a pull-string, on wwch

cain Russ/a and the IJJ ~ 'll ted States: A Compal'ative Analysis I 141

an operator could tug to i . combat readiness mt_ned1ate1y open the packet. In o e . n 1 was thus Increased by 18 p ratiOnal documents th 1 e y, greater. seconds, although in realit "t , e

Th . . y 1 was, most e sttuatton changed in man

effortds of the C00perative heade~ ~ey peecmts d~kr:~~g the 1970s. Under the guidance and was eveJo d d eru n1n a se f the qualtypef an mtroduced at the highest level of . ne od automation techniques

! o nuclear coinman and 1 . Play d b h . command and con trol. Imp t . con tro, Improving Y t e d s1gn M "k! . or ant roles m th . Lesh h k ers 1 1a1l Log inov v 1 e.u creation were c en o and t1 .1. , rctor l" onash I .. B l . d ' . 1e mJ ltary customers; far hal f . ev, gor MJzin, Yuri e. ov an en . Vladimir ruzbi . E o ommumcation Troo s And . . i1khavlenko Vladimir h t 1: nm, ugeny Evs6gneev Anatoli z kp leJ

- ' u ov .r 0 t t T . - m1en ov v t _ la tter wer tragically kill d . , ~san m rof1mov and Yuri Ryab . Tlu . tc o.r

. e m 1987 m fi mm. ee of th co~cerrung equipping the troop ofWa;s::c; t tdJsas ter.in Hungary, during a missio: an control systems. . ac countne with automat d

Th b e command e asic automation systems for

~he authorizing system Kazbek, the n~~~:~nd and control at the highest level included at command and control KSBU (k wn system Krokus, and the system fc

th t h ommandna or com-a ' at t e Central Command Post of the G ya ststema boevogo upravleniya). Beside

mand and control system was installed T . eneral Staff, the Signal automated com-group of Sokolov. With the hel of KS. his system was developed by the Leni

~hntrol of t~e airborne and sea-b;sed SN~U, ~~.:e~eral Staff exercised comman~g::~ ro~gh ?tgnal, SRF command and co~;n 1 ewise of the Armed Forces as a whole

descnbed In more detail . th rol was accomplished Th . A 1 k b k m e next section . ese systems are

1 . oo ac at the history of creation and ?wr~g. In contrast to the Kazbek and Krok development of KSBU will show the fol -

tJon In a rather short period of time b t us, developed and introduced into op trodu t' e ween 1978 d 19 era-Its d c rtn Into operation of this large, multi -lev 1 an 83, the development and in-

eve opment had begun in the e system dragged out for man

~::; ~erahtio1nSal sys.tem during sta~:r:~s~~~~e~~~ it was officially accepted as~ ~::s~ , ars a erge1 Akhromee e command of the chief of G se~ar~e elements of the system :e::~:~t on ~ombat watch only in 1985. Untilet~::1 a uc a long gestation period was d fc xpehnmenta1 use by the military. ' PProach b th ue, or t e most p t t

ing on intery d~ customer. Instead of introducing one K~B;l ~ not altogether successful Inents h. :e Iate versions of the system the mlt e ement at a time and rely-new re~urc resulted in significant additi~nal wo~~ :ry const~tly advanced new require-A '-'o 1rements were advanced etc Th. . . n expenmental sections, after wh1"ch ~ llleev t k h ' Is VIc1ou 1 l< BD d ~o c arge of the program Akh s .C1rc e was broken only when

an ell] d romeev pard c d b consider bi . aye a good relationship with its 1 d . onst era le attention to the f~rrcn \V ka e Influence upon the Soviet h. h genera esrgner, Semenikhin, who had . ea ened o 1 . rg command Th . ll}terest i n y m the last years of his life (h d. d . . e prestrge of Semenikhin 's

n command and control and because f:h 1e In 1990), mostly because of reduced o e general problems in the Soviet Union.

142 1 1}1: Nuc/eal' Command, Contl'o~ Coopel'alion

During the last decade, a high-level command and control system has been devel-oped within the framework of the Armed Forces global command and control project, named Tsentr (Center). 11 8 The highest level in that system is called Vershina (Summit). The projects are intended to enhance reliability and speed for preparing authorization, to automate collection of information and analysis of the operational-strategic envi-ronment, and to introduce some new means of C3 However, the political turbulence of the 1990s has somewhat slowed down the implementation of these projects.

(B) The SRF Command and Control System The SRF moved toward automation of command and control independently. De-

spite Semenikhin's closeness to the powers that be, and his sharp competition with the Leningrad group, all the command and control issues for the main service of the Soviet Armed Forces were decided by SRF officers and the LPI Design Bureau. The General Staff was involved in the details only when it came to protection against unauthorized use, plans of combat employment and integration of the SRF automated system with the highest level of command and control.

The history of how the creative collaboration between SRF specialists and civilian scientists grew and developed is interesting in itself. It shows that even under the bu-reaucracy it was possible- in a relatively short time - to create from scratch the basis of a theory of nuclear command and control, to develop and manufacture the necessary technical means, and to equip the troops with them.

It is a striking historical fact that the basic concepts of nuclear command and control, such as protection from unauthorized actions, remote monitoring of weaponry, retar-geting, the degree of automation oflaunching processes, efficiency, reliability, etc., origi-nated and obtained real maturity in the minds of the General Staff and SRF Research Institute specialists, but also, to a similar degree, in the purely civilian team of the LPI Design Bureau. More than that, not bound by the narrow confines of military tradi-tion, Sokolov and his young colleagues, Vladimir Petukhov, Boris Mikhailov, Leonid Vasil'ev, Anatoliy Greshnevikov, Vitaliy Mel'nik and others, provided many new ideas precisely for the command and control doctrine. These and other designers of SRF 0 systems are the subject of published memoirs by their SRF colleagues and veterans. 119

Their ideas were immediately discussed with the military in Leningrad, in Moscow, at the SRF Research Institute in the Moscow suburb of Bolshevo and then implemented at the design bureaus and factories of the Leningrad research, development and pro-duction network. Nobody, in essence, hindered the Leningrad team's work, moneY was regularly disbursed from above, and the appropriate results appeared soon nough.

Already in 1967-1968, the first serially produced levels of the Signal automated sys-tem had begun to enter service with SRF formations. Equipping the troops went froOl the bottom up. First, they created the equipment for the command posts of mediwn range missile battalions and regiments, and ofiCBM regiments (levels 6 and 5). At tbe same time, in the Design Bureau all the other levels of the Signal system were devel

cain Russ/a and IIJe United States: A Compal'atlve Analysts I 143

oped- for the missile division (level 4) ar (1 and 2). These types of equipment , .my evel3), and the SRF main staff (levels 1

were given to the troop 1 t . h d years each. For the training oft ff' s a er, Wit elays of several M roop o leers, courses wer t L

oscow, and at the strategic missile sites. e se up m enmgrad and The Signal automated system belongin t th .

1960s through the mid-1970 ) it d g o. ~first stage of automation (the late-. h s ' a owe transmission of 13 f d mgt e establishment of watch b k Ixe commands concern- . or com at wor modes for th .. o troops to vanous readiness lev 1 th b e system, the transitiOn d e s, e num er of the co b t 1 or er concerning the start and c n 11 t' f . . m a p an to use, and the

c h . a ce a wn o a missile lau h Th lOr gat enng up to 20 confirm t' d . nc . ere was a provision

. a Ions an reports m respo t . questwns concerning the stat f . '1 nse o prevwusly prepared

. e o miSS! e and troop read. d 1 accomplishment. mess, e ays, and mission

Durin~ that stage, missile launch installations th . other device, not included in th S. l emselves were equipped with an -

e zgna automated system Th. h remote command and control d . . . Is wast e so-called

. . an monrtonng system (SDUK . d' . upravlenzya t kontrolya) 120 D . h - szstema zstantswnnogo

. esigners w o worked o 1 h . . several variants. The SDUK 'd . . n aunc positiOns developed its

' on one SI e was Jomed t th k - by an underground cable - to th t , d o e roc et, and from the other 1 1 e au ornate command a d t I owest eve! of command post A d 1 h h . n con ro system at the

. n ' a t oug a vanant I t . 11 . . . mated system and the SDUK d . tl . e ec nca y JOinmg the auto-

Irec y was envisaged the . d tern use was assumed to be rna 1 h h ' . pnmary mo e of Signal sys-d. . . nua, w en t e crew earned t 1 h . IVIswnal or regimental command t fi . . ou aunc operatwns at the fi h . pos s a ter receiVmg the 0 d f b ore, t e first-stage automated sy t 11 d r er rom a ove. There-t 1 b s em was ca e a system oft ro , ut not yet of weaponry. roop command and con-

Equipment of the first generation was uit I . especially at the division level It t ql d e. ~rge .and required considerable power, t. I . sex erna efiCtencies co b' d . h 1' . IOna capabilities evoked c 't ' . f . , m me Wit lmited opera-t h . ' n Icism rom competmg d . d d' ec meal solutions to be us d . th . esigners an Id not allow the th SR . e In e mterests of the entire A d v

e F Itself, the benefit ofth f ' t d . rme l'Orces. However, in . e Irs mo est step wa d. 1 c

bona! experience not only revealed t h . I d f' . s .Imme late y lelt. Actual opera -. ec n1ca e lCienCie b t

questiOns of operational nuclear d d s, u gave a powerful surge to n comman an control Not . I SRF .

ow performed without the use f th . a smg e exercise was posts, as well as operation and o . e automated system. Duty crews at the command co . mamtenance personnel l't 11

ncernmg the system 's refinement ' I era y poured suggestions

The second stage of the SRF auto.mated c M system (mid-1970s through m'd-1980 ommand and control system - the Signal-acteristics. A mode f t . l s).- already possessed noticeably better char-! I o au omatic connectwn k d . eve and subsequently at t . . was wor e out m detail at the lowest

Were, as a rule, conducted~;~~~ rang~and lfn troop exercises. Actual missile launches Psts of ad ' . . e pus mg 0 a button at higher levels the d !VISion arm t h C - comman taff. Thi's b ' y or a t e entral Command Post of the SRF and G 1

ecame a type f t 1 . enera o n ua' crownrng many major activities of the military-

144 \ Cl: Nucleal' command, contl'ol, coopel'alion

. l ded the higher leadership of the . 1 The main designers willingly me u 11 nches. These lead-

industnal camp ex. . such remote-contra au . 1 d of the Ministry of Defense m w th this goal in mmd, regu ar

~~~~~~::elves, more than once, pus~edn~=o~~r~~:~uip~ent was specially installed at . l M automated comman a

level4 Stgna- . . t the Central Command Post. . . channels of the General Staff to. lmlta e and control system, comroumcauons . - that

Simultaneously, m the commabnd d - cab! 'radio, radio-relay and s~tclhteff' th . h ical t pes began to e use . . At theSRF mam sta m e

vanout~alply~:ight:Ued the stability of its work m peacteatgllens\or each missile division was essen 1 t d system ou k 1970s a daily accounting of automa e t' due to the measures underta en, was

d ' t d The combined monthly outage .lmde, t command and control system

con uc e . d dlze remo e . '1 b to a few minutes. A new stan ar e new third-generatlon Sl os y

r(eSdDuUceKd) code-named Pauk. (Spider), was created folrytphrotected against unauthorized ' . B It was extreme

the LPI Special Deslgn ureau. -121 V ubsystem was devel

launches. . s nal-M system., the yuga 5 _ le-w thin the framework of the 1g . 97 6 1978 and was u ed to create supp d \22 lt was introduced to the troop m 1 ~ o t~ at all levels via broadcast-o~ly

ope . ths for delivering orders to conunan pl s -SRF leveL the V'yuga eqUlp-mentary pa . . h els At a ower . '1 l cl radio and satellite coromumcatla~s:s ~n l- .1 in order to allow a remote rru Sl e aun 1 ment was electrically interfaced wlt tgna b to be conducted by radio from the. cen~er. d technical characteristics of the V'yug~ su -

In view of the acceptable operatlOna an . Bureau was accepted in 1982 fort e o-

~=~o~~~ ~:~:::~~~:~~e~;~~o~~o~=~~u~;:e: ~i::"'::~!o';,:i Additionally, Signal-M.~a~ inter;:s ~ade it possible to accelerate the pro~es~:

SRF rocket armies and dLvtslons. ea ons and to solve a number o o

Ope:ational re-planning for the use of nucbl~swoft~e LPl esign Bureau developed ~e

L t on mern er . to a unl-om::putational problems. . a er : 11 b . SRF computatiOnal centers m c . . tymg a aslC

. formation-calculatlOn ystem, 1 Ill' F command and contra fied network. k in the development of SR . . Before it,

The most important landmar Th' step was new in pnnclple. . p t ystem lS 1 regl-has been the creation of the en~e.; s . a a ~ommand post of a lower leve ;ef, a the all orders were passed to the ~lSSl es Vltem reception of launch comman s rom

1. In the Penmetr sys h ment or batta lOU. . dio directly at the launc er. 1 ed by center could be accomplished by ra from a simple listing of the systems de~~:foUI1d

Here, it is pertinent to ste.p awayf tl working with the troops, qUlC s i.n His engmeers, requen y d d control system

the Sokolov group. f 1' ble function of comman an d by the the key to solving the problemT:.re;a was the communication channels use alit ln

complex cOombat sl~u~~:s~utst~:di~g terminal equipment andh:~~:~:yd~~:~:ansmis~ system. ne cou d t but all this would come to naug super hardened comman pos s, sion channels were vulnerable.

C8 /n Russ/a and the United States: A Compal'ative Analysts 1 145

The experience of using the first generation of automated systems had shown that traditional cables, radio and even satellite communications channels could not support reliable command and control in nuclear warfare or in warfare with massive use of pre-cision-guided weapons. Other means would have to be found. Thus an idea grew for the creation of an independent system of channels for combat command and control, specially intended for extraordinary conditions. Naturally, this could only be radio chan-nels, and hence the system received the designation of radio-combat command and con-trol (radio-boevoe upravlenie - RBU). At the installation stage for the Signal-M system, several types of such channels appeared, the most important of which became Perimetr .

The Perimetr system was accepted into service and, since 1985, has been on com-bat watch. It will be described in the next section. Some types of radio -combat com-mand and control have been described in an authoritative Russian publication. 123 The Gorn system mentioned there is a group of Perimetr command missiles, based on the SS-20 ballistic missiles. The Gorn was on active duty for some time, but then it was retired because the United States and the Soviet Union agreed to destroy ballistic mis-siles of this class. Several publications provide additional information on the channels of radio-combat command and control. 124

Finally, the smaller size, weight, and power requirements of this second-generation automated system allowed its accommodation into all command posts of the then new types of missiles . The standardized regimental command posts for remote launch (otdel'nye starty) type missile regiments - that is, without any launch crews in the im-mediate vicinity of the launchers - as well as the mobile command posts of the mobile Pioner missiles (SS-20, in NATO terminology), were equipped with the Signal-M, Perimetr and V'yuga equipment. This period demonstrated that, in the nuclear forces, it is impossible to separate the missiles themselves from their command and control system; these, in principle, constitute a unified whole. Hence, the transition from com-mand and control of troops to command and control of nuclear weapons began in the 1970s. This transition was actually completed with the creation of the third -genera-tion automated system - the Signal-A.

Signal-A began to reach the troops in 1985-1986. The lowest level of this system -(seventh link) - was installed directly at the launchers, and was an important integral part of them. Without them, not a single new missile could have been put on combat duty. This made the SRF leadership still more attentive to the question of command and control. It is not surprising that in this period the deputies of the SRF commander in chief, and at times he himself, were frequent visitors at the Impuls Corp. Design Bureau, other design bureaus and production plants carefully considering the time-tables for testing and production of the automated systems.

The introduction of the Signal-A system began also from below, from equipment in tqe missile complexes (levels 7 and 5) up to the division level and higher. This process has continued even to the present time. The interfacing of the Signal-M and the Sig-nal-A resources is accomplished at a functional level.

146 l IJI: Nuclear command, Conrrol, cooperation

Signal-A allows retaxgeting of missiles remotely from the General Staff or from the SRF main staff. Several sets of the flight plans (poletnye zadaniya) are stored in the equipment of the seventh level (launchers). For retaxgeting missiles, it is sufficient to

indicate in the launch order the number of the corresponding combat plan. Upon re-ceiving the plan number, the automated sy tem gives the command to the missile for the selection of the required fligh path, which is carried out automatically within 1 0 to I 5 seconds, after which the nrissi le is launched. Recently, much has been written about

retargeting. 125 The Signal-A system comprehensively uses all existing types of "traditional" com-munications channels and the high-reliability channels of the radio-combat command and control system. It has its own highly secure encryption equipment developed by lmpuls Corp., has an improved design and is easy to use thanks to the special mini-computers used by the operators. The extension of this system is being accomplished according to plans, and designers and customers are already looking to the future, where it is proposed to have a Signal-A\ system. (If, of course, by that time peopl will not have decided to throw nuclear weapons and all of their exc Uent infrastructure onto the

rubbish heap.)

(C) IJevelopmmt of c.,..,.,..,a and Control oJS..rB=rated organizational rea-ase an . airborne SNF were caused al b e ~ommand and control for the sea-

deploy~d m rigidly fixed areas and were ;i~e y techm~al complexities. SRF sites were a techmcal point of view th . rally dug mto the ground Th r . e1r command a d l erewre from tlons channels within th n contra was based on cabl , . bl' e country Even r d' d e commumca ta Ished more easily along w 11. d 1 a 10 an satellite communicati'o - e - eve oped ns were es-nnes and strategic aviation as a b . d permanent routes. Conversely s b t M ' asic mo e of d t 1 ' u rna-cen er. oreover, a constant . b u y, patro regions remote f h b maJor o stacle f, 1' bl rom t e su merged boats was the th' k f or re Ia e delivery of orde t d only by VLF transmissions." ness o the water, which, as is known, can b;spe:et:;:~ Sea-Based SNF

When the first missile-ca. . . in the earl~ 1960s, their coa:::g ~n~n:~~~e (SSBs or S~BNs) joined the Soviet Na of con_venbonal submarine . The main dif ol was not Mferent in principle from t~ planmng the use of strategic 1 erence was the General Staff's m