\\jciprod01\productn\N\NYI\47-2\NYI204.

txt unknown Seq: 1 21-MAY-15 13:28

DETERRENCE: THE STATE OF THE FIELD

AUSTIN LONG

I. INTRODUCTION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 357 R
II. THE IMPORTANCE OF GETTING COLD WAR HISTORY
RIGHT . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 359 R
III. POST-COLD WAR DETERRENCE: SOMETHING OLD,
SOMETHING NEW . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 366 R
IV. THE PERENNIAL QUESTION OF CREDIBILITY . . . . . . . 370 R
V. CLANDESTINE CAPABILITIES AND DETERRENCE . . . . . 374 R
VI. CONCLUSION . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 377 R

I. INTRODUCTION
Following decades of development in the Cold War con-
text, the study of deterrence entered a new and less well-de-
fined era in 1991. During the Cold War, the focus of deter-
rence studies in the West was overwhelmingly along the Cen-
tral Front in Europe and, by extension, the nuclear conflict
that could erupt from a confrontation between the United
States and the Soviet Union. The questions raised by this cen-
tral challenge were extraordinarily difficult, but the outlines of
how to answer them became clearer over the course of the
Cold War.
After the Cold War ended, deterrence theory faced new
problems. First, the interest generated in deterrence by the
threat of World War III receded to some degree. Particularly
in the 1990s, it seemed that intervention had replaced deter-
rence as the central question for strategists as the United
States sought to use its military power more expansively. With
Russia struggling to emerge from the Soviet wreckage, and
with China only barely beginning its rapid economic develop-
ment, there simply seemed to be no “peer competitors” to de-
ter.1
Second, it was unclear how deterrence theory fit non-Cold
War contexts. While few believed deterrence was entirely irrel-
evant to different contexts, such as the post-Cold War Middle

1. This is not to say the U.S. government did not worry at all about the
emergence of a new peer competitor. See THOMAS SZAYNA ET AL., THE EMER-
GENCE OF PEER COMPETITORS: A FRAMEWORK FOR ANALYSIS (2001).

357
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 2 21-MAY-15 13:28

358 INTERNATIONAL LAW AND POLITICS [Vol. 47:357

East or South Asia, it was also not clear how well concepts de-
veloped along the Central Front applied to confrontations
along the India-Pakistan border, much less along the Israel-
Lebanon border. These contexts deviated greatly from the
Cold War, with much greater levels of low-level conflict and
more direct geographic proximity.
Third, new technologies and techniques of warfare
emerged which complicated the assessment of capabilities for
deterrence. Advances in many areas, particularly computers,
intelligence, and precision guidance, have produced qualita-
tive changes that had to be accounted for in deterrence the-
ory. Similarly, new actors, particularly non-state actors such as
terrorist groups, became increasingly important subjects for
deterrence theory.
Fourth, with the United States occupying a hegemonic
role after the collapse of its Soviet rival, the emphasis in the
study of deterrence has shifted from deterrence by punish-
ment to deterrence by denial. Rather than threatening unac-
ceptably painful retaliation to deter, the United States and its
allies could make even attempting to change the status quo by
force impossible, or at least prohibitively expensive. This was a
significant shift in how many strategists thought about deter-
rence along the Central Front, where denying a Soviet conven-
tional military advance seemed unlikely.
The result of the foregoing has been a proliferation of
new deterrence theories and concepts in the post-Cold War
period. In contrast to what was viewed as the relatively straight-
forward problem of deterrence during the Cold War (some-
times referred to as “classical deterrence,” making it sound
charming but perhaps outdated), the post-Cold War challenge
of deterrence has been viewed as baroque and difficult. Deter-
rence is thus alleged to need revitalization and evolution to
address the challenges of the post-Cold War context.
Yet the shift of attention away from Cold War deterrence,
though understandable, has been unfortunate, as it has led to
widespread acceptance of a highly stylized narrative as fact. De-
terrence during the Cold War was substantially more compli-
cated than many post-Cold War accounts suggest. This loss of
focus on Cold War deterrence is doubly unfortunate as, during
the past two decades, vast quantities of new evidence have be-
come available through declassification. Similarly, more is now
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 3 21-MAY-15 13:28

2015] DETERRENCE: THE STATE OF THE FIELD 359

available about Cold War era deterrence outside of the U.S.-

Soviet rivalry. While some historians have made fruitful use of
this material, relatively little of it has made its way into think-
ing about deterrence by political scientists and policy analysts.2
This Article does three things. First, it critiques two of the
problems of the stylized narrative of the Cold War: the nature
of the Soviet threat and the nuclear balance. Second, it criti-
ques two of the post-Cold War concepts and developments in
the field of deterrence—tailored deterrence and cross-domain
deterrence—while at the same time highlighting the Cold War
antecedents to these concepts. It does so in order to highlight
that there has been a great deal of continuity in the problems
of deterrence despite some significant changes. Third, it high-
lights two areas that neither Cold War nor post-Cold War de-
terrence thinking has grappled with successfully that might be
fruitful areas for further research and analysis. In particular, it
focuses on two issues: the importance (or lack thereof) of
credibility and the use of clandestine capabilities for deter-
rence.

II. THE IMPORTANCE OF GETTING COLD WAR HISTORY RIGHT

The stylized narrative of Cold War deterrence in the early

21st century is relatively simple and straightforward. The So-
viet Union is characterized as having been “status quo” and
“risk averse,” and the military balance between the two is char-
acterized as one of “mutually assured destruction” that was ac-
cepted by both sides.3 The central challenge for the United
States, as one example of this stylized narrative describes, “was
deterrence of an attack on the United States through the

2. There are exceptions that have made excellent use of historical mate-
rial. See, e.g., ELBRIDGE COLBY ET AL., THE ISRAELI “NUCLEAR ALERT” OF 1973:
DETERRENCE AND SIGNALING IN CRISIS (2013); MICHAEL GERSON, THE SINO-
SOVIET BORDER CONFLICT: DETERRENCE, ESCALATION, AND THE THREAT OF NU-
CLEAR WAR IN 1969 (2010); AVERY GOLDSTEIN, DETERRENCE AND SECURITY IN
THE 21ST CENTURY: CHINA, BRITAIN, FRANCE, AND THE ENDURING LEGACY OF
THE NUCLEAR REVOLUTION (2000).
3. See LAWRENCE FREEDMAN, THE EVOLUTION OF NUCLEAR STRATEGY (3d
ed. 2003); see also ROBERT JERVIS, THE MEANING OF THE NUCLEAR REVOLU-
TION: STATECRAFT AND THE PROSPECT OF ARMAGEDDON (1989); RICHARD NED
LEBOW & JANICE GROSS STEIN, WE ALL LOST THE COLD WAR (1995).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 4 21-MAY-15 13:28

360 INTERNATIONAL LAW AND POLITICS [Vol. 47:357

threat of unacceptable retaliatory damage.”4 This stylized nar-

rative is then typically contrasted with the complexity of the
post-Cold War world and the need for new ways of thinking
about deterrence.
The problem with the stylized narrative of Cold War de-
terrence is that it is at best a gross oversimplification and at
worse factually inaccurate. While a full account of Cold War
deterrence is far beyond the scope of this Article, it is nonethe-
less worthwhile to highlight two major flaws in the stylized nar-
rative that are of relevance to understanding the current state
of the field of deterrence. These are the nature of the Soviet
Union and the nature of nuclear deterrence.
First, the widespread characterization of the Soviet Union,
reflected in the 2002 U.S. National Security Strategy, as “espe-
cially following the Cuban missile crisis . . . a generally status
quo, risk-averse adversary,” is not necessarily wrong but glosses
over important details.5 The intentions of the Soviet Union
during the Cold War, which may appear obvious in hindsight,
were hotly contested at the time. It is only after the fact that
the status quo, risk-averse nature of the Soviets can be diag-
nosed.
Yet for policymakers at the time, Soviet intentions were
not so clearly risk averse or status quo oriented. Indeed, as
March Trachtenberg has described, the very nature of the sta-
tus quo between the United States and the Soviet Union had
to be carefully negotiated.6 U.S. policy for deterrence was fun-
damentally driven by a very different perception of the Soviet
Union than the one portrayed in the stylized narrative.
For example, in a National Intelligence Estimate from the
end of 1960, the U.S. intelligence community characterized
the Soviet Union as having “fundamental hostility toward the
West” and “high confidence in the growth of the USSR’s

4. James Scouras, Post-Cold War Nuclear Scenarios: Implications for a New

Strategic Calculus, in DETERRENCE AND NUCLEAR PROLIFERATION IN THE
TWENTY-FIRST CENTURY 39, 40 (Stephen J. Cimbala ed., 2000).
5. The White House, National Security Strategy of the United States
(2002).
6. MARC TRACHTENBERG, A CONSTRUCTED PEACE: THE MAKING OF THE
EUROPEAN SETTLEMENT 1945-1963 (1999).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 5 21-MAY-15 13:28

2015] DETERRENCE: THE STATE OF THE FIELD 361

power and influence.”7 While the estimate concluded that the

Soviet Union was not reckless, it predicted that Soviet policy
would be one of “persistent activism and opportunism.”8 Five
years later, following both the Cuban crisis and the ouster of
Soviet leader Nikita Khrushchev, a Special National Intelli-
gence Estimate found that the new Soviet leadership was more
confrontational towards the United States than Khrushchev
had been, though again not absurdly so.9
The same may prove true of rogue states (or states of con-
cern) in the early 21st century. For example, the allegedly im-
placably revolutionary nature of the Iranian regime in 2014
may one day seem as dated as similar characterizations of the
Soviet Union. This implies that deterring Iran may be possible
using tools similar to those developed during the Cold War.
More importantly, recent historiography of the Cold War
argues that the Soviet Union did not become status quo and
risk averse simply because of some insight derived from the
Cuban missile crisis. Instead, it is now apparent that in the af-
termath of the crisis, the United States and the Soviet Union
were able to agree on a mutually acceptable status quo regard-
ing the division of Germany specifically and the political order
in Europe generally. Germany would remain divided, with the
anomalous status of Berlin preserved, and would be rearmed
solely with conventional weapons. This agreed upon status quo
actually relied upon both superpowers to collude against their
erstwhile allies in East and West Germany to thwart reunifica-
tion and nuclear ambitions.10
This second point, that a mutually acceptable status quo
may be possible, is critically important as it suggests that view-
ing post-Cold War adversaries as fundamentally different and
more risk acceptant and/or revisionist than the Soviet Union
is potentially wrongheaded. While this may seem naively opti-
mistic, the corollary is that such a status quo may require simi-
lar collusion against the interests of allies.

7. DIR. OF CENT. INTELLIGENCE, NUMBER 11-4-60, NATIONAL INTELLI-

GENCE ESTIMATE, MAIN TRENDS IN SOVIET CAPABILITIES AND POLICIES, 1960-
1965 1 (1960).
8. Id. at 2.
9. DIR. OF CENT. INTELLIGENCE, NUMBER 11-11-65, SPECIAL NATIONAL IN-
TELLIGENCE ESTIMATE, SOVIET ATTITUDES TOWARDS THE US (1965).
10. See FRANCIS J. GAVIN, NUCLEAR STATECRAFT: HISTORY AND STRATEGY IN
AMERICA’S ATOMIC AGE 57–74 (2012).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 6 21-MAY-15 13:28

362 INTERNATIONAL LAW AND POLITICS [Vol. 47:357

A potential example of this is the negotiation over the Ira-

nian nuclear program. The United States may be able to live
with a level of Iranian nuclear capability that some allies (e.g.,
Israel and/or the Gulf States) find unacceptable. This opens
the door for a possible compromise with Iran that would estab-
lish a status quo acceptable to the United States (assuming
Iran, like the Soviet Union before it, is willing to compromise).
This sort of collusion may be less acceptable than it was in
the 1960s, and so a mutually agreed status quo may be impossi-
ble in some cases. Yet the lack of an agreed upon status quo is
fundamentally and conceptually a different reason for deter-
rence failure than inherent risk acceptance or revisionism in
an adversary. Indeed, an adversary cannot truly be revisionist if
there is not an agreed upon status quo to revise.
The second major flaw in the stylized Cold War narrative
is that mutually assured destruction was accepted by both sides
and formed the basis for deterrence. First, it is very clear based
on declassified evidence that the United States never accepted
mutually assured destruction as a desired status quo and in-
stead planned and invested heavily in options for nuclear first
use. In other words, the United States planned that in a war
with the Soviet Union, presumably one that began with a con-
ventional military conflict between the U.S.-led NATO and the
Soviet-led Warsaw Pact, the United States would be prepared
and likely required to be the first to use nuclear weapons.
For example, it is clear in the now-declassified Nuclear
Weapons Employment Policy from 1974 that through at least
the mid-1970s there were only two planning assumptions for
strategic nuclear war. The first was the worst-case scenario
where the United States had to retaliate against a massive “bolt
from the blue” attack against U.S. nuclear forces on day-to-day
alert with minimal warning. Yet this scenario was extraordina-
rily unlikely, so much of the planning emphasis was on the
second planning assumption, which was U.S. nuclear forces
that had “generated” (i.e., were at full crisis readiness) would
be launched without being damaged by a Soviet attack. In
other words, in any crisis where the United States had suffi-
cient time to ready its forces, it planned to use nuclear weap-
ons before a Soviet attack.11

11. See OFFICE OF THE SEC’Y OF DEF., POLICY GUIDANCE FOR THE EMPLOY-
MENT OF NUCLEAR WEAPONS (1974).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 7 21-MAY-15 13:28

2015] DETERRENCE: THE STATE OF THE FIELD 363

Moreover, this first strike would heavily emphasize target-

ing Soviet nuclear forces in order to limit damage to the
United States from Soviet retaliation. Combined with massive
investment in intelligence capabilities to track Soviet ballistic
missile submarines and mobile land-based missiles, this strike
had the possibility of eliminating much of the Soviet Union’s
long-range nuclear arsenal.12 According to Soviet assessments
near the end of the Cold War, only a relative handful of Soviet
warheads might survive a U.S. first strike. A memorandum
from the files of a senior Soviet defense expert noted that, af-
ter the United States modernized its nuclear forces, “[e]xisting
Strategic Rocket Forces are capable of hitting 80 enemy rear
area targets in retaliation . . . below the calculated level of re-
taliation required- 200 targets.”13 Some in the Soviet Union
worried that even these few surviving weapons might not be
usable due to the destruction of command and control sys-
tems.14
The reason the United States planned for nuclear first use
was the belief that Soviet/Warsaw Pact forces would over-
whelm NATO in a conventional conflict. While the accuracy of
this contention was debated during the Cold War, it was none-
theless conventional wisdom. Further, for reasons of both
politics and economics, NATO and especially the United
States were highly unlikely to engage in a sufficiently robust
arms build-up to overturn this conventional wisdom.15
NATO thus always relied on nuclear weapons to deter
conventional aggression (an early example of what is now re-

12. Austin Long & Brendan Green, Stalking the Secure Second Strike: Intelli-
gence, Counterforce, and U.S. Nuclear Strategy, J. OF STRATEGIC STUD. (forthcom-
ing).
13. Memorandum from the archive of Vitalii Leonidovich Kataev (on file
with the Hoover Institution Archive, Stanford University) (the memoran-
dum is undated but based on context is from the mid-1980s).
14. See JOHN HINES ET AL., SOVIET INTENTIONS 1965-1985, VOLUME II: SO-
VIET POST-COLD WAR TESTIMONIAL EVIDENCE 90 (1995) (interview with Alexei
Kalashnikov); see also id. at 134–35 (interview with Viktor Surikov).
15. See generally GAVIN, NUCLEAR STATECRAFT, supra note 10; FRANCIS J. R
GAVIN, GOLD, DOLLARS, AND POWER: THE POLITICS OF INTERNATIONAL MONE-
TARY RELATIONS, 1958-1971 (2004); AARON FRIEDBERG, IN THE SHADOW OF
THE GARRISON STATE: AMERICA’S ANTI-STATISM AND ITS COLD WAR GRAND
STRATEGY (2000); Brendan Rittenhouse Green, Two Concepts of Liberty: U.S.
Cold War Grand Strategies and the Liberal Tradition, 37 INT’L SECURITY 2 (Fall
2012).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 8 21-MAY-15 13:28

364 INTERNATIONAL LAW AND POLITICS [Vol. 47:357

ferred to as “cross-domain deterrence”). Yet the credibility of

U.S. strategic nuclear use on behalf of NATO weakened as the
Soviet Union developed the ability to retaliate against the U.S.
homeland. If the United States accepted mutually assured de-
struction as a reality, then nuclear weapons would essentially
cancel each other out, making the world safe for conventional
war (the so-called “stability-instability” paradox).16 This led to
a variety of efforts to bolster the “coupling” of the U.S. strate-
gic arsenal to Europe, including the extensive counterforce ef-
fort noted above.
One implication of the foregoing is that deterrence dur-
ing the Cold War was, for the United States at least, much
more about deterrence by denial than the stylized narrative
would indicate. The threat of punishment through the de-
struction of Soviet industry and urban areas clearly was impor-
tant, but the emphasis of U.S. nuclear planning and force
structure was on denying the Soviet Union the ability to retali-
ate effectively against nuclear use on military forces. Thus, the
current emphasis on deterrence by denial rather than deter-
rence by punishment is less stark a difference from Cold War
policy than it might seem.
However, U.S. efforts to deny Soviet retaliation came at a
price. It created incentives for the Soviet Union to adopt com-
mand and control systems and policies that increased the risk
of accidental war or inadvertent escalation. One example is a
“launch on warning” policy, where Soviet missiles were to be
launched when early warning systems detected a U.S. attack.
While apparently never fully implemented by the Soviets, the
same incentives would apply to post-Cold War adversaries and
might be true in non-nuclear contexts as well.17 In the post-
Cold War environment, the balance between a nuclear-armed
Iran and Israel might be subject to similar concerns as the pos-

16. See Glenn H. Snyder, The Balance of Power and the Balance of Terror, in
THE BALANCE OF POWER 184–219 (Paul Seabury ed., 1965).
17. There is debate about the extent to which the Soviet Union adopted
particularly dangerous command and control systems and practices. See gen-
erally DAVID HOFFMAN, THE DEAD HAND: THE UNTOLD STORY OF THE COLD
WAR ARMS RACE AND ITS DANGEROUS LEGACY (2010) (arguing that systems
designed to make Soviet retaliation semi-automatic were deployed). But see
JOHN G. HINES, ELLIS M. MISHULOVICH & JOHN F. SHULL, SOVIET INTENTIONS
1965-1985: SOVIET POST-COLD WAR TESTIMONIAL EVIDENCE 134–35 (1995)
(arguing that the systems were never fully deployed).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 9 21-MAY-15 13:28

2015] DETERRENCE: THE STATE OF THE FIELD 365

sibility of Israeli preemption combined with missile defense

might convince Iranian leaders that in a crisis they might need
to adopt a similar posture.18
Paul Davis termed this trade-off between the needs of ex-
tended deterrence and the need to create stability in first-
strike incentives the “devil’s dilemma.” If first-strike stability
was too high, then the stability-instability paradox would come
into play and conventional aggression might be possible. If
first-strike stability was too low, then accidental war, escalation
spirals, and arms races were possible.19
Glenn Kent and David Thaler extended this line of argu-
ment by noting that
[t]he most important conflict [arises] between the
objectives of enhancing first-strike stability, on one
hand, and extending deterrence and limiting dam-
age on the other; i.e. the more robust the Soviets be-
lieve first-strike stability to be, the less they might
hesitate to precipitate a deep crisis by engaging in se-
rious aggression, for example, in Western Europe.
Balancing between first-strike stability and extended
deterrence presents a problem in the planning of
strategic forces. . . . Indeed, one might argue that an
optimal amount of first-strike instability is possible:
that is, enough to deter the Soviets from generating a
major crisis, say by invading Western Europe, but not
enough to allow a major crisis to spiral out of control.
Whether or not such an optimum actually exists, the
concept provides the proper intellectual framework
in which to think about the trade-off between first-
strike stability and extended deterrence.20
U.S. conventional superiority at present has dampened the
devil’s dilemma, so there is little need for the development of
an optimal amount of first-strike instability. Yet U.S. conven-
tional superiority is not a divinely fixed constant—in some
contexts, it may recede significantly in the coming decades.

18. For a longer discussion of this possibility, see Austin Long, Prolifera-
tion and Strategic Stability in the Middle East, in STRATEGIC STABILITY 400–12
(Elbridge A. Colby & Michael S. Gerson eds., 2013).
19. See PAUL K. DAVIS, STUDYING FIRST-STRIKE STABILITY WITH KNOWLEDGE-
BASED MODELS OF HUMAN DECISIONMAKING 33–36 (1989).
20. GLENN A. KENT & DAVID E. THALER, FIRST-STRIKE STABILITY 5 (1989).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 10 21-MAY-15 13:28

366 INTERNATIONAL LAW AND POLITICS [Vol. 47:357

Indeed, some efforts to extend U.S. conventional military ca-

pabilities against potential adversaries intent on countering
the United States may ultimately have effects on stability and
deterrence. There are concerns, for example, that Air-Sea Bat-
tle, the U.S. Navy and Air Force concept for projecting power
into the western Pacific against significant potential resistance
from China (or others), may decrease nuclear stability.21
Moreover, as Daryl Press and Keir Lieber, among others,
have noted, potential adversaries of the United States may seek
to use nuclear weapons to offset U.S. conventional superiority,
leading to a new version of the devil’s dilemma.22 For exam-
ple, in a counterfactual analysis of the 1990-1991 Persian Gulf
War, Barry Posen describes a scenario in which Saddam Hus-
sein could have tried to use nuclear weapons to defend his
gains following the invasion of Kuwait.23 Similar scenarios are
possible for other potential adversaries such as Iran or North
Korea.

III. POST-COLD WAR DETERRENCE: SOMETHING OLD,

SOMETHING NEW
Post-Cold War deterrence has spawned a variety of new
concepts. However, as suggested above, there is more rather
than less continuity between Cold War and post-Cold War de-
terrence. This does not mean that these new concepts are not
useful. Rather, it means that there may be additional evidence
with which to evaluate them, since more is known (particularly
in terms of access to primary sources) about Cold War deter-
rence than post-Cold War efforts. This section briefly ad-
dresses two such concepts: “cross-domain deterrence” and “tai-
lored deterrence.”
“Cross-domain deterrence” is, according to one leading
definition, “the problem of countering threats in one arena
(such as space or cyberspace) by relying on different types of
capabilities (such as sea power or nuclear weapons, or even

21. For a good overview of debates surrounding Air-Sea Battle, see gener-
ally AARON FRIEDBERG, BEYOND AIR-SEA BATTLE: THE DEBATE OVER U.S. MILI-
TARY STRATEGY IN ASIA (2014).
22. See Keir A. Lieber & Daryl G. Press, The New Era of Nuclear Weapons,
Deterrence, and Conflict, 7 STRATEGIC STUD. Q. 3, 5–7 (2013).
23. Barry R. Posen, U.S. Security Policy in a Nuclear-Armed World or: What if
Iraq Had Had Nuclear Weapons?, 6 SECURITY STUD. 1, 2 (1997).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 11 21-MAY-15 13:28

2015] DETERRENCE: THE STATE OF THE FIELD 367

non-military tools such as market access) where deterrence

may be more effective.”24 While most authors writing on cross-
domain deterrence acknowledge that it is not entirely new,
they argue that the increasing complexity of cross-domain in-
teractions, particularly those involving cyber tools, merits a
new focus.25 The final section of this paper concurs in part
with this assessment and highlights some ways in which new
technologies require new thinking in deterrence.
Yet while the renewed focus on cross-domain deterrence
should be welcomed, more can be mined from earlier—but
hardly uncomplicated—experience. There has already been
some significant effort in this realm, most notably Vipin
Narang’s efforts to explore the effect of nuclear posture on
cross-domain deterrence of conventional war by examining
past cases.26 However, more can and should be done, particu-
larly in terms of exploring deterrence escalation by invoking
multiple cross-domain interactions.
For example, during the Cold War, the United States in-
vested massively in conventional strategic anti-submarine war-
fare (ASW) capabilities to hunt Soviet ballistic missile subma-
rines. This conventional threat to strategic nuclear forces was
part of the overall effort to make extended nuclear deterrence
of conventional attack plausible by holding Soviet nuclear
forces at risk.27 Thus, a conventional threat was directed at nu-
clear forces in order to make a nuclear threat to deter a con-

24. Deterring Complex Threats: The Effects of Asymmetry, Interdependence, and

Multi-polarity on International Strategy, UNIVERSITY OF CALIFORNIA INSTITUTE ON
GLOBAL CONFLICT AND COOPERATION, http://www-igcc.ucsd.edu/research/
technology-and-security/cross-domain-deterrence/ (last visited Oct. 31,
2014).
25. See, e.g., James Scouras, Edward Smyth & Thomas Mahnken, CROSS-
DOMAIN DETERRENCE IN U.S.-CHINA STRATEGY: A WORKSHOP REPORT (Johns
Hopkins University Applied Physics Laboratory) (2014).
26. See Vipin Narang, What Does It Take to Deter? Regional Power Nuclear
Postures and International Conflict, 57(3) J. OF CONFLICT RESOL. 478 (2013); see
also Vipin Narang, Posturing for Peace? Pakistan’s Nuclear Postures and South
Asian Stability, 34(3) INT’L SECURITY 38 (2010).
27. The efficacy of these efforts has only been publicly understood since
the end of the Cold War. A brief summary is available in Long & Green,
supra note 12. For more detailed discussion, see OWEN COTE, THE THIRD R
BATTLE: INNOVATION IN THE U.S. NAVY’S SILENT COLD WAR STRUGGLE WITH
SOVIET SUBMARINES (2003); CHRISTOPHER FORD & DAVID ROSENBERG, THE AD-
MIRAL’S ADVANTAGE: U.S. NAVY OPERATIONAL INTELLIGENCE IN WORLD WAR II
AND THE COLD WAR (2005); SHERRY SONTAG & CHRISTOPHER DREW, BLIND
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 12 21-MAY-15 13:28

368 INTERNATIONAL LAW AND POLITICS [Vol. 47:357

ventional threat more plausible—certainly a complex set of

cross-domain interactions.
Yet some scholars, most notably Barry Posen, argued that
such an effort would undermine rather than bolster deter-
rence, as it would force the Soviets into a “use or lose” scenario
with their submarine-launched weapons, thus yielding inadver-
tent escalation to nuclear war.28 It was entirely plausible that
both perspectives were true. Cross-domain interaction might
have promoted pre-war deterrence, but at the same time it
might have weakened intra-war deterrence. Unfortunately, in-
terest in these issues all but died out when the Cold War en-
ded. There is thus tremendous opportunity to use newly availa-
ble sources and interviews to explore these cross-domain de-
terrence phenomena.
“Tailored deterrence” as a concept emerged in the post-
Cold War period as an attempt to rectify the perceived failure
of Cold War deterrence theories to develop a mechanism for
evaluating the specific capabilities needed to deter particular
adversaries. Beginning with the 2006 U.S. Quadrennial De-
fense Review, the term tailored deterrence entered the lexicon
as an alternative to what was characterized as “one-size-fits-all”
deterrence. This one-size-fits-all description is presumably
based on the stylized narrative of Cold War deterrence with
the Soviet Union.29
However, “tailoring” deterrence to the Soviet Union’s spe-
cific concerns was a major component of Cold War deter-
rence, particularly in the era of nuclear parity. During the
1970s, first the Nixon/Ford and then Carter administrations
sought to improve nuclear targeting policy by generating op-
tions that were both more discriminate (and therefore more
plausibly used) and more tailored to specific Soviet concerns
(or, more accurately, hypothesized Soviet concerns) such as

MAN’S BLUFF: THE UNTOLD STORY OF AMERICAN SUBMARINE ESPIONAGE

(1998).
28. See generally BARRY POSEN, INADVERTENT ESCALATION: CONVENTIONAL
WAR AND NUCLEAR RISKS (1992).
29. U.S. Dep’t of Def., QUADRENNIAL DEFENSE REVIEW REPORT 49 (2006);
see also Elaine Bunn, Can Deterrence Be Tailored?, 225 STRATEGIC FORUM 1, 1
(2007); NATO DEF. COLL., NATO AND 21ST CENTURY DETERRENCE 11 (Karl-
Heinz Kamp & David Yost, eds., 2009).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 13 21-MAY-15 13:28

2015] DETERRENCE: THE STATE OF THE FIELD 369

military forces, leadership, and command and control.30 While

the actual efficacy of these targeting changes remains unclear
(there is evidence that the Soviets were more rather than less
concerned about the possibility of nuclear war after they were
implemented), it nonetheless underscores that Cold War de-
terrence was hardly “one-size-fits-all.”31
Indeed, the United States dedicated extensive effort both
inside and outside government to developing an understand-
ing of how to tailor deterrence to the Soviet Union. In the
1970s, for example, the RAND Corporation conducted a
lengthy set of analyses for the U.S. Air Force to create models
that would allow different combinations of military forces to
be evaluated for their impact on deterring the Soviet Union.
Moreover, it did so by specifying multiple models of Soviet
decisionmaking criteria, enabling a variety of plausible Soviet
motives and values to be included in modeling deterrence.32
Yet, despite the vast effort expended to develop an under-
standing of the Soviet Union, it is not clear that efforts to tailor
deterrence were particularly successful. The actual character
of Soviet motives and values remained contested in the West
throughout the Cold War.33 This led to uncertainty about the

30. William Burr, The Nixon Administration, the ‘Horror Strategy,’ and the
Search for Limited Nuclear Options, 1969-1972: Prelude to the Schlesinger Doctrine,
7 J. COLD WAR STUD. 34, 36–37 (2005); William Odom, The Origins and Design
of Presidential Decision-59: A Memoir, in GETTING MAD: NUCLEAR MUTUAL AS-
SURED DESTRUCTION, ITS ORIGINS AND PRACTICE 175, 176–77 (Henry D.
Sokolski ed., 2004).
31. See BEN B. FISCHER, A COLD WAR CONUNDRUM: THE 1983 SOVIET WAR
SCARE 5 (1997) (discussing the early 1980s KGB search for strategic warning
of nuclear attack); MARKUS WOLF, NOTE REGARDING THE RESULTS OF THE
CONSULTATIONS WITH COMRADE MAJOR GENERAL SHAPKIN, DEPUTY HEAD OF
THE KGB’S 1ST MAIN DIRECTORATE, AND TWO EXPERTS ON THE RYAN PROB-
LEM FROM 14 TO 18 AUGUST IN BERLIN 2 (Bernd Schaefer trans., 1984), availa-
ble at http://www.wilsoncenter.org/publication/kgbstasi-cooperation. Wolf
headed foreign intelligence for East Germany.
32. See DONALD EMERSON, CONTRIBUTIONS OF FORCE TO DETERRENCE: A
PROGRESS REPORT (1971); see also WILLIAM M. JONES, MODELING SOVIET BE-
HAVIOR AND DETERRENCE: A PROCEDURE FOR EVALUATING MILITARY FORCES
(1974).
33. See JOSHUA ROVNER, FIXING THE FACTS: NATIONAL SECURITY AND THE
POLITICS OF INTELLIGENCE (2011); see also RICHARD K. BETTS, ENEMIES OF IN-
TELLIGENCE: KNOWLEDGE & POWER IN AMERICAN NATIONAL SECURITY (2007)
(discussing the Team B controversy on Soviet intentions).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 14 21-MAY-15 13:28

370 INTERNATIONAL LAW AND POLITICS [Vol. 47:357

effect of targeting Soviet military forces and leadership on de-

terrence.
Newly available evidence, including post-Cold War inter-
views with Soviet leaders and archival evidence, suggests the
effect was mixed. For example, the improvements in U.S. abil-
ity to target Soviet nuclear forces described earlier were noted
by Soviet leaders as posing a serious risk to the Soviet Union’s
ability to launch a significant retaliatory second strike. As the
logic of the devil’s dilemma suggests, this may have produced
caution about initiating crises, as the possibility of escalation
seemed more plausible. On the other hand, it created (or at
least amplified) fears about the possibility of U.S. preventive or
preemptive war, particularly in the early 1980s, which exacer-
bated tensions.34

IV. THE PERENNIAL QUESTION OF CREDIBILITY

One of the central questions in deterrence both during
and after the Cold War has been the credibility of threats, par-
ticularly for extended deterrence. The first generations of de-
terrence theorists as well as policymakers strongly believed that
commitments for deterrence were strongly interdependent,
and therefore that reputation mattered. If one did not fight
for every commitment then the credibility of all commitments
would be weakened. The shadow of Munich and its alleged
lessons weighed heavily on this generation of theorists and
practitioners alike.35 Some have even characterized the Ameri-
can view of credibility and reputation during the Cold War as
almost an obsession.36
Yet over the past several decades the scholarly consensus
from both quantitative and qualitative research has shifted sig-
nificantly. The consensus is now much more firmly in favor of
the view that commitments are at least mostly independent
and reputation is of much less significance. Instead, targets of
deterrent threats apparently tend to evaluate the specifics of

34. HINES ET AL., supra note 14; see also Vojtech Mastny, How Able Was
“Able Archer”? Nuclear Trigger and Intelligence in Perspective, 11 J. COLD WAR
STUD. 1, 108 (2009).
35. YUEN FOONG KHONG, ANALOGIES AT WAR: KOREA, MUNICH, DIEN BIEN
PHU, AND THE VIETNAM DECISIONS OF 1965 (1992).
36. See Patrick Morgan, Saving Face for the Sake of Deterrence, in PSYCHOLOGY
AND DETERRENCE 125 (Robert Jervis et al. eds., 1986).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 15 21-MAY-15 13:28

2015] DETERRENCE: THE STATE OF THE FIELD 371

the situation, particularly focusing on the capability to carry

out the threat and the stakes for both parties in the crisis.37
However, the consensus in the academy has not migrated
to policymakers. Many in the policy world are already pro-
claiming linkage between the failure of the United States to
punish Syria for violating a “red line” on chemical weapons
use in 2013 and the Russian intervention in the Crimea in
2014.38 Even in the Syria crisis, U.S. National Security Adviser
Tony Blinken was alleged to have told President Obama that
“superpowers don’t bluff,” presumably because it would in-
deed undermine credibility in other instances.39
What explains this disconnect between scholarly and poli-
cymaker consensus? It could be a simple case of the widely
bemoaned failure of political scientists and other scholars to
leave the ivory tower and communicate with policymakers.
While plausible, there is likely to be more at work than a sim-
ple failure to communicate.
One possible explanation is that the focus of recent re-
search on credibility and deterrence has been focusing on the
wrong elements of the historical record. Much of the scholar-
ship has examined records of decisionmaking in crisis situa-
tions, where both sides have some significant commitment to
the resolution of the dispute. In these instances, typically re-
ferred to as situations of immediate deterrence, reputation
may matter little as both sides have already factored reputation
into their decisionmaking to initiate crisis (either explicitly or
implicitly).40
However, reputation may matter much more in pre-crisis
periods. An adversary’s reputation may affect decisionmaking

37. See, e.g., JONATHAN MERCER, REPUTATION AND INTERNATIONAL POLITICS

(1996); DARYL PRESS, CALCULATING CREDIBILITY: HOW LEADERS ASSESS MILI-
TARY THREATS (2007); Vesna Danilovic, The Sources of Threat Credibility in Ex-
tended Deterrence, 45.(3) J. CONFLICT RESOL. 341 (2001).
38. Matt Spetalnick, Obama’s Syria ‘Red Line’ has Echoes in his Warning to
Ukraine, REUTERS (Feb. 20, 2014), www.reuters.com/article/2014/02/20/us-
ukraine-crisis-obama-idUSBREA1J2C920140220.
39. Jon Swaine, How President Barack Obama became Convinced of Need to
Arm Syrian Rebels, TELEGRAPH (UK) (June 15, 2013), www.telegraph.co.uk/
news/worldnews/middleeast/syria/10122567/How-President-Barack-
Obama-became-convinced-of-need-to-arm-Syrian-rebels.html.
40. On the concept of immediate deterrence, see generally PATRICK MOR-
GAN, DETERRENCE: A CONCEPTUAL ANALYSIS (1983).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 16 21-MAY-15 13:28

372 INTERNATIONAL LAW AND POLITICS [Vol. 47:357

to initiate crisis. Keren Yarhi-Milo and Alex Weisiger have ar-

gued that reputation contributes to so-called general deter-
rence before crisis as well as the perception of interests on
both sides of a deterrence relationship. They highlight several
examples where reputation for resolve either increased or de-
creased the effectiveness of general deterrence.41
Reputation may also matter for reassurance of the targets
of deterrence by punishment. In order for deterrence to be
effective, the target must have some confidence that if it does
not seek to alter the status quo, then punishment will be with-
held. A reputation for aggression may make deterrence more
difficult, as the target may think punishment will be forthcom-
ing regardless of the target’s action.
Another way in which research on reputation may be mis-
placed is the focus on deterrence effects with regard to adver-
saries. Yet there are other audiences that reputation may af-
fect. Two that matter significantly in the context of extended
deterrence are allies and domestic publics, so it is worth ex-
ploring how these concerns affect policymaker decisions.
While it has been widely appreciated that extended deter-
rence must not only deter potential adversaries, but also assure
allies, there has been substantially less systematic evaluation of
the role of reputation in assurance guarantees.42 Assurance
and deterrence are parallel but may not be equivalent, as an
adversary may be deterred but an ally not reassured. Reputa-
tion and prior actions may therefore not matter much for de-
terrence, but it does not follow that reputation does not mat-
ter for reassuring allies.43
The second audience is the domestic public. Here there
has been a substantial body of literature on “audience costs,”

41. Keren Yarhi-Milo & Alex Wesiger, Revisiting Reputation: How Do Past
Actions Matter in International Politics, INT’L ORG. (forthcoming) (paper at 2-3,
11-28, available at https://wws.princeton.edu/faculty-research/research/
item/revisiting-reputation-how-past-actions-matter-international-politics).
42. See, e.g., JERVIS, supra note 3; GLENN H. SNYDER, ALLIANCE POLITICS R
(Robert J. Art et al. eds., 1997) (discussing alliances in a multipolar interna-
tional system before 1914).
43. There is emerging scholarship on this topic and more should be en-
couraged. For example, at the 2013 Nuclear Scholars Research Initiative
meeting, Gene Gerzhoy presented work on the role of assurance in West
German renunciation of nuclear weapons while Mira Rapp-Hooper
presented work on the role of assurance in nuclear alliance guarantees.
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 17 21-MAY-15 13:28

2015] DETERRENCE: THE STATE OF THE FIELD 373

following James Fearon’s seminal 1994 article arguing that

democratic leaders who fail to carry out threats will be pun-
ished by their electorate.44 Yet a recent empirical analysis calls
into question the role of audience costs in punishing empty
threats.45 However, there may be more subtle or general ways
in which leaders pay audience costs that are not related to spe-
cific threats, particularly if domestic opponents are able to
generate a narrative about the leader and/or his party’s weak-
ness.
The classic example is the narrative constructed around
the 1949 victory of the Chinese Communist Party over the Na-
tionalists. Republicans portrayed this “loss of China” as funda-
mentally the fault of President Harry Truman and the Demo-
cratic Party. This narrative was then exploited to show the
weakness of Democrats on national security generally in the
1952 presidential election, which was won by Republican (and
retired general) Dwight Eisenhower.46
Alliance and domestic audience concerns may also be mu-
tually supporting. According to Fredrik Logevall, concerns
about reputation, particularly domestically but also with allies,
were the major driver behind President Johnson’s decision to
escalate in Vietnam. While the Munich analogy lurked in the
background, Johnson was at least as concerned about his per-
sonal reputation and the reputation of the Democratic Party,
which had been punished by the narrative that blamed Demo-
cratic weakness for the loss of China.47 Policymakers may
therefore be right to worry about reputation, but for the
“wrong” reasons. Additionally, alliance and domestic political
concerns about reputation may combine to make threats more
credible to adversaries (at least in some contexts). More re-
search needs to be done to confirm or deny these hypotheses,
making it a fruitful area for research.

44. James D. Fearon, Domestic Political Audiences and the Escalation of Inter-
national Disputes, 88 AM. POL. SCI. REV. 577 (1994).
45. Jack Snyder & Erica D. Borghard, The Cost of Empty Threats: A Penny,
Not a Pound, 105 AM. POL. SCI. REV. 437 (2011).
46. See ROBERT DALLEK, HARRY S. TRUMAN (Arthur M. Schlesinger, Jr. &
Sean Wilentz eds., 2008).
47. FREDRIK LOGEVALL, CHOOSING WAR: THE LOST CHANCE FOR PEACE AND
THE ESCALATION OF WAR IN VIETNAM (1999).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 18 21-MAY-15 13:28

374 INTERNATIONAL LAW AND POLITICS [Vol. 47:357

V. CLANDESTINE CAPABILITIES AND DETERRENCE

In one of the best scenes in Cold War classic Dr.
Strangelove, the titular scientist berates the Soviet ambassador
to the United States for keeping secret the existence of a
“Doomsday Machine.” He concludes that “the whole point of
the doomsday machine is lost if you keep it a secret!” Sadly, in
the film the Soviet General Secretary had planned to reveal
the capability later (he was said to love surprises), but too late
to deter a rogue U.S. Air Force general.
Decades later, James Fearon came to a related conclusion
that emphasized the role of “private information” (such as se-
cret war plans or other capabilities) in causing war. If only
both sides in a potential conflict had full knowledge of each
other’s relative strength, war would be less likely. Yet certain
capabilities might lose efficacy if revealed, so there are incen-
tives either to deceive or simply not to reveal these capabili-
ties.48
However, there has been little work done to bridge these
two related insights. How can one incorporate clandestine ca-
pabilities into deterrence strategies if the revelation of those
capabilities will attenuate them? The revelation of conven-
tional or nuclear force structure in general terms posed no
real risk to their efficacy, so this problem was not widely con-
sidered during the Cold War. In contrast, during the post-Cold
War era, at least two sets of capabilities that might be impor-
tant for deterrence would be weakened if revealed in all but
the most abstract way. These are offensive cyber capabilities
and persistent intelligence, surveillance, and reconnaissance
(ISR) capabilities.
Cyber offensive capabilities can be powerful tools for de-
terrence in terms of potential countervalue and especially
counterforce operations. There are already indications that
cyber offensive capabilities can, for example, cripple industrial
control systems, as in the Stuxnet attack on Iran’s nuclear pro-
gram, and air defense systems, as in the Israeli strike on a Syr-
ian nuclear reactor in 2007.49 If an enemy knew about such a

48. James Fearon, Rationalist Explanations for War, 49 INT. ORG. 393 (Sum-
mer 1995).
49. For evaluation of these efforts, see Richard Gasparre, The Israeli ‘E-
tack’ on Syria – Part I and Part II, AIR FORCE TECHNOLOGY, Mar. 2008; Jon
Lindsay, Stuxnet and the Limits of Cyber Warfare, 22 SECURITY STUD. 365 (2013).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 19 21-MAY-15 13:28

2015] DETERRENCE: THE STATE OF THE FIELD 375

capability and was convinced it would be effective, it would re-

duce confidence in the ability to deny air attacks. Yet if an en-
emy knew about the capability in all but the most abstract way,
the enemy could begin taking steps to neutralize whatever se-
curity flaws it exploited.
Similarly, ISR capabilities can be critical in supporting de-
terrence. U.S. strategic ASW capabilities, noted above, de-
pended heavily on a variety of intelligence sources that, if com-
promised, would be lost. One critical example is the alleged
tapping of Soviet undersea naval communications, which was
lost when revealed by a Soviet spy who had worked for the Na-
tional Security Agency.50 Another is the capability that the
United States allegedly developed at one point in the late Cold
War to interfere with Soviet nuclear command and control
due to vulnerability in the Soviet strategic communications sys-
tem. The Soviets eventually discovered and fixed this vulnera-
bility—highlighting the inability to communicate such capabil-
ities for deterrent purposes without compromising them.51
In the post-Cold War environment the opportunities for
persistent ISR have grown, making them potentially even more
important for deterrence, yet they are just as vulnerable if re-
vealed. For example, modern tagging, tracking, and locating
devices would allow real time tracking of Iranian mobile mis-
siles if placed inside the missile or transporter. This would re-
quire a penetration of the Iranian missile complex, but, as
Stuxnet has demonstrated, this is a possibility that cannot be
dismissed. If the United States had such a capability it could
significantly affect the survivability of an important component
of the Iranian deterrent, yet this could not be revealed without
jeopardizing it. The same would be true for a cyber capability
to target these missiles (for example by corrupting guidance
or launch control software).
Further, the inability to reveal these capabilities would
likely include many if not all allies. This would mean that such
capabilities could not be used to provide assurance to allies. In
the Iranian example above, the United States might share such

50. See SHERRY SONTAG & CHRISTOPHER DREW, BLIND MAN’S BLUFF: THE
UNTOLD STORY OF AMERICAN SUBMARINE ESPIONAGE (2000).
51. Benjamin B. Fischer, CANOPY WING: The U.S. War Plan That Gave the
East Germans Goose Bumps, 27 INT. J. OF INTELLIGENCE & COUNTERINTEL-
LIGENCE 431 (2014).
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 20 21-MAY-15 13:28

376 INTERNATIONAL LAW AND POLITICS [Vol. 47:357

capabilities with the Israelis and its critical Five Eyes intelli-
gence partners, but could it share with the Gulf States or even
the broader NATO community? Even many parts of the U.S.
government are unlikely to understand fully these capabilities
because they are so closely held.
Finally, while the previous examples have all centered on
high technology, clandestine capabilities can be less techno-
logically sophisticated. Perhaps the best post-Cold War exam-
ple is Hamas’ massive development of tunnels in Gaza. While
the Israelis were cognizant of the existence of tunnels, they
were unaware of the full scope and the potential for offensive
use of the tunnels in a conflict. Hamas was able to inflict non-
trivial losses using these tunnels due to the surprise factor—
had they fully revealed this capability to the Israelis for deter-
rence purposes it would likewise have been much less effec-
tive.52
The role of clandestine capabilities in deterrence is thus
one of the major lacunae in post-Cold War deterrence. While
a significant amount has been written on cyber deterrence,
very little of it has addressed these specific issues, which are
frequently cross-domain (such as cyber v. conventional air de-
fense, or ISR v. nuclear or conventional mobile missile).53 Sub-
stantially more research is needed, though of course studying
clandestine capabilities is obviously difficult for scholars. From
a policy perspective, this underscores the need for systematic
evaluation of such capabilities within the military and intelli-
gence community, which do have access to information about
these capabilities. From a scholarly perspective, this under-
scores again the utility of the Cold War as more is known
about clandestine capabilities from this period thanks to de-
classification.

52. See, e.g., Isabel Kershner, In Tunnel War, Israeli Playbook Offers Few
Ideas, N.Y. TIMES (Aug. 1, 2014), http://www.nytimes.com/2014/08/02/
world/middleeast/an-old-playbook-leaves-israel-unready-for-hamass-tunnel-
war-.html.
53. An exception is CHARLES GLASER, DETERRENCE OF CYBER ATTACKS AND
U.S. NATIONAL SECURITY (2011). Yet Glaser devotes only a few paragraphs to
these questions.
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 21 21-MAY-15 13:28

2015] DETERRENCE: THE STATE OF THE FIELD 377

VI. CONCLUSION
Deterrence studies should be on the threshold of a new
golden age, as there is both broad recognition of the impor-
tance of deterrence in the 21st century and a wealth of newly
available information from the 20th century. The key to the
future of deterrence lies in combining these two broad trends
in order to use the newly illuminated past to understand the
present and future. This requires scholars and analysts to
move beyond stylized narratives about the Cold War while also
acknowledging that some categories of capabilities, particu-
larly those that must remain clandestine to be effective, are
genuinely novel.
\\jciprod01\productn\N\NYI\47-2\NYI204.txt unknown Seq: 22 21-MAY-15 13:28