The Myth of Cyberwar

The Myth of Cyberwar Erik Gartzke
Bringing War in Cyberspace
Back Down to Earth

A blitz of media, pun-
ditry, and ofªcial pronouncements raise the specter of war on the internet. 福-
ture conºicts may well take place in cyberspace, where victory or defeat could
be determined in mere “nanoseconds.”1 Secretary of Defense Leon Panetta has
even warned of a “cyber-Pearl Harbor.”2 Nor are fears of cyberwar abstract
speculation. Events such as the denial of service attacks against Estonian and
Georgian government websites, the Stuxnet worm designed to disable Iranian
nuclear centrifuges, and the recent hacking of U.S. military computer networks
seem to indicate that the era of cyberwar has already arrived.

Cyberwar can be viewed as the most recent phase in the ongoing revolution
in military affairs.3 This time, 然而, the threat is said to be directed at the
sophisticated technological civilizations of the West, rather than at desert in-
surgents or the leaders of rogue states with arsenals of inferior second world
military hardware. Joseph Nye expresses this emerging consensus, “Depend-
ence on complex cyber systems for support of military and economic activities
creates new vulnerabilities in large states that can be exploited by nonstate ac-
tors.”4 Following this logic, the United States appears destined to be “the

Erik Gartzke is Associate Professor of Political Science at the University of California, 圣地亚哥, and Pro-
fessor of Government at the University of Essex.

The author thanks Susan Aaronson, Tai Ming Cheung, Peter Cowhey, Peter Dombrowski, 尤金
戈尔茨, Florian Grunert, Jeffrey Kwong, Jon Lindsay, John Mueller, and Heather Roff and the anon-
ymous reviewers for comments and encouragement. Oliver Davies provided valuable research
协助.

1. Dan Kuehl, quoted by Grace Chng, “Cyber War: One Strike, and You’re Out,” Sunday Times
(新加坡), 七月 18, 2010.
2. Elisabeth Bumiller and Thom Shanker, “Panetta Warns of Dire Threat of Cyberattack,” New York
时代, 十月 11, 2012.
3. On the revolution in military affairs (RMA), see Alvin and Heidi Tofºer, War and Anti-War: 在-
vival at the Dawn of the 21st Century (波士顿: 小的, 棕色的, 1993); Andrew F. Krepinevich, “Cavalry
to Computer: The Pattern of Military Revolutions,” National Interest, 不. 37 (落下 1994), PP. 30–42;
Andrew F. Krepinevich, The Military-Technical Revolution: A Preliminary Assessment (华盛顿,
华盛顿特区: Center for Strategic and Budgetary Assessments, 2002); Eliot A. 科恩, “A Revolution in
Warfare,“ 外交事务, 卷. 75, 不. 2 (March/April 1996), PP. 37–54; Richard Hundley, Past Revo-
lutions, Future Transformations: What Can the History of Revolutions in Military Affairs Tell Us about
Transforming the U.S. 军队? (圣莫妮卡, 加利福尼亚州。: RAND, 1999); Michael O’Hanlon, “Why
China Cannot Conquer Taiwan,” 国际安全, 卷. 25, 不. 2 (落下 2000), PP. 51–86; and Mi-
chael G. Vickers and Robert C. Martinage, The Revolution in War (华盛顿, 华盛顿特区: Center for
Strategic and Budgetary Assessments, 2004). For criticism of the RMA, see Stephen Biddle, “As-
sessing Theories of Future Warfare,” 安全研究, 卷. 88, 不. 1 (秋天 1998), PP. 1–74.
4. Joseph Nye, “Cyber Power” (剑桥, 大量的。: Belfer Center for Science and International Af-
博览会, Harvard Kennedy School, 可能 2010), p. 4.

国际安全, 卷. 38, 不. 2 (落下 2013), PP. 41–73, 土井:10.1162/ISEC_a_00136
© 2013 由哈佛大学和麻省理工学院的校长和研究员撰写.

41

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 42

country most vulnerable to cyber-attack.”5 If powerful developed nations
largely immune to terrestrial onslaught can have their defenses disabled and
their factories idled by foreign hackers, then perhaps “Pearl Harbor” is an ap-
propriate metaphor. One should then heed the growing chorus of warnings
that the West is about to become the target, rather than perpetrator, of “shock
and awe.”

There is a signiªcant fault, 然而, in the theme of impending cyber apoca-
lypse: it is far from clear that conºict over the internet can actually function as
战争. Predictions about the nature or signiªcance of cyberwar generally commit
a common fallacy in arguing from opportunity to outcome, rather than consid-
ering whether something that could happen is at all likely, given the motives
of those who are able to act. Cyber pessimism rests heavily on capabilities
(方法), with little thought to a companion logic of consequences (结束).
Much that could happen in the world fails to occur, largely because those who
can act discern no meaningful beneªt from initiating a given act. Put another
方式, advocates have yet to work out how cyberwar enables aggressors to
accomplish tasks typically associated with terrestrial military violence. Ab-
sent this logic of consequences, cyberwar is unlikely to prove as pivotal in
world affairs, and for developed nations, 尤其, as many observers
seem to believe.

This article assesses the salience of the internet for carrying out func-
tions commonly identiªed with terrestrial political violence. War is funda-
mentally a political process, as Carl von Clausewitz famously explained.6
状态, 团体, and individuals threaten harm to deter or compel, 生成
inºuence through the prospect of damage or loss. Military force can also be ex-
ercised to maintain or alter the balance of power and resist or impose disputed
结果. The internet is generally an inferior substitute to terrestrial force in
performing the functions of coercion or conquest. Cyber “war” is not likely
to serve as the ªnal arbiter of competition in an anarchical world and so
should not be considered in isolation from more traditional forms of political
violence.7 The capacity for internet coercion is further limited by some of the
same factors that make cyberwar appear at ªrst so intimidating. For threats or
demands to prove effective, targets must believe both that an attack is likely to
follow from noncompliance and that the attack is destined to inºict unaccept-
able harm. 然而, as I detail here, the need to apprise targets of internet vulnera-

5. James Adams, “Virtual Defense,“ 外交事务, 卷. 80, 不. 3 (五月六月 2001), p. 98.
6. Carl von Clausewitz, On War (普林斯顿大学, 新泽西州: 普林斯顿大学出版社, 1976).
7. Hannah Arendt wrote, “The chief reason warfare is still with us is . . . that no substitute for this
ªnal arbiter in international affairs has yet appeared on the political scene.” Arendt, On Violence
(纽约: Harcourt, Brace, 1970), p. 2.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 43

bilities to make cyber threats credible contrasts with the secrecy required to
ensure an effective attack.

Given the inherent difªculty of credibly threatening cyberattacks without
also compromising operational effectiveness, it will be tempting for actors to
practice cyberwar rather than engage in coercive threats. 这里, 也, 然而,
key limitations exist regarding what can be achieved over the internet. It is one
thing for an opponent to interrupt a country’s infrastructure, 通讯,
or military coordination and planning. It is another to ensure that the damage
inºicted translates into a lasting shift in the balance of national power or re-
solve. Cyberattacks are unlikely to prove particularly potent in grand strategic
terms unless they can impose substantial, durable harm on an adversary. 在
许多, perhaps most, 情况, this will occur only if cyberwar is accom-
panied by terrestrial military force or other actions designed to capitalize on
any temporary incapacity achieved via the internet. Those initiating cyber-
attacks must therefore decide whether they are prepared to exploit the win-
dows of opportunity generated by internet attacks through other modes of
combat. If they are not willing and able to do so, then in grand strategic terms,
there are few compelling reasons to initiate cyberwar. The need to back up
cyber with other modes of conºict in turn suggests that the chief beneªci-
aries of cyberwar are less likely to be marginal groups or rising challengers
looking to overturn the existing international order and more likely to be
nation-states that already possess important terrestrial military advantages.
Conceived of in this way, the internet poses no revolution in military affairs
but instead promises simply to extend existing international disparities in
power and inºuence.

The remainder of this article is organized into ªve sections. After a brief re-
view of the burgeoning literature on cyberwar, the subsequent section lays out
the case for a logic of consequences. It is not enough to determine what could
happen in a world where so much is possible. The third section applies basic
insights about the nature of war to detail critical shortcomings of cyberwar as a
political instrument. The Pearl Harbor attacks admirably illustrate the inade-
quacies of conºict in cyberspace. Section four addresses additional implica-
tions and limitations of the main argument, while the ªfth section offers some
concluding remarks.

Panic over the Internet: The Literature on Cyberwar

The character of war has evolved, if not regularly, then certainly at various
points in history. 像这样, it is reasonable, even forward looking, for observers
to consider what impact each new technology might have on the nature of war

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 44

and peace. Innovations such as the stirrup, steam propulsion, the airplane, 和
the exploding shell transformed warfare. Other changes, such as the tele-
电话, high-rise construction techniques, and the advent of effective birth con-
控制, may have been less revolutionary than evolutionary in their military
effects. A large and growing literature seeks to alert—some might say alarm—
observers to the dangers of cyberwar. These studies shine much less light on
how cyberspace is destined to change the nature of political conºict than
on what harm it is possible to perpetrate over the internet.

scope and scale conditions of cyberwar—cyber pessimists

For many thoughtful commentators, the size and scope of the cyberwar threat
could be unprecedented. William Lynn III argues that cyberwarfare is indeed a
重大的, imminent threat.8 Those with a motive to launch an attack against
the United States will soon possess the capability to do so. 在这个意义上,
cyberwarfare is unique in that opponents who utilize the strategy are not lim-
ited by ªnancial or physical constraints. Lynn advocates a vigorous defense
as the most viable and ºexible strategy in cyberspace. 美国
can avoid large-scale cyber calamities through the collaboration of public, 在-
vate, and government-sponsored corporations. In a follow-on article, 林恩
outlines a ªve-pillared cyberspace defense strategy, “treating cyberspace as an
operational domain, like land, 空气, sea, and outer space; employing active de-
fenses to stop malicious code before it affects our networks; protecting com-
mercial networks that operate the critical infrastructure that our military relies
之上; joining with allies to mount a collective cyber defense; and mobilizing
industry to redesign network technology with security in mind.”9

Treating cyberspace as an operational domain is an excellent idea, but doing
so quickly reveals differences between internet conºict and warfare on land,
sea, in the air, or in space. Deterring or even defending against cyberattack
may prove difªcult, as others have argued, but it will prove much harder still
for an attacker to ªgure out how to beneªt from internet aggression, 除非
cyberattacks occur in conjunction with attacks in other domains.

James Adams anticipates Lynn in arguing for a comprehensive U.S. 网络-
warfare defense strategy.10 The United States is vulnerable to attack because

8. 威廉·J. Lynn III, “Defending a New Domain: The Pentagon’s Cyberstrategy,“ 外交事务,
卷. 89, 不. 5 (September/October 2010), PP. 97–108.
9. 威廉·J. Lynn III, “The Pentagon’s Cyberstrategy, One Year Later: Defending against the Next
Cyberattack,“ 外交事务 (九月 28, 2011), http://www.foreignaffairs.com/articles/
68305/william-j-lynn-iii/the-pentagons-cyberstrategy-one-year-later.
10. Adams, “Virtual Defense,” PP. 98–112.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 45

many smaller nations and private groups will seek to gain an advantage by
employing asymmetric warfare, whose impact would be felt not only in the
public sector, but also in private industry.11 The U.S. military has become in-
creasingly dependent on new technology to dominate modern battleªelds,
paradoxically rendering itself more vulnerable and prone to ever-increasing
incidences of potentially crippling cyberattacks. Richard Clark and Robert
Knake offer early examples of cyberwar.12 They claim that the United States
has been slow to orient itself to what they argue could be a national security
nightmare. Shane Courville also considers the United States to be highly sus-
ceptible to a catastrophic cyberattack, but he is pessimistic about the ability of
the United States or other nations to defend themselves.13 Cyber defense is
problematic given rapidly changing technology. 尤其, Courville notes
that little thought has gone into exactly who manufactures computer hardware
for the U.S. military.14

Kenneth Knapp and William Boulton point out that the limited barriers
to entry in cyberwarfare leave even great powers vulnerable to a constant
stream of virtual attacks.15 As they write, “[瓦]ide ranges of formidable cyber-
weapons have become more affordable and available. . . . An attacker can
build an [electromagnetic] bomb, designed to fry computer electronics with
electromagnetic energy, for as little as $400.”16 The authors blame advances in cyberwar technology for heightened losses in U.S. 行业: “500 U.S. companies showed an increase in reported ªnancial losses of 21 百分, 或者 $455.8 百万, in 2002.”17

Lorenzo Valeri and Michael Knights are also concerned about the vulnera-
bility of U.S. military and civilian infrastructure to cyberwarfare.18 These au-
thors speculate that terrorists will exercise offensive information warfare,
focusing on electronic commerce websites instead of national infrastructure, 作为
the former will be signiªcantly more accessible and might ultimately wreak

11. 同上。, p. 99.
12. 理查德·A. Clarke and Robert K. Knake, Cyber War: The Next Threat to National Security and
What to Do about It (纽约: Ecco, 2010).
13. Shane P. 考维尔, “Air Force and the Cyberspace Mission: Defending the Air Force’s Com-
puter Network in the Future” (Maxwell, Ala.: Center for Strategy and Technology, Air War Col-
lege, 2007).
14. 同上.
15. Kenneth Knapp and William Boulton, “Cyber-Warfare Threatens Corporations: Expansion to
Commercial Environments,” Information Systems Management Journal, 卷. 23, 不. 2 (春天 2006),
PP. 76–87.
16. 同上。, p. 79.
17. 同上。, p. 83.
18. Lorenzo Valeri and Michael Knights, “Affecting Trust: 恐怖主义, 互联网, and Offensive In-
formation Warfare,” Terrorism and Political Violence, 卷. 12, 不. 1 (春天 2000), PP. 15–36.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 46

the most havoc. Attacks on commercial industry will seriously damage con-
sumer trust in the internet, and might ultimately undermine government
plans to convert services to the digital domain. Valeri and Knights emphasize
that the skills needed to conduct offensive information warfare “are easily
available as the Internet and exposure to information and network technolo-
gies encourages increasing technological sophistication in society.”19

John Arquilla and David Ronfeldt seek to distinguish between actors and
their roles in internet conºict.20 Cyberwar generally involves conºicts among
organized militaries, while “Netwar” consists of internet conºict that includes
nonstate actors. As they explain it, “[时间]he term Netwar refers to an emerging
mode of conºict (and crime) at societal levels, involving measures short of tra-
ditional war.”21 Arquilla and Ronfeldt argue that “cyberwar” has become mis-
leading with the advent of newer technology. 然而, while Netwar and
traditional cyberwar differ in their respective forms, each is a potentially cata-
strophic threat.

the attribution problem and international law

It has been argued that one of the most important, and potentially menac-
英, characteristics of cyberwar involves anonymity. David Clark and Susan
Landau highlight attribution as a critical issue that is difªcult to overcome in
seeking to deter cyberattacks.22 As they argue, “Retaliation requires knowing
with full certainty who the attackers are.”23 Similarly, Martin Libicki is con-
cerned that the internet may create intractable difªculties in seeking to deter
would-be perpetrators.24 As he explains the problem, “The lower the odds of
getting caught, the higher the penalty required to convince potential attackers
that what they might achieve is not worth the cost.”25 Attackers are much
more likely to strike if they are unlikely to be targeted in return.

The focus is again on the potential for harm, rather than on exploring the
motives and operational logic of perpetrators. If internet anonymity is awk-
ward for targets of attacks, it is also a problem for initiators. Terrorists spend as

19. 同上。, p. 20.
20. John Arquilla and David Ronfeldt, “The Advent of Netwar: Analytic Background,” Studies in
Conºict and Terrorism, 卷. 22, 不. 3 (九月 1999), PP. 193–206.
21. 同上。, p. 194.
22. 大卫·D.. Clark and Susan Landau, “Untangling Attribution,” Harvard National Security Jour-
纳尔, 卷. 2, 不. 2 (行进, 2011), PP. 25–40.
23. 同上。, p. 2. Targeting improves with information about perpetrators, but blanket retaliation
could also prove effective, provided that punishment also touches the attackers. 同时,
no one ever possesses “full certainty.” The U.S. legal standard is “reasonable doubt.”
24. Martin C. Libicki, Cyberdeterrence and Cyberwar (圣莫妮卡, 加利福尼亚州。: RAND, 2009).
25. 同上。, p. 43.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 47

much time marketing their exploits as they do ªghting, bombing, assassinat-
英, 等等. Where anonymity protects an aggressor from retribution, 它也是
dilutes credit for the deed. Vandals often “tag” their handiwork—creating an
identity where none need exist—precisely because real anonymity means not
receiving credit for one’s handiwork. Internet vandals also brand their ex-
ploits, presumably because they wish to receive credit for their exploits, 甚至
at some risk to their anonymity.

Just as ongoing cyberattacks from unidentiªed sources do not give the target
a way to retaliate, they also fail to provide the target with the means to acqui-
esce. Demands from an anonymous cyberwarrior will tend to be ignored or re-
neged upon once vulnerabilities are identiªed and addressed. Demands might
also come from a source that did not, or was even incapable of, mounting a
cyberattack. As with the use of identifying symbols in war, it is in an attacker’s
interest to “brand” its actions to most effectively elicit concessions from a
目标. 的确, even if demands are complied with, an attacker will have dif-
ªculty obtaining sustained compliance, given the impossibility of demonstrat-
ing future capabilities.

Discussion of attribution problems in cyberspace also reºects a subtle but
telling shift in framing. Libicki’s simple calculus of deterrence, 例如, 在-
volves “getting caught,” something more often characteristic of crime than
战争. Some aspects of international relations involve anonymity. 间谍, 共-
vert operations, and certain kinds of political theft or murder function most ef-
fectively when the perpetrators are unknown, or indeed when the operations
themselves remain undisclosed. Strategic or tactical advantage can also stem
from anonymity and surprise in terrestrial military missions, though nations
and groups often sacriªce surprise and advertise their role in contests to ex-
tract concessions or tacit or formal admission of defeat. How does one surren-
der to no one in particular? The advantage of anonymity will persist for
peripheral forms of warfare on the internet, just as it has played a role in ter-
restrial competition and conºict. Most forms of political conºict, 然而,
encourage disclosing an initiator’s identity. Coercion requires attribution. 辛-
伊拉利, threats designed to elicit concessions or deter aggression are already
problematic in physical space.26 This “credibility problem” mirrors the attri-
bution problem and is perhaps equally likely to make internet aggression
problematic for initiators as for possible targets.

26. Robert Powell, Nuclear Deterrence Theory: The Search for Credibility (剑桥: 剑桥
大学出版社, 1990); and Barry Nalebuff, “Rational Deterrence in an Imperfect World,” World
政治, 卷. 43, 不. 3 (四月 1991), PP. 313–335.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 48

Several scholars also note that cyberwar creates an unparalleled legal envi-
罗门特, one in which even deªning the scope of activities has become prob-
莱马蒂克. There has not yet been sufªcient time, or perhaps the inclination,
to solidify the legal standing of cyber conºict in international law. 查尔斯
Dunlap notes that democracies,
尤其, have sidestepped attempts
to formulate a treaty covering cyber conºict.27 Indeed, it is inherently dif-
ªcult to legally deªne a process when capabilities have yet to be uncovered.
Cyberattacks could perhaps be looked upon as the legal equivalent to armed
attacks.28 Yet, the legal deªnition of armed conºict involves “signiªcantly de-
structive attacks taking place over some period of time and conducted by a
group that is well-organized,”29 a set of conditions yet to be adequately dem-
onstrated in cyberspace.

cyber salience—a balancing of perspectives

While the bulk of reactions to cyberwar emphasize dramatic dangers, 一些
studies offer a more balanced perspective. Tim Maurer contrasts the gloomy
picture provided by the bulk of writers with estimates of likely determinants
of loss of life associated with a cyberattack.30 Maurer lists the relative secu-
rity of civilian infrastructure, participation of nonstate actors, and the evolu-
tion of law regarding retaliation strategies. He concludes that loss of life from
cyberattacks will generally be slight. 的确, drawing on his estimates, Maurer
asserts that “a digital Pearl Harbor would cost fewer lives than the attack
70 years ago.”31

Wesley Clark and Peter Levin anticipate an inevitable rise in cyberwarfare,
one that will eventually involve broad sectors of society.32 Populations will
face “network-born disruptions of critical national infrastructure”—including
terrestrial and airborne trafªc, energy generation and distribution, 和
ªnancial system. The authors note, 然而, that the United States and other

27. Charles J. Dunlap Jr., “Perspectives for Cyber Strategists on Law for Cyberwar,” Strategic
Studies Quarterly, 卷. 5, 不. 1 (春天 2011), PP. 81–99.
28. 文顿·G. Cerf, “Safety in Cyberspace,” Daedalus, 卷. 140, 不. 4 (落下 2011), PP. 59–69.
29. Michael N. 施密特, “Cyber Operations in International Law: The Use of Force, Collective Se-
安全性, Self-Defense, and Armed Conºict,” in Committee on Deterring Cyberattacks, 编辑。, Proceed-
ings of a Workshop on Deterring Cyberattacks: Informing Strategies and Developing Options for U.S.
政策 (华盛顿, 华盛顿特区: 美国国家科学院出版社, 2010), p. 176.
30. Tim Maurer, “The Case for Cyberwarfare,“ 对外政策, 十月 19, 2011, http://万维网
.foreignpolicy.com/articles/2011/10/19/the_case_for_cyberwar.
31. 同上.
32. Wesley K. Clark and Peter L. 莱文, “Securing the Information Highway: How to Enhance the
United States’ Electronic Defenses,“ 外交事务, 卷. 88, 不. 6 (十一月十二月 2009),
PP. 2–10.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 49

nations are doing a great deal to mitigate the threat. 美国有
pledged a reported $30 billion by 2015 as part of the Comprehensive National
Cyber Security Initiative. 此外, Clark and Levin note lessons learned
from previous attacks. The most effective electronic security strategy must op-
erate under full disclosure. Experts in academic, 工业的, and governmental
sectors must quickly collaborate on a mitigation strategy. 然而, Clark and Levin
also acknowledge that “electronic security works best when it is autonomous,
adaptable, distributed and diversiªed.”33

Stephen Walt argues that a critical preliminary task is to separate out differ-
ent dangers grouped under the common rubric of “cyber-warfare.”34 For Walt,
cyberwarfare consists of four distinct issues: degrading an enemy’s military
能力, penetrating networks to shut down civilian infrastructure, 网络-
based criminal activity, and cyber espionage. These four issues help to frame
cyberwarfare as an evolving, nuanced set of issues, each amenable to its own
cost-beneªt analysis. Thomas Rid argues that cyberwar is not really war
because it fails to conform to conventional deªnitions of conºict.35 Rid’s chief
point—mirroring in an interesting manner Maurer’s argument—is that cyber-
war is not sufªciently violent or casualty-producing to be considered war. 作为
这样的, cyberwar is a misnomer. This perspective risks becoming a purely aca-
demic exercise, 然而, if cyber conºict eventually supplants military vio-
lence as the ultimate arbiter of international politics. Cyberwar does not need
to be war to make war obsolete. 反而, it must fulªll the existing functions of
terrestrial warfare if it is to rival the utility of existing forms of conºict. As I ar-
gue below, the internet is extremely unlikely to substitute for, or serve as an al-
ternative to, earthbound warfare.

Contrasting Vulnerability and the Nature of Threats

It has become an article of faith among those attentive to questions of security
or technology that cyberspace is a new domain where the old rules of warfare
no longer apply. Cyberspace could constitute a hidden back door, enabling op-
ponents of the Western-dominated world order to undermine the hard-won
terrestrial advantages of the established powers. History makes clear that tech-
nological innovations or new modes of organization eventually topple every

33. 同上。, p. 6.
34. 斯蒂芬·M. 沃尔特, “Is the Cyber Threat Overblown?“ 对外政策, 行进 30, 2010, http://
walt.foreignpolicy.com/posts/2010/03/30/is_the_cyber_threat_overblown.
35. Thomas Rid, “Cyber War Will Not Take Place,》 战略研究杂志, 卷. 35, 不. 1 (Feb-
鲁阿里 2012), PP. 5–32.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 50

hierarchy.36 The United States and other nations have already begun expensive
reforms designed to prepare for war over the internet, while some civil liber-
ties have been curtailed on the grounds that cyberwar may well constitute the
next existential threat.

然而, it is far from clear that the internet is transformational in military terms,
let alone revolutionary. Lacking information about whether developments are
radical or merely incremental, it may make sense to adopt a few guidelines
that will help to determine whether there is cause for panic. A reasonable level
of caution is usually provided by our own common sense. Most readers will
lock their doors at night, 例如, and refrain from handling large sums
of cash in a dark alley. Imagining what others could do to injure each of
我们, 然而, can quickly descend into paranoia. It is not reasonable to believe
that someone is intent on mischief simply because it is possible for them to
inºict harm.

Even in the safest of societies, 个人, 团体, and entire communities
are subject to an enormous variety of potential hazards. Much could be done
to impinge on each of us, even though few of these possibilities are ever exer-
cised, or experienced, with any regularity. The physical world hosts a multi-
tude of venues for extremely unlikely accident or disease. A small number of
people prefer to remain indoors rather than risk being struck by lightning or
struck down by botchulism. 仍然, individuals with these concerns may merit
more attention from psychiatric professionals than from military planners. 是-
ing vulnerable will be novel to no one living in our modern, highly integrated
世界. 的确, the capacity to hurt is so ubiquitous in densely populated por-
tions of the globe that blood would coat the streets if it were not true that rela-
tively little relationship exists between the capacity to attack and the actual
prospect that one will be invaded, assaulted, or otherwise done in.37

Just about anything is possible. Someone may have put poison in your Corn
ºakes at breakfast. Terrorists may have singled you out for vengeance, or you
might just become one of the unlucky few who are in the wrong place at the

36. On political order and technological change in world affairs, see Robert Gilpin, War and Change
in World Politics (纽约: 剑桥大学出版社, 1981); William H. McNeill, The Pursuit of
力量: 技术, Armed Force, and Society since
1000 (芝加哥: 芝加哥大学出版社,
1984); Martin van Creveld, Technology and War: 从 2000
到现在 (纽约: 自由的
按, 1989); Emily O. Goldman and Leslie C. Eliason, 编辑。, Diffusion of Military Technology and Ideas
(斯坦福大学, 加利福尼亚州。: 斯坦福大学出版社, 2003); and Keir A. Lieber, War and the Engineers: 这
Primacy of Politics over Technology (伊萨卡岛, 纽约: 康奈尔大学出版社, 2005).
37. The current fascination with vampires (in popular culture) and zombies (among academics) il-
lustrates the basic point. Aggression becomes endemic if harm has an intrinsic payoff (例如, 人类
beings are food). Vampires and zombies threaten not just individuals but the whole of civil society,
because interdependence/specialization requires that vulnerability not equal insecurity.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 51

wrong time. When a commuter steps outside to start her car or to catch
the bus, it is impossible to be certain that no truck will jump the curb and that
every asteroid will remain in its usual orbit. 但是, despite endless potential
for injury or death, few of us have chosen to harden our living rooms against
cruise missile attack or immersed ourselves in real-time plots from NASA
charting the trajectories of space detritus. In dealing with known unknowns,
we became comfortable with not being protected. California homeowners
typically do not carry earthquake insurance, 例如, even though “the
big one” is an eventuality. We do so because security is expensive; being in-
demniªed against unlikely events may literally not be worth the effort. 一
could buy that bulletproof vest listed on Ebay, but then how often would
it prove fashionable at the ofªce or in the classroom? The probabilities of eso-
teric catastrophe are by their nature minute. Unlikely events are unlikely, 和
so most of us go about our business, paying little attention to the potential
menace from the skies or, for that matter, 来自彼此.

Governments face similar realities. Many threats are conceivable, but rela-
tively few actually materialize. A holistic approach to security involves assess-
ing risks, and then allocating ªnite resources to address tractable threats,
making the largest improvements in protection or, 反过来, the greatest in-
creases in inºuence.38 Every dollar spent on national defense must be taken
from objectives such as improving education, building or repairing infrastruc-
真实, or paying down the debt. Only extremely afºuent (or paranoid) pop-
ulations pay the price of pursuing protection from the most exotic hazards.
更重要的是, protection is inevitably incomplete, and comes with its
own consequences, including other forms of insecurity. The risk of attack
is never zero, given that a potent defense or deterrent endangers the security
of others.39

If violence is ecological, why do human beings not live in consummate fear?
Most of us are safe because the multitudes capable of causing us harm have lit-
tle interest in doing so. 大部分情况下, violence does little that potential per-
petrators view as worth their while. Humanity is protected by an invisible

38. The distinction between security and inºuence is often missed or misrepresented in policy de-
bates. 例如, although discussions often center on protecting the American people (secu-
理性), much of U.S. defense spending actually purchases discretion abroad (inºuence).
39. On the security dilemma, see John Herz, “Idealist Internationalism and the Security Di-
引理,” 世界政治, 卷. 2, 不. 2 (一月 1950), PP. 157–180; Robert Jervis, “Cooperation un-
der the Security Dilemma,” 世界政治, 卷. 30, 不. 2 (一月 1978), PP. 167–214; Randall L.
Schweller, “Neorealism’s Status-Quo Bias: What Security Dilemma?” 安全研究, 卷. 5, 不. 3
(春天 1996), PP. 90–121; and Charles L. Glaser, “The Security Dilemma Revisited,” 世界政治,
卷. 50, 不. 1 (十月 1997), PP. 171–201.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 52

shell of indifference or inefªcaciousness. The stranger coming toward you on a
busy city street could swing out his arm, catching you under the chin. 他
could be carrying an Uzi, which in a ªt of rage will leave you and other
passersby on the pavement in a pool of intermingling ºuids. 然而, you are prob-
ably not going to die a violent death in the next minute, or at any other time.
The potential for violence is both ubiquitous and seldom realized because
striking out, though often possible, does little to beneªt the violator. The use of
force is costly, risky, and mostly unproductive. When we learn of violence, 我们的
natural inclination is to ask “why?” Like a police detective, we seek a motive.
Violence requires a cause. 什么时候, on occasion, we cannot identify the logic be-
hind a perpetrator’s choice of a target, the experience is remarkable, 和骗局-
siderably puzzling.

Most of us are capable of seriously injuring others, but for the most part we
fail to exercise our capabilities because there is no positive reason to strike. A
driver behind the wheel of an automobile can cause enormous damage or loss
生命的, and yet pedestrians walk in front of stopped vehicles at an intersection,
with little care or attention to the fact that a tap of a toe on the accelerator is all
it takes to end their lives. The mere capacity to harm is just not a very good
predictor of aggression, because the potential for harm is everywhere but sel-
dom practiced. Few of us are likely to be the target of attack, and so each of us
can greet the day with minimal anxiety, to say nothing of personal security, 不是
because we are thoroughly protected, but because causing harm is usually in-
convenient, unnecessary, or pointless for potential perpetrators. Attacking us
(or others) serves no purpose.

The internet makes it possible to interact with people just about anywhere
on the globe as easily, or even more easily, than conversing with the neighbors
next door. Initial attention to the mobility of cyberspace focused on the poten-
tial for good, but convenience also overcomes barriers to conºict. The supply
of targets for cyber acts of aggression is certainly huge relative to the supply of
perpetrators of physical violence. 按这样的方式查看, it is remarkable that
cyberspace has yet to become exclusively the domain of fraud, identity theft,
and other acts of predation, interspersed only with pornography and the occa-
sional Nigerian emailer looking to deposit millions in your bank account. 然而,
if the internet makes it easy to reach out and touch others, it in no way makes
those contacts profound. Casual attempts to undermine one’s welfare abound,
but it is with equal casualness that we ignore the bulk of spam or internet sites
marketing lapsed software. Predation continues unabated on the world wide
网络, but if it is easy to reach us, contact is all the more superªcial.

The bulk of internet trafªc is benign. Cyberspace has not made life more

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 53

dangerous for the multitudes. There are crimes on the internet, but it is unclear
whether these have increased the overall crime rate. Internet crime often sub-
stitutes for crimes that would have been committed in terrestrial space never-
theless. Perhaps even more to the point, much of internet fraud and cyber
violence is intrinsically tied to the physical domain; much of the harm initiated
on the internet eventually gets perpetrated in more conventional ways.

The safety from predation that mass populations achieve as a result of num-
bers and anonymity is denied political institutions and their representatives.
Organizations and certainly countries have personnel and property that can be
targeted with violence. How might this shift in the nature of the target affect
the risk of cyberattack? 再次, we must inquire not about what could
happen but why individuals, 团体, or nations might act.

Nations and organizations can be attacked through the internet, just as they
have long been attacked in physical space. The ease with which such attacks
can be perpetrated is an obvious, and much discussed, 现象. Physi-
cal space has always been an important barrier to conºict.40 Even today, 这
single best predictor of interstate conºict is contiguity.41 Lowering the cost to
transmit an attack, 然而, only increases its appeal if the attack is also capa-
ble of achieving the actor’s desired ends. Beyond numbers and anonymity, 这
main factor protecting individuals is even more potent for institutions; 如何
force will produce the desired change is often unclear. Cyberattacks can be
appealing as political acts only to the degree that they affect the decisions that
organizations and sovereigns make with and without cyber violence. 因为
understanding how cyberwar can (and cannot) inºuence politics is largely the
same as understanding how conventional military violence acts on politics, 我
turn next to a brief discussion of the nature of terrestrial warfare.

War and Peace in the Internet Age

我们. Defense Secretary Leon Panetta returned to the theme of a cyber “Pearl
Harbor” more than once. In his Senate conªrmation hearings, then Central

40. See Kenneth Boulding, Conºict and Defense (纽约: Harper and Row, 1962) concerning the
“loss of strength gradient.” Halvard Buhaug and Nils Petter Gleditsch show that the relationship,
though weaker in recent decades, remains in effect. See Buhaug and Gleditch, “The Death of Dis-
坦斯?: The Globalization of Armed Conºict,” in Miles Kahler and Barbara Walter, 编辑。, Ter-
ritoriality and Conºict in an Era of Globalization (剑桥: 剑桥大学出版社, 2006)
41. 保罗·R. Hensel, “Territory: Theory and Evidence on Geography and Conºict,” in John A.
Vasquez, 编辑。, What Do We Know About War? (纽约: Rowman and Littleªeld, 2000), PP. 57–84;
and Paul Senese, “Territory, Contiguity, and International Conºict: Assessing a New Joint Expla-
国家,” American Journal of Political Science, 卷. 49, 不. 4 (十月 2005), PP. 769–779.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 54

Intelligence Director Panetta noted, “I have often said that there is a strong
likelihood that the next Pearl Harbor that we confront could very well be a
cyber attack,” adding that, “This is a real possibility in today’s world.”42 No
event in the twentieth century has done more to realign U.S. public opinion
than the Japanese attack on Pearl Harbor, which mobilized the nation psycho-
logically for entry into World War II. The analogy may be apt, but almost cer-
tainly not for the reasons contemplated by Secretary Panetta. The situation in
1941 serves as a basis for comparison and a point of departure in examining
cyberwar. To understand why a cyber Pearl Harbor is not as threatening as it
声音, it helps to review what the air raids on December 7, 1941, were meant
to accomplish and what they achieved.

Before exploring Secretary Panetta’s Pearl Harbor analogy, 然而, I ªrst
discuss the nature of war and how key attributes of warfare function or fail via
the internet. States and nonstate actors make war to further their interests
when incompatibilities exist between those interests, and when alternative
methods of conºict resolution are deemed ineffective or inefªcient. 虽然
many conºicts are conceivable, most potential wars do not occur because the
participants recognize that threats or uses of force are futile. Only some acts of
violence are likely to achieve the objectives for which political actors strive. 如果
the futility of violence discourages the bulk of terrestrial conºict, it is an even
larger concern in cyberwar.

a brief look at the logic of war

The theory of war provides two basic mechanisms for the expression of politi-
cal interests through physical violence. 第一的, force can be used to punish or
compel, indirectly affecting the state of the world by harming an enemy to
make it do something that it would not do otherwise, or alternately discourag-
ing change by raising the price of aggression. Just about everyone from parents
to police and prime ministers seems to grasp the intuition behind creating con-
sequences to modify behavior. 第二, force can be used to conquer, 直接地
imposing one’s will by capturing and controlling inhabitants or resources con-
tained in a given physical space. One thing that the recent “Occupy” move-
ment shared with the U.S. military was the conviction that denial can prove
effective if exercised for a sufªcient period of time.43 The ability to conquer or

42. 我们. Senate Committee on Armed Services Hearing to Consider the Nomination of Hon. Leon
乙. Panetta to be Secretary of Defense, 六月 9, 2011, 华盛顿, 华盛顿特区, http://万维网. armed-
services.senate.gov/Transcripts/2011/06%20June/11-47%20-%206-9-11.pdf, p. 25.
43. The whole question of denial/conquest depends on whether a target is likely to acquiesce. 这
British have occupied Scotland for centuries, with some success. They occupied Ireland for just

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 55

compel can, 反过来, be used to achieve these objectives through actual force
(offense or defense), overt threats (deterrence or compellence), or the shadow
of war (例如, diplomacy).44

冷战期间, the two superpowers could easily have annihilated
彼此, along with much of the rest of the world. 然而, vast nuclear arsenals
remained dormant, and the prospect of mutual harm even led to the peculiar
stability of mutual assured destruction. 再次, the mere capacity to hurt tells
us relatively little about the actual advent of violence, even though the poten-
tial in the Cold War was extraordinary and mutual assured destruction, 由其
自然, meant that devastation could not be prevented. 的确, it was this mu-
tual vulnerability that many credit with the Cold War remaining cold.45

相比之下, Japan kept even its own diplomats in the dark about its plans to
attack U.S. bases in the Philippines and Hawaii because it could not share in-
formation about its intentions with U.S. ofªcials without fatally weakening the
effectiveness of such a plan.46 President Richard Nixon reportedly threatened
to restart the Christmas bombing campaign against Hanoi to expedite the talks
for ending U.S. involvement in the Vietnam War. The threat could be made un-
der any circumstances, but it was more credible given that Hanoi had recently
been attacked and the consequences of renewed heavy bombing were also
清除. Whether force is threatened or carried out depends on whether threats
can be made without degrading the instruments of military advantage. Reve-
lation of the ability to harm can prove sufªciently compelling to cause a target
to make concessions or to alter the target’s behavior. 反过来, if threatening
an enemy makes an eventual attack less effective, then the temptation may be
to strike rather than threaten.

Experts on cyber security have failed to draw the same conclusions from
the inability to protect that strategists drew from the Cold War. It is true
that mutual assured destruction may not exist in cyberspace, as it did in the
post–World War II terrestrial world. Whether the internet is a different kind of
strategic setting, and what this means in terms of what can and cannot be

about the same length of time with very little lasting political effect. Perhaps days or decades are
unlikely to succeed when centuries often prove insufªcient.
44. Geoffrey Blainey, The Causes of War (纽约: Free Press, 1973); and James D. Fearon, “Ratio-
nalist Explanations for War,” 国际组织, 卷. 49, 不. 3 (夏天 1995), PP. 379–414.
45. 约翰·J. 米尔斯海默, Conventional Deterrence, (伊萨卡岛, 纽约: 康奈尔大学出版社, 1983); 和
Kenneth N. 华尔兹, “Nuclear Myths and Political Realities,” American Political Science Review,
卷. 84, 不. 3 (九月 1990), PP. 731–745.
46. Branislav Slantchev, “Feigning Weakness,” 国际组织, 卷. 64, 不. 3 (夏天
2010), PP. 357–388. and Erik Gartzke, “Globalization, Economic Development, and Territorial
Conºict,” in Miles Kahler and Barbara Walter, 编辑。, Territoriality and Conºict in an Era of Globaliza-
的 (剑桥: 剑桥大学出版社, 2006), PP. 156–186.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 56

accomplished—offensively or defensively—remains to be shown. 虽然
the Cold War is remembered as the ideal deterrence environment, strategic
thinkers and government ofªcials struggled with how they could exercise
inºuence in such a world. The mere potential for imposing harm did not imply
that harm would be imposed, 甚至那个, when imposed or threatened, 已经-
tions would respond in an obliging manner. Few could doubt in retrospect that
citizens and leaders on both sides of the iron curtain felt vulnerable, 尤其
during the ªrst decade or so of the nuclear era. It does not follow, 然而,
that a heightened sense of insecurity was reºected in actual behavioral con-
ºict. Whether warfare in cyberspace will depart radically from previous pat-
燕鸥, or will mimic, in part or in whole, familiar patterns of conºict from
earlier eras, will depend on the degree to which the strategic logic of cyberwar
is able to accommodate the objectives of political actors in contemplating or
exercising coercion.

Nor do students of cyberwar seem much preoccupied with the implications
of Nixon’s Hanoi bombing campaign. The threatened use of force is intended
在这个, and in most other instances, to alter behavior through the prospect of
long-term damage. To the degree that damage can be quickly and easily re-
paired, there is not much leverage in raising such a threat. 反过来, 细节
harmful to attackers or to the effectiveness or potency of attacks are typically
concealed from an opponent, even when this information would signiªcantly
increase the credibility of coercive threats. Flight plans, bomb loads, and elec-
tronic countermeasures used by U.S. B-52s, 例如, were not shared with
Hanoi, because this would have compromised the capacity of U.S. forces to
carry out Nixon’s threat.

国家, 团体, or individuals with the ability to inºict harm must ask not
just how much can be inºicted at what cost, but also what is to be achieved
through force, and whether these ends are justiªed compared to the price and
availability of other, typically cheaper, mechanisms. 力量, or the threat of
力量, is useful as punishment to the degree that the harm imposed is substan-
tial and durable. Damage that can be quickly or easily undone will not do
much to deter or compel, but it will alert an enemy to vulnerabilities in its de-
栅栏, and certainly also antagonize an opponent, increasing the risk of coun-
terattack and general hostility. The threat or exercise of physical conquest can
be effective provided the perpetrator ªnds it worthwhile to engage in the cost-
liest form of politics. 这里, 再次, the simple ability to act aggressively is not it-
self sufªcient to predict aggression. The United States could probably conquer
Canada if it chose to, and yet Canada remains free and independent. 最多
状态, 团体, and individuals persist in peace because they can conceive of

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 57

no beneªt from force, even if violence, and victory, are feasible. The mere abil-
ity to cause harm over the internet does not sufªce to predict that cyberwar
will substitute for terrestrial conºict, or even that it will be an important inde-
pendent domain for the future of warfare.

warfare in cyberspace

Beyond questions of means and motive, two basic features make cyberwarfare
different from other types of conºict. 第一的, much of the damage contemplated
by cyberwar is in all likelihood temporary. The assumption among many cyber
pessimists that the potential for creating harm is sufªcient to make cyberspace
a suitable substitute for, or indeed alternative to, terrestrial conºict is incorrect.
Shutting down power grids, closing airports, or derailing communication
could be tremendously costly, but most damage of this type will be ªxed
quickly and at comparatively modest investment of tangible resources. 关于-
gardless, damage of this type is sunk. Losses experienced over a given interval
cannot be recovered whatever one’s reactions and so should not have much di-
rect impact on subsequent policy behavior. Harm inºicted over the internet or
any other medium will matter politically when it alters the subsequent balance
of power, or when it indicates enemy capabilities that must be taken into ac-
count in future plans. Because cyberwar does not involve bombing cities or
devastating armored columns, the damage inºicted will have a short-term im-
pact on its targets.47 To accomplish politically meaningful objectives, 网络-
attacks must contribute to other aspects of a more conventional war effort.
And to affect the long-term balance of power, 例如, cyberwar must be
joined to other, more traditional, forms of war.

Temporary damage can be useful under two circumstances. 第一的, compro-
mising or incapacitating networks might give an enemy valuable tactical, 或者
even strategic, advantages. An opponent that cannot shoot, 移动, resupply,
or communicate will be easier to defeat. 尽管如此, the advantaged party
must still act through some medium of combat to seize the initiative. Notions
that cyberattacks will themselves prove pivotal in future war are reminiscent
of World War I artillery barrages that cleared enemy trenches but still required
the infantry and other arms to achieve a breakout. Whether an actor can
beneªt from cyberwar depends almost entirely on whether the actor is able to

47. In the 1970s, the so-called neutron bomb promised high casualties with minimal damage to
physical structures. It might be tempting to think of cyberweapons as anti-neutron bombs, 因为
they damage infrastructure, leaving people largely unharmed. 然而, this is also a key limitation of
cyberwar, as damage is temporary and far from complete.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 58

combine a cyberattack with some other method—typically kinetic warfare—
that can convert temporary advantages achieved over the internet into a last-
ing effect. In the parlance of war, internet attacks produce a “soft kill” that is
valuable only when attackers prosecute follow-on attacks with traditional mil-
itary force or permanently weaken an enemy in some other way.48

The notion of a devastating surprise attack is a particularly baroque aspect
of cyberwar paranoia, and is certainly frightening to the degree that such sce-
narios are accurate. 然而, the idea of a surprise internet attack is misleading and
relies on a fundamental misconception of the role of internet-based aggression.
Modern warfare seldom allows any one element of combat to prove pivotal.
反而, it is the ability to combine elements into a complex whole that increas-
ingly distinguishes the adept utilization of force.49

The archetype of modern, combined arms warfare is the blitzkrieg, 在哪里
the lethality and decisiveness of conventional military violence is enhanced by
actions designed to disrupt the enemy’s military and civilian infrastructure.
An important element of blitzkrieg was the use of terror weapons, 例如
Ju 87 “Stuka” dive bomber, to sow panic, causing enemy populations to ºood
roads and railways, thereby crippling transportation grids needed by the de-
栅栏. 然而, fear is temporary and in the absence of substance, subsides. 这
Stukas were effective only as long as Germany held other military advantages
over its enemies. 相似地, unless Stuka attacks were accompanied by a
ground attack or an invasion, their role as terror weapons was largely redun-
丹特. Stukas contributed little to Germany’s effort to subdue Great Britain, 为了
例子. Stuka units experienced heavy casualties when engaged against a so-
phisticated air defense and were eventually removed from service in the Battle
of Britain. The hubris of Luftwaffe Commander in Chief Hermann Göring in
promising victory while exploiting just one domain (the air) was precisely that
he exaggerated the effectiveness of a new technology in isolation from other el-
ements of an integrated offense.

There is no reason to believe that cyberwar will be any more useful as an iso-
lated instrument of coercive foreign policy. An attack that causes temporary
harm will inevitably be followed by countermeasures and heightened vigi-
lance, as has happened, 例如, in Estonia in the aftermath of the 2007 在-
tacks. For cyber aggression to have lasting effects, a virtual attack must be

48. Nonlethal weapons have similar functionality. Immobilizing an enemy with sticky foam
works until it does not; nonlethal weapons are effective only if they are followed up by other ac-
系统蒸发散, such as physical restraint or redeployment of forces away from the immediate area.
49. Stephen Biddle, Military Power: Explaining Victory and Defeat in Modern Battle (普林斯顿大学, 新泽西州:
普林斯顿大学出版社, 2004).

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 59

combined with physical intervention. Knocking out communications or power
infrastructure could cause tremendous disruption, but the ability to quickly re-
cover from such attacks implies that the consequences for the balance of na-
tional power would be negligible. The need to follow virtual force with physical
force to achieve lasting political consequences suggests that the application of
cyberwarfare independent of conventional forms of warfare will be of tertiary
importance in strategic and grand strategic terms. If one cannot foresee circum-
stances where the terrestrial use of force is plausible independent of cyberwar,
then cyberwar is also unlikely to constitute a fundamental threat.

A second element of the logic of cyberwar has to do with inºuence. 相当
than attacking directly, an actor can use the potential to harm (deterrence or
compellence). The ability to shut down the U.S. energy grid, 说, 可能
used to compel U.S. ofªcials to refrain from aggressive policies or actions, 或者
persuade the United States to make diplomatic concessions. 然而, 问题
with the standard deterrence or compellence logic in the context of potential
网络攻击, as I have already pointed out, is that revealing a given set
of cybercapabilities heavily degrades their usefulness. Deterrence and compel-
lence are therefore marginal as “pure” actions in cyberspace. 的确, 担忧
that nations will not be able to deter cyber aggression amount to recogni-
tion that neither will cyber threats prove very effective as threats or induce-
评论. 再次, actions in cyberspace can be combined with initiatives in
physical space, but this just reinforces the fact that, rather than a distinct form
of conºict, cyberwar is basically tied to conventional forms of warfare.

Imagine for a moment that a foreign power has hacked into the communica-
tions systems of the United States or another major Western power. Imagine
further that this foreign power can disable cellular phone systems or military
radio networks more or less at will. The foreign power could threaten its target
with this capability, but obviously the leadership of the target state must be
skeptical of such a threat, because the foreign power could easily be blufªng.
Proof of the capacity to damage the target nation is needed, but such evidence
会, 反过来, jeopardize the potency or effectiveness of the cyberattack,
allowing the target to address some or all of its vulnerabilities or adopt
to countermeasures.

Contrast this scenario with the revelation in the 1990s that the United States
had developed radar-evading “stealth” aircraft. Knowledge by foreign powers
that a confrontation with the United States would necessarily involve the risk
of attack by stealthy ªghters and bombers in no signiªcant way lessened the
military effectiveness of these weapons systems, given that countermeasures
to stealth technology have been slow to develop. Stealth technology thus

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 60

serves as an excellent deterrent/compellent, because it can be used to coerce
an opponent without sacriªcing much of its military value. The “perishable”
nature of capabilities in cyberwar means that advantages generally yield time-
bound deterrent or compellent threats, and thus create little in the way of lev-
erage for states that have, or plan to invest in, cyberwar assets. Deterrent/
compellent threats work best when they are tied to capabilities that are not
much affected by enemy knowledge of the capabilities, whereas the opposite is
true for capabilities that are compromised by the revelation of forces, technolo-
吉斯, or attack plans.

Offensive cyber advantages are thus “use and lose” capabilities. Revealing
the capacity to harm via the internet typically also means tipping the enemy
off to vulnerabilities that can be remedied or compensated for, while inºicting
harm seldom has a durable effect on the balance of power. “Use and lose” ca-
pabilities cannot compel or deter, because convincing evidence of the capac-
ity to inºict harm is itself useful information in nullifying the threat. 如果
instead cyberwar is waged rather than threatened, then the temporary nature
of cyber harm dictates that an enemy follow internet attacks with more kinetic
行动. 否则, the use of cyber force is punitive, even provocative. 作为
这样的, cyberwar remains adjunct to terrestrial force unless some way is found
for cyberattacks to permanently alter the balance of power.

the myth of a cyber pearl harbor

The air strikes on December 7, 1941, against U.S. military instillations in
Hawaii and in the Philippines were an important tactical and even strategic
victory for Japan. 然而, the attacks, as is now widely understood, were a prodi-
gious failure in grand strategic terms, setting up a nearly inexorable path to
Japanese defeat and surrender.50 Ofªcials on both sides recognized this almost
immediately. When informed of the attack, Adm. Isoroku Yamamoto is said to
have offered this stark commentary, “I fear all we have done is to awaken a
sleeping giant and ªll him with a terrible resolve.”51 Leaders in Washington
could not be bothered even to give war in the Paciªc ªrst priority, focusing the
might of the nation instead on war in Europe.

The Japanese decision to strike at the United States was a calculated gamble,
balancing the imperatives of seasonal weather patterns and the impending de-

50. “From a strategic point of view, Pearl Harbor was one of the most spectacular miscalculations
in history.” Ian W. Toll, “A Reluctant Enemy,“ 纽约时报, 十二月 7, 2011.
51. Documentary evidence for the famous quotation is not available. Although widely referenced,
it may be apocryphal. It certainly summarizes Yamamoto’s views, 然而.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 61

cline in power projection capabilities for the resource-starved empire with the
realization that much was being staked on a complex plan linking the conquest
of oil ªelds in Southeast Asia with a temporary shift in the regional balance of
power in the Paciªc.52 With almost no indigenous sources of iron, rubber, 和
especially oil, Japan was dependent on foreign-held reserves to feed its facto-
ries and allow it to sustain its occupation of Manchuria and war in China. 这
我们. embargo led Japanese ofªcials to consider stark alternatives. Either Japan
must relent and withdraw its forces from East Asia, or it would have to cap-
ture oil-rich regions in the south. 这, 反过来, put Japan in direct conºict with
美国. The Japanese plan was to blunt U.S. naval and military capa-
bilities temporarily, long enough to prepare a defense in depth in the western
Paciªc and present the United States with a fait accompli, one that would com-
pel the Americans to submit to a compromise, a negotiated settlement that
would end the war.

The prospect of an impending and signiªcant relative decline in power and
optimism about the potential beneªts of a “bounce” giving Japan a temporary
advantage in the region were critical elements of Japan’s decision to go to war.
重要的, Japan underestimated the psychological impact that the Pearl
Harbor raid would have in mobilizing U.S. public opinion. Japanese com-
manders also overestimated the damage they could inºict on U.S. forces.53 The
surprise assault famously failed to catch the U.S. aircraft carriers in port. 它是
this inability to impair U.S. naval airpower that seemed to vex Yamamoto the
最多. Even allowing for Japanese optimism and error, 然而, Tokyo was re-
luctant to act against the United States prior to the end of 1940, when events in
Europe laid bare Dutch holdings in Asia and dramatically weakened the abil-
ity of British forces in the region to resist.

Tactical or strategic surprise is useful as a temporary force multiplier; an at-
tack such as that on U.S. and European forces in December 1941 could shift the
balance of power in Japan’s favor for a time, but the real value of a surprise at-
tack is what it allows an assailant to accomplish subsequently. An attacker can
exploit the effect of surprise to prepare a more effective defense or, alternately,

52. Louis Morton, “Japan’s Decision for War,” in Kent Roberts Greenªeld, 编辑。, Command Decisions
(华盛顿, 华盛顿特区: Center of Military History, Department of the Army, 2000).
53. Evidence of this is reºected in subsequent efforts by the Imperial Japanese Navy to realize
the original strategic objective. Midway, 尤其, was an elaborate trap, masterminded by
Yamamoto, to locate and destroy the remaining U.S. carrier force in the Paciªc. 看, 例如,
Gordon W. Prange, Miracle at Midway (纽约: 企鹅, 1983); Jonathan Parshall and Anthony
Tully, Shattered Sword: The Untold Story of the Battle of Midway (Herndon, Va.: Potomac, 2007);
Mitsuo Fuchida, Midway: The Battle That Doomed Japan (纽约: Ballantine, 1986); and Craig L.
Symonds, The Battle of Midway (纽约: 牛津大学出版社, 2011).

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 62

to prosecute additional offensive action against the target or others. 就其本身而言, A
surprise attack has limited utility, precisely because surprise fades in time.

Japanese war planners anticipated the temporal nature of advantages
gleaned from the Pearl Harbor attack. It was hoped that Japan would be able
to secure critical resources in the south, fortify its gains in depth, and wait out
the American onslaught.54 At no time did Japan seriously consider unlimited
war with the United States. 的确, Japanese planners recognized the impossi-
bility of directly defeating the United States.55 In the months after December 7,
the United States mainland was open to attack. Japanese forces landed in the
Aleutian Islands, and Japanese submarines shelled a few isolated coastal com-
社区, but there were never any serious plans to carry the war to the conti-
nental United States.56

Now suppose that Japanese ofªcials recognized from the outset that they
would not be able to target the U.S. carriers or other U.S. military assets with
permanent destruction. 反而, imagine (not very plausibly) that Japanese
dive bombers and torpedo planes were ªtted with special “delay bombs” that,
unlike delay fuses, would simply disable a ship for hours, 天, or possibly
weeks, rather than permanently, or at least for months or years. Faced with
this altered reality, Japanese ofªcials and military planners would have been
forced to contemplate a very different war, one that they would almost cer-
tainly have preferred not to initiate. 有效, Japan would have had to choose
to precipitate total war, as surprise attacks themselves would not do much to
diminish or delay a military response from the United States. The only value
one could anticipate from a surprise attack, 然后, would be if it facilitated a
follow-up invasion of the U.S. mainland. This is the basic shortcoming of
cyberwar. Because cyberattacks involve temporary “soft kills” of a target’s mil-
itary capabilities and civilian infrastructure, the point of the attack is largely
nulliªed if an attacker cannot reasonably be expected to accompany internet
aggression with terrestrial strikes designed to make permanent short-term
damage to a target’s security capabilities.

No foreign military force is capable of subduing the United States, now or in

54. Louis Morton writes, “Japan planned to ªght a war of limited objectives and, having gained
what it wanted, expected to negotiate for a favorable settlement.” Morton, “Japan’s Decision for
战争,” p. 110.
55. Admiral Yamamoto’s other famous quotation, which was meant as an ironic reference to total
war with the United States, 索赔, “To make victory certain, we would have to march into Wash-
ington and dictate the terms of peace in the White House.” The private letter from Yamamoto to
Ryoichi Sasakawa is quoted in Gordon W. Prange, At Dawn We Slept: The Untold Story of Pearl Har-
博尔 (纽约: 麦格劳-希尔, 1981), p. 11.
56. The invasion of the Aleutians was a feint to distract attention from the attack on Midway.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 63

the foreseeable future, with or without the assistance of a phenomenally suc-
cessful coordinated cyberattack. If cyberwar is unlikely to allow a foreign
power to permanently overtake U.S. or allied capabilities, and if temporary
damage is useful only when practiced in conjunction with more conventional
军事行动, then an opponent must plan and evaluate its use of
cyberwar in relation to its complementarity to terrestrial combat, not as a fully
independent method of force. If instead a cyberattack is carried out in which
conventional force is either ineffective or not contemplated, then an attack of
this kind fails to serve a meaningful grand strategic purpose, degrading nei-
ther the target’s longer-term capabilities nor its resolve.

Unless cyberwar can substitute for a physical surprise attack, there is no rea-
son to believe that it will be used in place of conventional modes of warfare.
Nor is it clear why an attacker would choose to strike over the internet, 除非
a conventional surprise attack is also planned and when it is expected that the
combination of cyber and terrestrial aggression will yield a decisive advantage
to the attacker. If it is difªcult to imagine a particular nation being attacked by
traditional methods of warfare, even with the beneªt of surprise, then it is
hard to see how that nation might be fundamentally threatened by warfare
conducted over the internet. 的确, the connection between internet ag-
gression and traditional forms of military force imply an unfashionable predic-
的: cyberwar should be particularly appealing to capable states confronting
weaker opponents. Rather than threatening to overturn the existing world or-
这, cyberwar may perpetuate or even increase current military inequality.

Additional Implications of Cyberwar

If cyberwar functions not as an independent domain, but as part of a broader,
coordinated military action, then the conventional military balance is the best
indicator of where the most important threats exist in cyberspace. 因此, 除非
someone believes that economically and militarily advanced nations are in
danger of physical attack from a foreign power, the threat of cyberattack can-
not be treated as particularly serious in political terms. Most experts view an
attack that subdues U.S. military capabilities and subjects the U.S. mainland to
a foreign power as remote, even fanciful, 例如. To the degree that pow-
erful states are immune to conventional attack, cyberattacks are at most a nui-
sance and not a fundamental threat.

讽刺地, the greatest threat posed by cyberwar may be to those states and
actors that are thought to be insulated, the same actors that are currently vul-
nerable to conventional terrestrial aggression. Cyberwar is not a revolution in

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 64

military affairs in strategic (军队) 条款, nor is cyberwar likely to prove
revolutionary in political terms, by threatening or transforming existing global
or regional power structures. 在这个意义上, cyberwar appears to be reactionary,
reinforcing the advantages of states that already possess signiªcant terrestrial
military advantages. The need to prosecute cyberattacks with more kinetic
forms of force, and the perishability of cybercapabilities in the face of rev-
elation, mean that nations with capable militaries are best equipped to exploit
the type of damage that cyberwar inºicts, even as they are better able to credi-
bly threaten cyberattacks and to “reveal and replace” a given capability to tar-
get an enemy’s cyber vulnerabilities. This new mode of warfare, most feared
by technologically advanced states, may pose greater grand-strategic chal-
lenges to the technologically backward or weak, something that has not been
considered previously.

This is not to say that the sophisticated computer-dependent nations of the
West are immune to attack. The United States and other advanced countries
are certainly vulnerable to internet aggression. 反而, the consequences
of this harm will prove ineffectual, because the ability to alter the balance of
power in regional or world affairs or to exploit capabilities is present only for
those nations and interests that already possess considerable international
inºuence. We would all like to live in a world where no one could harm us
even if someone fervently wished to do so. To the degree that this is not possi-
布莱, a useful second best is for others to possess disincentives to inºict injury or
死亡, even if they still can. An inability to exploit temporary opportunities
created by cyberwar is a disincentive to conduct concerted internet attacks. 在
对比, powerful nations continue to possess both the ability and interest
in intervening in the developing world, even if they do so only episodically.
Existing examples of cyberwar illustrate and reinforce this counterintuitive
结论. Attacks on Estonian websites, which appear to have originated in
俄罗斯, pitted a tiny nation against a considerable military and economic
力量. The ability of Russian leaders to exploit effects of the attack, not just on
the internet but through military and diplomatic pressure, ensured that the im-
pact was much more potent than if equivalent attacks were carried out against
Russia by nonaligned hackers. 相似地, the Stuxnet worm, which was appar-
ently designed and unleashed by the United States or Israel, was more ef-
fective because of the U.S. advantage in conventional and nuclear military
能力. The dynamic would have been far different if instead Iran had
been able to counter the Stuxnet attack with aggressive military action.57

57. For a discussion of the politics and technology behind Stuxnet, see Jon Lindsay, “Stuxnet and
the Limits of Cyber Warfare,” 安全研究, 卷. 22, 不. 3 (八月 2013), PP. 365–404.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 65

The Russian invasion of Georgia/South Ossetia is perhaps the clearest ex-
ample of the kind of combined terrestrial-cyber conºict conceived of here. 作为
one commentator described it, “This appears to be the ªrst case in history of
a coordinated cyberspace domain attack synchronized with major combat
actions in the other warªghting domains (consisting of Land, Air, Sea, 和
空间).”58 Because they allow for exploitation of the temporary effects of
surprise and asymmetry highlighted in other studies of internet warfare,
cyberattacks will be most efªcacious when the conventional military balance
already favors an attacker.59

An open question exists in any crisis about how far competitors are willing
to escalate, but an ability to counter cyberattack with other, more kinetic forms
of military violence serves alternately to deter or to facilitate the use of
cybercapabilities, giving those nations with terrestrial military power yet an-
other option that, even if available to their opponents, may prove extraordi-
narily dangerous to practice. As we see today with U.S. drone attacks and
Special Operations raids on foreign sovereign territory, the power to do much
more ensures that an opponent maintains a level of discretion in its response to
provocation. Few can doubt the reaction of the United States, 例如, 如果
Pakistan attempted to mount a commando raid in suburban Baltimore, 说, 到
assassinate a local resident. Nations that can physically punish others for
transgressions in any domain, electronic or otherwise, are better able to oper-
ate in all domains. Once one distinguishes between simple vulnerability and
actual threats, terrestrial capabilities become pivotal in determining who exer-
cises cyberwar.

Even if cyberattacks are available to weaker actors, their effectiveness will be
stymied where these actors lack the ability to prosecute advantages generated
by cyberwar, and where weakness in more traditional modes of diplomatic,
经济的, or military competition ensure that these actors are exposed to
countermeasures. The intractable nature of vulnerabilities ensures that cyber-
war will not fundamentally transform either war or world affairs. 尽管
a dependence on high technology, developed countries will ªnd that they
can better exercise cyberwar as a political tool. Attacks against prosperous
Western powers, if well publicized and the source of considerable anxiety, 将要
turn out to be less consequential. While other technological or social changes

58. 大卫·M. Hollis, “Cyberwar Case Study: Georgia 2008,” Small Wars Journal, 卷. 6 (一月
2011), p. 2.
59. For skepticism about the coercive effect of cyberpower, see John B. Sheldon, “Deciphering
Cyberpower: Strategic Purpose in Peace and War,” Strategic Studies Quarterly, 卷. 5, 不. 2 (Sum-
梅尔 2011), PP. 95–112. Sheldon’s perspective relies on induction from recent cases, rather than on a
theory of the role of cybewar as part of a modern integrated military.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 66

may well transform contemporary hierarchies, cyberwar will most likely func-
tion more to perpetuate than to undermine existing inequalities of inºuence.

the adjunct role of cyberwar

Because war on the internet is adjunct to more conventional forms of ªghting,
a cyberattack is extremely unlikely to prove pivotal in conºicts involving capa-
ble states or their partners. 仍然, cyberwar could be used by and against forces
in the ªeld, a valid and possibly important concern. A common approach to
evaluating the implications of new technologies for war and peace involves
the offense-defense balance. Proponents of offense-defense theory focus on
material or, alternately, on cognitive/informational factors that they believe
will lead to increased military aggression.60 Nations or time periods that expe-
rience or perceive offensive advantages will be associated with more war,
whereas the opposite is said to happen when innovations or circumstances fa-
vor the defense. There is considerable skepticism about the empirical validity
of offense-defense theory, as well as about the ability of researchers to isolate
the factors leading to offense or defense dominance.61 Even if there were noth-
ing controversial in the application of offense-defense theory, it would still be
challenging to draw conclusions about the impact of cyberwar on the general
appeal of warfare, given that cyberwar is new, and given that cybercapabilities
are not the only factors inºuencing the offense-defense balance. In evaluating
the impact of cyberwar, it would be valuable, 然而, to know whether the
internet systematically favors attackers or defenders.

Robert Jervis proposed a framework for understanding technology-induced
instability that seems well suited to explaining the externalities of cyberwar for

60. For perspectives and coverage of the debate concerning offense-defense theory, see George H.
Quester, Offense and Defense in the International System (纽约: John Wiley and Sons, 1977); 杰克
L. 斯奈德, The Ideology of the Offensive: Military Decision Making and the Disasters of 1914 (伊萨卡岛,
纽约: 康奈尔大学出版社, 1984); 特德·霍普夫, “Polarity, the Offense-Defense Balance, and War,”
American Political Science Review, 卷. 85, 不. 2 (六月 1991), PP. 475–494; Stephen Van Evera, “Of-
栅栏, Defense, and the Causes of War,” 国际安全, 卷. 22, 不. 4 (春天 1998), PP. 5–43;
Charles L. Glaser and Chaim Kaufmann, “What Is the Offense-Defense Balance and Can We Meas-
ure It?” 国际安全, 卷. 22, 不. 4 (春天 1998), PP. 44–82; and Karen Ruth Adams,
“Attack and Conquer?: International Anarchy and the Offense-Defense-Deterrence Balance,” 国际米兰-
national Security, 卷. 28, 不. 3 (冬天 2003), PP. 45–83.
61. For several different criticisms of offense-defense theory, see Jack S. 征收, “The Offense/
Defense Balance of Military Technology: A Theoretical and Historical Analysis,“ 国际的
Studies Quarterly, 卷. 28, 不. 2 (六月 1984), PP. 219–238; Jonathan Shimshoni, “Technology, 米利-
tary Advantage, and World War I: A Case for Military Entrepreneurship,” 国际安全,
卷. 15, 不. 3 (冬天 1991), PP. 187–215; and Yoav Gortzak, Yoram Z. Haftel, and Kevin Sweeney,
“Offense-Defense Theory: An Empirical Assessment,” 冲突解决杂志, 卷. 49, 不. 1
(二月 2005), PP. 67–89. For a rebuttal of some of these critiques, see Sean M. Lynn-Jones, “Of-
fense-Defense Theory and Its Critics,” 安全研究, 卷. 4, 不. 4 (夏天 1995), PP. 660–691.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 67

the broader question of conºict among states.62 As James Fearon makes clear,
然而, offense-defense theory also needs to distinguish between advantages
gleaned from initiating disputes and those from acting aggressively, 应该
war occur.63 Fearon further notes a more general tendency to confuse the shift-
ing offense-defense balance with changes in the balance of power. Offense
dominance implies that states are more likely to prefer attack rather than de-
fense regardless of the prevailing balance of forces. Weaker states seldom pre-
vail under any circumstances.

Imagine ªrst that cyberwar is defense dominant. This does not seem all that
likely, and in fact contradicts the prevailing view in the literature. 仍然, sup-
pose that information infrastructures are more readily defended than attacked.
In such as world, the balance of power would favor those states that could
most effectively orchestrate military command, 通讯, logistics, 和
intelligence through the internet or similar types of networks. Even if this
imagined cyber world is defense dominant, it does not follow that terrestrial
conºict is also defense dominant. The relative immunity of networks to attack
could lead to a reluctance to use conventional force, or it could increase incen-
tives to act aggressively, depending on whether secure networks are more
important for defenders or attackers. The standard military answer is that
command and control are more critical for the offense, as commanders have
more need to directly control their forces in the attack. 如果是这样, then perhaps
cyber defense dominance is actually destabilizing, because it increases the abil-
ity to attack and (轻微地) decreases the ability of defenders to prevail. 骗局-
versely, 如果, as many contend, cyberspace is offense dominant, then this should
tend to weaken offensive operations in the physical world, making terrestrial
conºict more defense dominant.64

This basic conception assumes dichotomous conditions of network vulnera-
bility that do not reºect actual wartime dynamism. It may take time to disable
网络. 如果是这样, then there is a ªrst-mover advantage that could prove more
critical than the defense dominance created by the heightened need among
attacking forces for effective command, 控制, 通讯, comput-
呃, and intelligence. Advantages may follow from starting early in cyber-
战争. Striking ªrst could also prove valuable if disabling an opponent’s
internet reduces the enemy’s ability to retaliate. Given that a state projecting

62. Jervis, “Cooperation under the Security Dilemma.”
63. 詹姆斯·D. Fearon, “The Offense-Defense Balance and War since 1945,” typescript, 大学
芝加哥.
64. I am grateful to Eugene Gholz for raising the issue of offense/defense cyber dominance.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 68

power beneªts from being able to choose the time and place of its attack, 这
ªrst-strike advantage might be more important for victory than the pacifying
effect produced by offense dominance in cyberspace.

Regardless of whether the internet increases or decreases incentives to at-
tack, cyberwar is likely to continue to favor the strong against the weak. 这是
not to say that cyberattacks will not have an effect, only that they are extremely
unlikely to be strategically decisive. A capability to address cyber threats is
有用, but planning for cyberwarfare must occur within the larger frame-
work of recognition that this new domain is evolutionary rather than revolu-
解说性的. There will not be a cyber Pearl Harbor, except possibly when and if a
foreign power has decided it can stand toe-to-toe with conventional U.S. 米利-
tary power.

cyberterrorism

The terrorist attacks of September 11, 2001, generated a sense of insecurity in
the Western world: How can governments protect their citizens in an age
where the enemy is concealed and where an attack may come at any time or
地方? The temptation has been both to treat terrorism as an existential threat
(because it is frightening) and to assume that the best response is a vigorous
防御. 然而, as I have argued, one of the most effective mechanisms of protec-
tion is not to remove capabilities, but to puncture resolve, foremost by ensur-
ing adversaries that their objectives will not be realized. A vault does less to
deter bank robbers than the presence of countermeasures (die packs, 编号-
bered bills) that deny criminals the fruit of their plunder, even when they suc-
cessfully complete the crime. Terrorism is a marginal business, 不是因为
airports and diplomats are too well protected or because guns or bombs are
hard to come by, but because most people, even if very unhappy, do not be-
lieve that bombings, hijackings, or assassinations will effect positive change.
Incapable of achieving their objectives directly, terrorists seek to mobilize fear
and to cause overreaction, mechanisms that mobilize the targets of terror to as-
sist in accomplishing terrorist objectives.

The resort to cyberwar by terrorists does not imply that cyberterrorism is an
important threat to national security, any more than the appeal of the lottery to
the poor or ªnancially desperate implies that the odds of winning are in-
versely tied to one’s income. Desperation leads to desperate measures, an indi-
cation in fact that such measures probably will not succeed. 在这个意义上, 这
rise of cyberterrorism may say more about rigidity and impotence between
agent and structure than about either in isolation. Cyberterrorism may be rela-
tively ineffective, not unlike terrorism generally. 尽管如此, terrorists may

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 69

adopt cyberwar even though internet attacks are unlikely to sway national
policies or public opinion. The adoption of a particular method of attack by
terrorists does not mean that their actions represent a critical threat, any more
than crime and corruption are considered issues of national security. Most so-
cieties treat the latter activities as distinct from national security, 不是因为
they are not important or fail to harm people, but because they do not directly
threaten the state. Unless attackers have the ability to prosecute temporary ad-
vantages through physical force, it is not clear whether cyberterrorism requires
an elaborate or concerted national security response.

Terrorism is a form of compellence. Lacking the ability to impose their will
on others, terrorists rely on the prospect of harm to inºuence a target’s be-
行为. 的确, because their ability to harm is limited, the terrorist relies on
心理学 (fear and uncertainty) to multiply the impact of relatively ªnite
capabilities on opposing populations or states. Cyberwar is arguably espe-
cially poorly suited to the task of fomenting terror. 尤其, 此外
the problems in credibly threatening cyberattacks that have already been dis-
cussed, it is difªcult to see how internet attacks will be able to instill the
quality of fear needed to magnify the terrorist’s actions. How terrifying is a
cyberattack? No one will be happy when the power goes out or when one’s
bank account is locked down, but attacks of this type evoke feelings of anger,
frustration, even resignation, not terror. Terrorism relies on generating a partic-
ularly visceral emotion (the “terror” in terrorist), one not likely to be effected
through the actions of cyberwarriors, at least not directly. The old journalistic
adage that “if it bleeds, it leads” implies the need for graphic trauma and lurid
imagery. The very attributes that make cyberwar appealing in the abstract—
the sanitary nature of interaction, the lack of exposure to direct harm, 和
strikes from remote locations—all conspire to make cyberterrorism less than
terrifying. White collar terrorists are probably not going to prove any more ef-
有感染力的, and perhaps may prove less, at shaping hearts and minds than the
traditional model.

This is even more the case with long-duration, low-intensity conºicts that
are a key component of both non-Western attempts at resistance and Western
efforts to protect the status quo international order. From the perspective of
the insurgent, asymmetric warfare has never been about attacking to diminish
an opponent’s strengths, but is instead focused on maximizing one’s own
strengths by targeting enemy weaknesses.65 Insurgency seeks out kinetic close

65. Mao, Zedong, On Guerrilla Warfare (纽约: 普雷格, 1961).

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 70

physical combat where sophisticated technology is at its least effective (和
decisive). Damaging the technology may draw an enemy into direct contact,
but it might also cause that enemy to withdraw and reschedule operations.
Mobility dominates every battleªeld for this very reason. Internet attacks in
the midst of close contact make little sense, as it is here that the comparative
advantage of cyberwar (distance and asymmetry) is least potent. The ability of
internet-dependent armies to perform in superior ways on existing dimen-
sions means that this is generally a process of leveling, not revolution.

cyber espionage

By far the most compelling scenario for the transformation of political conºict
through the internet, and the one that makes new headlines daily, involves the
use of the internet for espionage. As events such as the disclosure of an enor-
mous number of classiªed documents through WikiLeaks illustrates, it may
become increasingly difªcult for states to hide details of their capabilities and
plans from individuals, 团体, and other nations.

States have always sought information about prospective opponents. Suc-
cessful espionage creates signiªcant advantages, but also challenges. For most
of history, spying was physical. An agent had to enter enemy territory to ob-
tain information about the capabilities or intentions of a foreign group or
力量. The products of espionage were equally tangible. Spies brought back
文件, captives, tallies, or other materials designed to inform their mas-
ters and demonstrate the veracity of their claims. This made spying risky. Espi-
onage required an overt act that could itself precipitate war. Evidence of spying
could form the casus belli for an attack by a target of espionage against the per-
petrator. Even if it did not lead directly to a contest, where agents were found,
and what they were looking for, revealed sensitive information. For these rea-
儿子们, counterespionage is itself as much about spying as it is about preventing
espionage. 当然, captured spies also fared poorly, as international norms
offer none of the protections afforded to conventional combatants.

The internet makes it possible for the spy to telecommute.66 Information can
be collected without leaving the territory of the sponsoring state, making it
difªcult to deter or capture cyber spies. A spy’s afªliation can also be con-
cealed, so that the target is uncertain whether espionage indicates a prelude to
战争, threats from speciªc states, speciªc formal national objectives, 甚至
whether espionage actually occurred. 同时, cyber spies face their
own challenges. 的确, the detection of cyber spies may be easier than con-

66. Telecommuting and the paperless ofªce are two other myths of the internet age. 有可能的
that telecommuting may be no more successful for spies than for anyone else.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 71

ventional espionage, given computer forensics and the trail left when access-
ing ªles.

One of the perennial challenges for political decisionmakers in dealing with
any form of espionage is what to do with the information collected. Acting on
covert knowledge is tempting, but often this will also tip off the target and
lead to countermeasures. Even more fundamentally, the challenge to analysts
is to interpret the signiªcance of information, not something that the internet
makes easier, particularly given the quantity of materials likely to be involved.
Critical facts may even be obscured among masses of trivial details, 保护
not unlike the anonymity provided by mass humanity that shelters most of us,
something that is only possible given the mountains of information in the
internet age.

反过来, the single most dramatic impact of the cyber world on political
conºict may well come in the form of transparency. Nations may ªnd that
they can no longer sustain jealously guarded secrets. The phenomenon of
classiªcation that led large portions of government activity underground may
ªnd itself “outed,” not by alien spies or terrorists, but by groups devoted to the
idea that airing national secrets makes it more difªcult for governments to
connive against one another, or to scheme against their own people.

There is considerable reason to believe that reducing secrecy will led to a de-
cline in warfare, even if some nations are made worse off in terms of relative
力量. The conceit of sovereigns is that secrets can be kept indeªnitely. 这
has never been true. Code breaking in both world wars exposed German and
Japanese operational plans to exploitation by the Allied commanders. 相似的
results may have shadowed opponents during the Cold War. The problem of
course was that espionage that unearthed enemy secrets also had to be kept se-
卷曲, because there were advantages to knowing something that an enemy did
not know that the opponent knew. Espionage did not reduce the prospect of
war as much as it affected who was likely to win, because those in the know
were able to exploit relationships and win contests. 今天, nations may ªnd
it increasingly difªcult to imagine that their secrets are safe, even if they are.
In addition to “nonproªt” espionage that publicly discloses state secrets,
the ubiquity and success of “for proªt” spying in the internet age must make
nations consider the possibility that security is inherently porous. Secrets
are not sacrosanct, making it more difªcult to carry out the conspiracies so
often associated with coercive politics. War will still occur, but surprise in
war will be increasingly difªcult to achieve, in turn reducing at least part
of the motivation behind the use of coercive military force. With fewer sur-
采取, identifying bargains that competitors prefer to the costly exercise of
war becomes easier.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

国际安全 38:2 72

结论

In war, tactics must serve strategy and strategy must serve grand strategy. Stu-
dents of cyberwar have yet to explain how the internet can host meaningful
political conºict, precisely because it cannot serve the ªnal arbiter function
that has for millennia been the purview of physical violence. The tendency
among pundits of cyberwar has been to focus on tactics and possibly strategy,
showing that harm is possible without explaining how the harm generated is
likely to shape the product of political differences. In the absence of this logic
of consequences, the internet becomes an adjunct domain to more traditional
forms of warfare. Cyberwar is an evolving dimension of war and a source of
concern, but in grand strategic terms, it remains a backwater. A failure to focus
on grand strategy is an all-too-familiar by-product of the war on terror, 在哪里
the objective has been to harm and not be harmed, rather than to effect mean-
ingful changes to the disposition of world affairs.

It would be absurd to infer that there is no role for the internet in twenty-
ªrst-century conºict. The internet will be affected by conºict, just as is the case
with every other domain in which individuals, 团体, and societies interact.
的确, the real message for soldiers and politicians is that cyberwar involves
a broadening of the dimensions of warfare, rather than a narrowing of future
conºict. 在多数情况下, the internet is not a viable, free-standing venue for the
pursuit of national interests. It would be surprising if a country intent on at-
tacking another nation failed to carry out preparatory or simultaneous attacks
of target’s defense capabilities via the internet. It would be even more surpris-
ing if an aggressor successfully substituted cyberwar for conventional, tangi-
ble forms of conºict. This is the conceit of Nikolai Kuryanovich, 以前的
member of the Russian Duma: “In the very near future, many conºicts will not
take place on the open ªeld of battle, but rather in spaces on the Internet,
fought with the aid of information soldiers.”67 Mr. Kuryanovich continues,
“[A] small force of hackers is stronger than the multi-thousand force of the cur-
rent armed forces.”68 Surely, a country with thousands of soldiers and hun-
dreds of hackers would be inclined to use both.

By itself cyberwar can achieve neither conquest nor, 大多数情况下, 强迫.
Russian military planners obviously understood this in preparing to invade

67. Cited by Stephen W. Korns and Joshua E. Kastenberg, “Georgia’s Cyber Left Hook,” Parame-
特尔斯, 卷. 38, 不. 4 (四月 2009), p. 60. Korns and Kastenberg reference Brian Krebs, “Report: Rus-
sian Hacker Forums Fueled Georgia Cyber Attacks,” Washington Post, 十二月 16, 2008, http://
voices.washingtonpost.com/securityªx/2008/10/report_russian_hacker_forums_f.html.
68. 同上.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3

The Myth of Cyberwar 73

乔治亚州, not just with an army of hackers, but with tanks. 的确, the tanks
appear to have done more for Georgian insecurity than anything accom-
plished by information soldiers. The threat of cyberwar cannot deter or compel
particularly effectively either, except possibly in the short term, and only with
the consequence that an attacker will have forfeited the potential to exploit a
given set of vulnerabilities in the future. Cyberwarfare will most often occur as
an adjunct to conventional warfare, or as a stop-gap and largely symbolic ef-
fort to express dissatisfaction with a foreign opponent. It is best to discuss
cyberwar in these contexts, not as an independent, or even alternative form of
conºict, but as an extension of the logic already expressed in combined arms
battle. Because in most cases cyberwar cannot achieve the objectives that have
historically prompted nations to commit to tangible military violence, “cyber-
war” is only warfare in the context of terrestrial forms of interstate threats
or force.

Even the most successful forms of cyberwar (such as cyber espionage) 做
not presage much of a transformation. Just as innovations in artillery and
small arms made ªghting in close formation untenable, militaries, 治理-
评论, and societies will adapt. It would be ludicrous to suggest to modern
infantries that the massing of ªres would be best achieved if they stayed in for-
mation while on the march or in the assault. Contemporary ªeld commanders
have become comfortable with the idea that perimeters are partial or tempo-
rary, that air-land battle (and naval warfare for a longer time) necessarily in-
volves not fronts, but mobility; not frontal attack, but maneuver. 相似的
concepts will pervade discussions of cyberwar. Static security is insecurity. 它
does not follow, 然而, that being vulnerable means one will be attacked, 或者
that there is much that can be done to prevent aggression if it is initiated. Secu-
rity in the modern world, terrestrial and cyber, is a function more of the mo-
tives of opponents than of the ability to attack. Nations or groups that strike
through the internet in minor ways may be ubiquitous. Those that threaten
critical national security goals will be rare if for no other reason than that
cyberwar is not really war in grand strategic terms. 在这方面, the next
Pearl Harbor is much more likely to occur in Hawaii than in cyberspace.

我

D
哦
w
n
哦
A
d
e
d

F
r
哦
米
H

t
t

p

:
/
/

d
我
r
e
C
t
.

米

我
t
.

e
d
你

/
我
s
e
C
/
A
r
t
我
C
e
–
p
d

我

F
/

/

/

/

3
8
2
4
1
1
8
4
4
1
5
6

/
我
s
e
C
_
A
_
0
0
1
3
6
p
d

.

F

乙
y
G
你
e
s
t

t

哦
n
0
8
S
e
p
e
米
乙
e
r
2
0
2
3
下载pdf