(DRAFT: First Prepared, Summer, 1996, but still under revision).


Richard J. Harknett
Department of Political Science, University of Cincinnati

The United States Department of Defense's definition of information warfare is remarkable for its lack of distinctiveness. Information Warfare is described as "action taken to achieve information superiority in support of national military strategy by affecting an adversary's information and information systems, while leveraging and protecting our own information and information systems."1

This phrasing is amorphous and loose enough that the dynamic and strategic vision it describes are ones with which Sun Tzu or Attila the Hun could feel comfortable.2 While the assumption is that we are talking about the employment of high technology, the DOD definition does not require it. Information warfare is an intriguing concept, yet it is only useful either as a basis for strategy or as an analytical construct, if it captures dynamics that depart in significant ways from other forms of war.

This paper attempts to bring some further definitional rigor to the use of the term by examining the unique features of the Information Age itself, assessing how these characteristics may affect both war and deterrence of war. The primary objective of this paper is to discuss the dynamics associated with the pursuit of a strategy of infowar deterrence. Are they similar or distinct from nuclear or conventional deterrence? Before discussing deterrence of information warfare, however, we must have a clear understanding of this new form of war. I present the case that the Information Age is best understood in the context of a shift in forms of organization and, that reflective of this change, the term information warfare can be sharpened by focusing on the concept of connectivity as a national asset.

The paper is divided into three major sections. First, since we have yet to experience information warfare as it is defined in this paper, I deduce the central implications for future military conflict through examination of five general characteristics associated with the emergence of information technology. Two important consequences of the "Information Age" are noted: the empowerment of individuals and the emergence of a new form of organization--the information technology network. Second, I employ these observations to refine two concepts found in the current literature on information warfare--Cyberwar and Netwar. Finally, I examine these two forms of information warfare in the context of a deterrence model. Two conclusions are offered: first, that deterrence of net- and cyber-war will be extremely difficult and second, that deterrence models should not take the lead in guiding strategic thought about either new form. Military and societal connectivity are inherently contestable and thus difficult to protect through a strategy of deterrence.

The Information Age

It has become quite commonplace in the mid-1990s to speak of living in "the information age." The designation of a particular time of human history as an identifiable "age" is meant to highlight its distinctiveness from the past. It tends to signal that a transformation has taken place in the way human activity is conducted and/or defined. One of the central human activities that tends to be greatly affected when such transitions occur is warfare. In fact, the movement from one "age" to another has tended to be intertwined with developments in the ability to conduct warfare. In other words, transformations in the conduct of war have tended to bring about societal change and societal change has reinforced corrections on the battlefield.3 The dynamic force behind such transformations has, at many times, been advancements in technology. 4 Thus, it is not surprising that the growing ubiquity of personal computers and other information technologies is viewed not only as the basis for a new societal age but as the foundation for a new form of warfare as well.

Before discussing the impact information warfare may have on the ability to pursue strategies of deterrence, however, a clear understanding of the broader transformation must be developed. Analysis requires a foundation of definitional precision. Not unexpectedly, as with most new concepts, definitional rigor and consistency is currently lacking in the growing literature concerned with this new term.5 It is important, therefore, to denote first the broader context in which information warfare is set before discussing its specific dynamics.

The distinction of the late Twentieth Century as "the" Information Age (has there ever been "an" age without information?) rests specifically on the recognition that the micro-processing silicon chip has created a new level of combined computational and communicative power in the form of the networked computer. While individual computers in isolation represent powerful tools, it has been the connecting of these devices that has proven to be a truly significant advance. Computer networks support everything from local, regional, and national banking systems to telephones and transportation structures. Information age technology also includes fax machines, cellular phones and satellite television. While these technologies have important consequences, the progression of the Information Age will be dictated primarily by the growing significance and ubiquitousness of the networked computer. This trend is captured most vividly by the explosion of interest and usage of the Internet--the network of computer networks. Initially conceived of as part of an American defense plan to improve communications during a nuclear attack, the Internet has transformed computer usage.6 While the creation, accumulation, and manipulation of information has always been a central part of human activity (warfare in particular),7 the computational and communicative power of the networked computer has produced some distinctive consequences. By identifying these consequences we can analyze the potential changes in human practices that may emerge. There seems to be at least five inescapable features that can be ascribed to the Information Age. They revolve around the ideas of accessibility, availability, speed, affordability, and recursive simplicity.

Accessibility.  The networking of personal computers has led in turn to the networking of individual networks. The universe of these networks, what is now commonly referred to as cyberspace, carries information in all of its forms.8 This high (and ever growing) degree of connectivity has decreased tremendously the obstacles to information retrieval. Through the use of modems, individuals sitting at home can now connect into cyberspace to access depositories of information world-wide. To a certain degree, access to information in the past has been geographically dependent. The purpose of creating collective depositories of information, such as libraries and archives, was to enhance common access to public information. One's ability to access these depositories depended on being able to physically travel to the collection. While, there has existed the option to receive some of this information by the mail, such a process typically has required requests that are both narrow and specific. This is difficult to achieve unless one has come into contact with related resources, which likely only occurs through direct physical contact with the related material. A tremendous advance in information access is created when retrieval of information is not primarily dependent on being in geographic proximity to the information storage facility itself.9

Availability.  The act of housing information in central depositories to increase access to it has a number of important consequences. A system of information retrieval that depends on geographic proximity for access requires an enormous amount of duplication. Since many people may wish access to particular information, it must be distributed to all of the local depositories found in the system. If the information is not physically on-site it is not available for retrieval. 10 In this system, availability depends on how many copies can be made and stored in relation to how many information retrievers are at work. The difference between access and availability is important. Living near a library might mean one has access to a book, but if someone has already checked out the only copy, the book will not be available for some specified time. Cyberspace offers a significant advance in availability by creating the opportunity for multiple simultaneous retrieval of information. The limitation on availability is not dependent on how many copies of a particular instrument exists, but on how many users can access the particular site. This should prove a less vexing and less expensive problem. Increasing a computer network's ability to handle more users on-line will prove to be more cost effective than having to duplicate the actual source of information by the same number of users.

Aside from multiple simultaneous information retrieval, availability in the Information Age can also be discussed in sheer quantitative terms. By being less geographically dependent for access, one can simply have available much more information than in the past. The existence of cyberspace allows the retrieval of information that was physically unavailable because of geographic distance. In fact, one of the growing problems with the Information Age is how to manage too much information.

Speed.  A third feature of the Information Age is the enormous increase in computational and communicative speed. What is distinctive about this age is that vast sums of accessible information can be disseminated in seconds. The computational speed of computers has grown exponentially and so has network connectivity. When the ARPANET was first developed, modem speed was about one-tenth the speed of the slowest modems in the late 1980s.11 Other modes of information transmission, such as fax machines, cellular phones, and satellite television, now enable information-seekers to have up-to-the-minute or even real-time knowledge of events. The implications for the decision-making process of individuals and organizations are significant. The time it takes to collect initial information, analyze and process that information, make a decision and implement it has been tremendously compressed. The speed associated with information age technology means that more information can be handled in a faster manner. Previously, to speed up a decision-making cycle, information retrieval and analysis had to be shortened (since they were the most time-consuming). Information Age technology, while not eliminating this problem, makes it much less vexing. A large amount of information from multiple sources can now be retrieved without putting undo time pressures on the overall decision-making cycle.

Affordability.  The fourth distinctive feature of the Information Age is that the resource base required to exploit the advantages of this technology is relatively low and is decreasing rapidly. The average price for computing power has been decreasing. Graphic-based point-and-click software has now become standard with computers. This requires the user to know little more than what certain functions obvious icons (pictures) represent. Use and programming of VCRs now requires people to push a few buttons and speak in a clear voice to a voice-activated machine. Fax machines, cellular phones, VCRs, and even computers are following the same pattern of the digital hand-held calculator, which was a luxury item in the early 1970s and now are owned by elementary school students. For the median average income family, the cost of computers is no longer prohibitive.12 This is, of course, not to imply that there exist no barriers either financial or skill-based, but to suggest that such barriers are becoming less and less significant for general access. Information Age technology is affordable in the broad sense of that term. It is requiring the investment of less time for training and money.

Recursive simplicity.   The final general feature of the information age is the inherent recursive nature of computer technology, which supports a clear trend of ever expanding growth in access, availability, and speed with a simultaneous reduction in cost and skill barriers. Recursion can be identified as self-similarity in structure, where symmetry runs across scale; in essence, pattern inside pattern.13 Put another way, a recursive structure is one in which the whole is structurally identical to its parts.14 While recursive structures exist in nature (snowflakes and ferns show this quality), they were originally discovered as mathematically-based constructions. By identifying the principle of self-similarity across scale, mathematicians Helge von Koch and Benoit Mandelbrot, showed that fractal patterns could lead to infinite length in a finite space.15 These observations laid the foundation upon which early developers of computer software constructed large programs out of existing smaller ones. The recursiveness of software tremendously simplifies its development. Instead of thinking of each independent part of a design separately, base commands can be re-used in similar but broader commands again and again. Thus, one need only know a fraction of the levels (essentially its foundation) to understand and construct multi-level designs. The application of pattern within pattern as a structuring principle allows for exponential growth factors. The microchip has been recursively constructed. In a broad sense, the faster chip is simply more chips pressed into one central component. The ability (through greater miniaturization) to construct a chip simply by compressing more chips into the same space means that the basic design of the chip is not radically altered from its slower antecedent and that no new design is necessary every time one seeks greater computational speed. Because there is recursive simplicity to both computer software and hardware the trend toward greater speed, access, and availability can be predicted. In the mid-1990s, for example, the World Wide Web portion of the Internet had nearly 22 million pages of content and the Internet, itself, was doubling in size every three months.16 As with Koch's snowflake, the growth in these features may be infinite even though the principle device we use--the computer--represents a finite space.


These five general characteristics of the information age certainly promise to affect human relations in significant ways. There is a growing literature, particularly associated with military consequences, that argues that the information age amounts to revolutionary change. It is not the objective of this paper to discuss in detail the revolutions in military affairs (RMA) debate.17 However, there are two related implications that can be extrapolated from the five general characteristics discussed above that do portend revolutionary effects.18

Individual Empowerment.   Taken as a whole, having quick access to enormous amounts of available information at a reasonable cost has the potential to empower individuals.19 Current and emerging information technology provides the single person the ability to collect, organize, analyze, and disseminate the same amount of information that only forty years ago required a team of technicians and researchers.20 The computational power that is now found in the small personal computer sitting on the desktop (or lap) of one individual could be found only in computers requiring large institutional support (university and government mainframes) 25 years ago. This impact is noteworthy because it stands in direct contrast to the Orwellian vision of the relationship between technology and the individual.21 In George Orwell's 1984, advancements in computational and communicative power enhanced the power of the state over the individual creating the "Big Brother" effect.22 There is no doubt that current technological advancement can increase surveillance capability and could support (to a degree) the ability of the authoritarian state to control its populace. However, the flow of events in Eastern Europe and the former Soviet Union at the end of the 1980s revealed that control of information resources was beyond the power of the communist state. In an ironic twist, it was an information-supported domino effect that led in quick succession to the collapse of seven East European regimes and the Soviet Union itself.

While satellite television, radio, fax, and telephony can be used with effect to undermine authoritarian structures, they also remain important sources of centralized information control.23 The networked computer, particularly as a communicative tool, however, is the information age platform that to date seems most supportive of individual, rather than state empowerment. Part of this simply may be related to how the Internet and cyberspace have developed. Both are direct relatives of the United States Department of Defense plan in the 1960s to develop a communication system that could survive a nuclear attack from the Soviet Union. It was determined that the best communication system would be a distributed network of computers. This network would have built-in redundancy so that the loss of links would not disrupt the entire system. The network, therefore, would have increased survivability without a central control, relying on information traveling across a series of nodes, which would contain routing information and thus be able to reconfigure information if some node or link was lost. This nodal structure would help guarantee that information could travel from its source to its anticipated destination even if a significant part of the communication system had been destroyed.24

This approach eventually served as the foundation for a network of university and government small computer networks created by the Advanced Projects Research Agency of the U.S. Defense Department. This network of networks, known as the ARPANET, consisted of four nodes in 1969, thirty-seven in 1973 and 60,000 in 1989, when it was disbanded in favor of the Internet.25 What is important here is that defense requirements of survivability led to the adoption of a system that regarded the lack of centralized control as a structural prerequisite. In its early development, of course, the ARPANET relied on technology that required significant resources and technical skill to manage. Thus, institutional control was to a certain degree exerted by control over support for the network (rather than pure control of the network itself). The structural innovation of creating a system without a center aided in the development of practices that even the original designers did not highlight. For example, electronic mail, which is the one of the primary functions served by the Internet, was not listed among the original goals of the ARPANET.26 Innovations, like the graphic-interfaced World Wide Web, were hastened by the increasing ease and ability of individuals to acquire the technology and access the network.

The empowerment of individuals gained through information technology spans across political, economic, and social realms. Attention is now being directed toward assessing the impact the Information Age will have on the relationship between individuals and their governments, particularly in modern democracies, how business practices are shifting, and how "virtual reality" will redefine human to human contact.27

Revolution in organization.  While individual empowerment is important, revolutions are not made in isolation. Broader consequences, while related to and affected by individual empowerment, will appear to the degree to which the five characteristics of access, availability, speed, affordability, and recursive simplicity bring about organizational re-structuring. The true revolution afoot is organizational in character. The key question is whether the network will become the superior form of organization for the information age. These five characteristics seem to hold such potential.
German sociologist Max Weber argued that organizational form can both define and be defined by the political, economic and social structures of society.28 He felt that modern society, with its money economy as a foundation, was becoming increasingly affected by the emergence of bureaucracy as the dominant organizational structure. Weber argued that the advance of the bureaucratic form was tied to its superiority as a organizational type in meeting the demands of political, economic and social activities that required administration to be "discharged precisely, unambiguously, continuously and with as much speed as possible."29 The ideal Weberian bureaucracy has five distinguishing characteristics: a focus on position rather than personality; hierarchy of rational-legal authority; written system of rules; division of labor (specialization); and, a career system tied to the organizational chart. The visualization of bureaucracy is the steeped-pyramid shaped "org chart," where power flows from the pinnacle down to the base. Weber asserted that bureaucracy profoundly affects social dynamics. Loyalty is directed toward the position rather than the occupant of the position. The goal is to rationalize the processing of information and decision flows. The consequence of this structure, according to Weber, is the creation of efficiency and the promotion of de-personalization. Although he viewed bureaucracy as the superior form, Weber looked bleakly on its effect on individuals. He saw this organizational form spawning a "narrowed professional, publicly certified and examined, and ready for tenure and career," who valued systemic consistency and coherence over the "magical elements of thought."30

Weber, of course, did not argue that bureaucracy was the ideal form of organization, but that it was relatively superior to other forms, particularly those associated with feudal, patrimonial, and early plutocratic administrations.31 One of the significant managerial dilemmas created by vesting authority in hierarchically demarcated functions is that some efficiencies do exist in de-centralizing decision-making power, particularly for implementation purposes. Weberian bureaucracy, at its core, rests on the principle of hierarchical control and thus is not well suited to grappling with situations that require dispersion of authority.

The progress of the Industrial Age (presumably the "age" from which we are now transitioning), according to Weber, was supported by the routinization and rationalization created by bureaucratization. The market capitalist economy, in particular, thrives on business being conducted through reference to calculable rules and 'without reference to persons.' This is the goal of bureaucracy. But if we are moving toward a new "age" in which social, economic, and political activities are dramatically altered (ie., a revolution) might not a new organizational model begin to dominate. In line with Weberian logic, this organizational model will both produce and be produced by alterations in human practices across the spectrum of societal structures. The network stands in sharp contrast to bureaucracy and may represent the first effective challenger to this now "traditional" form.

The information technology network, as organizational form, is distinguished structurally by its lack of center (or put another way, the center is not central) and its reliance on high technology. The network is nodal in structure, flat in visualization and essentially anarchical. In all, its characteristics might be counterpoised with those underlying Weberian bureaucracy as follows:1


Information technology network

positional impersonality

re-emerging personality (cloaked)32

hierarchy of rational-legal authority

modified anarchy of technical skill-based authority

written code of rules

norm-based action; reciprocity

division of labor


organizational chart advancement

advancement through individual distinction

The information technology network differs significantly from bureaucracy in organizational form. International relations theory defines an anarchical structure as one lacking centralized authority over the basic units of the system. The information technology network, in its ideal form, approximates a modified anarchical structure33; that is, linkages between nodes may emanate from some center, but the functioning of the information technology network is not dependent on that central node.

The original thinking behind the development of the communications network that led to the Internet emphasized the need for a structure without center. From a security standpoint, a hierarchically structured system offers the enemy a prized target--the centralized authority. Whether it is a communication system or a political structure, elimination of centralized authority can bring about the collapse of hierarchically-based systems. During the Cold War, this was the logic behind first-strike nuclear attacks that sought to "de-capitate" leadership either by the direct targeting of leaders or their means to communicate with their defense/retaliatory forces. It was determined that one could reduce the vulnerability of the communication system by replacing the "Achilles heel" of a centralized command center with, paradoxically, the absence of a center.34 The information technology network as an abstract organization structure parallels the form envisioned in that original Defense Department program. The Internet is currently the best example of this form. While it still relies on infrastructural support from government and large institutions (like universities and research centers) collapse of one node or set of nodes does not dramatically affect functions and processes across the entire information technology network.35

In anarchical structures, authority is derived from the relative differences in power possessed by a system's units. Realist international relations theory holds that the distribution of territorial state power across the system impacts the functioning of the system. Similarly, in the information technology network, authority is not vested in hierarchically demarcated positions, but rather is derived from differences in the distribution of power. Power in the information technology network resides with those individuals that have greater technical skill, those who know how to manage and manipulate the technology best to achieve specified goals in the most efficient and effective manner. In a recent study, social science survey researchers attempting to assess the development of civic life in cyberspace ran afoul of a number of Internet users, who felt that the posting of a research questionnaire was inappropriate. Because of their technical skill, these individuals were able to regulate the communication traffic and suppress the survey, essentially deleting it from many of the usernet groups to which it was forwarded.36 This authority derived not from any formally granted position as regulators, but from their individual technical capacity. In the information technology network, functional-based positional authority is inconsequential (ideally it is nonexistent). The anarchical nature of the information technology network creates a structural-imperative to focus on technical skill. While a minimal level of skill is required for entry into the information technology network (on the Internet it is quite minimal for basic functions like e-mail communication), continued access and advancement require personal initiative in maintaining and enhancing one's own level of technical skill. The nature of the technology, noted earlier, supports a trend of continual change.
Survival and achievement in such an environment require consistent personal effort. Centralized functional controls will be taxed to keep up.
The information technology network is supported through norm-based regulation that rests upon the principle of reciprocity, rather than a formal legal code of rule. This is necessitated by two factors. First, the absence of centrally-designated authority inhibits the creation of law that is deemed legitimate by all units. Without a formal hierarchy of authority, the mechanism for establishing the regulation of the information technology network amounts to nothing more than what is ultimately most efficient. One begins to adopt or accept certain norms of behavior because general consensus clusters around more efficient practices and avoids practices proven to be inefficient. The process is self-reinforcing as individuals reciprocate actions taken by others with cooperation based on the expectation that they will receive cooperation in return.37 Second, since activity in the information technology network is dependent on personal skills (which are in flux) rather than bureaucratically designated job descriptions, regulation must be flexible. Since the nature of information technology--its recursive simplicity--encourages rapid development, formal regulation will be constantly out of sync with the actual processes and dynamics of the information technology network. Informal norm-based regulation is much more adaptive.38

The emphasis on individual technical skill in the context of a flat organizational chart promotes a fourth distinction between bureaucracy and the information technology network. To the degree that formal positions promote a division of labor and specialization, information technology networks promote "multi-tasking." For example, in the realm of information publication, the traditional division of labor between writer, editor, mass publisher is now combined in "desk-top" publishing. While the information technology network does not eliminate the need for specialized expertise, it does formally organize itself to allow for individuals to function across specialties and efficiently produce the high quality once only ascribed to bureaucracies.

These characteristics add up to a most significant consequence. The information technology network can provide individuals with access to shared global and local situational awareness. In Weberian terms, one of the great advantages of the bureaucratic structure is the management of large enterprises through steeped supervisory power. In a bureaucracy, centralized authority provides strategic direction through its grasp of the whole; that is, its global situational awareness. Operational and tactical direction is funneled through middle layers of authority and base production is conducted at the specialized local level, which only needs to be aware of its particular functions. The amount of time and effort required to provide everyone at the local level with strategic vision and the central authority with specialized local knowledge according to bureaucratic proponents made such a management approach unnecessarily inefficient. The bureaucratic structure, thus, serves as the means to disaggregate and disseminate knowledge. Information is distributed in parallel to bureaucratic functions. While this approach was a gain in efficiency over feudal or patrimonial organizations it does have its limitations. The bureaucratic reliance on standard operating procedures and the problem of centralized decisions/decentralized implementation ("the left hand does not know what the right hand is doing") are the most commonly noted problems with bureaucratic management.

The information technology network exploits the availability, accessibility, affordability and speed of modern information technology and organizes knowledge in a radically different fashion. In parallel with its flat structure, all individuals can effectively have their local situational awareness informed by a sense of the whole. It is technically possible to have the necessary amount of information traverse swiftly enough across a network in an organized fashion as to make shared situational awareness an efficient mode of operation.

The emergence of this shared situational awareness phenomenon is actually the key manifestation of a organizational revolution that could justify the designation of this period of human history as a definable "age." If human practices are going to change across the realms of economics, politics, and social activity, the dominant bureaucratic structure will have to be transformed or displaced. There are signs that this is taking place.

In the realm of economics, business practices have begun to shift and are producing a hybrid-organization--one that is still bureaucratic in form, but beginning to take on the characteristics of the information technology network. The "down-sizing" of major corporations in the 1980s-90s has been seen primarily as a bottom-line driven calculation to reduce business outlays. A large proportion of job loss has been at the middle management level--the hierarchical layer that served as the way-station between the activities of localized specialization and centralized strategic vision. While cost-cutting might be the primary goal, the process of down-sizing can not occur unless the organization remains at least as productive as it was before. Information technology is allowing large corporations to empower individual workers (thus necessitating less direct supervision), while ensuring central authority. Middle management is less critical in serving as a filter of information and as an intermediary level of control. Whether this flattening of the bureaucratic organizational chart leads to a flat organization can not be predicted. Weber himself posited that once established, bureaucracy was extremely difficult to displace. However, it would have been equally unimaginable to the feudal administrator that his hereditary claim to a position of authority would be usurped by a professionally paid and trained specialist who was awarded the position through examination and merit. If business competition can be won through organization along an information technology network, the dominance of bureaucracy will be severely challenged.

Weber assumed that change within the economic, political and social orders of a society will be greatly influenced by the economic organization of that society. The fact that the bureaucratic form greatly enhanced capitalist activity ensured the continuing growth and eventual ubiquity of this organizational structure. Recent changes in business practices that emphasize individual worker empowerment, reduction of hierarchy, and the communicative and computational power of technology are manifestations of movement toward information technology network characteristics.39 In the United States, the leading information technology state, the political order is also beginning to show signs of adjustment. There is an emerging National Information Infrastructure (NII), which comprises, "all of the interconnected information systems in the United States and all of the interconnection mechanisms used to make those connections work. This currently includes more than one half of all of the computers in the country, all of the telephone systems, all of the cable television systems, [and] all of the satellite communications systems."40

The American NII is ubiquitous and generally unnoticed by the average citizen. It supports, among other things, the national banking system, transportation (particularly air traffic control), food and water supplies, electrical power grids, medical and emergency systems, and communication.41

The social manifestation of an information technology network organization is beginning to take root in the actual electronic communications network. Electronic mail (E-mail) has become the dominant mode of entry into information technology. Studies indicate that e-mailing tends to be the first type of information computer technology function in which people engage and that it serves as the primary catalyst for more sophisticated use of information computer technology.42 The social and political communities developing on the "net" (there are four dominant e-mail systems globally--Internet, BITNET, UUCP, and FidoNet) are paralleling the actual physical structure of the communication system; that is, discussion and communication is being conducted without a centralized authority (although discussion groups usually have a "list owner" or "manager" as a center point of contact ). The effect is to increase the individual's power by allowing greater access to information and providing the opportunity to engage actively in discourse that is not bound by geography. The Internet is not isolating people in their homes as some have suspected it would. Greater access to information has helped to illuminate issues held in common and one's locality still has a great impact on one's interests and concerns. A recent study by the Rand Corporation's Center for Information Revolution Analyses concluded that many civic networks, which emphasize public interest issues, tend to revolve around local or regional concerns. Thus, rather than creating virtual communities that replace traditional face to face community organization, network technology can supplement and enhance the communitarian impulse. The report concludes that civic high-tech networks can support "interpersonal relationships, local community-building and social integration" in four ways. First, civic high-tech networks provide improved and new forms of communication. E-mail allows what is termed "multidirectional discourse"; that is, one-to-one, one-to-many and many-to-many communication. Second, these networks provide easier access to more and better information that in turn can promote greater access to employment, health and educational opportunities. Third, high-tech information technology can be used for intra-organizational communication, which can lead to cost-efficient links between organizations. This is particularly important for many non-profit public interest groups under constant financial pressure. Finally, civic high-tech networks promote shifts in the organization of social and political activities by providing greater ease of access to government sources and, at least in theory, promise to increase government responsiveness to community and individual concerns.43

If we are indeed entering the Information Age, we should begin to see manifestations of organizational change reflective of the characteristics associated with information technology. The appearance of network forms in civic and business activities is occurring and at least supports the abstract contention that such a transformation may be afoot. There is one institution, however, that has played a critical role in state transformation in the past and whose role in the organizational transformation from the bureaucratic form to a hybrid or pure network form will likely be critical--the military. Max Weber himself, while focusing on the economic foundations of societal orders, pointed to the military as the social institution, whose adoption and promotion of bureaucracy led to its dominant position as an organizational form. He wrote, "...the bureaucratic tendency has chiefly been influenced by needs arising from the creation of standing armies as determined by power politics and by the development of public finance connected with the military establishment."44

In modern society, the military is one of the most heavily bureaucratized institutions. To conceive of the military outside of the bureaucratic form seems difficult. Thus, one could hypothesize that the emergence of a network form of organization in the military would be noteworthy.

In the United States, the country with the leading emphasis on information technology in the 1990s, some initial movements toward an information technology network form are discernable. One major project underway within the U.S. Army is "Force XXI," the goal of which is to create the 21st-century army that is "digitized and redesigned to harness the power of information-age warfare." Brig.General Joseph Oder, director of the Digitization Special Task Force, argues for a restructuring that resembles much of what above is theorized as a networked form of organization. Oder begins with the assumption that, "land force dominance does not lie with the sequential application of each objective but with the synergy created through the simultaneous application of modernized systems." 45 In order to produce this synergy, information technologies must be integrated into the force across multiple systems so that there are common friend or foe identification mechanisms and procedures and general interoperability.46 This is to be accomplished through digitalization, which Oder defines as, "the application of information technologies to acquire, exchange and employ timely digital information throughout the Battlespace, tailored to the needs of each decider (commander), shooter, and supporter, allowing each to maintain a clear and accurate vision of the Battlespace necessary to support planning and execution."47

The major consequence of such digitilization goes well beyond a simple firepower force multiplier. It promises to provide "shared situational awareness."48 Thus, the individual soldier who comes in contact with the enemy may be able to not only inform his local unit of the enemy location, but higher level commanders as well. Supporting air and artillery may also be instantaneously informed and most significantly, the individual soldier may be able to integrate this new information through the use of hand-carried graphic displays to create for himself a complete picture of where he fits into the overall distribution of friendly and enemy forces. In such an environment, information dominance truly becomes the arena of great contest for it means that advantage in knowledge about location could prove decisive, given the greater precision and destructive potential of modern weaponry.49

However, such shared situational awareness presents some serious challenges for a highly bureaucratized institution in which hierarchy in command is a core principle. How does shared situational awareness impact command? How ubiquitous will it be? What appropriate filters will distinguish awareness provided the foot soldier, his commander, the theater CINC, and the Commander in Chief? How do you protect a system from enemy infiltration when it thrives on continuously flowing information that is well-organized, pervasive in scope and easy to access-- without interrupting the flow, reducing scope, and hindering access? These are the organizational and structural challenges facing the military as they contemplate warfare in the information age. The point highlighted by the U.S. Army's efforts is that the implications for organizational transformation seem to be under study.

If warfare in the information age is to be treated as a phenomenon that departs in significant ways from other forms of war and thus represents a revolution in military affairs we must be able to identify unique characteristics. One place to begin is with the characteristics of the information age itself. The preceding section points to five such characteristics: accessibility, availability, speed, affordability, and recursive simplicity. The implications of these characteristics is the promise of individual empowerment through the emergence of the information technology network as organizational form. The preliminary manifestation of this emerging form can be seen in business, civic, as well as military institutions. The emergence of the network as abstraction and institution gives distinction to the late 1980s as the beginning of a new age. But what does it mean to think about warfare in the context of the network? What does thinking in terms of networks do for threat perception and analysis? What are the target sets, when networks dominate?

Information Warfare

As discussed briefly in the introduction, there is as yet no consensus about the term information warfare.50 In March 1993, U.S. Chairman of the Joint Chiefs of Staff Memorandum of Policy Number 30 (MOP 30) identified seven different concepts to be grouped under the umbrella of information warfare of which some like Command and Control warfare (C2W), intelligence-based warfare, and psychological operations do not necessarily involve information technology. Defense analyst Martin Libicki concluded that, currently, the term is "an unfortunate catch-all phrase."51

There is always the possibility of creating more analytical confusion by introducing new terms to a developing debate. One area of consensus, however, does seem to exist around the idea that information warfare can be divided into two general categories: one which involves strictly military forces and one encompassing society at large.52 What is missing in the evolution of this definition is conceptual clarity based on an organizing principle. The concept of the network may be of some assistance.

If the network is what distinguishes the Information Age, in military terms it is the essence of the network--the idea of connectivity--that appears as the critical target. The information technology network functions on the basis of connections between nodes, which can be individuals at computers, several workstations, or small networks themselves. The strength of the network rests not on its individual nodes primarily, but on its degree of connectivity. If we apply the conceptual construct of network connectivity to the two general categories mentioned above, we may develop a clearer definition of information warfare: one in which connectivity on the battlefield and societal connectivity are the focus.

John Arquilla and David Ronfeldt have offered two terms that could be reworked to capture the idea of connectivity: netwar and cyberwar (again, rather than offer two new terms, I believe that debate is best forwarded by developing and improving upon existing terms). Arquilla and Ronfeldt define netwar as "information-related conflict at a grand level between nations or societies. It means trying to disrupt, damage, or modify what a target population knows or thinks about itself and the world around it." Cyberwar "refers to conducting, and preparing to conduct, military operations according to information-related principles....It means turning the 'balance of information and knowledge' in one's favor." It signifies a "transformation in the nature of war, [but]...does not necessarily require the presence of advanced technology."53

Netwar.  If we begin with the presumption that what is potentially revolutionary and distinctive about the Information Age is the emergence of the network organizational form and the attendant importance of connectivity, both of these terms can be sharpened. Arquilla and Ronfeldt place netwar in the context of ideational competition. As they present it, the target is information itself, or more specifically, knowledge. In this sense, the term seems to closely resemble what might be considered sophisticated propaganda warfare. Arquilla and Ronfeldt agree with the notion that the information revolution has more to do with organization than information, but fail to capture this in their netwar definition. The refinement offered here is that the term, while capturing ideational competition primarily refers to attacks on or defense of societal connectivity. That is, netwar refers to information-related conflict at a grand level, which involves attempts to destroy one's enemy's and protect one's own societal connectivity. The target range includes a society's communication, financial transaction, transportation, and energy resource network links. While degradation and protection of physical assets are important, the primary focus is on attacking and defensing the linkages upon which modern society at large relies to function. Again, there is no reason to develop new conceptions of warfare if all we are talking about is competing for the hearts and minds of people. Propaganda and psychological warfare cover that realm.

The concept netwar is useful only to the degree that it conveys a new form of warfare. To consider societal connectivity a useful target, a society must be dependent enough on these networks to make their loss important. Thus, nomadic, feudal or even moderately industrial societies that show little signs of network characteristics are not likely targets for offensive netwar operations. Arquilla and Ronfeldt posit that both cyberwar and netwar do not necessitate high-technology. In this they and others leave open the door for employing high-tech information warfare tactics against low-tech societies. While they are correct that information warfare is not simply about high-tech weapons, they overlook the point that the emergence of organizational change--the emergence of the network--is a result of the information technology revolution and thus, where that revolution has not taken hold, societal connectivity is likely to be low, making the use of netwar and cyberwar conceptions unnecessary.

The Rand Center for Information Revolution Analyses published a general scaling of worldwide connectivity for 1995. They used the basic presence of e-mail networks as the measure of connectivity. Employing a scale of 0 to 16, where 16 was high connectivity, they found that much of Africa had zero connectivity and most of the Middle East and Southeast Asia would receive no more than a 4 rating.54 While computer communications is not the only measure of societal connectivity, it is a critically important one. In terms of the pervasiveness of the network form, the presence of information technology is key. Where it is not present, capture or destruction of the single radio-television station can produce great effect. Here, the hierarchical form still dominates, decapitation is possible and traditional forms of conventional, propaganda and psychological operations are sufficient. While decapitation certainly can lead to the severing of ties between central authority and society at large, it is centralized control that is the target. In an emerging networked society, like the United States, the communications center (if one can be identified) is not central. Again, the whole point of the original plans that led to the Internet was to create a communication systems without a center. When this is present, new forms of warfare need to be contemplated. It is society's dependence on and connection to the network that must become the focus. The same holds for other societal resources. Thinking about the disruption or defense of the delivery of energy from a centralized power grid system or the support of financial transactions from a centralized banking system involve analysis of different dynamics than would exist if such systems were highly networked.
The conceptual usefulness of netwar is clearest when one recognizes that new forms of human organization necessitate new thinking about conflict. As society becomes more networked, new vulnerabilities and new strengths will emerge. Analyzing the ways such weaknesses and strengths can be exploited for national security purposes will fall under the domain of netwar strategy.

Cyberwar.  The increasing presence of the network form across society means that societal connectivity will begin to emerge as an important national asset itself. This will hold true for the networked military as well. The impact of information technology and the network form on the particular institution of the military means that on the traditional battlefield, military connectivity will be of greater significance. Building on Arquilla and Ronfeldt again, I would refine the term cyberwar to refer to conducting and preparing to conduct military operations against or in defense of military connectivity. This re-conceptualization broadens the term cyberwar beyond the narrow focus on command and control warfare. As the military organization becomes less centralized and hierarchical and more networked, it is the overall flow and quality of information and knowledge that must be contested, rather than the specific control over information.

Netwar, Cyberwar, and the Strategy of Deterrence

Deterrence is best conceived as a game of strategy, where a move by one side affects the next move made by the opposing side.55 In terms of offensive strategy, the focus of a challenger is to produce as much freedom of action as possible. This is accomplished by devising technical, tactical, and operational innovations that reduce (or ideally eliminate) the costs threatened by a deterrer's reaction. The deterrer's focus must be on confounding the challenger's search for acceptable alternatives by developing a comprehensive response that approaches a guarantee of inflicting unacceptable costs on an opponent who initiates an offensive challenge.

This competitive quest for strategic advantage between challenger and deterrer is what I have referred to as the strategic dynamic.56 It is the heart of a deterrence environment. The dynamic, itself, revolves around two important theoretical concepts: shared information and rationality.

Shared Information. In order for deterrence to function, both the challenger and the deterrer must possess specific knowledge about each other's:
national objectives; commitment to the issue in dispute; and relative military, political, and economic resources available to support such a commitment and set of national objectives.57

Incomplete or incorrect information about the challenger can lead to an insufficient deterrent, which either promises costs that are not considered prohibitive by the challenger or costs that do not cover the entire spectrum of military options open to a challenger. If the deterrer lacks clear information about the capabilities and tactics of an opponent, the constructed deterrence strategy will be vulnerable to the availability of countermeasures--weapons, tactics, or operational strategy--possessed by a challenger.

Information concerning the deterrer's objectives and capabilities that is shared with the challenger may be the most critical. While a formidable structure of retaliatory costs can be created, its effectiveness as a deterrent will depend on the proper communication of those potential costs to the challenger. The fundamental problem associated with this requirement of shared information is the environment in which it must take place. A challenging state motivated to the point of actively planning the use of force is unlikely to be receptive to threatening messages from the opposing side. The deterrer will have to provide information to resolve two general questions dominant in the challenger's decision-making process: Does the deterrer possess the political determination and ability to follow through on the deterrent threat? And, does the military capability to inflict the threatened cost exist?

Rationality.  Deterrence can fail for two reasons: the deterrer's strategy does not raise costs above expected benefits to be gained by military action; or, the deterrer's strategy threatens sufficient costs, but the challenger miscalculates.

Assuming that a country is not basing its deterrence strategy on bluff, a deterrent threat should encourage a challenger to rationally calculate all possible options. The deterrer should want a potential challenger to assess deterrent threats in an intelligent and consistent fashion. If constructed properly, the challenger will calculate that the expected benefits of the use of force will be muted by the costs inflicted by the deterrer's response. The problem, of course, is that the deterrence environment is not especially conducive to rational decision-making.

Effective rational decision-making is threatened by the conditions brought on by immediate deterrence situations. Both deterrer and challenger must manage decision-making in an environment of heightened tension, perceived time constraints and, ultimately, stress. These factors can degrade rational decision-making.

Since a decision-maker's susceptibility to stress, rationalization or other factors is partially a function of constraints on time, the challenger should be persuaded that a strategy of delay would be more beneficial than one of hasty action under the limits of incomplete information evaluation.58 The other conditioning factor that may lead to irrational decision-making is uncertainty. The potential for misperception (and its attendant promotion of irrational decision-making) increases under the condition of uncertainty. To promote a decision-making process that will lead the challenger to evaluate rationally the consequences of deterrent threats in a way that matches the expectations of the deterrer, uncertainty needs to be reduced. A challenger must be made aware that the deterrent threat, when employed, will inflict severe and specific costs.

Although pure certainty can not be achieved, deterrence strategies can promote greater or lesser degrees of certainty. The development of an unambiguous deterrent force structure that can, when employed in retaliation, inflict the costs that were threatened is essential for mitigating uncertainty and thus encouraging a rational preference to yield on the part of the challenger. Coming full circle in our deterrence logic, the reduction of uncertainty for the enhancement of rationality is inextricably tied to the requirement and problems of shared information.59

The intensity of the strategic dynamic associated with deterrence environments is greatly affected by the types of weapons that support deterrent threats. Nuclear deterrent threats have a degree of "reliability of effect" that make the costs associated with a nuclear response seem incontestable.60 Traditional conventional weapons, however, are susceptible to technical, tactical, and operational manipulation to a significant degree. The costs associated with conventional deterrent threats are generally viewed by opponents as contestable. The strategic dynamic is, thus, muted in the nuclear deterrence environment and exacerbated in a conventional one, making attempts to deter in the latter exceedingly difficult. The more contestable that deterrent costs appear to be, the more susceptible they are to challenge. But how is the strategic dynamic of deterrence impacted by a shift in focus towards connectivity? How do cyberwar and netwar affect the pursuit of deterrence and to what degree are threats to connectivity contestable?

Netwar and cyberwar should be understood as the two general forms of contest that may fall under the heading information warfare. Employing connectivity as their organizing principle, both forms can capture a variety of offensive and defensive strategies now being highlighted by security studies analysts and the American Pentagon. Strategies are being developed for conducting operations across command and control, electronic, intelligence-based, psychological, economic information, and computer systems spectrums.61  Information warfare has been also discussed in the context of the pursuit of a strategy of deterrence. This focus was spawned by two coinciding events--the Persian Gulf Conflict of 1990-91 and the end of the Cold War between the United States and the Soviet Union. The aftermath of both of these wars brought a renewed focus on conventional weaponry and their ability to provide a deterrence umbrella over extended U.S. vital interests. In assessing deterrence capabilities, generally it has been argued that nuclear deterrence, particularly its extended form, was practiced in order to counter the nuclear arsenal and conventional superiority (at least numerical) of the Soviet Union. Remove such an enemy and the threat the United States is likely to face will be from smaller conventional forces that may or may not have a weapons of mass destruction (WMD) capability. The Persian Gulf conflict revealed how advanced the United States was in high-tech conventional warfare as it defeated in one hundred days--with few friendly casualties--what at the time was considered the fourth largest land army in the world. It is both natural and logical to emphasize as the cornerstone for national strategy weapons and forms of warfare in which one dominates . In the case of the United States in the mid-1990s, emphasizing extended conventional deterrence rather than nuclear deterrence has the added advantage of lending moral support to the pursuit of nuclear non-proliferation policies. By reducing reliance on such weapons the United States can argue more persuasively for others to do the same or forego the option altogether.62  Dr. William Perry, before becoming U.S. Secretary of Defense, concluded that the success of high-tech weaponry and information-processing systems in Operation Desert Storm showed that this type of force, while "certainly not as powerful as nuclear weapons it is a more credible deterrent, particularly in regional conflicts vital to U.S. national interests....The United States can now be confident that the defeat of a conventional armored assault in those regions could be achieved by conventional military forces."63

The Persian Gulf conflict certainly revealed the great impact that improved precision-guided weaponry informed by better intelligence and battle damage assessment (BDA) can have on the conventional battlefield. Both intelligence and BDA act as "force multipliers;" that is, they enhance the destructive potential of conventional weapons and in doing so present a challenger with a formidable threat. But as the high-tech innovations begin to translate into organizational and doctrinal changes how will the ability to practice deterrence be affected?

Deterrence and cyberwar. Much of this paper has been concerned with developing some definitional rigor around the variety of issues and terms associated with early discussions of information warfare. In terms of analyzing deterrence dynamics, care must be taken to clearly denote the strategic conditions one is assuming. Cyberwar, the preparation to conduct and the conducting of military operations against military connectivity, represents an enhanced dimension of conflict. On the one hand, it re-focuses combat into a new area. As U.S. Army Chief of Staff, General Gordon Sullivan, explains it in reference to the abstract organizational chart, instead of focusing primarily on the boxes, as armies traditionally did, the twenty-first century army will direct attention toward the lines that connect those boxes. Connectivity is the key.

Under this strategic notion, the application of deterrence strategy would have to contend with the type of dynamic associated with traditional conventional weapons; deterrence strategy would have to overcome the problem of contestability. To menace the connectivity of an opposing military may represent a significant threat, particularly if that military is highly dependent on those connections. Two related problems, however, present themselves. First, the disruption to the military network would have to be substantial for it to be feared. Destruction of only a segment of the network would not preclude the rest of the military force from achieving its offensive goals and thus the deterrent threat directed toward connectivity is unlikely to be seen as prohibitive. Second, if one of the major objectives of moving toward a networked military is to provide it with greater sustainability through redundancy and economy of force, the deterrent threat directed at connectivity may suffer from a perceived low reliability of effect. The opposing state can hope to sustain attacks on its connectivity. Therefore, it will likely view deterrent threats directed at connectivity as contestable and, in the moment of decision, of questionable credibility.

Cyberwar is likely to be dominated by a contest for supremacy over the electromagnetic spectrum.64 The side that achieves such supremacy will have an enhanced ability to see, decide, and move at a pace that should overwhelm one's adversaries. This spectrum of conflict is better understood in the context of offense and defense than deterrence. Since the operations of networked militaries will be dependent on their connectivity, combat will, by necessity, involve attacks on connectivity.

The potential consequences of greater military connectivity is greater lethality. This is brought about on one level by having a better idea of where your opponent is and the capability to precisely hit him before he moves. The digital battlefield, with its shared situational awareness, promises to solve the traditional problem that "operational mobility has never matched the capability of intelligence to tell us what the enemy is trying to do."65 In the days in which the horse was the primary source of transportation, intelligence moved at about the same speed as the enemy. The digital battlefield, by creating theater-level integrated sensor-to-shooter capability (TLISTS), promises to allow detection and reaction that can outmatch the enemy's ability to move.66 This force- multiplier effect is amplified further when one begins to consider the value-added dimensions of the networked military.

Traditionally, the securing of flanks or reserves and general force protection required the deployment of combat troops. They had to be used in what is now termed, "operations other than war" (OOTW)67 primarily because one never knew with certainty where the enemy might strike. However, if the electromagnetic spectrum can be seized and "topsight" over the entire Battlespace provided, combat troops would not have to be dedicated to protect flanks and rear-areas not under potential pressure from the enemy.68  These combat forces could now be used in offensive operations. Thus, connectivity can not only create greater lethality, it can increase the number of lethal forces available to commanders.

Such a force should cause most opponents to take pause. Dominance in cyberwarfare could make the whole prospect of challenging a state seem prohibitively costly. In this regard, overall deterrence protection of vital national interests might be enhanced. The problem, of course, is that such dominance can be contested, both before and after war begins. Command, Control, Communication, Computer and Intelligence (C4I) are susceptible to disruption. The employment of computer viruses, electronic disinformation, or direct destruction of sensing equipment will become increasingly prevalent as the importance of connectivity increases. As the Gulf War AirPower Summary Report suggested, "the more sophisticated and expensive the information gathering system, the greater the premium opponents will put on disabling it....The pay-off for shooting down a state-of-the-art radar surveillance aircraft, for example, will surely attract efforts to do so."69

Cyberwar can not be understood in a static context. The Persian Gulf conflict, while considered a harbinger of cyberwar, may be a poor touchstone. In the future, opponents of high-tech networked militaries are unlikely to make the same mistake Iraq did by giving the United States a "free ride" to deploy and use its communications network.70  This, of course, does not mean that the ability to conduct and dominate in cyberwarfare should not be used to promote deterrence against attacks on vital interests. It simply must be recognized that cyber-deterrence suffers from the same inherent difficulties that are built into conventional deterrence. As long as the costs associated with a deterrent threat can be viewed by an opponent as contestable to a significant degree, deterrence is unlikely to hold under great stress. The dynamics associated with cyberwar would seem to support the conclusion that much more time should be directed toward the development of both offensive and defensive cyberwarfare capabilities, tactics and strategies, than on deterrence models.

Deterrence and Netwar.  In a deterrence context, cyberwarfare differs little from traditional conventional warfare. Both are dominated by a strategic dynamic driven by the presence of contestable deterrent costs. Deterrence is, therefore, fluid, in constant need of maintenance, and in the end prone to occasional breakdown.71  When applied to the concept of netwar, the utility of a deterrence model and the practicality of a deterrence strategy seem even more limited. Netwar focuses attention on societal connectivity. From an offensive strategy mind-set, societal connectivity can be attacked, disrupted, or destroyed on three different levels: the personal, the institutional, and the national.72

Societal connectivity can be targeted by affecting the electronic records of individuals. By changing or destroying those records one can alter who society thinks an individual is and how an individual interacts with others. These records include credit reports, medical histories, school transcripts, financial portfolios and bank accounts, as well as social security and law enforcement files. Such records support a digital persona. Manipulation of these records could effectively change one's wealth or even one's identity.73  The seriousness of this threat (it is not simply a potential nuisance) is compounded by the general public view of computer information. As Winn Schwartau emphasizes, the perception is that "computers don't lie....In Cyberspace, you are guilty until proven innocent."74  If anyone has ever attempted to challenge their credit rating report or deal with a government agency like the Department of Motor Vehicles, they can attest to this observation. Computer printouts are perceived as truth. You must prove the computer wrong. In most instances, the difference between the reality the computer describes and the one to which you attest can be explained away by data entry error. But what if, on a large scale, manipulation of the digital record were to occur in a sophisticated and purposeful manner? The social disruption could be astounding.

Again it is important to recall the very nature of the Information Age. The accessibility, availability, affordability, and speed of information management and the likelihood that all of these things will continue on their current trends creates the potential for many significant advances. The very nature of connectivity, however, breeds some serious potential second-level vulnerabilities as well. The idea of electronic privacy begins to sound like an oxymoron.

The information dependencies and foundations of societal institutions are equally vulnerable. Take the modern corporation, for example. As the financial strength of the corporation becomes increasingly tied to its ability to manage information more effectively than its competitors, information systems may become new arenas for economic conflict. Rather than send the fleet to open and manipulate markets, competitors (be they territorial states, other corporations, or disillusioned former employees) may place at risk network systems supporting economic activity.75   Why compete in expensive marketing contests, when disruption of R&D projects before production even begins might be possible with a well-placed computer virus?76  Instability in major corporations could have wide-ranging negative affects on a society in general.

The consequences of netwar--disruption or destruction of societal connectivity--may be brought about not only by specific attacks on personal electronic or institutional records. It may be conducted on a broader scale, where electronic connectivity is indiscriminately targeted. At its core the high-tech network relies on electronic circuitry that is extremely vulnerable to disruption by other magnetic fields. We have all experienced some form of "interference" like when one's conversation on a portable phone is cut out by static created as one passes by something as benign as an audio baby monitor one has set up in the kitchen. This tends to be a temporary phenomenon. There is, however, little technical expertise needed to manufacture enough of a magnetic field in a directed fashion so as to put at risk the circuitry of an individual computer or computer system. Weapons such as HERF guns (high energy radio frequency) and EMP/T bombs (electromagnetic pulse transformer) are not only conceivable but may already have been built.77  The use of such weapons to disrupt whole city transportation, communication, or financial transaction systems begins to blend into cyberwar, where political-military command and control may be also affected. In a strict use of our term, however, such disruption would be a side benefit.

The three levels of netwar do not fit well into a deterrence model. At its core deterrence theory rests on the principle of retaliation in kind; where, the cost inflicted in retaliation will minimally match the level of costs associated with the offensive attack.78  If an attack reduces no buildings to rubble, kills no one directly, but destroys information, what is the response? We tend to think about information as intangible, but the loss of information can have very tangible personal, institutional and societal costs. What credibly can be held at risk that would dissuade a state contemplating such an attack?

The quick answer is, of course, their connectivity. Two problems arise with this answer. First, it presumes that the attacking state depends on and values societal connectivity as much as the deterring state. The Information Age, however, means access to information warfare capabilities (both cyberwar and netwar) without necessarily having to be a high-tech networked society. The United States may have required an advanced technological infrastructure to produce the global positioning satellite system (which it exploited to great effect during the Persian Gulf war), yet now all one has to do is go down to Radio Shack to purchase a GPS monitor to access the system.79  Information age technology is inherently accessible and one can not presume the loss of connectivity will be viewed as prohibitive for a low-tech society. The dilemma, then, is whether you threaten physical destruction of national assets in response. In essence, do you treat attacks on societal connectivity as acts of conventional war?

The second problem with retaliating against connectivity is its potential negative effects on the deterrer. So far, I have treated connectivity as if it was a nationally-bounded asset. But, of course, the very nature of high-tech networks is that they begin to outstrip the usefulness of employing geographic territorially-defined boundaries as the unit of analysis. The connectivity of a "nationally" defined networked society does not respect geographic conventions. It is difficult to separate out disruptions in connectivity between national and global levels. This sort of electronic interdependence should in theory work to create disincentives against offensive attacks in the first place, but here again subjective estimates of value may make a difference.

The most significant problem of a "retaliation in kind" model of deterrence in the context of netwar is that the attack may not emanate from a state at all. The technology is such that small groups--terrorists, organized crime, hackers--now have a capability that once only belonged to states themselves. They can threaten societal-wide damage instantly. Deterrence requires that the opponent be identifiable. This may not be the case with netwar. In such an environment, general defense against attacks takes on greater importance than developing credible deterrence responses. Of course one can threaten to seek out "netwarriors" and promise great retribution, but there is a lot of room between threatening and finding them. Again, contestability as an analytical concept may prove helpful. The ultimate way to contest a promise of retaliatory costs is not to be identified as the source of an attack. The prospect of avoiding detection in netwar is at least high enough that most actors motivated enough to contemplate such warfare will be unfazed by promises of future discovery. It may be prudent from a deterrence perspective to consider netwar as being no different than traditional military attack on one's homeland and thus be ready to threaten an appropriate military response. However, the degree of ambiguity that is likely to revolve around this new form of combat will undermine the credibility of such an approach in the eyes of a motivated actor.


The nuclear context of the Cold War raised deterrence to the dominant position of strategic thought. Great power conflict was something not to be fought and won, but avoided. For the Cold War superpower competition, deterrence was the primary strategic approach under which offensive and defensive structures provided support. The enormity of the potential destruction associated with nuclear weapons required such a focus. This, in turn, led to the extension of the deterrence concept to help organize thinking about the protection of extended vital interests and homelands through reliance on conventional weapons; an extension of thought that generally misses the significant difference between nuclear and conventional deterrence.

Traditionally, however, a state has been able to avoid being the object of attack through the development of an imposing offensive capability and a formidable ability to defend. In such a context, offensive and defensive strategies dominate strategic discourse and deterrence is best viewed as a by-product. The nature of netwar and cyberwar lend themselves to analytical frameworks and a strategic calculus dominated by offense-defense models, rather than deterrence. The high degree of contestability likely to be found in netwarfare and cyberwarfare operations means that net- or cyber-deterrence will best be pursued as by-products of formidable offensive and defensive strategies. Attempts to simply roll information warfare into strategic approaches in which deterrence is the primary concern miss what is distinctive about this new form of contest--the contestability of connectivity. As we move further toward the Twenty-first century, ironically, it is the approach to war (offense-defense) that dominated the early part of the Twentieth century rather than its latter half (deterrence) that may be most useful for understanding Information warfare.


1 Note: This is an idealized form of the information technology network set against Weber's bureaucracy. It is not suggested that this organizational form exists now, but its characteristics do seem in line with the features discussed earlier concerning the Information Age. The Internet is used as an illustration of an early iteration of this idealized alternative structure, but the information technology network is theoretically applicable to many levels of social interaction.
1. Cited in Wayne Rowe, Information Warfare: A Primer for Navy Personnel (Newport: RI, Naval War College Center for Naval Warfare Studies, June 1995), p.3.
2. John Arquilla and David Ronfeldt, "Cyberwar is Coming!," Comparative Strategy Vol.12 2 (Spring 1993) point to the Mongols as a perfect example of a military force that exploited information superiority. A similar point is made in Norman Davis, "An Information-based Revolution in Military Affairs," Strategic Review Vol.24, 1 (Winter 1996):43-53.
3. William McNeill, The Pursuit of Power: Technology, Armed Force and Society (Chicago: University of Chicago Press, 1982); Martin van Creveld, The Transformation of War (New York: Free Press, 1991); Geoffrey Parker, The Military Revolution: Military Innovation and the Rise of the West, 1500-1800 (Cambridge: Cambridge University Press, 1988).
4. Martin van Creveld, Technology and War (New York: The Free Press, 1989); Stephen Rosen, "New Ways of War: Understanding Military Innovation," International Security, 13, no.1 (Summer 1988).
5. For a brief overview of the lack of consensus over the basic definition of information warfare itself, see Martin C. Libicki, "What is Information Warfare?," Strategic Forum no.28, May 1995 (Washington, DC: National Defense University, 1995). An extremely helpful bibliography on information warfare is included in Matthew J. Littleton, Information Age Terrorism: Toward Cyberterror unpub. Master's thesis, Naval Postgraduate School, December 1995.
6. The original U.S. Defense Department plan created what was known as the ARPENET. Its basic goal was to create a communication system that was both redundant and independent. For background see Katie Hafner and John Markoff, CyberPunk: Outlaws and Hackers on the Computer Frontier (New York: Simon and Schuster, 1991), pp.263-282 The link between innovation, warfare and societal organization again seems apparent.
7. The idea that information warfare is fundamentally new is contested by many who point to the emphasis that has always been placed on knowing where your enemy was, what his plans were, and the capabilities that supported his plans as well as the importance of denying the same information to your enemy. Authors have pointed to the writings of Sun Tzu as an example. John Arquilla points to the Mongol approach to war as another example. See Arquilla and David Ronfeldt, "Cyberwar is Coming," Comparative Strategy 12 (April-June 1993): 141-165.
8. The computer supports, of course, text, pictorial, verbal, as well as real-time and taped full-motion video forms of information. The ubiquity of computers by the late Twentieth Century in most of the developed world was such that most people had little recognition that they were interacting with these machines.
9. One recent advance that highlights the break with dependence on geographic proximity is in the field of medicine. Real-time access to the expertise of specialists has traditionally required travel to where they are. Rural inhabitants, in particular, have either had to live without such expertise or make significant efforts to get to particular urban centers. In 1996, the use of video computer connections between specialists in cities and local community doctors have allowed patients to be examined by and interact with a doctor hundreds of miles away.
10. This extends to information held by individuals as well. If the "depository" of information is an individual one, has typically needed access to that individual. While the telephone removed the access obstacle of geographic proximity, the person still was required to be on the other end. The advancement in e-mail communication, which promotes efficient asynchronous retrieval and dissemination of information will be dealt with later.
11. Katie Hafner and John Markoff, CyberPunk: Outlaws and Hackers on the Computer Frontier, (New York: Simon and Schuster, 1991), 265.
12. Spring Tech Guide, "Taming the Internet," U.S. News and World Report (April 29, 1995).
13. James Gleck, Chaos: Making a New Science (New York: Penguin Books, 1987), 103.
14. David Gelernter, Mirror Worlds: or the Day Software Puts the Universe in a Shoebox...How It Will Happen and What It Will Mean (New York: Oxford University Press, 1991), 54.
15. Gleck, Chaos, p.98-103.
16. Data from Dataquest cited in U.S.News and World Report, pp. 60, 68; "The Web Keeps Spreading," Business Week, (January 8, 1995), pp. 92-93.
17. Norman Davis, "An Information-based Revolution in Military Affairs," Strategic Review 24, no.1 (Winter 1996):43-53; Pat Cooper, "Information Warfare Sparks Security Affairs Revolution," Defense News June 12-18, 1995; Andrew Krepinevich, "Cavalry to Computer: The Pattern of Military Revolutions," The National Interest, 37 (Fall 1992):30-42; William Odom, America's Military Revolution: Strategy and Structure after the Cold War (Washington, DC: American University Press, 1993).
18. I tend to ascribe to Thomas Kuhn's sense of revolution as a conceptual shift that requires the "reconstruction of prior theory and the re-evaluation of prior fact." Kuhn, The Structure of Scientific Revolutions 2nd ed. (Chicago: University of Chicago Press, 1970), p.7.
19. For an overview of the impact of information technology on individuals see Lee Sproull and Sara Kiesler, Connections: New Ways of Working in the Networked Organization, (Cambridge, MA: MIT Press, 1991).
20. The ENIAC (electronic numerical integrator and computer) first demonstrated in 1946 required 1,500 sq ft of space and 17,468 vacuum tubes to operate. It could make 5,000 additions per second. U.S. News and World Report February 19, 1996, p.22.
21. The connection between futurist science fiction writing and technological development is an interesting one. Many advances in technology that have actually emerged can be found referenced in early sci-fi literature. However, the general vision of the technological society has been rather bleak. The Orwellian nightmare is a predominant theme. See, Charles Waugh and Martin Greenberg, Battlefields Beyond Tomorrow: Science Fiction War Stories (New York: Bonanza Books, 1987).
22. Orwell's narrative was greatly influenced by his perceptions of the Soviet Union and the growing power of the totalitarian state.
23. Note that in the counter-coup attempt by Soviet communist hardliners in 1991, it was Moscow television and radio stations that remained key targets.
24. For greater detail see, Peter J. Denning, " The ARPANET after Twenty Years," in P. Denning, ed. Computers Under Attack: Intruders, Worms, and Viruses (Reading, Massachusetts: Addison-Wesley Publishing Co., 1990), 11-19.
25. Denning, Computers Under Attack, p. 11
26. Denning, Computers Under Attack, p. 15
27. See Sproull and Kiesler, Connections; Robert Anderson, Tora Bikson, Sally Law, Bridger Mitchell, Universal Access to Email: Feasibility and Societal Implications (Santa Monica, CA: Rand Center for Information Revolution Analyses, 1995), 119-121. On this point also see J.D. Eveland and T.K. Bikson, "Evolving Electronic Communication Networks," Office: Technology and People, Vol.3, 1987:103-128.
28. Weber concluded that with regard to growing emphasis on bureaucratization, the "economy leads, it is not led." H.H. Gerth and C. Wright Mills, eds., From Max Weber: Essays in Sociology (New York: Oxford University Press, 1946), 54,
29. Weber, "Wirtschaft un Geseltschaft," part III, ch.6 translated in Gerth and Mills, p.215.
30. Gerth and Mills, pp. 50, 54.
31. These other forms rested primarily on "inherited avocational administration by notables" instead of "paid professionals." Weber in Gerth and Mills, pp.224-25.
32. I propose the word "cloaked" in order to capture the phenomenon reported on the Internet of people portraying themselves as different people, genders, ages etc.
33. Stephen Krasner, Structural Conflict: The Third World Against Global Liberalism (Berkeley: University of California Press, 1985).
34. The vulnerability of leaders, themselves, could never be entirely be removed. The solution practiced by the superpowers during the cold war was to have ready processes of pre-delegation of authority within their command and control system in case the leadership was lost. This reduced the gain to be achieved by eliminating leaders. In fact, since there were significant costs associated with decapitation (who would negotiate a peace, or might this lead to a massive spasm of retaliation) the incentive became negligible.
35. This is not to suggest that such disruptions are unimportant or that the entire net could not be attacked. The 60,000 computer system was disrupted on November 2, 1988 by a Trojan horse designed by Cornell graduate student Robert Morris. See, Hafner and Markoff, CyberPunk , p.253-346
36. Bonnie Fisher, Michael Margolis, and David Resnick, "Surveying the Internet: Democratic Theory and Civic Life in Cyberspace," Southeastern Political Review (forthcoming, 1996).
37. Kenneth Oye, Cooperations Under Anarchy (Princeton: Princeton University Press, 1986); Stephen Krasner, ed. Regimes (Ithaca, NY: Cornell University Press, 1983).
38. As in all anarchical systems, enforcement of regulations is dependent on the distribution of power. On the internet, the technical "gods" have the power "to censure and to censor those who offend them, and some of them may even rain flame down upon those who violate their netiquette." However it is important to note that ordinary users have the opportunity "to improve their status. If they master the arcana and practice netiquette, there appears to be nothing that formally forbids them from ascending Mount Olympus and joining the deities." Fisher, et.al., "Surveying the Internet."
39. Sproull and Keisler; articles
40. Fredrick B. Cohen, Protection and Security on the Information Highway (New York: John Wiley & Sons, 1995), 7.
41. Cohen, p.1
42. Robert Anderson, Tora Bikson, Sally Law, Bridger Mitchell, Universal Access to Email: Feasibility and Societal Implications (Santa Monica, CA: Rand Center for Information Revolution Analyses, 1995), 119-121. On this point also see J.D. Eveland and T.K. Bikson, "Evolving Electronic Communication Networks," Office: Technology and People, Vol.3, 1987:103-128.
43. On some of these efforts see U.S. Congress, Office of Technology Assessments, Making Government Work: Electronic Delivery of Federal Services, OTA-TCT-578 (Washington, DC: GPO, September 1993).
44. Weber, in Gerth and Mills, p.212.The idea that consolidation of the territorial state and its subsequent social evolution can be traced back to the need for a standardized and rationalized system of taxation to support national defense, which could be provided in a superior fashion by the creation of a state bureaucracy is found in the works of McNeill,
45. Joseph E. Oder, "Digitizing the Battlefield: The Army's First Step to Force XXI," Army May 1, 1994 Vol.44, 5:38.
46. This is a departure from the traditional "stovepipe" approach to acquisition in which common subsystems were not emphasized. See Cohen, Protection and Security, pp.25-26.
47. Oder, "Digitizing," p.38.
48. Oder employs this term in much the same manner as I do when theorizing about the implications of the network form of organization, although he as well as William Perry focus on how this aids in combat effectiveness rather than on how this phenomenon might impact organizational structure and dynamics. See William Perry, "Desert Storm and Deterrence," Foreign Affairs 70, no.4 (Fall 1991), p.69.
49. I choose the word "decisive" carefully. Of course, knowledge about enemy location has always been critical in warfare as the writings of Sun Tzu will attest. However, the impact such knowledge can have on the overall flow of battle now may be decisive in producing victory because reaction time is so compressed and the likelihood of a launched attack destroying its target based on this information is considerably higher than in previous eras. However, it is not a certainty and such an attack is still open to contest.
50. Vice Admiral Arthur Cebrowski, Director for Command, Control, Communications, and Computers for the American Joint Staff is quoted as saying that the services have yet to agree on what information warfare encompasses, but argues that since"the implications are so new, it is best that no central authority control its development." quoted in Patrick Cooper, Information Warfare Sparks Security Affairs Revolution," Defense News, June 12-18, 1995, p.1
51. Martin C. Libicki, "What is Information Warfare?," Strategic Forum no.28 (Washington, DC: National Defense University, Institute for National Strategic Studies, May 1995).
52. Martin Libicki and John Arquilla to date are the most often cited authors with regards to basic terminology and both make this distinction.
53. Arquilla and Ronfeldt, "Cyber war is coming!," pp.144, 146-47.
54. Anderson, et,al., Universal E-Mail, 152-157. Specifically, they measured nodes. "A node may consist of a single computer and user or an entire organization with many of both. The Matrix Information Directory Service (MIDS) tracks and maintains historic data on the size of these networks aggregated by country. The "interconnectivity" metric used here is a combined measure of MIDS data on nodes per capita per country for each of the four major computer systems that can exchange electronic mail." p.156.
55. Thomas Schelling, The Strategy of Conflict (New York: Oxford University Press, 1960).
56. Richard J. Harknett, "The Logic of Conventional Deterrence and the end of the cold war," Security Studies 4, no.1 (Autumn 1994): 86-114.
57. Shimshoni, Israel and conventional deterrence, p.10-16; Philip Green, Deadly Logic: The Theory of Nuclear Deterrence (Columbus: Ohio State University Press, 1966), p.185-188.
58. Ole Holsti, Crisis, Escalation, War (Montreal: McGill-Queen's University Press, 1972), p.233.
59. Greater certainty also carries a price. A country that specifies in detail its deterrence commitments is vulnerable to unexpected contingencies. The challenger may also be able to employ "salami tactics" that avoid the brunt of the deterrence strategy by gaining objectives in a piecemeal fashion. This problem is discussed in the following section under the broader heading of strategic dynamics.
60. I am indebted to Colin S. Gray for developing the phrase "reliability of effect" in discussing my concept of contestable costs. See Gray, "Nuclear Weapons and the Revolution in Information Warfare," in T.V.Paul, Richard J. Harknett and James J. Wirtz, eds., The Absolute Weapon Revisited: Nuclear Arms and the Emerging International Order (forthcoming, 1996). For more on the incontestability of nuclear deterrents see Harknett, "State Preferences, Systemic Constraints, and the Absolute Weapon," in that same volume.
61. Libicki, "What is Information Warfare?" and Rowe, "A Primer for Navy Personnel," work with these terms. Interestingly, Libicki separates them into to broad categories--against society and against the military, which is the same distinction captured by my use of netwar and cyberwar.
62. For more on how nuclear non-proliferation serves great power interests see Richard J. Harknett, "Nuclear Weapons and Territorial Integrity in the Post-Cold War World in Ken Dark, ed., New Studies in Post-Cold War Security (Dartmouth, forthcoming 1996).
63. William Perry, "Desert Storm and Deterrence," Foreign Affairs 70, no.4 (Fall 1991), 66.
64. Morris Boyd and Michael Woodgerd, "Force XXI Operations," Military Review November 1994:16-28.
65. Arthur DeGroat and David Nelson, "Information and Combat Power on the Force XXI Battlefield," Military Review (November-December 1995), p. 58.
66. Randall Bowdish, "The Revolution in Military Affairs: The Sixth Generation, Military Review (November-December 1995) http://www-cgsc.army.mil/cgsc/milrev/95novdec/bow.htm
67. Boyd and Woodgerd, "Force XXI Operations," p.18; DeGroat and Nelson, "Information and Combat Power," pp.59-60.
68. David Gelernter develops the concept of topsight, which captures the essence of shared situational awareness. See his Mirror Worlds, p.52; Arquilla and Ronfeldt also discuss the term in "Cyberwar is Coming!".
69. Eliot Cohen, "The Mystique of U.S. Airpower," Foreign Affairs 73, no.1 (January-February 1994), 115.
70. William Perry acknowledges that "many of the C3I systems used in Desert Storm could be degraded by foreseeable countermeasures." p.79. Cohen makes a similar point, "The Mystique," pp. 109-125.
71. Many conventional deterrence strategists assume that deterrence failure is ultimately necessary to strengthen deterrence. For an overview of this argument see Charles Allen, "Extended Conventional Deterrence: In from the Cold and Out of the Nuclear Fire?," Washington Quarterly 17, no.3 (Summer 1994): 203-233.
72. This builds upon and modifies an argument put forward by Winn Schwartau, who discusses information warfare broadly defined as being conducting on the personal information, corporate information and global information levels. See his Information Warfare: Chaos on the Electronic Superhighway (New York: Thunder's Mouth Pres, 1995).
73. On the vulnerabilities of automated teller machines for example see Paul Eng, "An Alarm Goes Off at the Cash Machine," Business Week (May 31, 1993), pp.39-40.
74. Schwartau, Information Warfare, p.17.
75. The ubiquitous nature of computer networks in the modern corporation led to an interesting shift in job elimination practices in the 1980s. Those with computer expertise were frequently given two-week severance, but escorted to the door the day of their announced dismissal, rather than allowed to remain. They were allowed back to clear out personal belongings after computer passwords and access were changed.
76. For an overview of such potentialities see Peter Denning, ed., Computers Under Attack: Intruders, Worms, and Viruses NASA Ames Research Center (New York: Addison-Wesley, 1990).
77. For more on these weapons and their effects see Schwartau, Information Warfare, pp.177-185. He refers to some early possible uses in the Gulf War as well as some instances in which the phenomenon of magnetic field disruption has disrupted or destroyed commercial and military systems.
78. The classic in this regard is Bernard Brodie, et.al., The Absolute Weapon (New Haven, CT: Yale University Press, 1946).
79. See, for example, "Who Knows Where You Are? The Satellite Knows," Business Week (February 10, 1992): 120-121.