[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[infowar.de] War 2.0
Weiterleitung hier mit freudlicher Bitte des Autors. Lesenswert.
RB
http://www.hoover.org/publications/policyreview/5956806.html
WEB SPECIAL:
War 2.0
By Thomas Rid
February 2007
Inventions can cast a seductive spell. Promising communication
technologies in particular may mesmerize even serious men: “Space will be,
to all practical purposes of information, completely annihilated,”
enthused a House Commerce Committee report published on April 6, 1838. Its
authors were enthralled by Samuel Morse’s recent invention, the telegraph.
One hundred and sixty years later, the internet similarly inflated
expectations in politics and commerce. After the bubble burst in 2001,
many disappointed entrepreneurs and investors recognized that the “new,”
transformed economy had been overrated and overheated. Just as the markets
overestimated the World Wide Web’s seemingly unlimited economic potential,
the U.S. defense establishment also was lured by a techno siren song, that
of network-centric operations. Widespread enthusiasm about the new,
“transformed” army’s seemingly unlimited military potential grew. But just
as many businesses in that digitalized age could not deliver profit, the
computerized force could not deliver victory. The Pentagon used its
technology-driven “transformation” project in a non-social way, to link
“sensors to shooters” in order to minimize reaction time. Its very ideal
seemed to have been to minimize the role of fallible humans. Only now, as
American soldiers are stuck in two mostly low-tech protracted guerrilla
campaigns in Iraq and Afghanistan, is the military’s high-tech bubble
beginning to burst.
The idea of network-centric operations initially was inspired by
developments in the IT-industry in the 1990s. But while today’s internet
industry is happily nurturing a new boom revolving around Web 2.0, the
defense establishment is haplessly managing counterinsurgency and
stability operations. Yet a closer look at the two seemingly separate
trends brings to light striking similarities. War’s changing character is
not only augmented by the emergence of the new media; the way the web and
today’s communication devices are used to organize lives also instructs
our understanding of how killing is organized. The argument put forward
here is that the web’s emerging organizing principles — including a social
as well as a technological dimension — increasingly govern the management
of violence. The new media consequently offer both a set of new metaphors
to understand the character of today’s wars and a socio-technological
platform that remodels the architecture of battle.
War’s true transformation has a face very different from the one
originally envisioned by the Pentagon’s civil and military leadership, in
which the force with the more expensive cutting-edge equipment would
prevail. Yet let there be no misguided enthusiasm: new means of
communication neither “annihilate space” nor disperse the fog of war; on
the contrary, the web makes warfare even more chaotic, messy, and deadly.
Just as the telegraph once did.
Web 2.0 at war
Marked most visibly by the technologically sophisticated first war against
Iraq in 1991, the U.S. Defense Department’s project of military
transformation was widely celebrated as a “revolution in military affairs”
of historical dimensions. Never before had an army acquired such
awe-inspiring technological superiority over virtually all possible
adversaries. Officers all around the world adapted the basic concept of
transformation, or “network-centric operations” in the military’s idiom.
But the movement threatened to turn into an inward-looking technology
exercise, with a narrow focus on high-tech projects such as
blue-force-tracker, an astronomically expensive system to monitor the
actual position of all American forces in real-time, or high-resolution
overhead imagery and even live video-feeds, beamed into command
headquarters by satellites and drones. Real-time signal intelligence from
the sky was to be instantly connected with massive firepower on the ground
to enhance the 21st-century warfighting machine’s efficiency and lethality.
This may be important, but it rather misses the point in modern war: the
enemy’s resort to asymmetric means of struggle, the significance of human
interaction and social contacts, of improvisation, endurance, commitment,
and trust. This is a lesson soldiers on the ground are painfully learning
in Iraq: “I would trade every satellite in the sky for one reliable
informant,” said Army Lieutenant Colonel Ross Brown, who commanded a
cavalry squadron in the 3rd Armored Cavalry Regiment south of Baghdad.
[1] The local insurgencies in Iraq, Afghanistan, and Lebanon, as well as
global militant jihadis, rely on tacit and trusted social networks, not on
attackable fiber-optic networks. As a consequence, the burgeoning but
introspective debate about transformation in today’s most advanced armies
has largely been replaced by a more down-to-earth debate about
counterinsurgency warfare. In December 2006, the Army announced plans to
cut its Future Combat System by $3.3 billion and to scrap the
transformational Land Warrior program. In the same month, for the first
time in more than two decades, a joint Army-Marine Corps publication on
counterinsurgency was issued; its lead author now commands America’s
troops in Iraq.
Yet the web is back. And once more the private sector is setting the
trend. Today some of the largest IPOs again pour money into web-based
companies’ pockets. Google epitomizes the web’s new bloom. In early 2007,
the firm’s market capitalization surpassed $150 billion, nearly four times
that of Lockheed Martin, the world’s largest defense contractor. This time
business models appear more solid, and profit is delivered. But there is
one persistent change: Individuals have moved to the center of attention.
Not only as buyers, but as vendors, advertisers, developers, designers,
and producers. Media professionals coined the phrase “Web 2.0” to describe
this new trend. Its effects have changed our lives: maps, diaries,
contacts, telephone calls, private financial transactions, photo
collections, videos, music, shopping, flea markets, dating, and even
mourning are today done electronically as much as manually.[2] An
increasing share of interactions both noncommercial and commercial are
between individuals, not between consumers and collective entities, like
companies or states. A historical precedent does not exist. Just like in
the peaceful metropolis with its wireless-capable cafés and parks,
informal social networks determine success on remote battlefields,
sometimes interacting through computers and cell phones, sometimes on
local houses’ front porches. That applies to the counterinsurgent, and
even more so to the insurgent.
Insurgencies, terrorist attacks, and urban operations, to be sure, are not
a video game fought virtually on screens. Suicide bombers are creating
real carnage, roadside bombs are ripping apart real human beings, and
nightly cordon-and-search operations are humiliating real families —
whether or not there is a nearby cell phone or internet access. Acts of
violence may or may not rely on fiber-optic networks. But whether in a
hierarchical army or in a decentralized insurgency with a scattered,
blind-cell setup, acts of violence are always organized by social networks
— and they target the adversary’s social coherence, his ability to
develop, execute, and maintain a political will. The web, ever more
focused on social networking, changes the equation on both sides.
Drawing analogies between two vastly different spheres is risky. But to
come to grips with a new or significantly altered phenomenon, we rely on
metaphors. To see the ways in which irregular warriors and traditional
officers use the web to their advantage, the internet industry itself
offers the richest and most useful set of analogies. The following
principles apply, to a varying degree, to both the insurgent and the
counterinsurgent. The descriptions are intended to be thought-provoking,
not as exhaustive and definite.
(1) In the media industry and in warfare, the initiative and innovations
increasingly come from small start-ups on the lower and middle-management
level, a norm that applies to Google, al-Qaeda, and the U.S. Army.
(2) Consequently, ordinary users must be treated as co-developers who can
come up with a new product or add a competitive edge to it, not merely as
consumers. Tactical battle guidelines and lessons-learned essays benefit
from user-developed suggestions and improvements in a way that is
analogous to the “patches” of open-source applications or Wikipedia’s
articles, called peer-production in the industry’s jargon.[3]
(3) User contributions based on open standards become decisive for
dominance on the marketplace as well as in the battlespace. Linux is the
media equivalent of the IED: successful beyond expectations, as “scripts”
or explosive designs can easily be accessed and adapted to each
application’s specific needs; successful tactics become commoditized.
(4) As a result, the distinction between the final product and its
development phase becomes obsolete, an effect that is known in industry as
the “permanent beta-version”: both counter-ambush tactics as well as
browser-based email platforms, to pick two examples, are permanently
updated and never graduate to a finalized version.
(5) It follows that the acceleration of development cycles becomes a way
to out-maneuver the competition, and to gain and maintain the initiative
over the adversary’s actions; software developers correspondingly adapted
their build-and-release management to embrace a more efficient “release
early, release often” philosophy.
(6) Simple technologies and systems with low adaptation costs have a
competitive advantage, called “loose coupling,” a term widely used by
programmers for friction-free linking of formerly incompatible IT-systems
through a common semantic framework. Such systems are more “adaptable to
the unexpected.”[4] This equally applies to insurgents and militant
networks that easily transfer their successful tactics and innovations to
other groups, a trend that sharply distinguishes them from technologically
sophisticated armies whose “interoperability” diminishes as their systems
grow more complex.
(7) Whether triggered by advertisements on obscure pages or by ambushes on
obscure highways, many small but numerous hits add up to significant
volumes that can have decisive consequences, an effect referred to in the
industry as “The Long Tail.”[5]
(8) As a result of the large numbers of contributors taking the initiative
on their own, in the business of software and warfare, finally, command
and control takes the form of syndication rather than coordination.
The transfer of these principles from software to warfare requires
abstraction — and possibly goodwill. Several of the analogous dynamics
have long dominated guerrilla movements and predate the internet by
centuries, which at first glace merely seems to offer new comparisons for
old phenomena. Popular uprisings, low-tech improvisation, and the steady
infliction of costs, after all, characterized most rebellions in history.
Yet even these perennial principles of war did not remain unaffected by
the arrival of the internet in theater.
Insurgency and counterinsurgency
To appreciate the relevance of social networks, and consequently the web’s
impact, some background in counterinsurgency theory is necessary. An
insurgency is a struggle for control of a political or economic space. A
government or a coalition of states tries to maintain the status quo, and
a sub-state actor or a group of nongovernmental challengers fight to
change the status quo. “Small wars,” “low-intensity operations,”
“asymmetric wars,” and “guerrilla wars” are interchangeable terms.
Insurgent groups can also have an interest — political, religious, or
economic — in a perpetuated state of ungoverned spaces, such as some
poppy-growing warlords in Afghanistan or martyrdom-seeking mujahids in
Waziristan. Then the counterinsurgent is contesting the status quo, and
the resistance may bitterly defend it. In any case, the counterinsurgent’s
political objective is to create a stable government system and the rule
of law. The insurgent, by contrast, may or may not have a clear political
war aim.
The classical theory of counterinsurgency, in a nutshell, is simple: the
counterinsurgent competes with the insurgent for the ability to win the
hearts, minds and acquiescence of the local civilian population. David
Kilcullen, an Australian counterinsurgency practitioner, wrote an
outstanding paper on the subject, “Twenty-Eight Articles.” Shortly after
its publication in 2006, it made the round in U.S. Marine leadership
circles, and got the author a job as consultant for the top U.S. commander
in Iraq. “In this battlefield popular perceptions and rumor,” Kilcullen
observes, “are more important than a hundred tanks.”[6] He is able to hark
back to decades of battlefield-hardened ideas and insights. The
communication revolution, however, adds a new quality to this classic tenet.
Centuries of colonialism were characterized by countless counterinsurgency
campaigns in alien lands around the globe, and many colonies’ struggles
for independence were long and bloody. So it comes as no surprise that the
classic works on guerrilla war were either penned by Europeans or their
former adversaries. The British had Charles Caldwell, a royal major
general, who took part in the Afghanistan Wars as well as the Boer Wars
and captured his insights in Small Wars[7]; T.E. Lawrence, an advisor to
insurgents against the Ottoman occupier in Arabia, published the legendary
book Seven Pillars of Wisdom, condensed “as stalking horses for beginners”
in 27 Articles[8]; Major General Frank Kitson, one of the crown’s best
practitioners of small wars, wrote Low-Intensity Operations.[9] Several
French officers published books on their experiences after the fourth
republic’s tough campaigns in Indochina and Algeria. The two best known
are Roger Trinquier’s tough Modern War and David Galula’s
Counterinsurgency Warfare.[10] Both were republished in 2006, one with a
foreword by Eliot Cohen, the other by Bruce Hoffman, two of America’s most
respected voices on war and terrorism. And then there are those from the
other side, the insurgency’s side. Their names bear the weight of history:
Mao Zedong’s innovative ideas on the role of the peasantry in guerrilla
warfare, Che Guevara’s notion of the foco, Vo Nguyen Giap’s writings on
the various stages of an insurrection.
Some of the classics’ tenets are applicable today, some are not.
Insurgencies remain protracted struggles. They are still characterized by
an asymmetric resource distribution and a specific set of tactics employed
by the weak against the strong. Guerrilla forces old and new are highly
mobile, mostly shun open confrontation, and prefer hit-and-run operations.
The insurgents usually are motivated by a superior cause, be it political
or religious, and they rely on the financial, logistic, and ideological
support of third groups. But here the differences already begin.
Colonialism’s wars of independence were focused on one country or a
limited region, and the local population was their most important base of
support. Today’s insurgencies may still target civilian populations, but
their support base is likely to be global; they operate in cities rather
than in deserts and forests; they may not have a strategy or a political
objective at all; and, once the larger movement is weakened, its remnants
may still have access to sufficient support, resources and weapons to
remain dangerous and transform into a terrorist organization. Al Qaeda
itself was born when two insurgencies joined forces: bin Laden’s Arab
radicals, who supported the Afghan mujahideen’s successful war against the
Soviet occupiers, and the Muslim Brotherhood’s and Ayman al-Zawahiri’s
unsuccessful uprising against the Egyptian government under Nasser, Sadat,
and Mubarak.[11]
In both classical and modern insurgencies the interplay between military
and civilian spheres is very complex. Mao wrote about the “relationship
that should exist between the people and the troops” and introduced his
famous analogy: “The former may be likened to water and the latter to the
fish who inhabit it.”[12] Consequently, as guerrilla fighters are hiding
within the population, it becomes very difficult to distinguish between
combatants and non-combatants. “No physical frontier separates the two
camps,” noted Trinquier. The line of demarcation between friend and foe
passes through “the very heart” of the nation, it separates villages, and
even divides families, he wrote. Intelligence about the insurgency is
notoriously hard to gather. It may not be classified secret but kept in
denied areas, physically or electronically. The closer the informal ties
between the insurgent and host society, the harder the insurgency is to
penetrate and the easier it becomes for the insurgent to “control the
masses.” Consequently, it is the aim of the counterinsurgent to isolate
the insurgency and undermine its standing in the population. The
counterinsurgent shares with the population an interest in stability,
security, and the rule of law. The rebels, therefore, can have an interest
in the opposite — terror, violence, and anarchy. The reign of chaos
demonstrates to civilians that the occupier, or the government, cannot
protect them. This lends credibility to the insurgent’s cause. In such a
situation, civilian tasks will have to be performed by officers to prevent
or break a vicious circle. Galula spells this out best:
To confine soldiers to purely military functions while urgent and vital
tasks have to be done, and nobody else is available to undertake them,
would be senseless. The soldier must then be prepared to become . . . a
social worker, a civil engineer, a schoolteacher, a nurse, a boy scout.
But only for as long as he cannot be replaced, for it is better to entrust
civilian tasks to civilians.
As Trinquier notes, only the “combination of political, economic,
psychological and military measures” will be effective in a stabilization
operation. This requires a unified command, or at least a unity of effort;
“directing the operation from beginning to the end,” in Kitson’s words, is
essential.
In modern counterinsurgencies the problem of unity of command is
confounded drastically. Actors with a critical role for an operation’s
success abound: multinational components of a military coalition,
governmental and non-governmental development agencies, international
organizations, local actors, the regional and international media. With
this plethora of actors, a unity of command it not workable any more, and
should be replaced by a “unity of purpose.”[13]
This has conceptual consequences. Today army doctrine in the largest NATO
countries elevates “support” to the same level as “offense” and “defense.”
Adding a third leg to this classical dichotomy is a radical move,
particularly if viewed against traditional strategic thinkers like Carl
von Clausewitz. Not to compel the insurgent to do our will, to paraphrase
the Prussian, but to get the political support of the population is the
primary aim in counterinsurgency; war, then, is not the continuation of
politics by other means, but by the same means. The two sides politically
compete for social networks with the population. French theorists beyond
the commissioned officer ranks score higher in contemporary military
thinking than many would expect. The U.S. Army’s and the Marine Corps’
2006 Field Manual on Counterinsurgency — COIN, for professionals —
introduces left-leaning sociologist Pierre Bourdieu’s concept of “social
capital” as one of four major forms of power in a society (the others
being coercive force, economic resources, and authority). An individual or
a group owns social capital if it has the power to utilize social networks
to advance his or its interests and goals. The counterinsurgent must
identify the individuals who have social capital and study how they
attract and maintain followers, according to the manual. Counterinsurgency
is a competition for trust, for informal networks, for “social capital.”
The web and mobile phones increase the return on social capital. As a
result, the counterinsurgent must “possess the training, capability and
will to fight on cognitive terrain,” as a U.S. Army journal put it.[14]
This cognitive terrain is landscaped and sometimes eroded by a corrosive
information environment.
The main texts on small wars and guerrilla campaigns may recognize the
effects of propaganda, mobilization, and political loyalties. But Mao,
Galula, and Kitson all fought and wrote before the arrival of the
internet. Even David Petraeus’s famed and much reviewed new American COIN
doctrine is largely silent on the internet’s role in the “long war.” The
insurgents, by contrast, are not. And neither are the mid-level U.S.
commanders who have to deal with the consequences. Today’s insurgencies
have moved financial transactions, recruitment, training, clandestine
communication, and even operational planning into a virtual hinterland —
beyond the control of counterinsurgents or other governments.[15]
“Classical counterinsurgency theory,” Kilcullen pointed out in Survival,
“has little to say about such electronic sanctuary.” It even has little to
say about the role of television in guerrilla war.
A wet environment
The first American experience with a new brand of counterinsurgency
warfare was Somalia. On December 9, 1992, an amphibious landing of U.S.
troops at an African beach near Mogadishu marked the launch of operation
Restore Hope. As the soldiers went into the hostile environment, they
expected, if anything, to meet enemy resistance. The Marines did not find
what they expected. “Get the f--- down, you wanna f--- me to blow your
f---ing head off,” one of the servicemen recommended to a young American
woman. She had her face in the dirt and a gun at her head in split
seconds. Donatella Lorch, a war correspondent, was awaiting the amphibious
landing along with a battalion- sized force of reporters.[16] In
retrospect Lorch defended the soldiers, saying that it was not the
Marines’ fault, but blamed it on “a bit of a lapse in communication.”
Robert Oakley, the U.S. ambassador with an important role in the entire UN
operation, had briefed a group of reporters at the U.S. military base in
Somalia announcing that an amphibious landing was scheduled for about
midnight. If the reporters would like to be present and cover the event,
they should cross the security gates without their Somali translators, he
advised. Nobody, though, had warned the approaching Marines. When they
came ashore, the beach resembled a movie set rather than a real strip of
African beach: Cameras, bright television lights, and journalists trying
to bring the clandestine landing to world attention.
The lesson was clear. The old media had become a permanent feature of the
battlefield. “There is no longer a question of whether the news media will
cover military operations,” an officer graduating from the Army War
College argued in his thesis. “As in Somalia, journalists will likely
precede the force into the area of operation; and they will transmit
images of events as they happen, perhaps from both sides of any conflict.”
Attempts to leash the media would not be feasible anymore; “efforts at
control are meaningless.”[17] Soon, joint and service doctrine would
acknowledge the new realities. By the mid-1990s the Pentagon began to
probe the embedding of reporters with military units in Bosnia. The Wall
Street Journal’s Thomas Ricks, for instance, traveled with the 1st Armored
Division into hostile territory. Because control of the media was not an
option any more, the Army reluctantly opted for cooperation.
Culturally and historically. the Marine Corps had a more relaxed attitude
toward reporters on the battlefield. As a light expeditionary force
composed of air, naval, and land components, the Marines have felt
themselves to be under constant pressure to legitimize their existence as
a separate service; they were culturally “paranoid,” in the words of one
of their generals, that they might one day be dissolved. As a result, the
Marines regarded their excellent reputation with the American people as
existential. The Corps, its officers are taught, exists not because
America truly needs it, but because the American people want it.
As a result, the Marine Expeditionary Force under General Walt Boomer, a
one-time public affairs officer, managed to get excellent coverage during
the 1991 Gulf War. The Army, by contrast, stonewalled. Chief warrant
officer Eric Carlson had devised a way of getting the news from the 1st
Marine Division to the rear in minutes. “We regarded [the media] as an
environmental feature of the battlefield, kind of like the rain. If it
rains you operate wet.” The smallest of the U.S. military services with
its specialization in amphibious operations excelled in this kind of “wet”
environment, and the Marines got a disproportionate share of the coverage,
notwithstanding that the Army in the northwest was doing the harder job
executing the famous “left hook,” one of the largest tank battles in the
history of warfare. Jamie McIntyre, the Pentagon correspondent of the
network that rose to global fame in the Gulf War, later used the same
comparison: “Wherever commanders go, they should plan for CNN. Like the
weather, we’ll always be there — just another feature on the battlefield
terrain.” What is most remarkable about McIntyre’s and Carlson’s
statements is that they were made in the early 1990s.
Communication between the soldiers at the front and their families at home
still had its classical features in 1991. It was time delayed, narrated
(not illustrated), and largely reliant on logistics provided by the
military. Soldiers received field mail, and they stood in long lines at
rear bases to call home. Neither the internet, nor its more sophisticated
second incarnation, had yet arrived on the scene. Mobile phones were still
the exception among the press corps, not to mention local civilians in the
area of operation. Videophones did not exist. Blogs were unheard of. Chat
rooms still evoked images of elderly ladies clustered around coffee
tables. When America went to war with Saddam Hussein the second time, the
information environment had undergone a drastic revolution of historic
magnitude. Two trends demonstrate the sea change.
First, soldiers became journalists. Abu Ghraib was only the peak event
that brought a general trend to popular attention. Enlisted soldiers carry
mobile phones, digital cameras, and they have access to the internet at
their bases. Most of them write emails to their friends and family back
home, occasionally attach a picture. Some publish their thoughts on blogs
to anybody who is interested and upload their images and videos. Chris
Missick, 24 years old and with the Army’s 319th Signal Battalion, was one
of those online chroniclers. He wrote the blog A Line in the Sand. Missick
candidly describes and questions the effects of his leisure activity:
Never before has a war been so immediately documented, never before have
sentiments from the front scurried their way to the home front with such
ease and precision. Here I sit, in the desert, staring daily at the
electric fence, the deep trenches and the concertina wire that separates
the border of Iraq and Kuwait, and write home and upload my daily
reflections and opinions on the war and my circumstances here, as well as
some pictures I have taken along the way. It is amazing, and empowering,
and yet the questions remains, should I as a lower enlisted solider have
such power to express my opinion and broadcast to the world a singular
soldier’s point of view?[18]
The spectrum of military weblogs, or milblogs as they are called in the
community, is as broad as American society and the soldiers and officers
that are recruited from it. Grey Eagle is the name the writer of
“afemalesoldier.com” has given herself. She describes herself as a mother
of two sons, serving in the 101st Airborne Division as a medic. Major
Michael Cohen, a doctor in a combat support hospital near Mosul, had his
blog shut down after he gave too many details of a suicide bomb attack
just before Christmas 2004. “Levels above me have ordered, yes ORDERED, to
shut down this website,” the doctor complained in his last posting.
The online diaries are only part of the larger picture. “Today, every
soldier has a cell phone, beeper, game device, or laptop, any one of which
could pop off without warning. Blogging is just one piece of the puzzle,”
said Lieutenant Colonel Christopher Conway, a DoD public affairs officer.
Indeed digital cameras and mobile phones have had a politically more
significant effect so far. On October 25, 2006 Bild, Germany’s most
popular daily tabloid, published five pictures that shocked the country.
They showed German soldiers posing with body parts in Afghanistan, with a
skull as a souvenir or a hood ornament. One soldier, his pants down, had
himself photographed with a cranium in a sexually explicit position. As in
the case of Abu Ghraib, the Bundeswehr’s scandal images were dated,
digitally photographed by low-ranking soldiers, and only unfolded their
devastating political impact after they spilled over into the mainstream
media.
Second, not only soldiers are reporting from today’s war zones; civilians
do as well. In Iraq, rising levels of violence force families to stay
indoors and deal psychologically with the death, robberies, kidnappings,
and explosions on their own. Writing, for many, becomes a way to deal with
the stress. “You have two choices—take a valium, or start a blog,” wrote a
24-year-old Iraqi woman who opted for the latter. The former computer
programmer chose the nom de plume “Riverbend,” and started the diary
Baghdad Burning. “That’s the beginning for me, I guess,” she commenced on
August 17, 2003, anonymously. Today she is one of the most prominent Iraqi
bloggers; her gripping first person accounts have made it into two books
and into the New York Review of Books. She covers many aspects and details
of civilian and political life in Iraq. Islamist militias sending
Kalashnikov bullets by mail to force some of Baghdad’s residents out of
their homes; women, even female Christians, increasingly wearing the
all-black hijab as a protective device; the popular reaction to Saddam
Hussein’s death sentence; and everyday issues like electricity supply. On
August 19, 2003, just hours before a large bomb destroyed the UN compound
in Baghdad and prompted the world body’s pullout, the young author
published a more representative account of her daily frustration:
Today a child was killed in Anbar, a governorate north-west of Baghdad.
His name was Omar Jassim and he was no more than 10 years old, maybe 11.
Does anyone hear of that? Does it matter anymore? Do they show that on Fox
News or CNN? He was killed during an American raid—no one knows why. His
family are devastated — nothing was taken from the house because nothing
was found in the house. It was just one of those raids. People are
terrified of the raids. You never know what will happen — who might be
shot, who might react wrong — what exactly the wrong reaction might be . . .
The weblogs’ lifespan and the frequency of contributions vary, as do their
political positions. In that respect they mirror the military’s blogs.
Some, like Iraq the Model, are pro-American. Two of the three Fadhil
brothers posting on that blog even met President Bush in the White House
during a sponsored tour of the United States. Others, such as the author
of A Star From Mosul or Sooni and dozens more hide their real names for
security reasons. Salam Pax, an architect and translator who started
publishing his diary online under Saddam, revealed his identity to a small
number of journalists; he got very high visibility in the American media.
Some reviewers in the mainstream media even called him the “Anne Frank of
the war in Iraq.”[19] In the insecure and terrorized environment of
wartime Iraq, it becomes increasingly difficult to work for professional
reporters. That resulting narrative void is increasingly filled by
detailed first-person accounts published online. Freely available blogging
software is currently being improved to allow publication in Arabic. As
the internet penetration in the Middle East increases, and news
consumption patterns change, the current dynamic will doubtlessly gain
momentum.
A Line in the Sand, the German skull affair, and Riverbend exemplify a
larger shift. Investigative journalists do not have to be present to have
an impact. The Columbia Journalism Review, the profession’s leading
periodical in America, jokingly but tellingly quoted Thomas Dworzak of
Magnum suggesting that Lynndie England, the female soldier in some of Abu
Ghraib’s scandal pictures, should have won a Pulitzer Prize. Indeed the
journal has a point. The new media, to paraphrase CNN’s McIntyre, will
even more, like the weather, always be there — just another feature on the
battlefield terrain.
But this insight still falls short of appreciating the real and
strategically more significant change. Understanding the media and the
internet as a permanent feature of the battlespace is progress, certainly.
It took decades for the U.S. defense establishment to fully realize the
implications. But this metaphor’s mindset is stuck in the first digital
age, in the old paradigm. In the first digital age the breakthrough was an
accelerated, instantaneous information flow: CNN’s Peter Arnett on the Al
Rashid Hotel’s roof in 1991, or his colleague Walter Rodgers in an Abrams
tank in 2003, broadcasting live. But their coverage was still traditional
broadcasting, a one-way street, delivered by a large company. The second
digital age, by contrast, is marked by interactive communication, a
two-way street, frequented by individuals. The consequences are profound.
In guerrilla warfare, the fish is not swimming in the ocean any more. It
is, if one likes to stick to the metaphor, rather using the water as part
of its fabric, more like a jellyfish. The information environment does not
stay external to the organization any longer, neither for the U.S. Army
nor for its enemies. It is flooding the hierarchy from the bottom up, and
enabling new forms of networked organizations. Consequently it is, in the
army’s case, more appropriate to look at the wet environment as an
innovative mode of internal communication, or, in the case of a militant
networked organization, as its lifeblood, or operating system.
U.S. military connects
The American armed forces are the most modern and best-equipped military
organization on the planet. They are also the world’s most intellectual
and prolific army; its officers fight and write. Each service has several
professional periodicals; the best ones are the Navy’s Naval War College
Review, the Marine’s Marine Corps Gazette, the Air Force’s Airpower
Journal, and the Army’s Parameters. In the 1950s, one such publication,
Military Review, explicitly mentioned the journal’s mission on each
edition’s first page. “The Military Review has the mission of
disseminating modern military thought and current Army doctrine concerning
command and staff procedures of the division and higher echelons and to
provide a forum for articles which stimulate military thinking.” The
quality of the articles in those journals varies, but often it is
stunningly high. Analytical, critical, constructive, they foster a culture
of open dialogue on issues of relevance to commanders.
So it comes as no surprise that some officers quickly saw Web 2.0’s value
as a publication platform. While the Marine Corps was traditionally better
at external public affairs, the Army took the lead in harnessing the
second digital age’s new gadgets for its internal purposes. The two prime
examples are CompanyCommand.com, which caters to the Army’s approximately
3,300 current company commanders, and PlatoonLeader.org, a highly
successful learning platform for the land force’s circa 12,000 platoon
leaders.
The websites’ purpose is to facilitate dialogue among junior leaders and
optimize the organization’s ability to adapt to an ephemeral operational
environment. The conversation takes place on “on front porches, around
HMMWV hoods, in CPs [Command Posts], mess halls, and FOBs [Forward
Operating Bases] around the world,” the welcome statement says. The front
porch is a hint at the website’s founding history. At the end of the
1990s, the two captains Nate Allen and Tony Burgess both commanded
companies in separate battalions of the same brigade, based in Hawaii.
They happened to be next-door neighbors and spent many evenings on their
Hawaiian front porches comparing notes. “How are things going with your
first sergeant?” they would ask, or: “How did your company live-fire work
out?” Realizing the positive impact of their peer conversations, the two
majors wrote a book and posted it on a website. Through this initial
publication they got in contact with another captain who proposed to model
a website on alloutdoors.com, an online switchboard for hiking and
survival advice, such as how to skin a squirrel. Allen and Burgess,
together with a dozen more captains, among them Pete Kilner and Steve
Schweitzer, decided to adapt the outdoor model and go ahead. “Such a site
for company commanders would replicate, in cyberspace, their front porch,”
as Dan Baum observed in The New Yorker.[20] The active-duty entrepreneurs
did not ask the Army for permission, nor for financial support, and they
registered their project with a .com address, not on the U.S. military’s
.mil internet domain.
Soon the Army discovered the value of the sites, and included them in its
official information network. Army Major General Peter Chiarelli, then
commander of the 1st Cavalry Division, started another site, CavNet
initially, that ran on the U.S. military’s “Secret Internet Protocol
Router Network,” or SIPRNET. The secure system is built on hardware that
is separate from the civilian internet, and only accessible from special
computers. SIPR’s downside is that it is not as widely accessible as the
internet. In Iraq it is available at the battalion level, but not at the
company or even platoon level.
CompanyCommand.com grew to 6,200 members in 2006, when the site was viewed
about a million times. “Today’s army is changing so fast, that people at
the high end don’t always know because they haven’t lived it,” said
Schweitzer, one of the site’s administrators. For the first four years the
internet site was entirely open to the public, and to the enemy. On
February 8, 2004, Pentagon correspondent Thomas Ricks published an article
in the Washington Post that prominently features the site. Overnight, the
traffic skyrocketed. Its founders — not the Army — then decided to limit
access to the site to professional U.S. soldiers, mostly captains.[21] The
closure, Kilner points out, made the online community much more powerful
and much more successful. “It’s not just information; it’s a personal
story, and commanders are able to connect with their peers who share their
knowledge. The forum fosters a powerful sense of shared purpose among
members.” It works not unlike MySpace: There are profiles with photos,
bios, and mostly information on a soldier’s professional background, all
focused solely on being a more effective company commander. “The
commanders see that the forum really is by and for them. They see others
who share their challenges and experiences; they see their own face and
those of their comrades, and that deepens their sense of professional
identity. The learning that occurs in and through the online connections,”
Kilner said, “has a real impact on the war.”[22]
The private initiative is today partnered with in the official Battle
Command Knowledge System, the Army’s institutionalized system of
chat-rooms and internet blackboards, and is supported by the United States
Military Academy and government grants. Yet the establishment understood
the site’s logic of a peer-to-peer culture that generates commitment. “We
don’t want to over-control,” said William Wallace, V Corps commander
during the invasion of Iraq and now head of the Training and Doctrine
Command, TRADOC. “There’s a certain amount of pride in these communities
in thinking that they operate outside the institution.”[23] The
institutional Army does not interfere with the operations of the
CompanyCommand and PlatoonLeader sites.
The Marines prove the Army right. In May 2006, 2nd Lieutenant Andrew
Schilling published an article in the Marine Corps Gazette which scored in
one of the magazine’s essay contests. “It is hard for a Marine to admit
when the Army does something better,” he writes, and goes on chastising
the Corps for not making better use of peer technology, such as
CompanyCommand and PlatoonLeader. The sites, the junior officer argues,
are “superior to anything the Marines have because they treat their users
as peers.”[24] This is not to say that the Marines do not use the internet
at all, but they do so outside the service’s control and awareness.
Schilling goes on describing how his platoon, just as many others in The
Basic School (TBS), used an internet site on a private server to keep
track of each other, provide study assistance, post photos, and to
organize the platoon’s activities. The young officers share their gouge
online — jargon for tips, templates, study guides and the like.
Navygouge.com is one of those sites.
The new media’s use on the battlefield is not limited to the internet.
“Most value is not created online,” said Kilner. “It’s in the chow hall,
it’s on the Humvee.” The online part merely amplifies the face-to-face
interaction. The CompanyCommand team, for instance, just installed a
podcasting capability and plans to equip commanders on their way to
Afghanistan with new iPods, fully loaded with video-podcasted interviews
with fellow commanders on their way out. Cell phones and private digital
cameras are used both for documentation and calling home. Commercial
gadgets, sometimes superior to the Army’s own equipment, are increasingly
used to gather and document intelligence. “Take pictures of everything and
even, more importantly, everyone. The right photo in the right hands can
absolutely make the difference,” one captain recommended online. One
particularly impressive example is the use of Google Earth by some U.S.
officers: The mapping tool is used in Iraq to document the personal
conversations between locals and soldiers. After a patrol returns back to
base, it becomes possible to add content to the map and document relevant
details of conversations with civilians and local leaders, and so create a
spatially and temporally mapped track-record of trusted or problematic
relationships that can be shared with other soldiers.
One effect is that today’s wars are the best documented operations in
history, in all their facets, including the ugly ones. And these may even
be documented by the troops themselves: by the Bundeswehr soldiers in
Kabul, by U.S. Army reserves in Abu Ghraib, and by British combatants in
Camp Breadbasket. And here again, the boundaries between the technology’s
external and internal use are blurring. War leaves a heavy psychological
footprint on its participants, on raided families or ambushed convoys. The
personal strategies to deal with this luggage are very different. Some
retreat and do not talk about what they saw and did their entire lives;
some do the opposite and publish blogs or leak images they find morally
repugnant to the press, even if they ware made for internal consumption
only. Just enter “Iraq” and “IED” as a search term on Flickr, a public
photo album, on YouTube, a public amateur video collection, or on MySpace.
But the positive effects of this development should not be overlooked; in
fact, they outweigh the negative ones. Junior military leaders recognize
this. “Our enemy is already using IT to his advantage. Information about
terrorist targets, schematics, tools and even how-to manuals are readily
available on the web,” Schilling wrote. “It is time for us to do the
same.” Others mirror this demand. “If we don’t mirror the insurgency with
our social networking and rapid transfer of knowledge, then soldiers’
lives are put at even greater risk. Insurgents watch our forces closely,
and learn what tactics are effective. We must do the same,” said Ron
Dysvick, who oversaw the design and implementation of the Army’s Battle
Command Knowledge System.[25] Arguably the American military — and some of
its NATO partners — have already begun to go down this road. Often,
though, without the explicit consent or control of its senior general
officers. The new counterinsurgency doctrine has recognized this trend:
Even the U.S. military is sliding towards a network organization as junior
leaders use cell phones and internet connections to solve problems and
resolve conflicts without going up the chain of command.[26]
The trend is best illustrated by the Army’s response to the remarkable
history of one the Iraq War’s most deadly weapons, the IED.
Sophisticated low-tech: IEDs
In Iraq,the American Army was forced to innovate tactically against its
will by a remarkably adaptive enemy. The insurgency’s most effective
weapon is the roadside bomb. Its fast rhythm of the changing tactics and
counter-tactics illustrates the altered organizing patters of war in the
second digital age.
The political context matters here. The administration’s stunning lack of
planning and its dismal management of what initially had been called the
“post-combat phase” is now well documented.[27] After the regime fell in
April 2003 and the “mission” was prematurely declared accomplished, an
absence of strategic guidance characterized the following months: a vacuum
for Iraqis to seize the initiative. Public buildings were looted, weapons
caches emptied, law and order, however repressive it was under Saddam,
broke down. When action was taken, the Iraqi Army dismantled and
de-Baathification enacted, it sent large numbers of trained and humiliated
fighters into unemployment. Humiliating entire families in nightly
“cordon-and-search” operations and, in some areas, interrogating nearly
all fighting-aged male Iraqis did not help to win the civilian
population’s hearts and minds either. A dangerous mixture of
disappointment with the new occupier, spreading anarchy and crime in a
society divided along sectarian lines, ready availability of huge
stockpiles of weapons, and a constant trickle of radical foreign fighters
entering the country through its unprotected borders, began to energize
the insurgency. The coalition’s lack of leadership and strategic vision
trickled down the military’s hierarchy. Clear orders were absent. Even the
commander’s intent, a senior leader’s concise statement about the purpose
of an operation and basis of any “mission command,” remained utterly
unclear. Yet the situation on the ground required tactical action. The
occupiers had to react; it was Auftragstaktik without the Auftrag, or mission.
Still, despite the confusion, the nascent insurgency faced the most
technologically sophisticated army in the history of warfare. The fighters
had to match technology with cunning. Two months into the occupation, one
of the classic weapons of an insurrection entered the stage: the
Improvised Explosive Device, or IED. The term was new, but the idea was
not. In Vietnam, booby-traps caused many casualties; the mujahideen in
Afghanistan even used bicycles filled with explosives against their Soviet
occupiers. Concealed bombs are the weak’s weapon of choice. Nothing
epitomized the Iraq war’s nature better than the insurgency’s signature
weapon. The IED resembles the insurrection itself: it takes many forms, it
is difficult to identify, and its sophistication has grown tremendously
since 2003.
The insurgency’s strategy followed an old rationale. T.E. Lawrence, better
know as Lawrence of Arabia, was a British military advisor to Arab tribes
during the First World War. The Ottomans had occupied Medina. An Arab
offensive in June 1916 on the city’s Turkish garrisons was squashed, and
the Arabs beaten back. Instead of repeating an open attack, Lawrence
recommended attacking the Ottoman supply lines: the Hejaz railway, the
Trans-Jordanian railway, and the Damascus-Aleppo connection. “Our ideal
was to keep his railway just working, but only just, with the maximum of
loss and discomfort to him.”[28] Iraq’s insurgents seem to have studied
the Arab Revolt well. One of the coalition’s must vulnerable points is
supply, particularly fuel supply. In August 2005, for example, the Army
1st Corps Support Command’s convoys suffered about 30 IED attacks per
week. Support units are not trained and equipped for combat, which makes
them easy and efficient targets. The long tail of small hits pushes up
costs and depresses moral.
Bombs are hidden behind signs or guardrails, concealed under rocks or
trash littered at the shoulders of roads all over Iraq. Some of the
devices were hidden in the carcasses of dead dogs, rotting in Baghdad’s
summer heat. Later, vehicle-borne IEDs were developed, with special
drop-mechanisms in the car’s belly — or the entire vehicle driven into the
targeted convoy and exploded there. During the summer of 2003, the
bomblets were still small, built out of mortar or single 152mm rounds. The
insurgency’s main targets were soft-skinned Humvees. As the coalition’s
armor improved, improvised mines grew bigger and more sophisticated. The
firepower of multiple heavy artillery munitions, stacked anti-tank mines,
or 500-pound Russian-made aircraft bombs was augmented with locally
available chemicals. In 2006, army units were even losing large numbers of
their once-invincible 63-ton, heavily armored combat vehicle, the M1A1
Abrams tank. One commander tells the story of a Bradley, a 35-ton fighting
vehicle, which was literally blown into the air and broken in half by an
IED; its crew was killed instantly. Not only is the bombs’ power stunning,
so is the insurgency’s ability to innovate. IED trigger mechanisms are an
example: Initially the mines were hardwired, and electronically ignited by
an observer. GIs learned to spot the wires and to take out the operator at
the end of the line. The insurgents then started to use cell phones,
garage openers, remote controls for toy cars, or hand held radios. The
army responded with Warlock, a frequency-jamming device. Insurgents
reverted to hardwiring the bombs, or mechanical triggers such as pressure
plates or even water hoses.
Ambushes are organized by cells. The insurgent network’s overall structure
is decentralized, which makes it difficult to penetrate small personalized
groups of trusted activists and gather intelligence. A typical IED cell
has 6 to 10 members with specific skills, innovative bomb builders,
someone to transport and place it, a spotter to watch advancing patrols, a
triggerman, and often a cameraman. (Many counterinsurgency missions in
Iraq and Afghanistan equally embed a combat-camera team to counter the
spin of those images[29]). More than 100 such cells reportedly operated in
Mesopotamia in 2006. U.S. intelligence officers believe they receive some
guidance from the larger insurgent networks and foreign terrorist
organizations, such as Ansar al Sunna or al Qaeda in Iraq.[30] Their
success is stunning. More than 45 percent of the more than 3,000 U.S.
fatalities in Iraq have been caused by roadside bombs; they have maimed or
wounded more than 11,000 Americans. And their lethality is on the rise: of
100 recent fatalities, 67 were inflicted by IEDs.[31] The psychological
impact on soldiers on patrol cannot be underestimated, even if they are
not hit. The weapon’s low price — in early 2006 the street price for a
152mm artillery round was $100 to $200 — its adaptability, and its
simplicity likely make it a permanent feature of the battlefield of the
future.
“We have a very adaptive enemy,” said Brigadier General Joseph Votel,
director of the Army’s IED Task Force, in a closed door briefing on the
threat situation to the Senate’s Armed Services Committee on November 1,
2005.[32] Votel told the senators that this enemy is able to buy and learn
to use the bombs’ components via the internet. Senator John Warner, the
committee’s chairman, called it “astounding” that insurgents were able to
make IEDs using information available from open sources and commercial,
off-the-shelf technologies. Three months later, in February, the
congressional research service published a report on the IED threat and
available countermeasures. “The Iraqi insurgents make videos of exploding
U.S. vehicles and dead Americans and distribute them over the Internet to
win new supporters,” it said. The videos demonstrate that the Americans
can be hit, and that it is easy to do.[33] Mirroring the executive’s as
well as lawmakers’ threat perception, the Pentagon’s Joint IED-Defeat
Organization, led by Montgomery Meigs, a retired four-star Army general,
was outfitted with an impressively large budget of $3 billion.
Cells advise their technical skills on the internet, with manuals
outlining how to build explosives, as well as video documentation of
attacks. “The Internet has changed the nature of warfare,” said Lieutenant
Colonel Shawn Weed, an Army intelligence officer based in Baghdad.
“Someone can learn how to build a new bomb, plug the plans into the
Internet and share that technology very quickly.” This is an apt
description of peer-to-peer networks, operating on the basis of open
source knowledge, where the cells’ activists are really co-developers.
It’s essentially like Wikipedia, just less easy to access.
Half a year later, on May 10, 2006, Lieutenant General Karl Eikenberry,
then commander of the Combined Forces Command in Afghanistan, was asked at
a press conference whether the IED threat had migrated from Iraq to
Afghanistan. Eikenberry said there was “no conclusive evidence” of a
migration of foreign fighters. But he said that his command was observing
a steady increase in the sophistication of roadside ambushes in both
theaters. In the first eleven months of 2006, the number of IED attacks in
Afghanistan rose to 1,297, up from 530 during the same period in 2005.[34]
“These are the kinds of skills that, very frankly, in today’s information
age can be gleaned from the internet,” Eikenberry said, “and be improved
by a force that’s operating against you over time as they continue to
adapt their own tactics.”
Not only have development cycles accelerated, various splinter groups
loosely coupled their tactics. Generally technical superiority is seen as
a battlefield advantage of modern armies. F-22s and B-52s certainly are
farther advanced that AK47s or 152mm shells. But the complexity and costs
of modern weapons systems are also a disadvantage; their coupling is tight
and technologically complex. Once a force is equipped with blue force
tracker and trained and set up to inter-operate jointly, it becomes very
difficult to just plug in a partnering coalition force. Insurgencies do
not face this problem. Barriers to entry, to use, and even to improved
technology are low if the machinery is simple and standardized. Knowledge
and techniques can easily be moved from one organization to another. RSS
feeds are an example from today’s internet. The technology is now a
worldwide standard. IED tactics are an example of today’s insurgency, and
they are spreading as well. Locally from Shia dominated neighborhoods in
Baghdad to Sunni quarters, and globally, from Chechnya or Iran to Iraq and
from there to Afghanistan. Traditionally insurgent movements operated in
their own countries, the FLN in Algeria, the IRA in Northern Ireland, the
FARC in Columbia. Direct cooperation and the direct exchange of knowledge
was very rare. Although some imitation of tactics took place before the
internet, Web 2.0 has significantly increased the transnational character
of guerrilla warfare.
The enemy’s sheer tactical innovation speed outperforms the Army’s
traditional learning and adaptation routines. The first official doctrinal
document on how to react to the bombs, Field Manual Interim (FMI)
3-34.119, “IED Defeat,” was published in September 2005, literally years
after the bombs’ debut in theater. Learning and adaptation cycles
accelerate rapidly, as insurgents inadvertently operate according to the
software industry’s “release early, release often” paradigm. Innovative
applications are considered imperfect by definition and are constantly
being improved by their users, thereby pushing up innovation speed. “There
is a constant cycle of new technologies, counter technologies, and
counter-counter technologies,” Weed pointed out in an interview with
Aviation Week & Space Technology.[35] The Army recognized that the
adversary’s use of new technologies like digital cameras, mobile phones,
and the internet accelerates this dynamic, and seems to heed junior
officers’ advice to mirror this structure. General Wallace, the head of
TRADOC, highlighted the internet-based learning tools at the army’s
disposal and singled out SIPRNET, which offers a “collection of the
current techniques being used to emplace and employ IEDs and indeed some
techniques associated with how one might defeat those IEDs or at least
identify them.”[36] This thought has entered the most authoritative
contemporary doctrinal document: “Learning organizations defeat
insurgencies,” Petraeus’s counterinsurgency manual says, “bureaucratic
hierarchies do not.” Insurgent networks, however, have an organizational
advantage over the hierarchical military they face.
The enemy’s new operating system
The most spectacular internet-based propaganda operation in the history of
terrorism was staged on May 11, 2004. A video showed Nicholas Berg, a
civilian from West Chester, Pennsylvania, dressed in an orange jump suit,
bound, sitting on the ground, masked men behind him. What follows is a
gruesome beheading: Berg’s startled expression, a knife sawing through his
neck, screams, blood. The now-notorious five-and-a-half-minute video first
appeared on a website of the militant group Muntada al-Ansar al-Islami,
initially identified by a Reuters journalist. CNN and Fox News, among
others, immediately downloaded a copy of the video. Within less than 90
minutes, the file allegedly disappeared from the site; al-Jazeera was not
able to obtain it from there. But the video spread to better-known forums
and mirrored sites, and within 24 hours was downloaded approximately half
a million times. The international media, talk shows, and editorial pages
voiced outrage for weeks. President Bush condemned the beheading. The
26-year old American victim and the alleged perpetrator, Abu Musab
al-Zarqawi, instantly became household names.
The execution illustrates a general trend: An interface between the
mainstream “old media” and the radical “new media” is evolving. After a
terrorist attack has been committed, several organizations and groups
usually claim responsibility and “ownership.” The problem for journalists
is to recognize the authentic one. While the intelligence services assess
the authenticity of these claims behind closed doors, the mainstream
media’s experts do so publicly. Yassin Musharbash, a Jordanian-born German
journalist, is one of these experts. He writes for Spiegel Online,
Germany’s most widely read online paper. Musharbash gives an example of
how the radicals have adapted to the need to be authentic. In summer 2005,
al-Hisba was one of the preferred sites to find official communiqués. One
day the site’s operators demanded a registration including a valid email
address. They even offered the possibility of accreditation for
journalists. Eventually, in early 2006, al-Hisba published a list of all
Western media organizations who used the site as a source for their
reporting, complete with logos. It included the Associated Press, CNN, the
Swiss News Agency, ABC News, and others.[37] The forum later even offered
RSS feeds, so that communiqués from Iraq would be “pushed” directly onto
journalists’ desktops.
But a focus on propaganda, or “public affairs,” as modern armies prefer,
is too narrow. The overall number of radical websites has grown rapidly.
Gabriel Weimann, a long-time Israeli observer of the field, counted 12
eight years ago. Today the number is hovering around 5,000.[38] In
practice that figure is impossible to determine because many of the sites
are too ephemeral to be indexed by search engines. While some radical
websites strive for media interest and cultivate their media readership,
others have an extremely short life-span and try to avoid general
publicity. This highlights the web’s multiple uses: it facilitates not
only public affairs or funding of radicalized organizations, but also
recruitment, engineering, training, and syndication.
Recruitment. While recruitment of qualified soldiers is increasingly
difficult for Western armies, the opposite is true for radicalized
Islamist groups. The web’s anonymity has stunning consequences: Junior
officers discuss TTPs (tactics, training, and procedures) and may even
counsel higher ranking fellow officers, and ordinary bloggers criticize
and speak out despite oppressive regimes. Similar effects exist among
radical groups: While in the real world very few people would admit
membership in al Qaeda, in the jihadist internet, sympathizers abound,
even if only a small fraction of online-jihadis have connections to the
organization’s hard core. Recruits can easily be contacted in chat- rooms
or by email, even anonymously, or they might decide to take action
independently. “The radicalization process is occurring more quickly, more
widely and more anonymously in the Internet age,” the U.S. government’s
National Intelligence Estimate 2006 concluded, “raising the likelihood of
surprise attacks by unknown groups whose members and supporters may be
difficult to pinpoint.”[39]
Engineering. On September 1, 1992, Ahmed Ajaj, a Palestinian operative,
was caught at New York’s John F. Kennedy airport carrying two books of
handwritten notes for explosives, six printed bomb-making manuals, and
instructional videotapes. His faked Swedish visa had aroused suspicion.
Ramsi Yousef, who traveled with Ajaj in the same plane, was temporarily
arrested because he had no visa, and then released. In February 1993
Yousef, who learned his craft in an Afghan al Qaeda camp, carried out the
World Trade Center bombings. Such a scenario would be highly unlikely
today. Throughout the 1990s al Qaeda operatives wrote and used the
Encyclopedia of Jihad, a multi-volume manual that covers explosives, small
arms, grenades, mines, espionage, reconnaissance, sabotage, interrogation
and counter-interrogation tactics, infiltration, tank ambushes, first aid,
etc. Written probably in Afghanistan and Sudan, the field manuals were
initially handled restrictively and confidentially. After the 2001
offensive against the organization’s strongholds in Afghanistan, the
papers that once filled suitcases were first conserved on CD-roms, and
probably in 2003 made available on the internet.[40] Today dozens of
different versions, continually updated and improved, are on offer. Even
films that explain the use of a suicide bomber’s belt can be obtained. Not
all peer-produced explosive user guides or poison recipes offer workable
information, just as not all Wikipedia articles are of high quality. But
even if sometimes flawed, open-source terrorism works.
Training. Ali Abdelsoud Mohammed was a worldly, charismatic al-Jihad
activist, who in the early 1980s was sent to the United States by
al-Zawahiri to infiltrate the CIA. He did not succeed in his original
mission, but Mohammed married an American, enlisted in the U.S. Army, and
was trained as a Special Forces trooper in Fort Bragg, North Carolina. In
the late 1980s he traveled to Afghanistan and trained al Qaeda’s far-flung
recruits and fighters there (bin Laden supposedly took Mohammed’s first
class); later he even worked with Hezbollah in Lebanon.[41] Today,
training manuals do not have to be smuggled out of the United States. The
Islamist network attempted to replicate the Afghan or Sudanese training
experience online. “Mu’askar al-Battar” is the best known training journal
published online; it means the training-camp of the sword. Different
issues feature discussions of light weapons, rocked-propelled grenades,
and urban ambushes, as well as organizational tips on how to form a cell
structure or increase physical fitness. “Oh Mujahid brother, in order to
join the great training camps you don’t have to travel to other lands,”
outlines the inaugural issue published in early 2004. “Alone, in your home
or with a group of your brothers, you too can begin to execute the
training program.”[42] The objective was to delegate responsibility and
initiative to entrepreneurs on lower levels, in what resembles a
terrorist’s crude version of Auftragstaktik. Each of the magazine’s
editions provided email addresses, encouraged submissions, and offered
personal assistance if needed.
Syndication. Just how much command and control remains in the hands of top
leaders is a matter open to dispute, even within the movement itself.
Bruce Hoffman argued that al Qaeda has both, a hard core able to retain
command and control of larger operations in a top-down fashion, and a
network of homegrown activists who take the initiative from the bottom
up.[43] While the first view is controversial, the second is not: Google
does not know how many entrepreneurs syndicate their services and
integrate them into their businesses; the insurgency does not know how
many individuals are using their tactics; and al Qaeda does not know how
many observers became supporters or how many supporters took up arms and
are combatants. The organization’s boundaries are so flexible that Beam’s
notion of the “unity of purpose” is helpful, with the unifier being the
jihadist ideology. The anonymous global community discusses the radical
methods, its self-understanding, and its future strategic direction.
Theological questions can be asked, and will be answered swiftly. Thanks
to al Qaeda and the internet, the ancient Ibn Taimiyya of Damascus is
today again one of the most widely read Arab religious scholars.[44]
Today’s jihad is fought with the methods and weapons of the 7th century
and the 21st century simultaneously. Al Qaeda is also a giant global think
tank, as indestructible as the internet itself.
STRG, ALT AND DELETE?
The trends described here have troubling consequences. They profoundly
affect social networks, the centers of gravity in today’s wars: within
insurgent groups, within the increasingly complex setup of
counterinsurgent actors, and, most centrally, with the neutral civilian
population — both in theater and at home. The interactive media have, at
first glance, two contradictory effects. They infuse both volatility and
stability: On the one hand, they make public support in the metropolis for
a protracted counterinsurgency campaign in far-away lands more volatile,
even if the political stakes are high. A permanent stream of bad news and
gruesome images from a protracted guerrilla campaign threatens to erode
even strong public resolve. And the U.S. military in Iraq is one of the
most isolated occupiers in history. There are no bars, no suqs, no
brothels — mostly for cultural and security reasons, but technology adds
to the problem: a chatting, cell-phoning, photographing, and
video-blogging occupying force maintains tight social connections to its
distant home communities, while local ties fail to develop. On the other
hand, though, the new media stabilize the insurgency and militant
movements. Even if the local population’s support for the insurrection is
waning as a result of a successful counterinsurgency campaign, radicals
are still able to make instant use of propaganda-driven operations to gain
moral, financial, organizational, and operational support from global
audiences. America faces some of the most adaptive and entrepreneurial
adversaries it has ever encountered militarily
What, then, can and what cannot be done? First the no-goes. It is neither
possible nor desirable to shut down internet in theatre, to black out
mobile phone networks, or to strip soldiers of their civilian
communication devices. Such methods are essentially illiberal strategies.
They send a wrong message, and they are economically and socially
unsustainable in a protracted counterinsurgency campaign. More important,
however, they ignore the new media’s positive potential.
For governments, perception management, or public diplomacy, has become
infinitely more complex. But simultaneously its prospects have never been
better. The use of government-controlled old media, such as the State
Department’s Cold War-style TV channels and foreign language radio
services is, for the most part, a waste of resources: more than $640
million are appropriated for “international broadcasting operations” for
2007. Such broadcasts are stuck in the old, one-way-street paradigm, and
cannot compete for credibility in an increasingly crowded media market in
the Middle East. Instead, programs that create durable social contacts and
linguistic skills through dialogue should receive more support, beyond the
traditional educational exchange schemes. Taxpayers’ money should instead
be used to aggressively promote internet and mobile phone penetration in
conflict-ridden areas, thus creating a platform for engagement of civil
society. Finally, governments should appreciate that the enemy is forced
to communicate openly, and adapt its intelligence services accordingly;
they need to be more open to recruits from minority groups with the
appropriate linguistic and IT skills.
For the military, the consequences are even more significant. As long as
the right ingredients are available — frustration, ideology, and the
know-how to take action — militant Islamic fundamentalism will be “just
another feature of the battlefield terrain.” The expectation to win,
consequently, should be replaced by a more humble hope to be successful;
the objective of extinguishing terrorism should be abandoned. But the
repercussions are not only conceptual, they are also organizational: The
U.S. military is culturally disposed to destroy the enemy rather than to
create stability in an alien environment. To change this, career
incentives and promotions schemes should be reorganized to place more
value on the specific skill-set required of a successful counterinsurgent.
The insurgent is not hampered by a bureaucratic hierarchy, so the
counterinsurgent should be able to fast-track the most talented
individuals into leadership positions. The “I had 500 cups of tea with the
locals, and now I’m out” problem also needs to be addressed. The practice
of rotating an entire unit out of an area after its commanders built
rapport with civilian leaders, and thereby cutting established social
contacts, should be overhauled. Unit cohesion should be weighed against
social cohesion with civilians in the war zone.
Development agencies should face the fact that they are engaged in a
counterinsurgency campaign if they operate in Afghanistan or Iraq. Just as
soldiers do “social work,” in Galula’s words, external nongovernmental
actors do “military work”: if their performance and their projects are
successful, international organizations and NGOs not only create
stability, they undermine trust and undercut social networks between the
insurgency and the civilian population. So it should come as no surprise
that aid workers are often seen as targets by the insurgents. Here, again,
the new media may alter the rules of the game in the future. Both long
distances between villages, such as in Afghanistan, as well as security
threats in an urban environment, such as in today’s Baghdad, limit the
mobility of civilians. The interactive media could in future — given the
necessary preconditions — be used for schooling, education, or political
participation, and thereby limit the insurgents’ impact on the population.
In many areas this is still unrealistic today, but it will not be so in
the future.
Finally the mainstream media should welcome the trends described here. The
new media are a new source for journalists and editors. Blogs give the
print media an alternative to quoting officials and insert fresh
perspectives, particularly if the security situation in a war zone makes
free reporting difficult. Independent specialists and experts could
validate the authenticity of blogs, even if their authors remain
anonymous. It is accepted practice to quote any Washingtonian “senior
official” without identifying him or her publicly; a similar convention
should be developed for civilians in war zones. The risk they take is much
higher than that of a White House leaker. Even anonymous interviews via
Skype may become an option.
The web’s emerging organizational patterns have not diminished the
significance of old, traditional businesses and their products. But
management, communication, supply-chains, R&D, production, administration,
marketing, customer relations, and competition itself are subject to
fundamental changes — changes that come with both great risks and great
opportunities. The same applies to the management of violence.
Notes
[1] Greg Grant, “Behind the Bomb Makers,” Aviation Week & Space Technology
(January 30, 2006).
[2] Seema Mehta, “Grief, comfort meet on MySpace,” Los Angeles Times
(January 24, 2007).
[3] The term peer-production was first introduced in a now classic article
by Yale University’s Yochai Benkler, “Coase’s Penguin, or Linux and the
Nature of the Firm,” Yale Law Journal 112:3 (December 2002); see also
Yochai Benkler, The Wealth of Networks: How Social Production Transforms
Markets and Freedom (Yale University Press, 2006).
[4] Doug Kaye, Loosely Coupled: The Missing Pieces of Web Services (RDS
Press, 2003), 131. The concept was originally introduced by the
sociologist Karl Weick in 1976, who applied it to educational
organization. Karl E Weick, “Educational Organizations as Loosely Coupled
Systems,” Administrative Science Quarterly 21:1 (1976).
[5] The term was coined by Wired’s Chris Anderson. His argument is that
the total commercial volume of low-popularity items can beat the volume of
high-popularity items — a logic that is easily transferable to an
insurgency’s low-intensity attacks and the occupier’s high-intensity
response. Chris Anderson, The Long Tail: Why the Future of Business Is
Selling Less of More (Hyperion, 2006).
[6] David Kilcullen, “Twenty-Eight Articles. Fundamentals of Company-level
Counterinsurgency,” Military Review (March-April 2006).
[7] C.E. Caldwell, Small Wars: Their Principles and Practice (University
of Nebraska Press, 1896, 1996).
[8] T.E. Lawrence, “The 27 Articles,” Arab Bulletin (August 20, 1917);
T.E. Lawrence, Seven Pillars of Wisdom (London: Bernard Shaw, 1926).
[9] Frank Kitson, Low Intensity Operations: Subversion, Insurgency,
Peace-keeping (London: Faber, 1971).
[10] Roger Trinquier, La Guerre moderne (Paris: La Table ronde, 1961);
David Galula, Counterinsurgency Warfare: Theory and Practice (Praeger,
1963); David Galula, Pacification in Algeria 1956-1958, MG-478-1 (Rand
Corporation, 1963, 2006).
[11] Lawrence Wright, The Looming Tower. Al-Qaeda and the Road to 9/11
(Alfred A. Knopf, 2006).
[12] Mao Zedong, On Guerrilla Warfare (London: Cassell, 1965), 93.
[13] The idea of a “unity of purpose” was brought up by a right-wing
American supremacist, Louis Beam — not because such an organization is
more practical, but less vulnerable. Kilcullen prefers a “common problem
definition.” David Kilcullen, “Counter-Insurgency Redux,” Survival 48:4
(Winter 2006); see also Louis Beam, “Leaderless Resistance,” Seditionist
12 (February 1992).
[14] Robert R Tomes, “Relearning Counterinsurgency Warfare”, Parameters
(Spring 2004).
[15] Audrey K Cronin, “Cyber-Mobilization: The New Levée en Masse”
Parameters (Summer 2006).
[16] Donatella Lorch, quoted in Thomas Rid, War and Media Operations. The
US Military and the Press from Vietnam to Iraq, (London: Routledge, 2007).
[17] Charles W. Ricks, The Military-News Media Relationship: Thinking
Forward (Army War College, 1993).
[18] John Hockenberry, “The Blogs of War,” Wired (August 13, 2005).
[19] Peter Maass, “Salam Pax is Real,” Slate (June 2, 2003).
[20] Dan Baum, “What the Generals Don’t Know,” New Yorker (January 17, 2005).
[21] Thomas E. Ricks, “Soldiers Record Lessons From Iraq”, Washington Post
(February 8, 2004)
[22] Pete Kilner, interview with author (January 27, 2007).
[23] Sandra I. Erwin, “Washington Pulse,” National Defense 90:627
(February 2006).
[24] Andrew P. Schilling, “Peers,” Marine Corps Gazette 90:5 (May 31, 2006).
[25] Ron Dysvick, CEO of Triple-I, quoted in Greg Slabodkin, “Army Lessons
Learned,” Federal Computer Week (July 17, 2006).
[26] U.S. Army/U.S. Marine Corps, Counterinsurgency, FM 3-24 (December 2006).
[27] Michael Gordon and Bernard Trainor, Cobra II. The Inside Story of the
Invasion and Occupation of Iraq (Atlantic Books, 2006); Thomas E. Ricks,
Fiasco. The American Military Adventure in Iraq (Penguin, 2006); Rajiv
Chandrasekaran, Imperial Life in the Emerald City (Knopf, 2006).
[28] T.E. Lawrence, “The Evolution of a Revolt,” Army Quarterly and
Defence Journal (October 1920).
[29] Cronin, “Cyber-Mobilization: The New Levée en Masse.”
[30] Grant, “Behind the Bomb Makers.”
[31] Brad Knickerbocker, “Relentless toll to US troops of roadside bombs,”
Christian Science Monitor (January 2, 2007).
[32] Defense Daily 228:20 (November 2, 2005).
[33] Clay Wilson, Improvised Explosive Devices (IEDs) in Iraq: Effects and
Countermeasures (Congressional Research Service, 2006), 1.
[34] Anthony Cordesman, “One War We Still Can Win,” New York Times
(December 13, 2006).
[35] Grant, “Behind the Bomb Makers.”
[36] William Wallace, quoted in Ann Roosevelt, “Army TRADOC works IED
issues,” Defense Daily 228:46 (December 16, 2005).
[37] Yassin Musharbash, Die neue al-Qaida. Innenansichten eines lernenden
Terrornetzwerks (Köln: Kiepenheuer & Witsch, 2006), 111.
[38] Gabriel Weimann, Terror on the Internet: The New Arena, the New
Challenges (United States Institute of Peace), 2006.
[39] U.S. Director of National Intelligence, “Trends in Global Terrorism:
Implications for the United States,” National Intelligence Estimate (April
2006).
[40] Musharbash, Die neue al-Qaida.
[41] Wright, The Looming Tower, 184-188.
[42] MEMRI, Dispatch 637 (January 6, 2004).
[43] Bruce Hoffman, “From the War on Terror to Global Counterinsurgency,”
Current History (December 2006).
[44] William McCants, The Militant Ideology Atlas, Combating Terrorism
Center (U.S. Military Academy, November 2006).
Thomas Rid is a Tapir Fellow at the Institut français des relations
internationales (Ifri) in Paris, at Johns Hopkins University’s School for
Advanced International Studies (SAIS), and at the Rand Corporation.
Previously he worked at the German government’s foreign policy think tank
Stiftung Wissenschaft und Politik and the American Academy in Berlin. He
is author of War and Media Operations. The U.S. Military and the Press
from Vietnam to Iraq (Routledge, 2007). Christofer Burger’s inspiration,
Marc Hecker’s counsel, and Pete Kilner’s fact checking, significantly
improved the text. The author thanks them.
---------------------------------------------------------------------
To unsubscribe, e-mail: infowar -
de-unsubscribe -!
- infopeace -
de
For additional commands, e-mail: infowar -
de-help -!
- infopeace -
de