The Archive Machine: the semi-automated systems that are interminably gathering data for storage and subsequent analysis, in search of insights into the state of national security, identifying threats, mapping networks, listing targets.

War Fever: the hot language of warfare, the ‘liveness’ of its imagery, the political discourse that makes space for the Archive Machine to do its work, to make it necessary, and then to extend it.

Over the past two terms of the Obama administration, the term “War on Terror” has fallen out of favour in discussions around the geopolitical position of the United States. The failings of the term to address the nature of this conflict, its problematic ‘boundless’ connotations, as well as its associations with the George Bush era are the obvious reasons for this. However, the War on Terror is not just written into the popular consciousness as history. The term occupies an ambiguous temporal space. It is not simply something that has happened, which can only be spoken of in the past tense. For the majority of those who experienced it in North America and Europe, the War on Terror was largely experienced as the battle for mass-psychology, and more specifically, about how we spatialise safety and security. Consequently, whether or not people believe that the war is ongoing matters a great deal. This ambiguity is continuously leveraged both by the mass media and in governmental politics. In the media, we find War Fever in reports about immigration, or in 24-hour breaking news coverage of terror attacks (which in fact may have little to do with the initial concerns or actors of the War on Terror). In governmental politics, War Fever is still invoked on both the left and the right as a means to foster public appreciation for the development and extension of observational apparatuses, racial profiling, and militarised policing tactics. This is presented as simply a minor side-effect of the complex systems that account for high-stakes national security and absolutely necessary when combatting a network-enemy, not a State-military-enemy. [0]

‘Chess is indeed a war, but an institutionalised, regulated, coded war, with a front, a rear, battles. But what is proper to Go [1] is war without battle lines, with neither confrontation nor retreat, without battles even: pure strategy, while chess is a semiology. […] The difference is that chess codes and decodes space, while Go proceeds altogether differently territorializing or deterritorializing it (make the outside a territory in space; consolidate that territory by the construction of a second, adjacent territory; deterritorialize the enemy by shattering his territory from within; deterritorialize oneself by renouncing, by going elsewhere…).’ [2]

The principles of the War on Terror may be ill-defined and ambiguous, but there are Archive Machines at work: documenting and logging; attempting to know the events of the war, if possible, before they actually happen. National security must operate on the basis of precognition, when an attack is seemingly ever-imminent. These Archive Machines structure, politicise, and ‘make queryable’ the environmental information of the State’s operations and the behaviours of the body of citizens, such that intelligence, trends, and projections can be produced. Whether or not the War on Terror is over, these apparatuses of security have become embedded in the daily tasks of the present State, and will remain so in the future. In this decades-long convergence of technological scope and high anxiety, they have entered a state of latent hyperactivity.

‘No, the technical structure of the archiving archive also determines the structure of the archivable content even in its very coming into existence and in its relationship to the future. The archivization produces as much as it records the event.’ [3]

There is a paradox we must engage with when attempting to understand the mechanics of the War on Terror: despite its covert, seemingly unknowable nature, it is surely the most documented war ever. There are two faces of the Archive Machine: not only one that gazes outward, wanting to understand exterior terror threats, but also a face that gazes inward, self-monitoring, which seeks to manage inter-institutional flows of information. On this face, every observation has its subsequent administration. Thus, the bureaucratic machineries produce their own archives, just as computers quietly generate logs of their operational status, recording the way they work. The meta-media has a greyness, ‘an unremarkableness that that can be of inestimable value in background operations’. [4] They prescribe operational frameworks, record-keeping, and relational hierarchies, and thus they provide a lens through which we can visualise the internal mechanics of security. Once-ignored documents suddenly take on new politically-charged contexts. In hindsight we might call them omens.

The underlying systems designed to make sense of these chaotic politics produce a quiet heat. This quiet heat: an uncomfortable ambient condition; one that becomes normalised; an inescapable environmental quality that affects rather than directs; it modulates. It is the side-effect, the by-product, of background processing. It should be loud, it is only quiet because it is kept discreet, contained. Quiet heat is the affective quality of computation during wartime, the product of the Archive Machines and the conditions of living with War Fever.


The Phoenix Programme, or in Vietnamese, Phung Hoang, ‘named after an all-seeing, mythical bird which, condor-like, selectively snatches its prey’. [5]

Phoenix: a mythological bird that dies in a spectacular flame, only to be reborn from its ashes.

Phung Hoang operated in Vietnam from 1966 to 1973. It was a CIA-led programme designed to identify potential actors within the Viet Cong (VC) and disrupt the communist Shadow Government in operation in South Vietnam. It aimed to answer the question: who is the enemy? Who knows who the enemy is? Disposition was suspected to change by time of day. Resistance movements in US-designated SECURE territories became active by nightfall. It was possible for any person in the village to play some role, however small, in the VC Infrastructure.

Phung Hoang sought to identify these people, record them, neutralise them. From a Phung Hoang manual:



On 1 June 1971, NGUYEN THI LAN, alias THANH LE and a 37 year old female, was arrested in Phuoc Hiep Village, Mo Cay District, Kien Hoa Province. She was arrested as a result of a specific targeting operation conducted by the National Police Field Force (NPFF). The operation was ordered by the District Chief. After her capture, she was sent to the Province Headquarters, where she was detained in the Province Compound. INSTRUCTIONS: Using the instructions contained in TAB A, initial submission of a Neutralization worksheet, and TAB B, submission of an alias/alternate job, fill in a Neutralization Worksheet based on the information contained in the above paragraph. [6]

The Phung Hoang Management Information System. Objectives:

‘Maintain files containing biographic data on identified members of the Viet Cong Infrastructure; Provide accurate statistical information on the neutralisation of the VCI; respond to inquiries by interrogating the database, thus eliminating the need for manual extraction and aggregation of data.’ [7]

Information is to be collected and notated by hand onto reports, then recorded into the PHMIS for later report generation on request.

The Phoenix Programme was just one of many Pacification Programmes that operated in Vietnam at the time (such as the Hamlet Evaluation System [8]), most of which were orchestrated from the Military Assistance Command Vietnam (MACV) headquarters in Saigon. Every aspect of the pacification programmes revolved around the practice of documentation: gathering data, recording the processes of the gathering of this data, producing new data through computational means, storing the data in databases, and reporting on the results of the data, et cetera. It was, according to many accounts, a laborious process. More information was always required, and more information meant more sifting, seeking ways to tabulate, and the work of interpreting subjective written text, the conversion of analogues to binaries. The life of a VC suspect was rendered as omitted space on card, or as magnetised space on tape, becoming a collection of categories and identifiers, a constituent datapoint in thousands of datapoints. Algorithms make these individuals outliers, or perhaps they blend in, ascribe to the mean. Better blend in, or maybe, better stand out.

This image is labelled ‘Miss Tran Thi Minh Huong, key punch operator, transfers programmed information from work sheets to IBM cards.’ In the image, there is a woman sitting at a desk. Her chair has armrests, and there is some fabric draped over the back of the chair. There is a keyboard on the desk. The keyboard is placed in front of a larger machine, which makes holes in the punch cards according to what she writes with the keyboard. To her left, there is an open binder. There is a sheet of paper with rows of information on it. She is using a ruler to underline the current row she is transcribing to punchcard. To her right, there is a window. The blinds are partially open. Around her, the wall tiles are patterned with a matrix of dots.

In order to arrive at the stage where data would be computed, it had to be prepared. This was a manual process: essentially all of the information coming back from the field was done by hand, mostly in the form of completed quantitative questionnaires or handwritten qualitative notes. Information was gathered from South Vietnamese government officials. Were grudges settled? Convenient names handed over for investigation?

Workers at the MACV headquarters had the task of interpreting this data, making it fit within parameters allowed for by the IBM computers. They worked as translators: translate the messy observations and chaotic politics of commanders and advisors, and make it fit into machines, scrawled handwriting, circles punched out of card or not. On card, on tape, in memory. Information became structured data, was archived, stored, retrievable at a later date.

‘He’d passed the cutting and pasting stage and begun making new headings for each of the nineteen thousand duplicate cards and arranging them according to place names mentioned, so that someday — not soon!— it would be possible to look at this information as it related to district, village, or city. Why hadn’t it been kept to these categories to start with? And why should he care? As with CORDS/Phoenix, officers had ventured out, asked questions, made notes, gone on to other posts. He longed to trip on a clue and follow it to some ravaging discovery — Prime Minister Ky spied for the Vietcong, or an emperor’s tomb hid millions in French plunder — but no, nothing here, all worthless; he sensed it with his fingers on these cards. Not only was the data as trivial and jumbled as those of CORDS/Phoenix, but also their time had passed. These three-by-five-cards served only as artefacts. In this they held a certain fascination.’ [9]

The label of the image says ‘Preparing the computer to supply some information from the memory bank is part of Miss Nguyen Thi Nhin’s job.’ In the image, we see a Vietnamese woman sitting in a chair. She is in front of a computer. The computer is an IBM 360. It has no screen. It is operated with dials and buttons. Information is displayed through rows of indicator lights. She is turning a dial on the machine. She is writing while she does this, onto a spreadsheet. Above the computer, there is the word THINK. The walls tiles have the same dot pattern as seen in the photo of Tran Thi Minh Huong. There is a fluorescent light illuminating the space. She has a telephone beside her. There is a clock above a filing cabinet. There are pages taped to the side of the filing cabinet. The pages look like they have a list of codes, or telephone numbers on them. Perhaps this is information that Miss Nguyen Thi Nhin requires regularly as she does her work.

The pacification data Miss Tran Thi Minh Huong and Miss Nguyen Thi Nhin are working on would be further processed. It could then be sent to Washington, where it might feature in Defense Secretary Robert McNamara’s reports to the press, or in briefings to the President on the progress of the US Armed Forces in Vietnam. It could be used to inform policy, to make a decision to send, or not to send, supplies to a specific location in Vietnam. It could indicate that Vietcong activity is becoming more influential in a certain district, or maybe that a key Vietnamese worker is in fact a VC informant. It could demarcate villages to be razed to the ground, its inhabitants relocated. It could map out Vietcong supply lines, routes for defoliation. Stories distilled, become numbers, statuses, locations, which taken together may add up to a neat categorisation. The process for establishing the veracity of this categorisation is unclear.

Official Government of South Vietnam data: 41,000 Vietcong suspects were assassinated in the Phung Hoang program between the years 1968-1972. [10]

‘Saigon, the center, where every action in the bushes hundreds of miles away fed back into town on a karmic wire strung so tight that if you touched it in the early morning it would sing all day and all night. Nothing so horrible ever happened upcountry that it was beyond language fix and press relations, a squeeze fit into the computers would make the heaviest numbers jump up and dance.’ [11]


‘If we have to use force, it is because we are America. We are the indispensable nation. We stand tall. We see further into the future.’ [12]

‘According to [the theory of raison d’état], the actions of the State, which is responsible for the life of the country and thus also for the laws obtaining in it, are not subject to the same rules as the acts of the citizens of the country. […] Raison d’état appeals — rightly or wrongly, as the case may be — to necessity, and the state crimes committed in its name (which are fully criminal in terms of the dominant legal system of the country where they occur) are considered emergency measures, concessions made to the stringencies of Realpolitik, in order to preserve power and thus assure the continuance of the existing legal order as a whole.’ [13]

‘Months of tracking ended in a matter of seconds. There was little reaction in the operations center. No cheers. No high fives. The Task Force was too professional for that. Frog and I shook hands. He was smiling. We were one team. We’d earned a victory, evident by the trucks smoldering in the monitor.’ [14]

September 30: Anwar al-Awlaki becomes the first US citizen to be killed in a US drone strike. Also killed in the strike was Samir Khan, editor of Al Qaeda in the Arabian Peninsula’s (AQAP) propaganda magazine Inspire. Al-Awlaki was in a car travelling east. The drone was piloted remotely from Camp Lemmonier, Djibouti. The strike location was across the Gulf of Aden, in the Yemeni province of Al-Jawf, east of the capital Sana’a. Al-Awlaki’s 16 year old son, who was searching the region for his father, was killed in a second strike two weeks later.

Born in New Mexico on the 30th April 1971, but raised in Yemen, al-Awlaki returned to the United States on a scholarship to study mechanical engineering at Colorado State University. [15] He spent a winter fighting for the CIA-backed Mujahideen in the final years of the Afghan-Soviet war. On his return to the States, he prioritised his faith over his studies, becoming an Imam at a mosque in San Diego, and later in Falls Church, Virginia. In these years, al-Awlaki had come to the attention of the FBI and the Joint Terrorism Task Force due to links with a charity allegedly associated with Hamas, as well as through his contact with a man implicated in the 1993 World Trade Center bombing. These investigations concluded that al-Awlaki posed no threat to US national security. After September 11th 2001, al-Awlaki was in the spotlight once again: two of the hijackers had once attended sermons at al-Awlaki’s mosques.

After further extensive government investigations and numerous alleged attempts at converting him to be an FBI informant, al-Awlaki left the US in 2002, eventually finding his way back to Yemen in 2004. In the following years, al-Awlaki became the focal point of the US Government’s efforts on stopping AQAP. He was imprisoned in 2006 by the Yemeni government, allegedly under political pressure from the US. Upon his eventual release in 2007, he began to use the internet to remotely preach in an increasingly radical manner to his followers. His Google-hosted blogspot website had significant numbers of followers and lengthy comment threads following posts on topics ranging from Islamic rules on cheese to outright calls for Jihad. Thousands subscribed to his Facebook profile. He used Youtube to publish lengthy video seminars, and email logs of the Fort Hood shooter Nidal Hasan and the Christmas Day Bomber Umar Farouk Abdulmutallab show evidence of limited correspondence with al-Awlaki. He was placed on the US Government’s ‘kill list’ in 2010. Since his death, his name has frequently arisen in terrorist investigations, notably in the investigation into the Tsarnaev brothers following the Boston Marathon Bombings in 2013.

Documents leaked by Edward Snowden suggest that the National Security Agency may have played a vital role in the surveillance of Anwar al-Awlaki. Two programmes were used in Yemen in the years preceeding al-Awlaki’s death. GILGAMESH: a false cell-tower appended to a drone geolocates targeted sim cards in mobile phones. SHENANIGANS: which ‘utilizes a pod on aircraft that vacuums up massive amounts of data from any wireless routers, computers, smart phones or other electronic devices that are within range.’ [16]

A 2012 investigation by the Washington Post revealed the existence of the Disposition Matrix, ‘a database that officials describe as a next-generation kill/capture list’. [17] Names of individuals, organisations, and their associated operative networks are continuously collected through various US Defense departments and surveillance agencies. The Disposition Matrix can be queried for specific parameters, identifying various individuals, their level of activity, their location, if their capture is a possibility, and if they pose an ‘imminent threat’. The President can subsequently sign off on kill or capture actions, which are often carried out in secret drone operations, and with limited congressional oversight.

The Disposition Matrix and the drone programme, like Phung Hoang in the Vietnam war, are the quiet counterparts to the weaponised aesthetics of military power. Their ‘surgical precision’ in targeting specific individuals rather than swathes of terrain is the antithesis of the blunt force of Shock and Awe and the cluster-bombing of Vietnamese jungles.

The US Military have various ways of speaking about drone strikes, dependent on circumstances: a kinetic strike, any weapons release that targets an individual; targeted killing, when specifically seeking out and firing on a high-profile political figure or known terrorist; signature strike, when an individual is observed during a sortie and is subsequently targeted after the drone operators believe them to be behaving according to the ‘pattern-of-life’ of terrorist. The pattern-of-life is a construct facilitated by the Archive Machine. Through extensive observation of its subjects, it quietly produces spatio-temporal objects which become a cumulative reference of individualised normality, or ‘good conduct’. Deviations from this reference are regarded as suspicious, and perhaps worthy of intervention.

A sighting of an individual described as fitting the profile of Osama Bin Laden (date unknown). As observed from a Predator drone:

‘The four figures were making their way up the jagged slope. It looked like three of the men were guarding the fourth. The fourth man was much taller than the others, wearing dark brown robes with white sleeves and headdress. He used a long walking stick to help him climb. The others were dressed in black and carrying AK-47s. They walked at a respectful distance from the taller man. Two of the guards flanked the taller man and the third walked in front of him. None got within five yards of the man.’ [18]

Reported upwards through chain of command. Proposed strike denied citing an incident of mistaken identity. A few years previous, in 2002, a man thought to be Osama Bin Laden was killed in a strike near Khost in Afghanistan: ‘The peasant, Mir Ahmad, was searching for scrap metal in the detritus of war in the region along with two companions when the missile struck. All three were killed.’ [19] War Fever permits human error as an inevitable by-product of warfare. It seeks to reduce this error with the computational capacity of the Archive Machine. Civilians killed in drone strikes are referred to as Collateral Damage.

Collateral Damage: >‘US military euphemism, developed in regard to nuclear targeting policies in the 1970s, for civilian casualties. Now defined as “unintentional or incidental injury or damage to persons or objects that would not be lawful military targets in the circumstances ruling at the time.” […] Such damage is not unlawful so long as it is not excessive in light of the overall military advantage anticipated from the attack”.’ [20]

‘We might think of war as dividing populations into those who are grievable and those who are not. An ungrievable life is one that cannot be mourned because it has never lived, that is, it has never counted as a life at all.’ [21]


Norbert Wiener gives the example of a thermostat as a quintessentially cybernetic device: it conceptualises environmental conditions as information, and then manages this information through feedback loops. The thermostat tries to maintain a balanced temperature, to regulate an environmental instability, through the process of continuously measuring temperature and adjusting the flow of a hot or cold medium (air or water) in order to balance it at a specified value. The temperature (input) is processed, a decision is made based on the relation between this temperature and the desired value (algorithm), there is the subsequent regulation of flow (output), then this regulation in turn starts to affect the temperature and the cycle continues once again. [22]

The Archive Machine has its physical locations. For all the perceived ephemerality of wirelessness, its sites of construction, its infrastructures and its resting places are firmly rooted in the earth. The Archive Machine requires its repositories. It requires its information taxonomies. It needs to be able to search, store, and retrieve on command.

The National Security Agency’s Utah Data Center was reported to have the Borgesian ambition to mirror the internet, to provide an explorable snapshot of all internet traffic, to act as a container for the incomprehensibly vast quantities of digital information intercepted by the agency and its collaborators. XKEYSCORE, PRISM, ELEGANTCHAOS, PINWALE, BOUNDLESSINFORMANT: mass surveillance programs variously developed to collate and analyse internet traffic metadata; observe the online activity of individuals through their use of Google, Facebook, Yahoo, and others; to intercept intra-satellite communications data such as text messages and emails. All programmes are unprecedented in their scope.

The devices that form the repositories of the Archive Machine generate a quiet heat. The quiet heat necessitates a management system. Complex climate control systems keep the space cool, at an optimal 21 degrees celsius, to protect the delicate machinery. The movement of the air through these discreet, contained spaces is engineered. The devices are arranged in rows and columns, a logical, rational spatialisation, an architecture of national security. The conduits between them allow for circulation of air and the movement of the Archive Machine’s minders. These minders, civilians, external to the military and the agency: they have no relationship with the specific data that is contained within, they simply observe the infrastructure, they record faults, they replace, they monitor, they survey the stability of the Archive Machines.

‘Software and ideology fit each other perfectly because both try to map the tangible effects of the intangible and to posit the intangible cause through visible cues.’ [23]

War Fever legitimises the extension of the Archive Machine, just as the Archive Machine provides the raw materials that form the rhetoric of Truth in War Fever. The quiet, cool, grey computation of the Archive Machine is thus the counterpart to the noisy, hot aesthetics of War Fever. With their respective frameworks, they both abstract bodies, such that behaviours become statistical information, allowing them to become knowable, to approach the rationalist fantasy that human error vanishes through proceduralism. In tandem, they seek to remove the viscerality of warfare, to predict and control the chaotic politics of the battlefield. In doing so, they change the characteristics of warfare: it almost disappears completely from the surface, it facilitates its operation on a continuum, a perpetual and mutating state of targeted conflict that is much less about one military against another, than it is as a series of discrete violent events. Sites for intervention, for the analysis of error, become obfuscated.

However, the Archive Machine, in its practice of self-documentation, contains the materials for its own critique. These materials, shielded as they are, can seemingly only ever become artefacts, objects of future histories, released eventually, perhaps to be declassified a quarter of a century after their creation, should the materials in question cease to have an impact on the conduct of the State. Data leaks and journalistic investigations can and do provide vital critical pathways of engagement, but when dealing with the colossal scale of the Archive Machine and the powerful discourse of War Fever, other strategies are required to bolster these efforts. Our failure – or our inability – to engage with these opaque structures should not lead us to apolitical thinking, or to inaction, but to an investigation of their recent history [24]. We can try to understand the evolving teleologies of the Archive Machine, to speculate on its present configurations as an extension of its histories. In doing so, we can not only deconstruct the hidden mechanisms of the Archive Machine, but witness the (ir)rational logic of War Fever, and disentangle the two such that their latent affordances become visibly isolated, and thus, subject to our observations.