Andreas Graae

The dissertation investigates how military UAV’s (unmanned aerial vehicles), or so-called drones, are represented within the aesthetic field as a “drone imaginary,” reflecting radical changes in the history of warfare. Using the imaginary as a conceptual framework, the drone is analyzed as a cultural construct fueled with ideological and political imagination, including, above all, promises of liberation from the burdens and vulnerabilities of human lives and bodies in war. The main goal of the dissertation is to critically analyze how the drone imaginary builds on a fantasy of the perfect weapon which is, essentially, cruel. Drawing on Lauren Berlant’s thoughts and ideas, my claim is therefore that the social and cultural imagination of drone warfare follows a logic of cruel optimism. This means that the object for these desires, the drone, becomes an obstacle for its own flourishing by actively impeding the goal it promises to fulfill. In other words, my aim is to show how the popular attachment to drones is formed by fantasies and imaginations that are “cruel” in so far as they compromise themselves and obstruct their aims through a negative feedback loop, which constantly negates the promises these very same machines seem able to deliver on regarding a higher and safer mode of warfare. Each of the chapters in the dissertation contains examples that demonstrate how these fantasies and promises in turn prove to be flawed or imperfect. Using the realm of aesthetics as prism, the analyses expose the darker side of this drone imagination focusing on its inherent cracks and frailties that altogether undermine the legitimacy as well as soundness of the fantasy of the drone as a new wonder-weapon. For instance, the analyses show how figurations of drone automation is uncannily non-human; how drone invincibility also entail trauma; how dreams of total vision become blurred by immensity; and how the myth of surgical precision ends up as carnage. Thus, each chapter specifically examines one of these drone figurations in order to show how they are con-figured into the larger drone imaginary. Based on strategies of close-reading in combination with a cross-disciplinary conceptual approach, the dissertation offers new insights to the rapidly growing field of academic drone research. While this field has, however, so far mostly focused on the political, juridical, and ethical aspects of drone warfare and less on imaginary, literary, and aesthetic constructions and configurations vibrating beneath these debates, the dissertation contributes with an alternative cultural drone imaginary.

Kapitlerne i denne bog har alle handlet om, hvordan forskellige udgaver af kunstig intelligens kan anvendes militært i dag. Samtidig rejses det mere diffuse spørgsmål om, hvordan disse teknologier så vil forme fremtidens kampplads. Dette spørgsmål er på én gang animerende futuristisk – og oldgammelt. Lige så længe der har været krige i verden, har der også været forestillinger om, hvordan fremtidens krige ville se ud. Spørgsmålet om karakteren af fremtidens krigsførelse har optaget politikere, militære ledere, diplomater, forskere, journalister, kunstnere, filmskabere og forfattere, og det har altid været uløseligt kædet sammen med udbredelsen af nye teknologier. Opfindelser som sværdet, armbrøsten, maskingeværet, flyvemaskinen, ubåden, atombomben eller for den sags skyld computernetværket, radaren og missilsystemet har hver især ændret forudsætningerne for sejr eller nederlag på den militære kampplads radikalt. Af samme grund har teknologisk overlegenhed ofte været et strategisk mål i sig selv, ligesom forudsigelser af kommende teknologiske nybrud og disses betydning for udfaldet af egne og modstanderes kampkraft har været stærkt efterspurgt blandt militære strateger (Freedman 2017, xix).
Med opblomstringen af kunstig intelligens og maskinlæring er sådanne prognoser på én gang blevet både lettere og vanskeligere at fremskrive. Mens algoritmer ved hjælp af eksempelvis reinforcement learning kan forudsige udfaldet af fremtidige handlinger og risikoscenarier, gør den stigende kompleksitet og uigennemsigtighed i udviklingen af dybe neurale netværk og ”sorte bokse” det stadigt sværere at opnå indsigt i, hvordan de reelt fungerer, og dermed hvad der lægges til grund for de opnåede resultater og anbefalinger. Der opstår med andre ord et behov for ”forklarlig” kunstig intelligens, som bidragyderne til kapitel 5 beskriver som en mulighed for at ”oversætte det abstrakte, matematiske maskinsprog til et praktisk, militært kommandosprog” (se kapitel 5).
Et sådant transparent kommandosprog lader dog vente på sig i det endnu spæde teknologiske udviklingstrin, vi for øjeblikket befinder os på. Men som denne bog har givet talrige eksempler på, handler kløften mellem teknologien og mennesket om andet og mere end sorte bokse. Den hænger også sammen med fremkomsten af en ny type hybrid kampplads, hvor kunstig intelligens bliver stadigt tættere forbundet med ikke blot fjernstyrede, automatiserede og semi- eller fuldautonome våbensystemer, men også ikke-kinetiske aktiviteter som eksempelvis cyberspaceoperationer, informations- og påvirkningskampagner, psykologisk krigsførelse og finans- og handelskrige. At skabe et realistisk billede af, hvordan denne netværksforbundne, hybride og distribuerede kampplads vil forme sig længere ude i fremtiden, er dog langt fra lige til.
Som den britiske militærhistoriker Lawrence Freedman udtrykker det, ”findes der ikke længere nogen dominerende model for fremtidens krig, men blot et mudret koncept og en vifte af spekulative muligheder” (Freedman 2017, xx). Og netop det spekulative element er da også ofte fremherskende i den stadigt hurtigere voksende litteratur om AI og dens betydning for den militære arena. På den ene side står teknologioptimisterne, der drømmer om total informationsdominans og beslutningsoverlegenhed på en fuldt integreret kampplads, hvor sensordata flyder frit og friktionsløst på tværs af platforme og operative enheder; på den anden side dommedagsprædikanterne, der frygter en fremtidig præget af dehumaniseret, biometrisk overvågning og autonome maskiner, der ikke blot diskriminerer bestemte befolkningsgrupper, men på sigt vil true mennesket på dets eksistens. I dette lettere polariserede billede er det oftest teknologer og militære praktikere, der repræsenterer første gruppe, mens forskere, journalister, kunstnere og science fiction-forfattere udgør den sidste mere kritiske position. Fælles for disse to karikerede positioner er imidlertid det spekulative aspekt – altså hvad AI-teknologien potentielt ville kunne gøre, hvis og såfremt – hvilket understreger den usikkerhed og uvished, som sådanne spekulationer om fremtiden altid bygger på.
Denne bog har så vidt muligt afholdt sig fra at bevæge sig ned ad den spekulative vej for i stedet at fokusere på, hvordan kunstig intelligens finder udtryk i form af militære AI-baserede teknologier, systemer og strategier, der allerede nu er under anvendelse eller udvikling. Her til slut må det dog være tid til at gøre en undtagelse og – med bogens bidrag og science-fiction-litteraturen i hånden – foretage et feed-forward (som det hedder i algoritmernes verden), det vil sige et spring fremad for at se nærmere på, hvilke ”output” militær AI kan få på lidt længeres sigt. For selvom spekulationerne er mange, lader der nemlig til at herske en basal mangel på kreativ forestillingskraft, når fremtidens krige skal beskrives. Det er derfor ofte de samme populærkulturelle science fiction-scenarier, der bliver skævet til i visualiseringen af fremtidens krig. Men hvilke billeder og fortællinger, der helt nøjagtigt bidrager til at skabe vores fælles forventninger til fremtidens militærteknologier, og dermed også til dels til denne bogs teser om fremtidens militære AI, er ikke altid klart.
Det skal vi i det følgende kaste et afsluttende blik på ved at genbesøge nogle af de temaer og dilemmaer, som er gået igen i denne bog. Først skal vi se på, hvordan fiktioner, simulationer og billeder er blevet et stadigt mere udbredt redskab for militæret. Dernæst zoomer vi ind på nogle af de populærkulturelle forestillingskomplekser, som bogens ideer om fremtidens kunstige intelligens er formet af. Og endelig samler vi op på nogle af de vigtigste analytiske pointer fra bogens kapitler og ser på, hvad de i lyset af de omtalte fiktioner kan fortælle os om næste generation af militær AI. Med bevidstheden om at sådanne fremsyn har det med at vise sig både upræcise og fejlagtige (ikke mindst fordi den slags spådomme per definition altid er fiktioner) skal det følgende således ikke læses som en skråsikker prognose for, hvordan udviklingen af AI vil forme fremtidens krige. Snarere skal det læses som et forsvar for forestillingsevnen – og en påmindelse om, hvordan vi er drevet af fiktion i vores analyser, forståelser, frygt og syn på AI. For som Freedman minder om i bogen The Future of War, er historien skabt af folk, som ikke anede, hvad der ville ske som det næste. Men hvordan disse folk forestillede sig fremtidens krige, har utvivlsomt formet normerne for og udfaldet af disse krige, da de endelig kom (Freedman 2017, xix).

I dette kapitel giver vi et bud på, hvordan implementering af kunstig intelligens i våbensystemer både kan og formentlig også vil transformere situationsforståelsen på den militære kampplads, herunder hvilke muligheder, men også hvilke nye problemer, denne transformation potentielt vil medføre. Selvom AI og automatisk måludpegning allerede er under udvikling og anvendelse i en række våbensystemer – såvel i land-, vand- og luftbaserede systemer – vil vi for begrænsningens skyld fokusere på særligt én type kampplads og ét militært domæne; nemlig luftrummet. Luftrummet er det mest enkle domæne at operere i, da det er fri for forhindringer, når man ser bort fra andre fly eller missiler. Her er den teknologiske udvikling med AI-baseret luftforsvar og ubemandede fly (UAVs) også længst fremme. Luftrummet er desuden særligt interessant set med danske øjne, eftersom Danmark fra 2023 modtager de første F-35 kampfly, som er verdens nyeste og foreløbigt mest avancerede kampfly. Flyene, der er produceret af amerikanske Lockheed Martin, har indbygget en række avancerede sensorer og anvender AI og maskinlæring til sensorintegration, selvforsvar, måludvælgelse og databehandling (Masuhr 2019). F-35 er ligeledes omdrejningspunkt for et antal projekter – hvor det mest omtalte bærer det spektakulære navn “Skyborg” – der har til formål at udvikle ubemandede luftfartøjer, som kan operere integreret med og som støtte til F-35 kampflyet.

This article investigates how robotic swarming entails new modes of sensing that not only blur the lines between human, insect and machine – and hence between natural and prosthetic sensoria – but also transcend beyond what the human sensory apparatus can possibly process and comprehend. Based on examples and analytical insights from the popular Netflix series Black Mirror the article analyzes how robotic surrogate bees capable of pollinating the world’s  ora in order to prevent an ecological catastrophe is depicted in the episode “Hated in the Nation”. Thus, the article not only demonstrates how artificial intelligence and robotic swarms can expand, refine, or disrupt the sensory capacity of human perception – it also shows how the relationship between natural and artificial swarming is troubled by historical, social, and technical ties that altogether shed new light on the idea of machine autonomy.

In the age of drone warfare, soldiers are no longer confronted with their opponents face- to-face on the battlefield. Instead, the practice of killing by remote control have shaped a new kind of war experience: Namely, one in which the drone operators are at the same time far and close, geographical distanced and emotional present; an experience of ‘voyeuristic intimacy’ as Derek Gregory has pointed out. But, as the paper suggests, it is also a ‘split experience’ producing a kind of bipolarity and manic paranoia. From a cultural perspective, the paper explores how this split, or even ‘shattered,’ experience is represented in Showtime’s television series Homeland as a specific ‘drone imaginary.’ Focusing on especially the fourth season of the series, in which the bipolar CIA agent, Carrie Mathison or ‘The Drone Queen,’ is in charge of a drone bastion in Afghanistan, the paper propose seeing The Drone Queen as an embodiment of this experience. In her manic endeavour to see ‘the bigger picture,’ but without missing any crucial details, she becomes a symptom of a larger political culture after 9/11, a culture marked by paranoia and control, in which drones are at least one of the most significant examples.

Karantænens adskillelse af raske fra syge har til alle tider været det foretrukne kontrolredskab til sygdomsbekæmpelse. I disse år er dens funktionsområde dog udvidet til at omfatte kontrol ikke blot af smitte, men også af mennesker. I
den spektakulære epidemifilm World War Z (2013) indgår for eksempel en scene, som er emblematisk for karantænens betydning i den senmoderne katastrofeimagination.

This book investigates drone technology from a humanities point of view by exploring how civilian and military drones are represented in visual arts and literature. It opens up a new aesthetic ‘drone imaginary’, a prism of cultural and critical knowledge, through which the complex interplay between drone technology and human communities is explored, and from which its historical, cultural and political dimensions can be assessed. The contributors to this volume offer diverse approaches to this interdisciplinary field of aesthetic drone imaginaries. Sprouting from art history, literature, photography, feminism, postcolonialism and cultural studies, the chapters provide new insights to the rapidly evolving field of drone studies. They include historical perspectives on early unmanned aviation and aerial modes of vision; they explore aesthetic configurations of drone swarming, robotics and automation; and they engage in current debates on how drone technology alters the human body, upsets available categories, and creates new political imaginaries.

Surveillance, Form, Affect. An International, Multidisciplinary Conference
December 7-9, 2016​

Centre for Popular Culture in the Humanities, The Education University of Hong Kong

Abstract
Surveillance performed by Unmanned Aerial Vehicles (UAVs) or drones is often associated with scopic or (post-)panoptic regimes of technologically enhanced gazing. For instance, the latest camera technology equipped on US Reaper drones, called ‘The Gorgon Stare’, metaphorically links drone surveillance to the myth of the Gorgon’s omnipotent and lethal vision (turning the object of its gaze into stone). However, the predominant ocularcentric focus on vision in popular cultural drone imaginaries and academic discourse might have overshadowed the significance of other, less human, senses and affects in relation to drones; not least their intrinsic relationship with insects. How have drone technologies transformed the human sensory experience towards a trans-human or even non-human mode of machine surveillance? And in what way can the concept of the swarm—information swarms as well as literal drone swarms—be used to understand these new forms and affects of surveillance? Based on an analysis of the German author Ernst Jünger’s novel The Glass Bees (1957) this paper will discuss such questions and how they relate to the cultural history of surveillance. As a war documentarian, totalitarian thinker, essayist preoccupied with photography and technology, but also as an eager entomologist and a speculative writer, Jünger’s strangely accurate drone imaginaries may serve as a prism to render an intensified sensory regime of non-human surveillance. In contrast to many other literary representations on future surveillance and control societies, Jünger is not a dystopian. On the contrary, a troubling but also highly interesting quality of his writings is his radical devotion to the world of machines. This rarely neutral account of surveillance technologies may serve to contextualize the contemporary political regime of drone warfare.

The Body of War: Drones and Lone Wolves, Lancaster University, 24–25 November 2016

Abstract:
“The kamikaze: My body is a weapon. The drone: My weapon has no body.” The line between suicide bombings and drone bombings is sharply drawn by Grégoire Chamayou in his Theory of the Drone. While the first indicates ultimate bodily sacrifice, the latter marks distance and absence of a human body in battle – two political and affective economies often opposed in academic debate on drones. However, the lines might be more blurred than that. As Derek Gregory has convincingly claimed, the human body is not quite absent in drone warfare. Contrarily, the very distance of the operator has formed what he calls an invasive, irruptive, “voyeuristic intimacy”. Such intimacy (embedded in the distance) is heavily represented in Showtime’s tv-series Homeland (2011-) in which the protagonist, CIA-agent and drone commander Carrie Mathison, is constantly torn between the two poles (as she is, in fact, suffering from bipolar disorder). Based upon Gregory’s idea of intimacy, the paper explores the bipolar phenomenology of drones—particularly in the fifth season of Homeland (opening with the episode: “The Drone Queen”)—negotiating the political and affective economies of the body in late modern remote warfare.

The dissertation investigates how military UAV’s (unmanned aerial vehicles), or so-called drones, are represented within the aesthetic field as a “drone imaginary,” reflecting radical changes in the history of warfare. Using the imaginary as a conceptual framework, the drone is analyzed as a cultural construct fueled with ideological and political imagination, including, above all, promises of liberation from the burdens and vulnerabilities of human lives and bodies in war. The main goal of the dissertation is to critically analyze how the drone imaginary builds on a fantasy of the perfect weapon which is, essentially, cruel. Drawing on Lauren Berlant’s thoughts and ideas, my claim is therefore that the social and cultural imagination of drone warfare follows a logic of cruel optimism. This means that the object for these desires, the drone, becomes an obstacle for its own flourishing by actively impeding the goal it promises to fulfill. In other words, my aim is to show how the popular attachment to drones is formed by fantasies and imaginations that are “cruel” in so far as they compromise themselves and obstruct their aims through a negative feedback loop, which constantly negates the promises these very same machines seem able to deliver on regarding a higher and safer mode of warfare.

Each of the chapters in the dissertation contains examples that demonstrate how these fantasies and promises in turn prove to be flawed or imperfect. Using the realm of aesthetics as prism, the analyses expose the darker side of this drone imagination focusing on its inherent cracks and frailties that altogether undermine the legitimacy as well as soundness of the fantasy of the drone as a new wonder-weapon. For instance, the analyses show how figurations of drone automation is uncannily non-human; how drone invincibility also entail trauma; how dreams of total vision become blurred by immensity; and how the myth of surgical precision ends up as carnage. Thus, each chapter specifically examines one of these drone figurations in order to show how they are con-figured into the larger drone imaginary.

Based on strategies of close-reading in combination with a cross-disciplinary conceptual approach, the dissertation offers new insights to the rapidly growing field of academic drone research. While this field has, however, so far mostly focused on the political, juridical, and ethical aspects of drone warfare and less on imaginary, literary, and aesthetic constructions and configurations vibrating beneath these debates, the dissertation contributes with an alternative cultural drone imaginary.