Real Wars on Virtual Battlefields: The Convergence of Programmable Media at the Military-Civilian Margin [1. Aufl.] 9783839412404

The book analyzes the multifarious exchange of algorithmic technologies and concepts between the military and the media

182 74 5MB

English Pages 412 [416] Year 2015

Report DMCA / Copyright

DOWNLOAD PDF FILE

Table of contents :
CONTENTS
Table of Illustrations
1 Introduction
1.1 Remilitarization – Coming to Terms with the Notion of War in Contemporary Western Societies
1.2 Programmable Applications within Media Deployment Policies
1.3 The State of Research and the Politics of Documentation
1.4 Programmable Media as Socio-Cultural Filters
1.5 Digital Games as a Subset of Programmable Media
1.6 Towards an Integrative Model of Programmable Media Analysis
1.7 Technical Preliminaries
2 Building Blocks for a Model of Programmable Media Analysis
2.1 Critical Considerations on the Current State of Game Studies
2.2 Towards Programmable Media Analysis
2.3 Reading (Digital) Games as Code
2.4 Socio-Cultural Implications of Analyzing Programmable Media
2.5 From Technology Semiotics to Programmable Media Analysis
2.6 Program Code Rhetoric in Academic Literature
2.7 The ‘Next Generation’
3 Epistemological Effects of Programmable Media in Perspective
3.1 Towards an Epistemology of Program Code
3.2 The Compatibility/Standardization Paradigm
3.3 The Computability Paradigm
3.4 A Program Code Model of Authenticity Effects
3.5 Nature in the Age of Algorithmic Representability
3.6 History as Simulation
3.7 Simulation/Emulation
4 Programmable Media at the Civilian/Military Margin
4.1 Before ‘Information Warfare’ (IW)
4.2 The Doctrine of Information Warfare
4.3 War and (Media) Literacy
4.4 New Rules of Play – Algorithms as Role Models of Military Self-Description and Organization
4.5 A Concluding Look Back
5 From the War in the Gulf to the Global War on Terror – A History in Games
5.1 The Methodological Problem of the ‘Gulf Wars’
5.2 The Global War on Terrorism in Digital Games
5.3 Delimiting Wars by their Media Representations
5.4 Encyclopedicity
5.5 Applied Forum Analysis – An Ethnographic Perspective on Programmable Media
6 Peripheralization and Universalization. Macro Effects of Programmable Media in Military-Civilian Contexts
6.1 Historiographical Effects of Digital War Games
6.2 A Comparative Look at COTS Games in Military Applications
6.3 Grids as a Ubiquitous Dimension of (Digital) Warfare
6.4 Politico-Military Rhetoric and Code(s) in Non-War Game Contexts
6.5 Inserting Military Iconography into Media Usage Ecologies
6.6 Psychophysical Coercion of the Interface
6.7 The Implied Economization of War and Programmable Media Systems
6.8 Composing Universal Aesthetics and Narratives of War
6.9 Virtual Wars – Real Battlefields
Bibliography
Recommend Papers

Real Wars on Virtual Battlefields: The Convergence of Programmable Media at the Military-Civilian Margin [1. Aufl.]
 9783839412404

  • 0 0 0
  • Like this paper and download? You can publish your own PDF file online for free in a few minutes! Sign Up
File loading please wait...
Citation preview

Stefan Werning Real Wars on Virtual Battlefields

2009-09-21 15-03-22 --- Projekt: transcript.titeleien / Dokument: FAX ID 02ea221438957174|(S.

1

) T00_01 schmutztitel - 1240.p 221438957182

Stefan Werning (Dr.) is currently working as user-generated content supervisor at Nintendo of Europe. His research fields include the history of technology, »new media« phenomena and the theorization of trans-media products and practices.

2009-09-21 15-03-22 --- Projekt: transcript.titeleien / Dokument: FAX ID 02ea221438957174|(S.

2

) T00_02 seite 2 - 1240.p 221438957190

Stefan Werning Real Wars on Virtual Battlefields. The Convergence of Programmable Media at the Military-Civilian Margin

2009-09-21 15-03-23 --- Projekt: transcript.titeleien / Dokument: FAX ID 02ea221438957174|(S.

3

) T00_03 titel - 1240.p 221438957198

Bibliographic information published by the Deutsche Nationalbibliothek The Deutsche Nationalbibliothek lists this publication in the Deutsche Nationalbibliografie; detailed bibliographic data are available in the Internet at http://dnb.d-nb.de

© 2009 transcript Verlag, Bielefeld zugl. Dissertation, Universität Bonn All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by any electronic, mechanical, or other means, now known or hereafter invented, including photocopying and recording, or in any information storage or retrieval system, without permission in writing from the publisher. Coverlayout: Kordula Röckenhaus, Bielefeld Cover illustration: The cover illustration is based on imagery from the game Patton vs. Rommel (1987) and is used by kind permission of the designer, Chris Crawford. Edited by: Stefan Werning Typeset by: Stefan Werning Printed by: Majuskel Medienproduktion, Wetzlar ISBN 978-3-8376-1240-0

Distributed in North America by

Transaction Publishers Tel.: (732) 445-2280 Rutgers University Fax: (732) 445-3138 35 Berrue Circle for orders (U.S. only): Piscataway, NJ 08854 toll free 888-999-6778

2009-09-21 15-03-23 --- Projekt: transcript.titeleien / Dokument: FAX ID 02ea221438957174|(S.

4

) T00_04 impressum - 1240.p 221438957206

CONTENTS

Table of Illustrations 8

1 Introduction 11 1.1 Remilitarization – Coming to Terms with the Notion of War in Contemporary Western Societies 11 1.2 Programmable Applications within Media Deployment Policies 17 1.3 The State of Research and the Politics of Documentation 23 1.4 Programmable Media as Socio-Cultural Filters 29 1.5 Digital Games as a Subset of Programmable Media 44 1.6 Towards an Integrative Model of Programmable Media Analysis 49 1.7 Technical Preliminaries 52  2 Building Blocks for a Model of Programmable Media Analysis 53 2.1 Critical Considerations on the Current State of Game Studies 53 2.2 Towards Programmable Media Analysis 63 2.3 Reading (Digital) Games as Code 89 2.4 Socio-Cultural Implications of Analyzing Programmable Media 95 2.5 From Technology Semiotics to Programmable Media Analysis 106

2.6 Program Code Rhetoric in Academic Literature 123 2.7 The ‘Next Generation’ 129  3 Epistemological Effects of Programmable Media in Perspective 131 3.1 Towards an Epistemology of Program Code 131 3.2 The Compatibility/Standardization Paradigm 135 3.3 The Computability Paradigm 142 3.4 A Program Code Model of Authenticity Effects 159 3.5 Nature in the Age of Algorithmic Representability 182 3.6 History as Simulation 196 3.7 Simulation/Emulation 199  4 Programmable Media at the Civilian/Military Margin 203 4.1 Before ‘Information Warfare’ (IW) 206 4.2 The Doctrine of Information Warfare 210 4.3 War and (Media) Literacy 215 4.4 New Rules of Play – Algorithms as Role Models of Military Self-Description and Organization 226 4.5 A Concluding Look Back 257  5 From the War in the Gulf to the Global War on Terror – A History in Games 259 5.1 The Methodological Problem of the ‘Gulf Wars’ 259

5.2 The Global War on Terrorism in Digital Games 263 5.3 Delimiting Wars by their Media Representations 291 5.4 Encyclopedicity 306 5.5 Applied Forum Analysis – An Ethnographic Perspective on Programmable Media 308   6 Peripheralization and Universalization. Macro Effects of Programmable Media in Military-Civilian Contexts 325 6.1 Historiographical Effects of Digital War Games 325 6.2 A Comparative Look at COTS Games in Military Applications 328 6.3 Grids as a Ubiquitous Dimension of (Digital) Warfare 335 6.4 Politico-Military Rhetoric and Code(s) in Non-War Game Contexts 340 6.5 Inserting Military Iconography into Media Usage Ecologies 346 6.6 Psychophysical Coercion of the Interface 347 6.7 The Implied Economization of War and Programmable Media Systems 352 6.8 Composing Universal Aesthetics and Narratives of War 370 6.9 Virtual Wars – Real Battlefields 379

Bibliography 391

Table of Illustrations

Illustration 1: UML Example - 22 Illustration 2: Sample query for VBS1 on Emule - 48 Illustration 3: Print ad for Battlefield 2 - 56 Illustration 4: Screenshot from Pacific Theater of Operations II - 69 Illustration 5: Particle fountain screenshot - 78 Illustration 6: Screenshot from Super Metroid - 90 Illustration 7: Ballistics in the game Full Spectrum Warrior - 97 Illustration 8: Oriental setting in Delta Force: Land Warrior - 102 Illustration 9: Fill patterns in MacPaint - 110 Illustration 10: Gunstar Heroes screenshot - 121 Illustration 11: Mapping of Gagne’s onto Keller’s model - 140 Illustration 12: Screenshot from Excalibur - 148 Illustration 13: Screenshot of Nuclear Weapons Effects Calculator - 158 Illustration 14: Hex-code map of Metroid - 176 Illustration 15: X für Grün - 186 Illustration 16: DARWARS usage example - 237 Illustration 17: War on Terror official website - 265 Illustration 18: Wikipedia entry on the 2003 invasion - 269 Illustration 19: Desert Strike Campaign 3 Map - 276 Illustration 20: First to Fight Fansite Toolkit - 280

Illustration 21: Cafepress merchandise for Desert Combat - 283 Illustration 22: Target zones in Soldier of Fortune II - 285 Illustration 23: 3d02 catalogue excerpt - 301 Illustration 24: Screenshot from Tetris - 314 Illustration 25: Cellfactor invitation - 322 Illustration 26: UMS army formation screenshot - 326 Illustration 27: UMS battle screenshot - 327 Illustration 28: Air Force Delta Storm replay screenshot - 330 Illustration 29: Map view of Knights of the Desert: The North African Campaign of 1941-1943 - 336 Illustration 30: Map view of Patton vs. Rommel - 337 Illustration 31: John Deere American Farmer Packshot - 341 Illustration 32: DIA Kids website - 346 Illustration 33: Online description of the Nike ipod accessory - 347 Illustration 34: AFM Games portfolio - 360 Illustration 35: America’s Army Gaming Championships logo - 362 Illustration 36: Billboard ad in Ghost Recon: Advanced Warfighter 2 - 367 Illustration 37: Banner ad for Massive Inc. - 368 Illustration 38: Final level of Painkiller - 372 Illustration 39: Texture viewer of the America’s Army editor - 375 Illustration 40: WarPaint2 screenshot - 381

1 INTRODUCTION “Es soll sich kein Staat im Kriege mit einem andern solche Feindseligkeiten erlauben, welche das wechselseitige Zutrauen im künftigen Frieden unmöglich machen müssen: als da sind, Anstellung der Meuchelmörder (percussores), Giftmischer (venefici), Brechung der Kapitulation, Anstiftung des Verraths (perduellio) in dem bekriegten Staat etc.“ (I. Kant, sixth preliminary article, Zum Ewigen Frieden, Ersttext 1795)

1.1 Remilitarization – Coming to Terms with the Notion of War in Contemporary Western Societies Immanuel Kant’s thoughts on the potentiality and conditions of “eternal peace” appear remarkably topical, even when read against only the publicly known conditions of contemporary (Western) warfare. The “poisoners” that Kant vehemently opposed as a tactic of warfare are still a powerful topos in the discourse on the proper conduct of war, made manifest in the fear of chemical and even biological weapons. Yet, digital computer and videogames like Command & Conquer: Generals (EA Pacific: Electronic Arts, 2003), released only weeks before the US invasion of Iraq, already begin to include these controversial elements of warfare into their programmed rule ecologies, fundamentally re-shaping their interpretation by making them systemically manageable, localizable and communicable, regardless of the popular rhetoric of ‘singularity’. This is only one example of the technological convergence between (commercial) programmable media and military technologies and its profound socio-cultural implications that constitute the central theme of this book. Not only is the “mutual confidence” [wechselseitige[s] Zutrauen] between politico-military entities characteristic of previous notions of warfare, but programmable media, both as tools and weapons, are an emerging key technology in managing the distrust that has taken its place within the logic of ‘information warfare’.

11

Real Wars on Virtual Battlefields? Noam Chomsky’s claim of a New Military Humanism from his eponymous 1999 book has been discarded as populist (cf. e.g. Bauerlein, 2005), but it illustrates how (non-fictional) media coverage of military intervention in the Kosovo conflict effectively revitalized war topoi from the Second World War by coining a new ‘humanitarian’ rhetoric. Similar to several other studies written since the spill-over of programmable media applications into mainstream usage, like e.g. Samuel Weber’s Targets of Opportunity: On the Militarization of Thinking (2005), Chomsky’s argument more or less explicitly cautions against the potential for ‘re-militarization’ in contemporary Western societies. The most obvious methodological problem with this lies in the fact that the notion of (re-)militarization suggests continuity with earlier forms of increased military embeddedness in society, which is not sustainable and ignores the impact of programmable media technologies in this regard, something which will be scrutinized in this book. A brief look at a few diverging opinions, such as Faisal Devji’s dictum of the “global war on terror as de-militarization” (Devji, 2006: 30), indicates the complications implied in the term if applied to the current situation. Devji understands the “transformation of war into policing” and the ‘criminalization’ of Al Qaeda as a form of de-militarization, since military rhetoric in its traditional sense is gradually supplanted by other concepts like criminal law, which stem from the perceivably ‘civilian’ sphere. While the notion of militarization, understood as a historically localizable idea, would provide a distorted image, programmable media, or rather practices and structures derived from program code properties, do arguably form a technical as well as epistemological infrastructure for blending military and civilian applications, while also constituting an arena to recursively ‘process’ (to use programming terminology) and re-negotiate these concepts in various, contingent configurations. However, while both the militarization hypothesis and the ongoing debate about digital games abetting violence on various levels might provide feasible and oft-chosen vantage points on the topic at hand, neither will play a major role in this book. Negotiating the perceived “gap” between civilian and military “society” in the United States has long been a motor of technological but also societal innovation which has often trickled down into other national militaries, e.g. very visibly with the end of the Second World War, after which the US could not demobilize as rapidly as in prior conflicts. (Cf. “Holsti Ole R. A Widening Gap between the U.S. Military and Civilian Society?: Some Evidence, 1976-96“ International Security, 23:3, Winter 1998-1999: 5) The Center of Military History (CMH) even dedicated a lengthy study to the integration of the Armed Forces into US society from 1940 to 1965, concentrating

12

Introduction

mainly on issues like segregation equality policies that were defining for contemporary US society as a whole. While such a study is not likely to be undertaken for the current situation, surprisingly similar topics are being constantly re-negotiated, e.g. on the official online forums of influential digital war games. A more comprehensive look at this corpus in Chapter 4b will demonstrate how this discourse is framed in various ways by the contingencies of programmable technologies. Related concepts like the notion of the ‘interwar’, which has been revitalized in the wake of September 11, 2001, will also be touched upon in later chapters. While this book will consider mostly patterns of media usage and deployment inaugurated by the US Department of Defense (DOD), other military forces shall be selectively set against this backdrop. For instance, the German military [Bundeswehr], after commissioning an advertising game (Helikopter Mission, 1995) which tapped into then familiar-digital game syntagmas in the mid1990s, came to reject the use of programmable media for instructional or recruiting purposes (cf. e.g. Deutscher Bundestag, 2002), but still, in specific areas like aerial rescue operations (SAR), utilizes digital games, which, being browser-based and created using the Adobe Flash authoring suite, resort to a very different technological setup and, as will be elaborated below, epistemology. (Cf. e.g. http://www.rettungsflieger.bundeswehr.de/sar/game/start.htm) The re-thinking of military-themed digital games in Germany is part of a larger process of ‘catching up’ on embedding programmable technologies into the infrastructure of the Bundeswehr (cf. e.g. NI33), which hints at the general tendency that, consequently, digital games will have to be localized within a broader context of programmable applications. The interrelation of war and (mass) media is a fairly well-studied field in media studies and, to a lesser degree, in historical and political analyses. For instance, Müller and Spangenberg mention several influential approaches in their 1991 article in a key anthology which represents to some extent the contemporary on-topic discourse in germanophone academia. (Cf. Müller/Spangenberg, 1991) The title of the anthology already exhibits the momentum of neologisms and the metaphorical or rather generally linguistic cultivation of media technologies within the corpus of scholarly literature; this applies particularly to the German discourse, although recent prominent anglophone contributions similarly utilize rhetorical figures like the ”citizen-soldier”. (Cf. Stahl, 2006). A common theme in those interpretations is the assumed instrumentalization of media channels, as evidenced e.g. already in Herman/Chomsky’s intuitively termed ‘manufacturing consent’ hypothesis. (Del Fabbro, 2002: 3) The hypothesis focuses on the inequality of power, assessing media

13

Real Wars on Virtual Battlefields? effects from that perspective. While Herman/Chomsky do not systematically unfold the technological contingencies of the media genres they analyze, their conceptual model of five layered “filters” being inserted into the flow of information intrinsically resonates well with principles of programmable media where ‘filtering’ is a key procedure. While this context was probably not implied in the original formulation of the hypothesis, the paradigmatic effects of programmable media frequently produce similarly quasi-technical readings and rhetoric, which will be more closely investigated below, thereby demonstrating the need for further differentiation of terms like ‘instrumentalization’. The widespread military adoption of programmable media, most prominently digital games, has been accompanied by other, only superficially unrelated technological developments. One striking affirmation of the topos of ‘war as a videogame’ is the recently increasing reliance on remote-controlled drones (UAVs), not only for surveillance but also for combat operations, (cf. e.g. Barami, 2004: 5) motivated by the almost paradoxical obligation of the US Army to fight wars without being ‘allowed’ to lose soldiers. Bahar Barami describes that regular-sized UAVs had already been in use for some time when the 2003 invasion of Iraq fostered research into miniaturization and provoked a shift to portable, more cost-effective devices. (5) Apart from the logic of UAVs as virtually unlimitedly reproducible, semi-autonomous and programmable agents, which exhibits several overlaps with programming concepts like ObjectOriented Programming (OOP) that will be further investigated in the second chapter, practical implications need to be taken into account, like the new archetype of soldiers controlling a UAV remotely from their hometown, participating in an on-going conflict while staying entirely embedded in their ‘civilian’ life world [Lebenswelt], e.g. having a family and other consistent relationships. The term [Lebenswelt] is hereby used in congruence with the common definition attributed to Alfred Schütz and Thomas Luckmann. (Cf. Fritz, Jürgen “Lebenswelt und Wirklichkeit” Jürgen Fritz/ Wolfgang Fehr (Eds.) Handbuch Medien: Computerspiele Bonn: BPB, 1997: 6/7) These forms of psychological displacement are already, albeit not as drastically, implied in the daily usage of military-themed digital games, and practices like these will be a recurring topic in this book. As Barami continues, “robotics entered a new market phase with the events of 9/11 and the U.S. military operations in Afghanistan and Iraq”. (5) What began as a civilian process, with the time-critical assembly of ‘search and rescue’ robots looking for survivors in the ruins of Ground Zero, soon expanded to include combat operations overseas, using e.g. a modified version of a product offered by the consumer-oriented company iRobot. (Cf. e.g. http://

14

Introduction

findarticles.com/p/articles/mi_qa3723/is_200608/ai_n17171882) The ‘re-programming’ of ‘civilian’ or commercial applications has by now trickled down into military usage of programmable media, especially, but not exclusively, with regard to digital games, and will be another important trajectory especially in the fourth chapter. One possible hypothesis could hold that programmable media, by virtue of their socio-technological contingencies, will replace or at least override the previous ‘economic’ paradigm of warfare with new rationales based on programmable technologies. I will illustrate this point briefly using an example Noam Chomsky repeatedly quoted in his lectures, indicating how the momentum of economic reasoning in military contexts transcended political camps; one of the more recent occasions was a lecture at MIT on the Middle East crisis held in 2000, which supports the claim that this type of rhetoric is still applicable. (Cf.http://web.media.mit.edu/~nitin/mideast/Chomsky _lecture.html) Hence, the US intervention against the Sandinistas in Nicaragua, which Chomsky himself considers an act of ‘state terrorism’ covered by geopolitical power struggles, was officially and, to a degree, also publicly considered a “sensible policy” even by left-wing liberals like Michael Kinsley, since it allegedly complied with “costbenefit analysis”. (20) The discursive power of rhetorical figures and semantics derived from the field of economics has been the product of a particular political, historical and also technological context, promoted not least by the ‘dual use’ of mathematical game theory both in economic and military contexts. Below, the effects of programmable technologies like digital games and simulations taking this place will be analyzed from various angles, leading e.g. to the perhaps surprising finding that economic rationales have indeed not been abolished, but are, on the contrary, being re-read and recombined within the framework of programmable systems as the new benchmarks of consensual military organization. The convergence of aesthetic, rhetorical and logical properties between digital games and military training simulations shall be illustrated using a brief case study in the form of the promotional videos of VirTra Systems, a company producing high-quality simulation environments for military purposes. (Cf. e.g. http://www.vi rtra.com/real/IVR4G_product_video_DSL.ram) The unique selling points listed in the video, e.g. “realistic video and sound”, “Real Time Scenario Creation” (01’40’’), as well as various ‘gameplay modes’ such as “Convoy and IED training” (01’42’’), are already exactly in line with those of many commercial digital war games. The claim that VirTra simulators use “Hollywood”-level actors (05’37’’) illustrates how, just as with digital games, the film industry still serves as a frame of reference for military training simulations; back cover claims like “the most cinematic and intense combat moments” (Call

15

Real Wars on Virtual Battlefields? of Duty 3) support this hypothesis. Even the military/civilian gap is effortlessly bridged by implementing both usage contexts as simulation “settings” advertised in the video (1’12’’) and, thus, homogenizing them within the programmed rule ecology of the system. A closer look at the semantics reveals how the technical framework of the application pre-structures interpretation, as with the separately available “low-light” scenarios, (1’47’’) dissecting lighting as a customizable parameter instead of an ‘organic’ element of the simulation environment. Furthermore, the described logic of military tool development coincides with ongoing trends in digital game development, most notably with its modularity paradigm aiming at maximum integration of components. One example is the advertised VirTra “drop-in recoil kits”, (2’25’’), designed as an add-on device to enhance the haptic realism of training weapons; this already suggests a characteristically modular approach to notions like authenticity which are similarly influential in user discourse revolving around military-themed digital games and will be revisited in the third chapter. The training concept is described as a “loop of interaction” (2’55’’), specifically through the implementation of a ‘threat fire belt’ delivering electric shocks to simulate being hit in the simulation; the ‘feedback loop’ model is again very similar to (earlier) models of Human-Computer Interaction, specifically in the case of digital games. (Cf. e.g. Friedman, 1995 or Newman, 2002) Finally, the promotional video is heavily rooted in the logic of universal computability. As evidence of the accuracy of its ballistic models, VirTra advertises the fact that Federal Ammunition resorts to their models to “create their ballistic tables” (the current ballistics catalogue can be accessed at: http://www.federalcartridge.com/ ballistics/default.aspx). Reciprocally, game companies refer to those tables to translate them into the physics model used in their games. The tabular representation thereby inherently dictates an interpretive ‘grid’, offering a consistent taxonomy that inherently lends itself to algorithmic expression within a programmable media framework, and even closely resembles the formal structure used in many digital games, thus becoming increasingly iconic. It is evident that a ‘translation’ of data between both functional contexts, rather than the hermeneutically organized presentation of information, is one case of a technologically induced practice stemming from the complex interplay between ‘real’ and ‘virtual’ weapons industries. Many of the aspects introduced in this brief example, like modularity, systems architectures with global variables like lighting and the scope of computability, will later be covered in dedicated paragraphs.

16

Introduction

1.2 Programmable Applications within Media Deployment Policies While digital war games have recently received a lot of scholarly attention from various perspectives, the goal of this book will be to analyze how programmable media in general affect military selfdescription and its overall disposition towards media technologies; this undertaking necessitates both historical and, to a certain degree, cultural contextualization of the few prominent digital game projects with military participation, like America’s Army or Kuma/War, which are usually taken as representative examples. Since the coverage of media infrastructures bridging the gap between military and civilian technology partners has increased greatly in the past few years, with e.g. the U.S. Army creating a dedicated digital games office (the TRADOC Project Office for Gaming; cf. NI-29), tracing the repercussions of program code in key areas related to the military-civilian divide like e.g. economic structures will be a central aspect of this argument. While the proposed notion of ‘programmable media’ will be unfolded in detail in the following chapter, it should be clarified early on that its main frame of reference is the concept of objectedoriented programming (OOP), a programming paradigm or style which has become immensely popular and characterizes both topical programming languages (such as C++) and scripting languages (such as Perl or PHP), thereby accounting for the technological makeup for most applications subsumed as ‘programmable media’ in this book. Heeding Kittler’s criticism of the troubled relation of the humanities with technologies, (e.g. Kittler, 2006) the first step will thus be to derive sustainable models of analyzing media texts by interpreting and adapting structural properties of the technologies used to create them, such as object classes and instances, inheritance, encapsulation and polymorphism in the case of OOP. As a second step, these concepts can then be used as a conceptual nexus for historically comparative inquiries, for example into the early history of war games and their materiality, similar, for example, to the information science concept of the “discourse network” Kittler attempted to illustrate the shifting logic of differentiating between ‘data’ and ‘noise’ under various historical conditions. (Kittler, 1985) However, while Kittler’s close-reading of programming technologies focuses on the 1980s, this book will draw on more recent developments in the field which characterize the current media landscape, and which occasionally require the rethinking of scholarly positions on programming. In his essay on “technological determinism in military history”, George Raudzens illustrates how earlier military technological

17

Real Wars on Virtual Battlefields? breakthroughs were accompanied by the belief that they would not only strengthen a just cause, but help prevent war itself (as e.g. with the Gatling gun in the American Civil War etc.). (Raudzens, 1990: 403) Raudzens’s thesis that weapons need to be considered as pieces within a larger “system of armaments and institutions” (405) rather than as “isolated devices” can be usefully adapted to the consideration of programmable media, including games and training simulations, as weapon technologies. An important effect that is implied in Raudzens’s historical synopsis is the association of specific weapons technologies with particular forms of social and governmental organization and the function of these technologies in modifying or even destroying these forms. This aspect is embedded e.g. in terms like “chariot aristocracies”, referring to a dominance of chariot-producing cultures in the Middle East that was abolished with the availability of cheap iron for weapons and armor. (Raudzens, 1990: 405). These thoughts will be guidelines for assessing patterns of military media deployment and at the same time they will indicate how these strategies of using programmable media frame military thinking, similarly producing systems of governance visible in military operations, e.g. in Iraq and Afghanistan, something which makes the earlier notion of ‘instrumentalizing’ media intrinsically problematic.

1.2.1 MILITARY THEMES AS A CONSTANT IN PROGRAMMABLE MEDIA PRODUCTION In the fourth chapter of his De Re Atari from 1982, Chris Crawford and his co-authors outline the hardware capabilities of early Atari 400TM and 800TM home computers, describing e.g. also character animation devices which were later termed ‘sprites’ but at the time were referred to as “player-missile graphics”. (Crawford et al., 1982) ‘Player’ and ‘missile’ were highly specialized functions for predefined character animation, the latter being especially “useful as bullets” as the user manual unequivocally suggests. Since horizontal motion was easier to implement than vertical motion and since the Atari computer provided “hardware collision detection”, most quickly computable between ‘player’ and ‘missiles’, military-themed patterns of thinking were arguably built into contemporary digital game hardware to some extent and encouraged specific types of gameplay. Another potential reason for the ubiquity of military themes in digital games is the fact that simulating destructibility has to date been the most feasible canon of psycho-visual techniques to produce ‘interactive’ environments, a key selling-point of digital games

18

Introduction in general. The third chapter of this book will present a case study on how aspects like destructibility or natural phenomena have quasi-’organically’ evolved throughout digital game history by virtue of specific program code features and how they modified the corresponding popular interpretations of ‘authenticity’, culminating in the equation of ‘destructibility’ and ‘interactivity’ as communicated in game magazines, advertisements as well as postmortems and other technical documents. This technological momentum of militarythemed games can plausibly be factored into the conceptual shift within the digital games canon from abstraction towards ‘mimetic realism’. While, formerly, war games sought to capture as well as define the ‘essence’ of military operations by coining as concise a symbolic order for it as possible, more recent war games partially seek to reverse this process by trying to achieve a more direct, mimetic correspondence of player action and played action (as e.g. in Quick-Time Events pioneered by games like Dragon’s Lair and the Shenmue series) and pre-structure the game as a programmable system into several discrete elements, like artificial intelligence, terrain deformation or character animation. All these building blocks are represented by dedicated components within a game engine, and the complexity within these elements is gradually increased by constantly adding new functions to a base of reusable working code; at the same time, this techno-epistemological setup is stabilized by implementing more and more algorithmic links between these elements, like AI routines dynamically reacting to terrain deformation or to environmental audio sources. Thus, the contingencies involved in writing program code apparently fundamentally alter the spectrum of conceivable and manifested media formats. This syntagmatic shift in digital war games is accompanied by recent nonwar game franchises often inscribing the game as an intermediary narrative layer within the game, as e.g. in the Japanese Yu-Gi-Oh! or Card Hero series. While this represents a move away from the chess paradigm of symbolically representing a war with a minimum number of interrelated ‘algorithms’, the new style of writing digital games has already retroactively affected board games and other earlier media genres where game pieces have become more and more detailed and lifelike and even programmable elements like ‘artificial intelligence’ are inserted into the logic of physical game tokens in pioneering games like King Arthur. (Rainer Knizia: Ravensburger, 2003) (cf. De Boer, 2004) The insight that game designer and theorist Chris Crawford allows into the development of early (war) game design and which will repeatedly be used as primary sources throughout this book reveals that military themes were among the most commonly projected applications for emerging computer technologies at universities, both

19

Real Wars on Virtual Battlefields? on mainframe and home computers; (Crawford, 2003 (2): 187/88) Crawford’s own initiation with digital games design occurred through a joint project to port the board game Blitzkrieg to a computer and use it as an AI programming challenge. Thus, the early affinity with representing military conflict might in part result from early designers’ personal interests. William Higinbotham, the constructor of Tennis for Two (1958), also developed weapons technologies for B-28 bombers and worked at the Manhattan project at the Los Alamos National Laboratory on the US nuclear weapons program (cf. e.g. Pias, 2002: 9). Trip Hawkins, later to found the biggest publisher of digital games to date, Electronic Arts, initiated an accredited degree in “Strategy and Applied Game Theory” at Harvard and subsequently devised an acclaimed simulation of nuclear war that was even considered at the annual conference of the International Peace Research Institute in Stockholm. (Cf. Retro Gamer 24, 2006: 62) In this context, computer and video games (digital games) often appear as a byproduct of research guided by the logic of military technological development. While the interpersonal networks and the resulting flow of knowledge and ideas provide an important and often overlooked subtext, the focus of my argument will be on the technologies and emergent practices and patterns they produce. Regardless of Crawford’s personal interest in war gaming, the programming languages used for his early games like Tanktics (1976) were originally created for scientific calculations (Crawford, 2003 (2): 194), which represents an important contingent constraint on the spectrum of digital game creation; ludic models of warfare using these types of functions and calculations necessarily required and produced a ‘scientific’ understanding of war that correlated with earlier notions of ‘scientificating’ war, starting with the First World War. As will be elaborated later on, similar technological contingences, such as shifting programming paradigms and middleware (cf. e.g. Cass, Stephen “Mind Games” Spectrum IEEE 39:12, 2002: 40-44), continue and slightly modify this framing impact on programmable media design.

1.2.2 THE CONTINGENCIES OF PROGRAMMABLE MEDIA TECHNOLOGIES In her description of the use of commercial digital games by the Singapore Armed Forces, Gwenda Fong makes the case that, while digital games and military simulations supposedly differ in their approach to realism, their convergence is motivated by a “common set of enabling technologies”. (Fong, 2004: 269) However, all the specific technologies mentioned, like “artificial intelligence”, “user interfaces” and “networking”, equally apply to many other forms of pro-

20

Introduction grammable applications. An important aspect of this book will thus be to identify and contextualize cases of technological convergence and assess their socio-cultural impact; according to Fong, the availability of PC-based real-time 3D rendering techniques in commercial games provided one such breakthrough for the military applicability of digital games. Not only with the advent of the computer have notions of warfare been directly related to their respective technologies of representation. Similar to how tabletop war games in the 18th century both mimicked and shaped the military media utilities of the time (like e.g. maps, tokens as proxies for military units etc.), contemporary digital games set in 20th century conflicts draw on other amalgamations of military and ludic technologies, allowing e.g. the control of joystick-controlled surveillance drones in specific sections of some games. (Cf. e.g. Pias, 2001/02: 39-50) Although in academic literature digital games are often described (mostly for political reasons) as an entirely novel media form, the comparison with older war game systems and military practices is a useful exercise in discerning recurring criteria, like the belief in computability, which are exemplarily elaborated on in the third chapter. For instance, Chris Crawford mentions War in the East (~1974), designed by war game mastermind James F. Dunnigan for Simulations Publications Inc. (SPI), as the “first ‘monster’ game” for its sheer size, requiring four map sheets and roughly 2000 counters. (Crawford, 2003: 13) This criterion of relative scale and detail, metonymically referencing the perceived complexity of the Eastern Front in WWII in an imposing way by strategically violating representational conventions of board games, is but one parameter which can be just as usefully applied to digital games. Accounts of digital war games often gloss over rather than foreground technological continuities compared with older, e.g. board-based forms of war games, which are usually not understood as ‘technical’ media at all any longer because their technological makeup (position markers made from cardboard etc.) is by now widely and comprehensively understood, as well as, more importantly, publicly reproducible and, thus, naturalized. Some issues pertaining to board-based war games such as the compatibility of board pieces or the reusability of rule systems are plausible points of contact with programmable media that will be worked out as systematic topics later on. Furthermore, the hexgrid as a common smallest unit of spatial organization can, in many ways, be understood as equivalent to a data type in programmable media and, in many cases, makes for a more useful common denominator than a single binary digit (or ‘bit’), which has often been consensually suggested as ‘unit of operation’ by media and cultural scholars. The most common data type in OOP is the ‘class’, a specified format of

21

Real Wars on Virtual Battlefields? storing data pertaining to an object which, thus, inherently defines the limits of meaningful, i.e. algorithmically usable, information on that particular type of object; the sample ‘button’ class in the example e.g. is resizable and potentially movable since its scale variables (xsize, ysize) as well as position variables (xposition, yposition) are given.

Illustration 1: UML Example

In digital strategy games, hex-grids were often adopted and subsequently ‘mathematicized’ to make the topography manageable for the AI. However, other data structures quickly emerged with new computational opportunities; for instance, rectangular grids had been popular in the early years when multiplication and division were not yet supported by the hardware, but with the rise of realtime strategy games in the mid-1990s and the presumably ‘unrestrained’ unit movement, the screen resolution effectively became the dominant (and increasingly evident, i.e. unreflected) grid and data type. In other genres such as 3D shooters, data types determining how game engines handle spatial organization historically led to similarly characteristic restrictions. The Doom engine, which did not compute true 3D coordinates, could not differentiate between rooms above each other and, gameplay-wise, used a 2D plane internally. Both hex-grids and data structures in war games predetermine the understanding of war since they are adopted as unquestioned interpretational filters of information (for the purpose of complexity reduction) by the player. The re-combination of data types into data structures in program code is but one example which can and will be analogously applied to the segmentation of the game rules in the following chapters.

22

Introduction In military practice, actual large-scale war games are gradually being supplanted by programmable simulations, but still play an important, however continually renegotiated role. (Cf. e.g. NI-39) Those ‘games’, or rather collective training scenarios, are rather understood today as a dramatic ‘play’ [Schauspiel], a signal of strength both towards the indigenous populace and towards potential enemies. Rather than serving tactical purposes like before, as a simulation or feasible training constellation, the massive real-world war games yield one contingent and emergent instantiation which translates into a complex ‘message’ that, similar to a piece of program code, is fed into both one’s own and enemy information ecologies according to the information warfare paradigm. The rooted nature of this paradigm in the logic of programmable media or digital games and its retroactive impact will be further investigated in the fourth chapter. At the same time, military conflicts or conflict types in military-developed games like America’s Army are similarly iteratively re-played by numerous semi-coordinated players as a kind of ritualized, dramatic performance, which is partially pre-defined by the game’s programmed framework and serves a symbolic rather than practical function. Play sessions being recorded, discussed in forums and re-played effectively de-singularize the events they depict (the numerous renderings of the D-Day scenario in games like Medal of Honor: Allied Assault, Call of Duty or D-Day are a prototypical example) and recursively update the common interpretive frame concerning real-world military conflict, as will be exemplified later, e.g. with regard to forum analysis. These examples should serve to demonstrate the interrelatedness of more and less topical media genres and the necessity of taking a step back in order to aim at a broader perspective.

1.3 The State of Research and the Politics of Documentation In recent years, the amount of news articles, company presentations, studies and other material on instances of technological convergence between the military and civilian sector available both online and in subscription-based on-topic publi-cations has become almost impossible to cover, regardless of the fact that many of these texts never catch much public attention. Some examples of more or less mainstream sources include StrategyPage.com (http:// www.strategypage.com/dls/articles/20030122.asp), CNN (http:// archives.cnn.com/2001/TECH/ptech/11/22/war.games/index.htm l), WIRED News (http://www.wired.com/news/technology/0,1282, 65403,00.html) and dedicated publications such as Military Training

23

Real Wars on Virtual Battlefields? Technology. (http://www.military-training-technology.com/article. cfm?DocID=665) Unfortunately, the current discourse on war games often still reverts to the question of whether programmable media, particularly digital games, foster violent behavior, desensitize the viewer towards images of violence and/or promote weapons handling skills among a mass audience. However, research conducted by the military on use of digital games for training purposes demonstrates that the intended goals and, by inference, the implied ethical problem “is not […][teaching] to handle or fire a weapon, but in learning to lead squads, units, or larger forces”. (Bonk/Dennen, 2005: 12) Thus, military rationales conceptually converge with patterns found in the civilian sphere, most notably the business world, as will be elaborated in the final chapter; in this sense, ‘leading squads’ is not an objectionable but instead a desirable quality in contemporary society. Moreover, preemptive moralization has been an obstacle in the scholarly analysis of military-related media genres, since it usually narrows down the discourse (cf. e.g. “war film is something different from even an approximately faithful rendering of the reality of war”; Büttner, 2004: 81) by dwelling on the impossibility of ‘representing’ war. A possible reason for the apparent public neglect of war games and programmable media technologies might be that the developments described do not fit the current frame of reference of the average news reader, e.g. demanding technical background knowledge and contextual information. In Germany, much of the critical on-topic research is indebted to the logic of Friedrich Kittler’s writings on war and the media. Many of these authors, Wolfgang Coy, Georg Christoph Tholen, Manuel Köppen and Hartmut Winkler, to name but a few, are also geographically located in the proximity of Kittler, having worked or still work in or close to Berlin. This ‘monolithization’ produces valuable insights and has prepared a specific ‘playful’ perspective on the repurposing of media in war times, but it has also necessarily placed other aspects on the periphery. Contributors from the field of ‘computer game studies’, like Hartmut Gieselmann and David Nieborg, often write from a journalistic or production standpoint and focus on comprehensive factual descriptions; some of those texts, however, do not question their own methodology, e.g. concepts like the pursuit of realism (cf. e.g. Graaf van der, Shenja & Nieborg, David B. “Together We Brand: America’s Army.“ LevelUp: DigitalGames Research Conference. Ed. Marinka Copier & Joost Raessens Utrecht: Universiteit Utrecht, 2003: 324-38 http://www.gamespace.nl/content/NieborgVander Graaf_TogetherWeBrand_2003.pdf: 7/8) In her writings, Swedish media scholar Eva Kingsepp concentrated on representational

24

Introduction conventions in WWII games set against examples from war films. (e.g. Kingsepp, Eva “Immersive Historicity in WWII Digital Games” HumanIT 8:2, 2006 http://www.hb.se/bhs/ith/2-8/ek.pdf) Kingsepp relies on the close-reading of selected games and established ideas, like Bolter & Grusin’s notion of ‘remediation’ and Baudrillard’s ‘simulacra’ and ‘hyperreality’, to substantiate her claim that WWII games feed into the desire for ‘reenacting’ history through immersion. This scholarly tradition often presents its object of study as a radically new phenomenon and as representative of the “military-entertainment complex”, a term prominently featured e.g. in (Lenoir, 2000), who traces it back to Bruce Sterling in Wired, 1993. (Cf. 292) Methodologically original work like (Pias, 2001) and (Pias, 2001/02) often pursues a linguistic approach, e.g. working productively and playfully with the descriptive language, and is based on a thorough understanding of the history of technology, yet usually contemplates only historicized scenarios like the Second World War or the Korean War. In comparison, this book will attempt to incorporate both recent technological and politico-military developments with special emphasis on the military activities in the Persian Gulf since the early 1990s and to bring together the numerous heterogeneous facets characterizing military media deployment from a ‘programmable media’ angle. Despite the wealth of information, the argument at hand can certainly not claim to give a comprehensive overview, especially since many potentially valuable sources are possibly classified. Even the reports and articles from military publications used in this book might have been revised for a non-military target audience in some cases, especially since many recent game applications (co)developed by military institutions are instrumental in the agenda of the Global War on Terrorism. Yet, military sources like the Marine Corps Modeling and Simulation Management Office (MCMSMO) and publications like Military Training Technology (http://www.military-training-technology.com/) offer a wide range of coverage for an arguably diffuse ‘target’ audience. (Cf. e.g. http://www.tecom.usmc.mil/techdiv/downloads.htm) While this might simply be standard policy for general research institutions like the Institute for Creative Technologies (ICT), the programmable media paradigm that will be elaborated below and the resulting ‘information warfare’ suggest an implied decentralized, strategic ‘deployment’ of these ‘sanitized’ texts for a number of reasons, some of which I hope to delineate later in this book. A very basic motivation might be to stage the idea of military games as ‘natural’ in popular discourse by aligning it with conventional forms of news coverage. Another reason might be to legitimate the massive financial support for institutions with ties to the mili-

25

Real Wars on Virtual Battlefields? tary, like the ICT, by emphasizing outreach and public awareness. In the following chapters, I will outline other, more subtle and, therefore, influential effects of this alleged policy of transparency, such as inserting partially overlapping units of information into the user information ecologies, thereby quasi-’programming’ society conceived of in terms of program code. In the fourth chapter, I will attempt to contextualize this as a system of interlocking decentralized strategies, understanding these texts as ‘inter-active media’, i.e. as media artifacts designated for inter-active usage, remediation and re-distribution within active user communities; the different use of the term compared to the commonplace notion of interactivity in digital game studies shall be indicated by the hyphen throughout this book. This strategic spread of information coincides with the logic of programmable media in numerous ways; most basically, the documents in question are regularly taken up and multiplied by journalists. (Cf. e.g. Amy Harmon’s inquiry into military use of games for the New York Times; NI-30) Thus, the texts can be said to be ‘compatible’ with the information infrastructure of contemporary news media and, therefore, are exponentially distributed either in their original form, as text segments, or even decontextualized quotes. This form of assumed information policy is closely related to the economic concept of viral marketing, i.e. choosing the formal elements of media artifacts in order to, in program code terms, maximize the compatibility of the data type with the defined methods in the given system or program and, thereby, the interaction within a news infrastructure; as will be shown later on, the plausibility of using biological imagery like the virus to conceive of and describe information brokering can again be traced back to the conceptual nexus of program code that is at the center of this book. The recently increasing tendency towards handbooks and anthologies on war gaming and simulations (cf. e.g. http:// www.hyw.com/Books/WargamesHandbook/Contents.htm) similarly serves to channel disparate interpretations by providing a framework for ‘compatible’ discourse. It is certainly possible to apply this model also to earlier examples of media deployment, like the propaganda ‘machinery’ of the National-Socialist regime, and historical analogies are frequently interspersed throughout the book. Scholarly research on the topic often pursues an episodic approach; for instance, Roger Stahl introduced the topic with an account of how the US military designed the control interface for an unmanned reconnaissance device (“Dragon Runner”) after the PlayStation2 controller, assuming that many soldiers would at least be vaguely familiar with it. Stahl uses this episode to prepare his thesis of the “blurred distinction between the soldier and the citizen” (Stahl, 2006: 113); yet, the episodic approach here circumvents

26

Introduction important related aspects, such as the question as to why and how the PlayStation2 controller, just like most digital game interface devices, produced at least roughly congruent mappings from game to game. While all buttons are freely assignable as far as the game software allows, ‘standardization’ usually emerges as a byproduct of software design and particularly paradigmatic button mappings recursively ‘program’ their users, creating reusable ‘scripts’, so that users are able to interact with most game software intuitively, which, as a constant practice, reinforces both the sense of becoming and the desire to become a quasi-extension of the algorithmicallydefined game rule ecology. Standardization through the reusability of program code will be explored as a general cultural momentum in the third chapter, especially because it shapes the collective production imaginary of how to structure interaction in a digital game. Thus, a closer look at which developments elude public attention is instrumental in making the constraints of one’s own disposition and expectations towards this complex field visible; for example, one usually ignored case in point is the technology transfer from military applications to location-based public entertainment systems, such as theme parks. (Cf. e.g. Katz, 1994) Taking this transfer of mostly network-related algorithms and hardware (110) into account provides a number of vertices to explain the usually disregarded formal convergence of digital games and theme park attractions, like Fightertown™ (cf. e.g. http://www.ibiblio.org/GameBytes/issue18/ misc/ftown.html), as media genres. The fact that the Distributed Interactive Simulation (DIS) technology used for that purpose can link commercial-quality PCs and “high fidelity flight simulators” in one network (112) further encourages the interchangeability of code and concepts on differently scaled platforms and, as will be fleshed out in the third chapter, creates a common epistemological context. With regard to this heterogeneous research situation, it appears productive to consider on-topic, especially academic, literature on digital war games as primary sources and para-textual layers of information to obtain a more comprehensive overview of contemporary conceptions of war in a particular cultural context. German scholars, for instance, already started writing about digital games, mostly as a youth culture phenomenon, in the late 1980s with a special focus on war games, although the term was inconsistently defined at that time. (e.g. Fritz, 1988) This body of rather pedagogical work exhibits a usually implied but still influential notion of the user being ‘programmed’, (cf. e.g. the title of a volume written for the Federal Center for Political Education (bpb), Programmed for Playing War [Programmiert zum Kriegspielen]) which anticipates ‘cybernetic’ models of player-computer interaction developed much later, since program code and basic code literacy were much more visible at the

27

Real Wars on Virtual Battlefields? time; however, the programming references usually stayed on a rhetorical level. Thus, the history of scholarly discourse can be an important subtext since, as in the case of Germany, the continuities are still visible e.g. in the authors and the still noticeable dominance of pedagogical topics. Hartmut Gieselmann produced a book-length account of digital war games in 2002 (Gieselmann, 2002) that sums up several arguments from the German research context. Gieselmann bases his distinction between game, reality and simulation on the psychoanalytical ideas of Freud and D. W. Winnicott, which focus on playing as a child’s activity; (11-14) in this sense, games act as an intermediary stage (termed a “third reality”) for children to negotiate between pleasure principle and reality principle and to explore the latter. Gieselmann attempts to link the “creative” aspect of play derived from Winnicott to digital games, using e.g. the “omnipotence” of the child to imbue trivial things with “magical” significance, to explain the popularity of MODding, i.e. modifying a commercial digital game with tools often provided with the game software. (Cf. e.g. http://www.moddb.com/mods) The model of programmable media sketched in the following chapter and the subsequent applications to other spheres of media usage and parasocial practices will present a different approach to MODding as ‘mental simulation’, exploring the spectrum of technological feasibility given by the game as program. Distinguishing between simulation and game, Gieselmann draws on Baudrillard, who describes simulation as a “third order simulacrum” which does not mediate between reality principle and pleasure principle but substitutes itself for reality (unlike “imitation” and “production”); instead of pursuing these established differentiations further, I will rather undertake a close reading of the specific technologies used to investigate the constant renegotiation of ‘simulation’, ‘game’ and other related discourse terms. Gieselmann references another early German scholar of digital games, Ralf Streibl, whose 1996 project on war in digital games cautioned against an impending militarization of society and pointed out the potential of digital war games to be instrumentalized by the military. Gieselmann’s account of war game developments and theorization characteristically leaves central issues like “realism” unquestioned, arguing e.g. that the game Conflict: Desert Storm II: Back to Baghdad (Pivotal games: SCi Games, 2003) reinforced established concepts of the enemy [Feindbilder] “due to its high level of realism”, which allegedly establishes “a close connection with the real military”. (Gieselmann, 2002: 17) After critically reviewing the notion of ‘realism’ and related concepts against the backdrop of a consistent model of ‘programmable media’ which will be established in the following chapter, it appears that, instead, military concep-

28

Introduction tions of the enemy, most notably the rather amorphous ‘Global Terrorism’, are fundamentally shaped by the enabling technologies used to model them in both digital games and simulations. This implied understanding of the enemy in program code terms simultaneously leads to corresponding self-descriptions and assertions which, through the venue of commercial digital games, ‘trickle down’ into public conceptions. The use of technical terminology and rhetoric is often observable in academic texts without adapting the context of the terms; for instance, Lev Manovich in his “Cinema as a Cultural Interface” uses terms like “[inheritance]“ and “interface“ in their program code sense, but rather casually (cf. Manovich, 1997: 5). The momentum of program code rhetoric and imagery within scholarly discourse will be investigated more closely in the following chapter. Thus, for reasons to be elaborated further below, the corpus of scholarly literature used for this thesis comes less from the still dynamic field of digital game studies, but rather from histories of technology, sketched by authors like Lewis Mumford, Arnold Pacey, Marshall McLuhan and Claus Pias. Even more importantly, the focus will be on close-reading technological documents and scholarly papers from areas like computer science and engineering as primary sources and vertices of recurring dispositions towards technologies that could be characterized as ‘programmable’ in a historical perspective.

1.4 Programmable Media as Socio-Cultural Filters Resorting to games as models of specific areas of society and culture is not a recent phenomenon, but one which has recently picked up momentum and shifted towards digital games as ‘role models’. One such example is Phyllis L. Speser’s interpretation of technology transfer, also an important aspect of my argument, as a game. Speser structures her book roughly according to the key elements of a game, conceptualizing technology transfer according to symbolic ‘game pieces’ and ‘game boards’. (Phyllis L. Speser, The Art and Science of Technology Transfer Hoboken, NJ: John Wiley & Sons Inc., 2006: VII) This technique of conceptualizing technology transfer as a game is applicable especially in the area of programmable media, and the re-use and differentiation of algorithms and components, as well as the creation of tailored business models from this technological setup, will play a key role later in this book. The fact that Speser’s book is written from a policy perspective (cf. e.g. xvii) suggests that the model of the ‘game’ is intuitively plausible in these

29

Real Wars on Virtual Battlefields? contexts and thus translates into actual corporate as well as political decisions. The increasing relevance of digital games as a form of cultural expression and performance [Performanz] can be seen in a number of episodes, like the withdrawal in 2006 of Badr Hakeem, a SaudiArabian professional gamer, from an E-Sports championship after he found that he was due to compete against an Israeli player. (Cf. NI-31) In an interview, Hakeem argued that the alleged ongoing aggression of Israel against the Palestinian people prevented him from playing against his Israeli opponent. Thus, the act of publicly playing against each other appears to be perceived as a heavily politically loaded one, which is complicated by the fact that both were to compete in the digital soccer simulation Pro Evolution Soccer, since soccer, due to its high visibility in mainstream media, has itself a tradition of being a proxy ‘battlefield’ for geopolitical constellations. While the disposition towards games is highly culturally specific, as e.g. Huizinga exemplified in his Homo Ludens, the explosiveness of the case above stems from the peculiar politico-economic constellation of digital game tournaments, something which is also, at least partially, attributable to the properties of program code, as will be argued in the final chapter. The fictional content of the games thereby does not represent an obstacle and quasi-’programmable’ media genres, most notably ludic forms like digital games, have even been implicitly instrumental in making the boundaries between fact and fiction in actual governmental and military practices more and more permeable. For instance, the US Ministry for Homeland Security started an initiative in mid-2007 encouraging science-fiction authors to contribute ideas for future threat scenarios that might actually shape the development of proactive defense measures. (Cf. NI-32) While the media artifacts in question, i.e. the contributions, are written texts, the public contest, or ‘challenge’ (as e.g. in the DARPA ‘Grand Challenge’; cf. http://www.darpa.mil/grandchallenge/), as a type of defense and media policy can actually, as will be laid out in this book, be regarded as informed by the paradigm of programmable media. Instead of previous centralized, top-down research directions, perfectly encapsulated e.g. by the SDI initiative proposed in 1983 and mainly pushed forward by US President Ronald Reagan as its key advocate, these new types of policies actively incorporate a large amount of more or less relevant bottom-up input with an intended, inherent fuzziness. A temporary analogon could be the move of chess AI algorithms from pre-computing as many steps as possible starting from the current constellation to using large movement databases as ‘ideal types’ which are obtained from observed real sessions and tagged as to be algorithmically usable; while this shift

30

Introduction was initiated by the availability of more computer memory, the contest is made feasible by the availability of larger semi-connected user communities as ‘hardware’ to run on. At the same time, this process clearly reflects common programmable media practices, e.g. looking up a topic online by using a search engine algorithm and selecting from a display of probabilistically derived potentialities (Google Suggest; http://www.google.com/webhp?complete=1&hl= en), to the point of being at least implicitly re-modeled by those technologies, just as efficient Google users develop a mental model of the expectable results. Extrapolating from this analogy, the political process in this case appears to be conceptualized as an algorithmically expressible system or a computer program, which currently represents the tacit benchmark for effective processes; this assumption will be explored in detail in Chapter 4a with reference e.g. to military decision-making processes and training constellations. Historically, war and games have always been discussed in similar terms, more or less intricately linked in both popular and academic discourse, with their status within society being constantly re-negotiated. For instance, Henry Jenkins points to the martial terminology used for sports event coverage and the original semantization of chess as a generic battle between two diametrically opposed forces. (Cf. Jenkins, 2003) Johan Huizinga, an early theorist of the interrelations between ludic activity and cultural achievements, argued that play and games were originally tools to contain the use of violence, substituting war for a virtualization of ‘agon’ (to use Caillois’ terminology). (Spreen, 2001: 36) The “enclosure of violence” within “sacred rule”, however, can also be understood as legitimizing or naturalizing war (by ‘defining’ it in the first place). In his comprehensive account of the cultural functions of play, Johan Huizinga gives an insightful anthropological overview of war and ludic activity. (Huizinga, 2003: 89-) In fact, although Huizinga’s personal pre-WWII experience of war is almost inconceivable from a contemporary perspective, his account of how war cannot be understood as an “agonistic [and thereby legitimate] function of society” (90) in many ways fits the description of military simulation games. In other words, playing war games as a collective everyday practice precisely eliminates the ‘play’ aspect and dissolves this connection. I will not pursue this anthropological argument with respect to games in the traditional sense much further, although my model of programmable media convergence also explicitly takes into account the level of everyday practices and experience. Marshall McLuhan described both weapons and games as ‘media’, i.e. “extensions of man”, yet in separate chapters of his Understanding Media. Accordingly, the respective chapter on games, may-

31

Real Wars on Virtual Battlefields? be coincidentally, contains surprisingly numerous examples of war games (e.g. McLuhan, 2003: 234/235). Following McLuhan, the increasing tendency towards specialization in the sphere of work is supposedly being countered by games allegedly propagating generalist actions. While this still applies in many areas of games, the case of programmable media introduces additional complications; as I will argue later, many digital games, by virtue of their technological makeup, profoundly and sometimes even deliberately conflate the spheres of work and recreation, whilst McLuhan still employs these spheres as a binary opposition for building his argument. Intuitively referring mostly to sports-type games, he points out that the main aspects that disqualify war from being a “true game” is that the rules are not known to and accepted by all participants, and that the ‘audience’ involvement is too high, wryly hinting at collateral damage, something which has become all the more topical since the 1960s. (McLuhan, 2002: 240) Again, McLuhan’s comments are useful methodologically for their formal rigidity, e.g. by taking into account the necessity and potential functions of the audience. However, McLuhan understands games as a culturally necessary form of inhibition against “automatism”. (241) It remains to be shown that digital games, by virtue of the programmability, often precisely reverse this notion. Later on, he describes how social changes lead to the interpretation of the ‘obsolete’ or historicized social norms and ‘rules’ as a game, (239) especially in the case of radical and short-term transformations where the overcome norms are displayed as man-made and unrelated to the new circumstances of social reality (following Huizinga’s criterion of games being detached from ‘reality’ in terms of rules and implications). McLuhan’s hypotheses are useful methodological snapshots because they suggest treating ‘games’ not necessarily as material objects but as structures that can equally be applied to construing social transformations. This technique will prove especially useful below while assessing the impact of programmable media on contemporary warfare and its politicoinstitutional context. The direct political implications of digital games, subscribing implicitly to Raudzens’ notion of technologies as influencing political structures, have already tentatively been explored in academic literature; one influential recent approach is Stahl’s proposition of a third sphere of identity, the “virtual citizen-soldier” as a product of the blurring between military and civilian spheres. (Stahl, 2006: 113) Accordingly, Stahl reads the differences between the TV coverage of the 1991 and 2003 Gulf Wars as an indicator of a new ‘embedded’ citizen/viewer. (115) Thus, starting with premises similar to this book, Stahl rather employs a politically-oriented approach. For

32

Introduction instance, he follows Arquilla and Ronfeldt in calling the net war paradigm “more ‘total’ than total war”, (114) thus concentrating on the macro level and utilizing concepts from political and military theory; furthermore, identity politics, like e.g. the double status of the “net war citizen” as both “object” and “subject” of war, are a key aspect of his argument. (114/15) Stahl utilizes Paul Virilio’s ‘acceleration’ hypothesis to sketch the ‘citizen-soldier’ figure who is characterized by the need for constant action rather than reflection, (126) a condition he finds similar to the situation of a game player. This “temporal collapse” allegedly even permeates political semantics as in the Bush administration’s mantra that the United States is “running out of time” in the Global War on Terrorism. (120) Finally, Stahl differentiates between interactivity and participation “in the democratic sense“ (115) and considers the restructuring of “the civic field” (118), hinting at the potential political implications of net war strategies. The ‘soldier-citizen’ model is a valuable tool for potentially mapping these forms onto the currently existing political system based on terms like democracy and citizenship. However, rather than seeking to reconcile these political concepts with the net war ramifications, I will attempt to analyze the enabling programmable technologies as structural role models for socio-economic, as well as military, organization.

1.4.1 WAR AND LUDICITY IN OTHER MEDIA GENRES Politico-military awareness of the cultural significance of ludic activities and symbols often exceeded concurrent public sensibilities. A 1967 military document, the PsyOps Policy No. 36 on the “use of superstitions in psychological operations”, details the massive collaboration of toy producers who supplied suits of Ace of Spades cards for US operations in Vietnam. (Cf. http://www.psywarrior. com/superstition.html) According to Viet Cong symbolism, the Ace of Spades is considered the “death card”; therefore, it was deployed to instill terror and encourage desertion. Since most Vietnamese playing cards are imported from China and their markings differ from those on Western cards, this fatal association arguably already stems from negative experiences with French colonial forces since the mid-19th century, which brought Western cards with them. With reference to the recent use of playing card decks in Iraq, Joyce Goggin contextualizes the use of the Most Wanted playing cards as a medium, both in terms of their material properties (“small and portable”), and their usually positive connotation as a ‘toy’. (Goggin, 2005: 1) Combined with these properties, their inherent horizontal and vertical differentiation (i.e. colors and values)

33

Real Wars on Virtual Battlefields? makes card decks ideal tools for ‘implicit’ learning, e.g. of mathematical tables or vocabulary lists. Extrapolating from Goggin, playing cards can be argued to anticipate a few structural properties of programmable media effectively, such as the segmentation of thematic complexes into ‘units’ and the symbolic encoding, although the information is ‘hard-coded’ into the card and the rules for their combination must be externally agreed upon. More importantly, playing cards can be understood as a form of ‘inter-active media’, a concept all-too-eagerly applied exclusively to digital games at the moment. In this sense of the term, it is not the cards as objects which provide feedback or ‘inter-action’ but their portability (and, thus, exchangeability with others), and their repurposing potential (by altering the external rules) enables usage practices markedly different from reading a book or watching a movie (although, as will be exemplified later, books can also be considered ‘inter-active’ media under specific circumstances). The use of playing cards in Iraq in 2003 reflects this idea: the cards were used both by soldiers for playing poker and also for decentralized, semi-directed educational purposes, e.g. for memorizing names and faces of Iraqi politicians, both by the soldiers themselves and the Iraqi population. The cards were later sold at online auctions as collectible items, which further increases their ‘inter-active’ potential, capitalizing on another intrinsic property as a ‘medium’, i.e. their limited availability. While the game of chess will not be vital to the argumentation in this book, the relationship between war and games proper has already been negotiated by means of, among other specific media technologies, the chess computer. (Cf. e.g. Spreen, 2001: 37) Already in 1950, Norbert Wiener and Claude Shannon interpreted the chess computer not simply as a showcase for AI research, but also as a “machine for taking away the restraints of violence” [Entgrenzungsmaschine der Gewalt]; while the game of chess, from a cultural history perspective, is credited with the canalization of violent outbreaks, chess computers were at that time considered as a “field of possibilities” for optimizing military strategies, influenced by the growing popularity of von Neumann’s and Morgenstern’s gametheoretical thoughts. Contemporary programmable media like digital games have taken over many of those functions without provoking the same critical stance. Scholarly extrapolations, e.g. from a poststructuralist standpoint, as in the case of Spreen, are often rather speculative and figurative, e.g. predicting the interconnection of machines into a “global, automatic network” and cautioning against the “genealogical interface of hypermodern computer and media technology and weapons technology and the atomic bomb”. (37) Departing from a close-reading of technologies, this book is designed to formulate a complementary position which, in programmable

34

Introduction media terms, is more ‘compatible’ with other research strands and is, thus, better suited as a quasi-algorithm used to process and structure the constant ‘input stream’ of related news items. The use of a quasi-ludic rule system within the military, independent of actual war games, can be extended to sports. For instance, the German variation of dodge ball [Völkerball] builds on the hierarchical differentiation between the ‘king’ [König] and a number of in-field players. After all in-field players are hit by the opposing team, the ‘king’ enters the field and becomes ‘vulnerable’; the respective player must be hit three times to win the game. While this setup already implies a martial narrative, the practical implications of the rules on the players are especially noteworthy and, occasionally, coercive, e.g. since players must decide on a ‘king’ who is usually the most agile player beforehand. Furthermore, the act of throwing a ball to hit other players is uncommon in ball games and, especially when team members have been ‘killed’ or ‘imprisoned’ (depending on semanticization), the pressure of being repeatedly targeted has strong psychological implications on the players, something which has even been attenuated, for instance, by including a ban on head shots in the official rules. Baseball similarly encapsulates a particular, historically localizable notion of military conflict (in this case the American Civil War), and the connection was even institutionalized by having troops play the game as part of their regular exercise regime. (Wenner, 1989: 52) This conflation of spheres like military training, sports and also its media coverage was propelled by the fact that by the mid-19th century the commercialization and institutionalization of sports began, which, among other things, effectuated the increasing standardization of rule systems. Similarly, Marshall McLuhan analyzes how particular sports (or games), like baseball, embody a particular socio-cultural order, an order which, however, is simultaneously co-shaped by the contemporary media technologies. Baseball, in his reading, is rooted in the “one-thing-at-a-time” mentality (McLuhan, 2002: 239) and work delegation principles of the “mechanical age”, which is popularly associated with the assembly line as a socio-technological frame. He furthermore speculates that (then) new technologies of representation like “the TV image”, synonymous with electrification, “alienate [media users]” from ‘outdated’ sports like baseball since they establish new organizational paradigms which are reflected in different sports. Put more abstractly, the contingencies of representational technologies, imposed by a specific media technology like television (or the computer), arguably influence both cultural dispositions towards certain sports and the formal aspects of sports as media genres. In brief, the fact that baseball is even less ‘compatible’ with

35

Real Wars on Virtual Battlefields? the socio-technological premise of programmable media technologies than with television could explain why it is comparatively seldom represented as a digital game. On the contrary, sports reflecting program code principles, such as de-centralized planning and interlocking local ‘scripts’, like American Football would be further propagated by the increasing spread of programmable media technologies; while this hypothetical interrelation cannot be empirically validated, the assumption itself is intuitively plausible, as will come out more clearly in the following chapters. The same pattern holds true for the medium-specific limits of representation imposed on warfare themes. MOUT-type operations and decentralized semiautonomous units, which are currently discussed extensively among military strategists, coincide with program code paradigms like Object-Oriented Programming (OOP) much more than the quasi-linear progression of massive tank divisions that characterize earlier military practices (and older programming syntagmas) and, for this as well as other reasons to be sketched below, are disproportionately more common in recent military-themed digital games. Finally, war toys are a ludic media genre that has embodied as well as shaped specific ‘war mentalities’ throughout its long history; in fact, the current historicization and scholarly neglect of toys in favor of digital games produces critical blind spots, while, before the pervasiveness of programmable applications, digital games were often perceived as a continuation of the debate revolving around war toys. (Streibl (2), 1997: 2) In February 2007, Märklin, a big player in the toys market founded in 1859, officially switched back to military semantics, adding a new product line called Metal Military Mission by Märklin (4MFOR). (http://www.4mfor.de/) Märklin has been one of the biggest producers of model railways and adopted digital control mechanisms for miniature trains already in the 1980s. Integrating the company name into the franchise is a particularly strong commitment; moreover, the label ‘4MFOR’ strongly references military abbreviations (e.g. KFOR) and CI design. Interestingly, the move towards military models followed a takeover by British investor Kingsbridge. (Cf. e.g. http://www.net-tribune.de/article/020207 -141.php) The Märklin model catalogue, juxtaposing extremely precise information on the original vehicles (e.g. dimensions given in millimeters) with model properties, can be read as a continuation of programmable media characteristics extending into toys production; apart from what Lev Manovich termed ‘database logic’, encyclopedic and systemic thinking are clearly visible since all the physical models are built on the basis of digital 3D wire-frame models, such as the one used in the above advertisement. These traits are even more clear-cut in other popular types of toys like trading-card games.

36

Introduction There are at least two similar ‘official’ games endorsed by the US military, Mission Command (cf. e.g. http://www.boardgamegeek. com/game/12425) and Daring Eagle (cf. e.g. http://www.boardga megeek.com/game/12424), a combined board and card game, both produced by the Army National Guard. The case of long-standing brands or game syntagmas like the Märklin models or also the 2008 re-issue of the iconic board-based war-game Risk can be read as a useful indicator of changes in the concurrent popularly communicated notions of warfare. Risk (Parker Bros., 1957) was originally released in 1957 and, to some degree, encapsulated the Cold War idea of ‘global domination’, despite being released in France and featuring a game board and board pieces referencing the Napoleonic era; the later revision was titled Risk: Black Ops (Hasbro, 2008) and most notably introduced a new resource system based on capitals and cities as well as more differentiated objectives other than global domination. The emphasis on cities within the rule system clearly reflects the recently increased attention towards urban conflict theaters and the training focus on MOUT operations that will be discussed later in connection with concurrent digital games. Even more importantly, the more localized and even shifting objectives take into account the fundamentally different premise of contemporary warfare, in which clear-cut victories or even the promise of ‘global domination’ have no place. The modifications to the rule system are interpreted as designed to accommodate new playing styles which are characterized by “shorter time commitment” (c.f. e.g. http://www.blog.newsweek.com/blogs/ levelup/archive/2008/02/04/is-risk-black-ops-from-hasbro-the-ne xt-big-hit-for-ea-casual.aspx) rather than to reflect changes in the perception and interpretation of warfare. As this paragraph attempts to show, considering digital games at the expense of other media genres is a pitfall that needs to be avoided; on the contrary, a comparative look at the enabling technologies and their impact on practices and interpretation of the military/media industry can be instrumental in situating digital war games within a broader socio-historical context.

1.4.2 MILITARY THEMES AS STRUCTURAL CHALLENGES OF REPRESENTATION From a comparative angle, even decidedly pre-digital media coverage of wars exhibits a tendency to transgress inherent representational limitations of media genres. Already Johann C.L. Hellwig (1743-1831), weighing his understanding of war against the visualization techniques offered by chess as a contemporary game and medium, found that the “need for [c]ommunication” [‘Bedürfnisse’

37

Real Wars on Virtual Battlefields? nach ‘Communikation’] among game units (meaning e.g. the feeding of horses instead of communication in the currently familiar sense) overstrained the rule ecology provided by chess. From a current perspective, these categories of war induced by Hellwig from observation appear perfectly congruent with the aforementioned logic of OOP, a key technology in most programmable media production including digital games. This notion of war-related themes ‘challenging’ representational means and conventions overlaps e.g. with Manuel Köppen’s broad investigation of the shifting position of the ‘observer’ in war scenarios since the 19th century. (Cf. e.g. Köppen, 2005: preface) In this perspective, though, war is found to challenge sensorial perception and is, accordingly, related by Köppen to other arenas of modernity like traffic and the general habitat of the modern metropolis. A more complex example of this phenomenon is Ernst Friedrich’s War against War [Krieg dem Kriege], a book published in 1924, which consists mostly of extremely graphic photographs of the trenches and victims of the First World War, condemning warfare in the name of capitalism and exploring the functions and new possibilities of documentary photography at the time. While War against War certainly defies most literary syntagmas, it even starts out with a quasi-’inter-active’ device (in the sense laid out above): a list of people or nations that would possibly like to ban the book. Friedrich achieves both the integration of the potential use and abuse of his text into the, however pictorial, narrative and the introduction of his book as a ‘tool’ [Werkzeug], an ‘inter-active’ medium, rather than as a literary work [Werk]. (Friedrich, 1991: 6) The direct address to the reader and the constant multilingualism (with all texts and annotations given in English, German, French and Dutch) support this claim. Furthermore, War against War strongly exhibits the belief that the newly discovered medium of war photography, more precisely the feasible production of photographs on the battlefield, could have a sufficient public appeal which might be deterring enough to prevent future wars altogether. The idea of an older medium, in this case the written word, being ‘[insufficient]’ (22) for the “[painting]” of modern warfare and its atrocities, was, at least for some time, a new thought, inspired by the shock those never-before-seen images instilled in contemporary society. The photographs are often characteristically set against recurring metaphors of the time, such as the “field of honor”, thus attempting to dismantle them through the raw force of a new medium. Ernst Friedrich’s use of photographs as pacifist ‘propaganda’ can be considered as an opposing strategy to the instrumentalization of photography that Susan Sontag analyzes in her Regarding the Pain of Others; both types of strategies, however, consider ‘li-

38

Introduction near’ or ‘static’ media photography as ‘material’ for repurposing and recontextualization, which could be summarized as ‘implied interactivity’, i.e. formal parameters of both the photographs and their media contexts which encourage remediation and circulation. Thus, implied inter-activity would denote the potential of formally ‘hermetic’ media, such as texts and images, to produce non-standard usage patterns that could be labeled ‘inter-active’; in programmable media terms, it would refer to the propensity of data types to respond to a given set of interlocking algorithms within a program. This differs notably from the understanding of texts as ‘interactive’ by allowing for multiple, recursive re-readings and rather refers to formal elements that pre-structure collective forms of media usage and the parasocial organization of media users, just as War against War spawned a ‘movement’ of activists advocating the use of the thennew medium for pacifist purposes. While this alternative notion of ‘implied inter-activity’ will be further explored below using fan site toolkits for digital games as an example, its most fundamental structural prerequisite is the easy (technical) modifiability and segmentability of the media text as source, which is distinctive from digital media (this would be partially congruent with Lev Manovich’s notion of ‘modularity’; Manovich, 2001: 51) and programmable systems. For instance, online video networks like youtube and metacafe host numerous military tribute videos, ranging from sentimental to belligerent in affect, which reassemble recorded TV footage by using programmable editing tools like Adobe Premiere. The independent production of military tribute videos is surprising given the chorus of critical voices against allegedly increasing military autonomy in Western societies and the assumed critical stance of independent media production in particular. Earlier accounts sometimes reduced independent media production exclusively to alternative news programs with a strong investigative mission, such as Undercurrents (cf. Taylor, Lisa, Willis, Andrew (Eds.) Media Studies: Texts, Institutions, and Audiences Boston: Blackwell, 1999: 141), which masks out potential external influences that could account for occurrences like the tribute videos, while some recent textbooks like Democracy and New Media, co-edited by Jenkins, Thorburn and Seawell in 2003, already provide a critical counter balance to the democratic promise of the independent media. While most of these accounts disregard formal elements and technological contingencies of programmable tools and media artifacts, the criterion of technically feasible modifiability and segmentability could yield a complementary perspective. Most videos, accordingly, are collages of still images, generic text objects and roughly cut news footage usually underscored with a continuous

39

Real Wars on Virtual Battlefields? piece of music. The second chapter will establish and support the hypothesis that this form of media production can be understood as inherently ludic, with the source material as playing field and the media tools as board pieces or tokens. Although there is usually no definite goal, the paradigmatic quality of ‘official’ military video syntagmas produces a mixture of free-form play (or Paidia according to Roger Caillois) and mimicry of these syntagmas (Ludus). For instance, one video labeled Bomb Saddam and described as a “tribute” to the soldiers serving in Iraq offers a collage of extremely short shots from war documentaries and news broadcasts arranged using the formal repertoire of music videos, e.g. exhibiting a very precise synchronicity between video and sound track, but also closely resembles official recruiting material of the US Army and other military institutions. (Cf. http://www.youtube.com/watch?v= wNWtFNS6K08) To conclude, the main goal of this book will be to provide a framework for tying together the various forms in which programmable media, most notably digital games, appear in contemporary Western societies and, as an outlook, to provide a common ground for discerning culturally specific ruptures.

1.4.3 FROM LUDICITY TOWARDS DIGITAL GAMES While the shift towards algorithmic expressibility which characterizes digital games and simulations has already been present in previous forms of games, the ‘mathematicization’ of games has not been constantly linear across historical periods and cultures, as e.g. an article by Christoph J. Scriba on the British mathematician John Wallis (1616-1703) exemplifies. One of the two technical applications by Wallis, Scriba, chooses to illustrate his hypothesis of the increasing mathematization of technology since the 17th century as a game or, more precisely, a “ring puzzle”. (95) The basic idea behind the mathematical modeling of the game was optimization; the other application accordingly models the statics of a roof as an optimization problem. The game’s principles were formalized through a customized mathematical notation (105/107); this ‘algorithmization’ turned the solution of the puzzle game into a ‘routine’ task (110) and, thus, mostly eliminated the notion of play by subdividing the larger goal into subtasks and, more importantly, presented a (theoretically) scalable model (103) that could accommodate any change in complexity. Thus, the tools for conceptualizing games like the symbolic mathematical notation by Wallis fundamentally alter the perception of the game as well as its cultural ascriptions and functions; by providing a scalable ‘algorithm’, Wallis eliminated the cri-

40

Introduction terion of difficulty or challenge that characterizes games in a cultural historical perspective. As will be elaborated below, playing strategies used in digital games like ‘powergaming’ clearly revitalize and exploit this understanding of games and, thus, blur the boundary between the logic of play and work-related rationales. Mathematical game theory as laid out by Morgenstern and Von Neumann was not so much read critically as a foundational text projecting a particular social ‘ideal type’ but rather as a quarry of inspiration in very diverse areas, such as, most recently, e.g. the design of socially conscious AI behavior for more efficient educational software tools. (Cf. e.g. Glass, A., Grosz, B. “Socially Conscious Decision-Making” Proceedings of the Fourth International Conference on Autonomous Agents 2000 http://citeseer.ist.psu.edu/ 529200.html) For that reason, it is useful to take a step back and identify some convergent aspects between game theory applications; for instance, the case of e-learning tools exhibits a highly formalized model of learning, stemming at least in part from technological contingencies, which, as will be elaborated below, constitutes one often overlooked point of contact between military and civilian technologies. A key vantage point within mathematical game theory is the model of interdependent ‘rational choice’ (cf. e.g. http://plato.stan ford.edu/entries/game-ethics/), which can be understood as a behavioral archetype that envisions society from the desire for economically predictable actions and, by analogy, e.g. also informs the notion of the learner in e-learning applications. As Pias indicates, the “classification” of problems already depends on the tools employed in its solution, leading to the increasing assessment of military problems according to “calculation time, number of steps of an algorithm, and memory size needed”. (Pias, 2002: 2) Thus, considering digital games and military simulations as a continuation of game theory assumptions, the notion of behavioral standardization, ‘laboratory conditions’ that are never met outside of mathematical models, is implied in both military training and the recursive projection of effective player strategies while playing a digital game. Moreover, Von Neumann and Morgenstern explain the usually counterintuitive concept of multiple attributions [Zurechnungen] as the solution for a game-theoretical problem in economics by comparing it with the consensual idea of a “behavioral standard” in actual social organization, i.e. by tapping into the level of observable, personal experience. (Von Neumann/ Morgenstern, 1973: 40) This analogy is carried even further, e.g. by inversely mapping the observed stability conditions for a given behavioral standard in a society onto the mathematical definition. (41) Von Neumann and Morgenstern continue to back the viability of their analogy (42) and,

41

Real Wars on Virtual Battlefields? consequently, deemphasize its bidirectional character; that is, not only is it exclusively an ‘ideal type’ of society that fits the mathematical model, but the consecutive widespread adoption of the theory since its proposition in 1943 has recursively shaped the projected societies. Concerning the ‘purity’ of their method, the authors interestingly admit that a “verbal explanation” of the rule systems they describe, while not universally applicable, is often possible, although keeping “mathematical rigidity” throughout appears more consistent. (279) This peculiar and unsteady relationship between observation (e.g. of social processes) and mathematical modeling, i.e. formalization, or between analogy and mode of description, constitutes the most relevant aspect of mathematical game theory for this argument, since it re-occurs on various levels of programmable media genres like military simulations, e.g. in the approximative development of physics models and the analogous common notion of ‘realism’ which is constantly and fluidly renegotiated, both in playing and designing digital games, as well as in refashioning them for training or ‘other’ purposes. While mathematical game theory is often treated as a historicized phenomenon and is closely linked to individual figures like Von Neumann/Morgenstern but also Herman Kahn and others, it is worthwhile to juxtapose it with more recent technological developments and to read it as a conceptual spectrum which is still influential today within an entirely different context. Most prominently, Claus Pias has been analyzing mathematical game theory along these lines, highlighting for instance the often neglected affective connotations of its practical application, like the “humiliating potential” of the early computers, such as the ENIAC used in the resolution of the Korea War (Pias, 2002: 1) Pias gives an illustrative example of his ‘linguistic’ approach in the same passage as his later elaborated thesis (inspired by Günther Anders) that computer games “declare their players simply as [unzurechnungsfähig]”, an attribute that would roughly translate into “certifiably insane” in English, but which, in German, has a second, literal meaning which is ‘not being computable’. In many ways, this book subscribes to Pias’ larger project of an “epistemology of the digital computer”, (1) although my argument focuses less on linguistic and affective inferences but on technologies and practices. Not only digital games but also specific game-related technologies are being used as tools for interpreting and recursively affirming contemporary society and culture. In October 2006, a team of computer scientists at Northwestern University, Illinois announced the development of a software tool (News at Seven), funded by the National Science Foundation (No. 0535231), that automatically gen-

42

Introduction erates a news format from online news sources using the Source Engine from the popular game Half-Life 2. (Cf. http://destroyer.cs. northwestern.edu/about.php) One reason for choosing the Source Engine might have been its focus on believable character representation, using e.g. facial animation and animation blending; consequently, some of the modifications of the original engine were routines for randomized and fixed-interval peripheral motions blended with the current bone animation state. An important aspect of News at Seven is the implied belief that news programs’ syntagmas can and should be sufficiently formulaic to be automatable, a belief which is apparently consensual enough to ensure public funding. Even ‘relevant’ information from the blogosphere and additional images are automatically collected and displayed by the News at Seven software. Thus, visual elements of representation familiar from digital games enter the daily practice of daily news watching, which is fairly constitutive, at least in Western democratic ‘information societies’. Complementarily, game footage from games like Brothers in Arms: Road to Hill 30 is increasingly being used in documentary TV formats. (Cf. e.g. http://www.game spot.com/xbox/action/brothersinarms/news.html?sid=6140685) By automating the selection and presentation of information, the cultural status of news formats is, however unnoticeably, critically challenged, since this status was, at least partially, based on the fact that the news formats presented a personalized, criticizable and accountable instance, instead of a set of configurable system parameters and filters. Another, earlier example which demonstrates the constant exchange of technologies and concepts between war games and digital games in general is the origin of digital role-playing games (DRPGs) from military games and simulations, a development which was anticipated already in the tabletop war games and role-playing games of the 1970s. One of the most prominent cases was the shift of influential developer STRATEGIC SIMULATIONS INC. (SSI) from producing military simulations to creating formative DRPGs based on the popular Tactical Studies Rules (TSR) franchise Dungeons & Dragons. (DD-SSI-1) Arguably, the major innovation in the still amorphous DRPG genre introduced by SSI was the implementation of AI opponents built on “list[s] of priorities” (83), i.e. weighing a simple set of possible actions against a very basic analysis of the given situation; this AI concept, which, in heavily modified form, is still in use today, even allowed for emergently “concert[ed]” AI behavior when applied to multiple agents in the same scene. At the same time, it permanently shifted the formal conventions of the genre towards tactical battles, most notably since the program code was inspired by and even partially reused from earlier military simulations the company had

43

Real Wars on Virtual Battlefields? worked on. DRPGs like Wizard’s Crown introduced disproportionately detailed combat rules, including e.g. injuries and bleeding, which had been adapted from the company’s six-year experience with military strategy games. The same case exemplifies the reciprocity of technology exchange with the company’s immensely popular Panzer General (SSI: SSI, 1994), re-importing DRPG conventions (and, thereby, algorithms) like a persistent experience system for all units which provided for gaining prowess depending on a unit’s performance in battle; (86) this system has subsequently become institutionalized in both genres. To conclude, one further key objective of this book will be to identify changes in the actual implementation of military operations and self-organization based on programmable media technologies. Historically decisive examples of the relationship between media representations and military strategies are e.g. the versatile role of optical telegraphy during the Napoleonic Wars and the use of chronophotography or aerial (reconnaissance) photography during World War I (Köppen, 2005: 1). This is especially relevant since, due to the aforementioned ‘inherent ludicity’ hypothesis, which will be elaborated in the following chapter, programmable media occasionally reference and re-interpret these representational conventions. For instance, the loading indicator screen in Full Spectrum Warrior decontextualizes the logic of satellite photography by using the continuous zoom into a satellite image of the respective environment as a loading indicator, i.e. playfully tying the depth variable of the satellite zoom state to the progress variable of the loading routine provided by the game engine. Thus, through quasi-ludic extrapolation, elements and visual syntagmas, like satellite views, are detached from their original functional frame by being algorithmically expressed and transcoded into a different program context.

1.5 Digital Games as a Subset of Programmable Media Most research on digital games has highlighted the ludic aspects and treated its objects primarily as games; focusing instead on the common denominator of program code as proposed in this book allows for meaningful cross-references to military-related findings in other media genres, like the increasing reliance on quantifiable (and quasi-algorithmicizable) information in current TV news coverage. Many, especially brief news reports close by detailing how many people were affected (e.g. killed, wounded or rendered homeless) in the respective event. While statistics also used to be given in earlier news accounts (cf. e.g. German TV coverage of the Turkish invasion

44

Introduction of Cyprus in 1974: http://youtube.com/watch?v=hgmJKhXJ0bM), the markedly shorter coverage of each event in current news formats puts far more emphasis on the numerical data, even for comparatively ‘peripheral’ or local events like traffic accidents. By using them as indicators of quantifiable information or ‘variable names’ in programming terms, labels such as ‘killed’ or ‘wounded’ are increasingly naturalized and even implicitly arranged into a hierarchy, with e.g. people killed bearing more strongly than wounded or even homeless, with the final implied ‘severity’ constant being theoretically computable by adding and averaging those variables. Accordingly, news items like car bombings or natural disasters are conceptually ‘homogenized’ by being made expressible using the same ‘data type’, i.e. a set of parameters that sufficiently describe every conceivable type of event and recursively modify the boundaries of conceivability. These categories are, in turn, fixated in actual programmable media structures like the MIPT Terrorism Knowledge Database, categorizing terrorism-related incidents by tactic, target type and even binary flags like ‘suicide’ and ‘US Attack’; according to this logic, the destruction of the World Trade Center is filed as “Al-Qaeda Attacked Business Target“ (Sept. 11, 2001, United States)“. (Cf. http://www.tkb.org/Incident.jsp?incID= 7757) While this overemphasis on numerical data can be read as either a symptom or a cause of the increasing pervasiveness of programmable media, it documents a turn towards algorithmically processable information on behalf of a critical public which will be further investigated in the third chapter. In part, this strategy is ‘imposed’ by the unmanageable flood of information and the decreasing tendency to generate (or, from the recipient’s side, accept) narrative coherence. Quasi-programmable applications like the MIPT database, but also much more commonly used cultural techniques like electronic program guides (EPG), increasingly permeate everyday life and arguably require the shifting of focus away from digital games towards a broader notion of inherent ludicity in programmable media objects. For instance, car navigation systems tentatively transform driving a car into a ‘game’ by cross-linking physical input and on-screen symbolic manipulation, i.e. the congruence of projected route and actual route as represented by lines and icons. Consequently, navigation systems already spawned apparently counterintuitive but nonetheless perfectly consequential applications, like e.g. a patent for a navigation system offering digital games that use geodata (cf e.g. http://www.freepatentsonline.com/6401033.html), which explicate the intrinsic ludic potential. From the beginning, i.e. before they became naturalized and reduced to their practical purpose in public discourse, 3D navigation systems were much more

45

Real Wars on Virtual Battlefields? immediately linked to digital games. (Cf. e.g. http://www.wired.com /cars/energy/news/2004/06/63842) One of the most convincing recent takes on formal game analysis, which implicitly understands games as programmable applications in an attempt to reconcile literary studies and computer science methodology, proposes ‘unit operations’, i.e. “modes of meaning-making that privilege discrete, disconnected actions over deterministic, progressive systems“ (cf. Bogost, Ian Unit Operations: An Approach to Videogame Criticism Cambridge, MA: MIT Press, 2006: 3), as the principal analytical currency. With concepts like ‘fungibility’, describing the conditions of successful interplay between ‘units’ of programming, criticism and analysis, Bogost unfolds a neo-structuralist approach to game analysis exhibiting a number of methodological overlaps with the approach chosen for this book; however, while the author focuses on digital games and game criticism, the argument at hand requires a more macro perspective and, thus, will only reference Bogost, as well as related authors, to demarcate the model of programmable media analysis at hand. In a recent ‘meta review’ of scholarly analysis focusing on the entertainment and military sector, Peter Bürger argues with Ekkehardt Jürgens, at least from a German perspective, that the critical review of “media products” and “media structures”, i.e. the aesthetic assessment of individual media texts or genres on the one hand and economic, political and technological transformations on the other, is still separated. (Bürger, 2007: 4) Thus, instead of defining programmable media as a tangible subset of media phenomena and discussing them on the object level, I will attempt to use programmability and the ‘inherent ludicity’ hypothesis as a theoretical framework to identify continuities and ruptures between technologies and social constructs that can be read as quasi-programmable. The benefits of this approach can be illustrated e.g. by taking a closer look at peer-to-peer (P2P) tools as programmable media producing similar effects (e.g. in the handling of military-related material) to digital games. A sample query in the popular tool Emule, which allows users to open folders on their computer for others to download from, demonstrates the abundant availability of military-related documents, most prominently combat and leadership training manuals as well as historical overviews. The often rigid nomenclature rules, including e.g. numbering titles in a series, hint at collecting practices as one result of the technical infrastructure provided by Emule. Apart from the aforementioned ‘encyclopedic’ genre of textbooks, other media genres, like e.g. dioramas and military models, as well as instruction manuals for constructing dioramas, are offered and di-

46

Introduction verse historical periods covered (from Roman antiquity to the Second World War). The frequent sharing of instructions on how to build military dioramas suggests an increasing network integration of hitherto isolated practices like building and painting military models comparable to developments in user-generated programmable media content, like the collective authoring of digital game modifications (MODs) that will be covered in more detail later on. Since the users’ file lists and profile data are made available to other users, P2P usage must be considered as at least potentially a social activity where e.g. a user’s file repository constitutes a powerful element of identity politics and social distinction; other Emule functions, such as the chat options, ‘preferred source’ lists etc., support this claim. In this constellation, collecting military-related material can be interpreted as pre-structured by quasi-ludic principles embedded in the software tools as programmable media. One quasi-ludic element is the momentum of encyclopedic completeness implied in the regular and differentiated file nomenclature and consecutive numbering reminiscent of popular video game structures, most prominently the Pokemon franchise and derivative games. The example of Emule and P2P tools, while pointing at an often overlooked arena of brokering military-related information, can be plausibly approached by using the ‘implied interactivity’ hypothesis, even though, for want of empirical data and a long-term perspective, its historiographical implications can only be touched upon as part of the overall topic. One such implication is a divisible and positivist notion of history as a not necessarily linear, but discretely segmentable, compound of phases produced by the contingencies of serialized file labeling and the need for discrete categories, whether they are precise periods of time, nationality or weapon types (cf. e.g. file names like “Roman Military Clothing 100BC200AD”). Broadly read against the ‘collective intelligence’ mode of media usage proposed by Pierre Levy, (cf. Levy, 2001) both P2P tools and independent media production lead to a quasi-ludic ‘reverse engineering’ and collective ‘mental simulation’ of distributed knowledge about military simulations, both concepts which will be revisited and expanded later on, which can e.g. tentatively be seen in discussions on the public accessibility of genuine military game software and modifications. For instance, numerous documents were shared using Emule (at least at the time of writing this book) that address the assumed availability of the Virtual Battle System 1 (VBS1) modification for the commercial game Operation Flashpoint, commissioned and used by the US Army.

47

Real Wars on Virtual Battlefields?

Illustration 2: Sample Query for VBS1 on Emule One user clearly lists the respective technical differences between the off-the-shelf game and the military version, including the afteraction review (AAR) function, an observer module for the instructor (similar to the freely maneuverable camera that many multi-user games played in E-Sports tournaments have consecutively taken up for broadcasting their game sessions more efficiently) and a slightly more detailed soldier model, which has, in the following, been independently re-created and published as “Infantry Model v2.0”. Thus, there is apparently an interest in making the ‘technological advancements’ and resources of the military MOD available as public domain modules. Furthermore, making the VBS1 available to the public can arguably be interpreted as a hacker’s ‘rite of passage’, indicated e.g. by the overall rhetoric and the multiplicity of proposed ways of getting the hacked game to work, something which would support the assumption of quasi-ludic principles as a pattern of Human-Computer Interaction (HCI). Yet another document shared through Emule alludes to a specific case of abusing the interest in VBS1 by substituting the desired file with a ‘camouflaged’ Hollywood movie rip, thereby prototypically replicating one of the foundational strategies of the information warfare paradigm, namely ‘disruption & destruction’ by inserting ‘faulty’ data into another information ecology to produce system-immanent dysfunction, something which will be elaborated upon in the fourth chapter. The file list generated by ssearching for VBS1-related files creates a contingent ‘historical’ snapshot or canon of how users exchange strategies to hack the software, share allegedly hacked files and source code or discuss whether the modifications made to the original game justify the efforts in the first place. In this instance, Emule is used as a communicative platform that is ‘programmable’, most basically by

48

Introduction enabling the recursive use of different search terms, and, as a ‘historical’ document, exhibits many features of the ‘database logic’ sketched by Lev Manovich in rarely encountered purity. At the same time, it technically enables and, thereby, encourages the ‘interactive’ distribution and repurposing of ‘hermetic’ digital data, like military textbooks, causing a shift from the formerly singular instances like War against War to common practices of collective media usage.

1.6 Towards an Integrative Model of Programmable Media Analysis The key to understanding programmable media convergence at the military/civilian margin will be an integrative and flexible model of programmable technologies based on the examination of key technologies in an attempt to transfer the basic structures and patterns and apply them to other areas like social organization and technology transfer. Many texts written from a technology studies standpoint align themselves with one of the diametrically opposed ‘determinist’ vs. ‘contextualist’ camps. Technological determinism, often negatively connoted as naïve and positivist due to its origin and heyday in the late 19th century, identifies technological advances as the driving force behind historical processes. Inversely, contextualism and the ‘social construction of technology’ (SCOT) hypothesis, popularized e.g. by Lewis Mumford and, more recently, authors like Bruno Latour and Wiebke Bijker, focus on the socio-historical circumstances that shape technological development (a concise summary can be found in: Erik Baark, Lightning Wires. The Telegraph and China’s Technological Modernization, 1860-1890 Westport, CT: Greenwood Press, 1997: 7-9) For instance, Mumford argues that the development of bow and arrows by the Paleolithic man was first obscured by an over-reliance on stone tools and their implicit cultural functions, which drew attention away from technological advances in the use of wood and cords. (Mumford, 1977: 138) He points out that this weapon was the “first real machine” and exhibited in its makers a “remarkable capacity for abstract thinking”; actually, not only does the design of a bow and arrow require abstract thought but, as McLuhan highlights and as will be elaborated in the fourth chapter, it also fosters the development of these mental capacities. This extension of the initial assumption, which Mumford himself does not explicitly make, could be supported by another observation concerning the bow he describes: the use of a hunting bow as part of an archaic stringed instrument used to produce music. (139) In his account, technological and artistic “advances” occurred in paral-

49

Real Wars on Virtual Battlefields? lel without any implied hierarchical order. The ambivalence of being both a weapon and a tool of cultural production, characteristic of the bow, can still be found, to an even more extreme extent, in current programmable media systems, an instance of convergence which is often overlooked due to the incompatibility of artistic and military discourses. Later, Mumford hints at other instances of inspiration in “art and sexuality” (148) derived from weapon development. Although it might appear as an ‘easy way out’, I would argue against the mutual exclusivity of technological determinism vs. contextualism. Baark points out that contextualist approaches tend to dwell on the “micro-level of society”, i.e. on the set of values and cultural imaginary that constrain the development of technologies at laboratory-level at a very early stage. Thus, by differentiating the level of analysis and occasionally overlaying multiple examples and readings, both approaches can and will be utilized in this book as ‘mainsprings’ or ‘algorithms’ that continually readjust the emergent perspective on programmable media and, ideally, keep it dynamic. Methodologically, an attempt to delineate an integrative model of programmable media analysis should take into account earlier media theories which directly work with the respective technological setup, such as Michel Chion drawing on the formal contingencies of filmmaking and the motif of the constantly unwinding projector in his writings on sound and music in film. Due to the structure of scholarly discourse, Chion’s simple but fundamental observations, such as the fact that the advent of sound films inherently necessitated the stabilization of projection speed (cf. Chion, 1994: 16/17), are often overlooked due to their technical and de-mystifying argumentation, but they nevertheless profoundly shaped the formal conventions of the medium and are immediately transferable to programmable applications, as will be exemplified below in e.g. considering the shift towards game engine design and reusable components. Similarly, Genette in his Paratexts: Thresholds of Interpretation departs from a close reading of production technologies and technological constraints on the formal spectrum of print media, pointing to aspects like typesetting, cover design and even the material properties of paper in books. Even though Genette focuses primarily on deliberate design choices such as e.g. the layout of a poem on a page, this approach is more useful for an integrative model of programmable media than many texts discussing digital games specifically. Consequently, the operative vocabulary Genette introduced, like the distinction between peritext and epitext based on the ‘distance’ to the main text, is much more ‘compatible’ and ‘transcodable’ (to revitalize programming terminology), i.e. designed as a ‘tool’, when applied to more recent technologies than many scholarly texts focusing on more recent media phenomena. There-

50

Introduction fore, rather than covering as many available studies on digital games as possible, the argument presented in this book will draw mainly from a critical re-reading of established theorists like Marshall McLuhan and Benedict Anderson by substituting the respective paradigmatic media genres (like e.g. the map and the museum in Benedict Anderson’s writing) with programmable technologies, but especially from the close-reading of technical documents, military publications and also scholarly contributions as primary sources. The main field of reference will be the level of program code, which, in the context of game studies, has been discussed with slightly increasing enthusiasm in recent years. In media studies, program code has an even longer history as an explanatory variable and object of study. This book, however, will occupy a middle ground between code in its concrete, hermeneutically unintelligible form (e.g. Montfort, 2006) and code as a highly abstract concept (cf. e.g. Hagen, W. “Der Stil der Sourcen. Anmerkungen zur Theorie und Geschichte der Programmiersprachen” M. Warnke, W. Coy und G. C. Tholen (Eds.) HyperKult Basel: Stroemfeld/nexus 1997: 33-68 or Winkler, H. “Das Modell. Diskurse, Aufschreibesysteme, Technik, Monumente - Entwurf für eine Theorie kultureller Kontinuierung” Pompe, H.; Scholz, L. (Eds.) Archivprozesse. Die Kommunikation der Aufbewahrung Köln 2002: 312) by focusing both on principles imposed by computability and on manifestations of code in actual production processes, i.e. modular design, game engines and elements of programming styles. Applying the model of programmable media to military strategy and self-assertion in the Middle-East since the early 1990s, I will attempt to explore the opportunities of understanding programmable systems as role models for policy-making and identification. The image of digital game or simulation users as becoming ‘part’ of the technical system has been a consensual aspect in most accounts of digital games to date, but was seldom systematically elaborated upon. For instance, Claus Pias inverts the common subject/object relation of man and technology, claiming that the hardware and software “design the player after their counterpart [Ebenbild]”, (cf. Pias, ComputerSpielWelten: 5/6) an idea which is encompassed in terms like ‘usability’, which already imply the player as “a second program” whose “output” is “requested” by the game. Pias already maps digital game genres onto the originally technical procedures they semanticize, e.g. generating time-critical selection chains (action games), retracing links in a database (adventure games) or optimizing a configuration of given values (strategy games). To interpret instances of programmable media convergence along the military/civilian boundary, I will extrapolate from this notion and con-

51

Real Wars on Virtual Battlefields? sider concrete effects, like the spread of ‘program code literacy’ in the following chapters.

1.7 Technical Preliminaries Since the book at hand is intended to cover a wide range of sources, many of them only available online, such as military online periodicals, technical documents on programmable media development and forum excerpts, all relevant texts have been converted into PDF format using the freely available tool FreePDF XP and archived for future reference. Due to the constraints of working with texts that do not have a consistent pagination due to their web-based origin, the respective pages from the PDF version have been indicated; they should be reproducible using the same tool, or at least give a sufficiently precise orientation. The bibliography lists only works that have been repeatedly used or are otherwise especially relevant to the topic at hand; similarly, the lists of other source types like designer diaries, news items and military publications include only the most relevant texts for reasons of clarity. All other sources are given with complete bibliographical details within the main text when necessary. Finally, all translations of quotes are provided by the author and occasionally, to preserve and point out selectively the specificities of German as opposed to Anglophone discourse, both versions are given.

52

2 BUILDING BLOCKS FOR A MODEL OF PROGRAMMABLE MEDIA ANALYSIS

2.1 Critical Considerations on the Current State of Game Studies The goal of this chapter will be to formulate a model of digital games as a subset of programmable media which is directly built on a close reading of the enabling technologies, most notably ObjectOriented Programming (OOP), and allows for tackling the complex issue of media deployment at the military-civilian margin. In order to position my approach in relation to existing notions of digital game theory, the first step will be to identify various developments within game studies as a ‘discipline’, a task which is often skipped despite the still expanding momentum of digital game studies and the sheer volume of scholarly output in that area. Just as C.P. Snow argued for a conceptual incompatibility between natural science and humanities research in 1959 (e.g. Snow 1998: 1-22), there seems to be little convergence e.g. between European and Anglo-American game studies discourse. The cultural specificity of theory formation can only be touched upon here but will be briefly exemplified in the final, concluding paragraph of this book. For instance, surprisingly little Japanese theorization on digital games exists, which contrasts starkly with the historically strong cultural dominance of that media genre in Japan. The contingencies of language in the process of transnational theory formation have not been systematically analyzed so far except for in rather historical, factual accounts. (G.M. Hodgson mentions language-transcending ties between German and American economists and a resulting widespread academic bilingualism during the 19th century; cf. Hodgson, How Economics Forgot History: The Problem of Specificity in Social Science London: Routledge 2001: 27) One particular facet of the German approach to digital games is the rejection of technical concepts in favor of other less clearly defined understandings of the object of study; for instance, in a 2004

53

Real Wars on Virtual Battlefields? anthology providing an overview of the German state of the discipline, Gunnar Sandkühler representatively argues for a “softer”, “hermeneutic” point of view, employing the methods of “cultural history” or, more specifically, “the history of mentalities”. (Sandkühler 2004: 220f.) A recently introduced alternative to using the term ‘digital games’ or even computer- and videogames is to use ‘programmable media’ as an analytical nexus. The term has been featured mostly with reference to literary forms beyond the print medium. (Cf. e.g. Glazier, Loss Pequeno Digital Poetics: Hypertext, Visual-Kinetic Text and Writing in Programmable Media or Gendolla, Peter/Schäfer, Jörgen, The Aesthetics of Net Literature. Writing, Reading and Playing in Programmable Media. 2007) Glazier argues that the materiality of electronic writing altered the perception of writing itself; thus, programmable media appear as a framework for an updated epistemology of “poesis”, exemplified in peculiar relationships between e.g. a web ‘page’ and the materiality of the book. I will pick up this useful thought with reference to ludicity, focusing even more on actual technologies and the structures they impose. A working concept from the anthology by Gendolla and Schäfer is Noah Wardrip-Fruin’s notion of “instrumental texts” and “textual instruments” (cf. preface) which exhibits a few overlaps with the notion of inter-activity sketched earlier and hints at classificational ‘grey areas’ between games and texts and offering the (musical) instrument as a bridging metaphor. In the following discussion, however, I will leave this metaphor in favor of concepts derived from the actual technologies. Thus, although rarely undertaken explicitly, an examination of the repercussions of the perceivedly ‘challenged’ notion of textuality in theoretical areas relating to the study of military conflict and organization appears more useful than examining the other already well-trodden paths in the fledgling field of game studies.

2.1.1 CONSIDERING THE IMPACT OF NON-LINEAR TEXTUALITY ON ESTABLISHED CULTURAL THEORIES The the overall trajectory of this book therefore is not to follow the formation of ‘game studies’ as an independent discipline (or to contribute to existing notions of game studies at all), but to apply the specific qualities of programmable media, as summarized in this chapter, to established approaches from various other disciplines, or conglomerates of disciplines, and to re-assess their suitability for the task at hand. For instance, Frans Mäyra is quoted as arguing for the need to “[rewrite] established theories of art, culture and

54

A Model of Programmable Media Analysis learning”, without specifying, however, how this rewriting can or should be approached. One tentative example illustrating this argument might be Clifford Geertz’s widely acknowledged and applied technique of ‘thick description’, which does not play a significant role in the topic of war games, but will be referenced later in the chapter on Gulf War games that assesses community dynamics and their rootedness in the technological framework provided by digital war games. The ethnographic tool of ‘thick description’ (Geertz 2002: 63ff.) describes an attempt to grasp the complex inner workings of a foreign social entity by describing, in detail and with descriptive language, instances of everyday life. One of the benefits of this approach is the built-in definitorial fuzziness which helps to avoid generalization and categorial reflection on cultural processes. However, the contingencies of the ‘linear’ medium or proxy of written language inevitably shape the outcome of an interpretive process using ‘thick description’; for instance, implicitly adopted literary conventions have a profound impact on the emerging texts. Briefly put, the paradigm of literary forms like novels or short stories acts as a filter for ethnographic research using this very prolific device. Since the print paradigm is gradually being substituted by other media forms, this filtering function is gradually taken over by less linear text forms like hypertext and -media genres which rely on programming like digital games. In this regard, conceptualizations of programmable media like the one proposed in this book can make a valuable contribution and, at the same time, be theoretically integrative, instead of operating in an assumed vacuum (as often seems to be the case, especially in early examples of computer game theory). What this contribution might look like in the case of Geertz and the ethnography of ‘virtual’ societies, in which programmable media syntagms are substituted for textual ‘thick descriptions’, will be tentatively explored in the fourth chapter. Another example which I shall not pursue very far theoretically, but which is applied and transferred later on in this book, is Gerard Genette’s useful concept of the paratext taken from structuralist narratology. (Genette 1997) In brief, paratexts are understood as “conventions and devices” in a book - like the title (page), preface and dedications - that shape the readers’ predisposition and “mediate” their relation to the text in manifold ways. While every text has to have a paratext (4), no particular paratextual type is obligatory. Genette explicitly includes non-textual forms of paratext like illustrations and “factual” pieces of information (like the gender or date of birth of the author), or even “material” forms like the tactility of the cover, which, whilst not containing meaningful statements in themselves, trigger specific associations. (7) First of all, the concept

55

Real Wars on Virtual Battlefields? of paratext provides a starting point for programmable media analysis that tends to be overlooked. Digital games, especially in the German academic context, are often readily understood as texts, so that discussing the delimitation of the text proper and potential peri- or epitexts (and problematizing the conditions of this process) can be a valuable exercise. A suitable example is (print) advertisements, which, despite their ‘traditional’ mediality, can exhibit a number of intricate links with the game ‘itself’, as the following magazine ad for Battlefield 2 demonstrates. The in-game graphics, iconically representing the game’s concept of realism, are used for the print ad and juxtaposed with digital retouching, added textfields and information, like the publisher’s logo, all of which blurs their epistemological status and glosses over the de-realizing effects of lacking visuals for the in-game experience. By mimicking a fictitious news format, the repetitive sequence of levels or ‘sessions’ that are not narratively tied together in the game can be interpreted as played-out news events. This complicates the interrelation of ‘reality levels’ by connoting the in-game events as news events, which are per definitionem recognized as mediatized versions, but which nevertheless draw their attributed authenticity from the culturally codified status of news media as ‘documentary’.

Illustration 3: Print ad for Battlefield 2 The print ad even references syntagms taken from the emergent game narrative on a micro-level, e.g. by implementing name tags

56

A Model of Programmable Media Analysis ‘floating’ over the players’ heads, just as in the proper game; this is done for a number of purposes. Occasionally, the fictitious nicknames simply mimic the martial rhetoric chosen by many ‘real’ players (e.g. “Hellswrath” in the lower left corner). Other tags exhibit wordplay (the soldier lying on the floor is called “Zzzzzz”, ambiguating his playing status), or playfully imitate sociolect and orthographic variation typical of player communities (e.g. the lonely soldier in the upper right corner of the screen is called “NoMat3s” = “no mates”). Even recurring in-game situations, e.g. the inexperienced soldier in the foreground shooting into the ground, are alluded to and probably easily ‘de-coded’, not only by experienced players, creating a ‘bond’ with the ad content through inter-action and thereby intensifying the advertising value. Since many inexperienced players have problems with the view control, the respective player is termed NewBie, a label denoting a novice player. Apart from following this textual notion of paratext, it might be worthwhile to think about para-segmentation, like the ‘para-games’ of programmable media themselves. In his thoughts on time in computer games, Jesper Juul describes the ‘loading screens’ in between two environments/episodes etc. as interrupting ‘play time’ whilst narratively maintaining the coherence of event time by indicating the loading process through the term ‘loading’ or an appropriate animation. (Juul 2002: 136) In order to think about loading screens as part of in-game narrative in the first place, it might be useful to have a pre-filtering notion of paratextuality with regard to digital games in order to functionally differentiate elements like loading screens, menus, credits sequences etc. as part of a programmable media product, rather than simply subsume it as part of the game narrative. It is evident that an elusive term like ‘interactivity’ is exceedingly difficult to define, let alone relate to my understanding of ‘programmable media’ as a unit of media theorization. Still, there are numerous attempts to do so within game studies literature. (Salen/Zimmerman 2003: 58; Friedl, Markus Online Game Interactivity Theory Hingham, MA: Charles River Media 2003: 58; Crawford, Chris The Art of Interactive Design San Francisco, CA: No Starch Press 2003:3f. and various other sources, often pointing out the general hype surrounding the term and its often irrationally positive connotations) Pierre Levy in his Cyberculture takes into account these complications and offers a definition of “interactivity as a problem” (Levy 2001: 61) that sets aside the ‘ideal type’ of a passive consumer, e.g. using a hypothetical example of not only watching TV but also recording, archiving and reassembling parts of the program. David Buckingham harshly but adequately asserts that the definition of a game is “tied up with the operation of

57

Real Wars on Virtual Battlefields? institutional power” in academia. (Cf. Buckingham 2006: 11) For that reason, I will tie my own concept of programmable media more closely to patterns induced from the respective technologies. It would certainly be a worthwhile undertaking to identify trajectories in the field of game studies as it currently presents itself in order to determine how disciplinary ‘bias’ produces different understandings of digital games; for the purpose of this book, however, I will concentrate on hitherto marginalized approaches which nevertheless provide a conceptual grounding for the model I wish to propose. The dominant methodological points of origin, namely film studies, literary studies and various forms of cultural and communication studies will only be used selectively. Buckingham even observes other vantage points, such as psychoanalysis, art history and cognitive psychology (Buckingham 2006: 11), which might serve the purpose of achieving a more differentiated scientific historical [wissenschaftsgeschichtliche] meta-perspective on the field, which, however, is not a goal of this book. Thus, instead of trying to reconcile my own approach with any of those institutionalized concepts, I will rather consider them and their research interests as layers of analysis which often still appear to build on an ‘essential characteristic’ of digital games such as narrative, ludic qualities or other focal points such as the psychological notion of “Flow”. Jesper Juul in the 2005 Handbook of Computer Game Studies, for instance, attempts to present reasons for both the ludology and narratology angle, using more recent, experimental narrative forms like William S. Burroughs’s Naked Lunch as a potential conceptual connector. (Juul 2005: 224) Juul finally acknowledges that although e.g. recent narrative strategies tend to use traditional temporal orders in more challenging ways, demanding more of the reader, the “emphasis on interpretation and ontological unstability” which makes a story more interesting would make a game “unplayable”. By means of this approach, Juul still tries to find conceptual overlaps, taking both narratology and ludology as a priori categories which, to some degree, mutually exclude each other. Instead, as suggested above, it might be more useful to see both schools as well as additional viewpoints as cumulative ‘layers’ of analysis, not as essentially different objects, and to hint at productive conflicts as e.g. in considering parts or elements of a digital game in narrative or ludic terms respectively. A salient example to demonstrate the benefits of this ‘layered’ analysis is the nature of missions in games like Axis & Allies that use randomly generated topography structured by fixed, prefabricated frame narratives or, inversely, games like Söldner: Secret Wars that map randomly generated mission descriptions onto a fixed topography obtained from large-scale satellite data. Contrary to oft-

58

A Model of Programmable Media Analysis cited examples (in Juul’s case Tetris or text adventures), those missions cannot simply be considered as inherently narrative or systemic, but it can be fruitful to treat them analytically as one or the other alternately in order to identify interdependencies, e.g. to analyze multiple randomly generated plots/topographies in narrative terms to arrive at a spectrum of narrative possibilities which were not hermeneutically created but may (and will) be hermeneutically read. Another game with a similar mechanism, Soldier of Fortune 2, creates missions based on formulae such as ‘destroy object x’ or ‘liquidate target y’ (which, in terms of program code, are identical and simply based on the infliction of a preset value of damage on a target, irrespective of whether it results from the player’s weapon parameters or environmental damage). In this case, the topography and other ‘ludologically’ relevant factors such as room size and distribution, positioning of secondary objects etc. are interpreted against this (contingent) backdrop. Along these lines, narrative and narrativity can be studied as a concept, e.g. by identifying patterns of how and why players narrativize systemic elements in a game. The ‘layer’ approach is a preliminary example of deriving an analytical tool from a technological configuration. While a layer is not necessarily a technical term, understanding it in a technical sense (as used in programmable applications like the image editing software Adobe Photoshop or the authoring environment Adobe Flash) can prove beneficial. Thus, layers (of images, as in Photoshop) can be blended and combined in multiple ways that translate well into ‘conceptual’ layers. For instance, a layer in Photoshop can be used as a cut-out frame for another layer; several tutorials demonstrate e.g. how to use alpha layers as a conceptual ‘proxy’ for that purpose (cf. e.g. http://www.cs.wisc.edu/graphics/Courses/559f20 07/Main/Alphatutorialphotoshop). Lev Manovich is one prominent author who proposes a similar argumentative approach; however, the technology-centric model proposed herein attempts to utilize layers and other technology-induced figures less as a rhetorical device than as an integrative perspective which is necessary for a complex issue such as contemporary digital war games. In an attempt to bridge the gap between ludic and narrative elements, Eladhari and Lindley simply presuppose a continuum of narrativity usable for localizing virtually any digital game-like application, from multi-path CD-ROMs to board game adaptations. (Eladhari/Lindley 2002: 293) While the authors argue – quite plausibly - that digital games can mediate between both poles by various means (without pointing to concrete examples) instead of using genres to flesh out the continuum sketch, it would be more useful to apply this dualism to lower levels also, rather than the entire game, e.g. the macrostructural distinction between subsequent phases (levels, cut-

59

Real Wars on Virtual Battlefields? scenes, interface sections etc.) and their transitions in the game. These thoughts will frame the considerations of a program code model of digital games and other programmable systems that follows below.

2.1.2 THEORY BY DESIGN Many influential texts focusing on digital games in recent years pursue a design-informed approach. This coincides with the increasingly analytical activities of game designers, resulting in the figure of the designer-analyst embodied by personalities such as Chris Crawford and Will Wright. This shift is largely in accordance with e.g. the methodological role model provided by film research in the 1920s, which saw an increased intermingling of journalistic, technological, scientific voices and rhetoric in the theoretical discourse surrounding film in mutually beneficial interplay. Arguing from a production standpoint, Bernd Kreimeier already made a case for using ‘design patterns’ in 2002 in order to remedy what he found to be the lack of a “shared vocabulary”. (Kreimeier, Bernd 2002: 1) He proposed using a formalist approach to digital games modeled on Christoph Alexander’s ‘pattern language’, which, from an academic perspective, appears rather anachronistic and is for precisely that reason an interesting theoretical building block. The desire to “[share] and [expand] knowledge” about digital games legitimated analytical techniques that literary and film studies had abandoned decades ago. Since Kreimeier’s ideas have been frequently quoted and expanded upon (e.g. in Bjork/Holopainen 2005; Natkin/Vega 2004 and several studies with a rather technical focus), even in purely academic contexts a revitalization of formalist thought can be identified as an early trace of the impact of programmable media on media theorization. Kreimeier separates his “content patterns” from software engineering patterns which had already been utilized in game programming (1) and process patterns used in project management, but concedes a number of similarities. Patterns as “semi-formal tools” (2) exhibit the desire to organize effective communication concerning game design and to ensure reproducibility of successful patterns, i.e. to make elements in the production process compatible; since compatibility of digital data is an important prerequisite, especially in programmable media production, this appears to be a logical step. However, the increasing interest in ‘conceptual compatibility’ within game studies discourse is significant; the topic will be further explored in the next chapter. Moreover, the prospect of being able to ‘re-use’ successful patterns and elaborate on them, through extrapolation, transforms design

60

A Model of Programmable Media Analysis (and theory) by patterns into an optimization problem, similar to the mathematicization of games mentioned in the introduction, which again hints at the formative impact of programmable technologies. Jesper Juul consequently summarized the game patterns approach (Juul 2005: 106ff.) in a useful way as “a process of iterative refinement”, taking “choke points” as functionally differentiatable topographical patterns in shooting games as an example. Chris Crawford’s propositions as to game genres from a design perspective are therefore closer to the ‘game patterns’ approach and obtained by extrapolation from technological rather than ‘ludological’ reasoning. One of his genre criteria is the presence/absence of an active opposing agent (i.e. one or more interrelated objects rather than a fixed constellation reacting to player input) which allows for the distinction between conflict- and puzzle-type games. While this binarism might be too rigid to be very workable by itself, it does not devalue the general principle. A more hands-on element introduced by Crawford in the same manner is “pattern recognition”. (Crawford 2003: 49) Contrary to patterns identified by Bjork/Holopainen, this exhibits a higher level of abstraction and is derived from observing a digital game’s internal structure, not its symptoms. A war gamerelated example of this pattern can be found in large-scale strategy games like the Close Combat series, where a complex strategic constellation cannot directly be assessed by the player by calculating the respective offensive/defensive values and ranges of movement computed by the game algorithms, but by recognizing recurring visual patterns of troop deployment and movement that signal a local, temporary advantage or disadvantage. This approach is comparable to the game-theoretical principles of dissecting a complex game state into sub-games and relating those rather intuitively to the macro situation. Crawford consciously uses terminology from war game design vocabulary like “color” and “dirt” to make his point. ‘Color’ in this sense denotes an added facet of realism in a game, e.g. a specific weapon property or method, like a weapon overheating after repeated use, which adds a distinct ‘flavor’ to the rule system. ‘Dirt’ is a somewhat complementary concept describing the reduced clarity or ‘purity’ of the rule system achieved by adding ‘color’ arbitrarily. (331) Combining both terms, we can begin to analyze e.g. the interrelatedness of internal rules, i.e. how a game creates complexity by interlocking mechanisms without adding too much ‘dirt’. Thus, an abstract phenomenon like a rule ecology apparently calls for visual metaphors that offer ‘intuitive’ access to an otherwise complex field through complexity reduction, not unlike the ‘pattern recognition’ category Crawford proposes for game analysis. Yet, by suggesting a ‘color/dirt ratio’, Crawford implicitly assumes the quantifiability of

61

Real Wars on Virtual Battlefields? his categories indebted to the technological makeup of the medium, which may or may not be a desirable feature as a theory element. Another example is the interaction forms model proposed by Manninen and Kujanpää in 2005. (Manninen/Kujanpää 2005) The central question of this approach is how players use ‘interaction forms’ such as “non-verbal communication and perceivable actions” in online (war) games to solve communication problems. Manninen defines “interaction forms” as “perceivable manifestations of ingame actions […] that can be used for interaction” and as “instantiated via audiovisual content”; this slightly fuzzy definition refers to visual cues in a game, e.g. a character waving or bowing, that have a (non-computable) communicative function. Manninen’s account of in-game functionality designed or re-purposed for communicative acts also exhibits strong traits of positivist, technicist assumptions. Basically, he aims at improving “usability” in games, conceding that the “full expression of interaction” is not yet available, implicitly assuming a hypothetical ideal state of human-computer interaction (HCI) which, it is speculated, can be reached through technological progress. The rhetoric used by the authors appears very technicist, e.g. considering facial expression as a “carrier of information forms”. The gameplay technique of calculating a position in the game from spatial information such as grid coordinates and compass directions is described as “runtime conversion” (7), assuming the player as an algorithmically expressible instance. While these terms are likely (at least in most cases) to have been chosen on purpose, the phenomenon itself is a useful lead that will be followed in a dedicated paragraph below. The type of rhetoric which players resort to when verbalizing their in-game experiences is also visibly dependent on the technical framework of the game and the ideal type of ‘military precision’. “Team member Aston Blade shouted for a medic. I responded immediately and requested him to acknowledge his position on the map. After visual acknowledgement I was able to give him an estimated duration of my journey…“ (Quoted from Manninen/Kujanpää, 7)

The implied actions (respond, request, visual acknowledgment, estimated duration) exhibit strong overlaps with the logic of program code, e.g. ‘scanning’ the map to calculate the default time a distance would take to travel). The impact of programmable technologies on user discourse will be analyzed in greater detail in chapter 4b. The particularities of design-informed game analyses outlined above are occasionally even directly translated into a technical framework during the research process. For instance, the Game Ontology Project (http://www.egl.gatech.edu/gamedesign/), which also

62

A Model of Programmable Media Analysis attempts to create a common vocabulary for game analysis, uses a Wiki to organize its information; while this would qualify as ‘database logic’ in Manovich’s terms, it more concretely employs a technically circumscribed and contingent ‘middleware’ for the theory formation process similar to the technologies it implicitly refers to. To conclude, design patterns seem to be among the most prominent tools geared towards actual applicability in game studies discourse. Put as a working hypothesis, design patterns conceptually anticipate the logic of program code, which is the smallest unit of operation in digital game production. Kreimeier’s four-part structure of describing a pattern (by name, problem, solution, consequences; cf. Bernd Kreimeier 2002: 2) is a useful example. The ‘consequences’, i.e. the “costs and benefits of a solution”, essentially store the relationship between the pattern as an object and related patterns/objects according to the model of Object-Oriented Programming (OOP), which is the technical foundation of contemporary digital games and will be further elaborated below. Consequently, the first example pattern mentioned by Kreimeier, the ‘proxy’ (3), which denotes a player taking over another in-game character, is conceptually equivalent to a ‘proxy pattern’ in software design terminology, i.e. a stand-in reference to another code object which can serve a number of optimization purposes. Kreimeier himself suggests that patterns might readily translate into pseudo-code if the pattern language overlaps with a corresponding software engineering language. (3) The fact that Kreimeier illustrates how a different pattern language, in this case Church’s FADTs, “can be expressed as patterns” (5), i.e. ‘transcoded’ in programming terms, is another piece of evidence to support the overlap of patterns with program code. The notation scheme is thereby comparable with a data structure that is the basis for determining transcodeability. Furthermore, nested patterns (like the “weenie” and “weenie chain”; cf. 7/8) closely resemble corresponding programming structures, e.g. nested functions or classes. Finally, from an ex-post historical perspective Alexander’s notion of pattern languages has already strongly shaped computer science and specifically the idea of OOP (cf. e.g. http://www.math.utsa.edu/sphere/ salingar/Chris.text.html#COMPUTER) which retroactively stabilizes the hypothesis at hand.

2.2 Towards Programmable Media Analysis In order to approach the complex interplay of institutions, technologies, agendas and media products revolving around contemporary digital war games, an integrative model of analysis is necessary.

63

Real Wars on Virtual Battlefields? Models of game analysis which do not use a pattern-oriented approach often suggest a number of aspects to be studied; in some cases (e.g. Konzack 2005) these are elements of the game, while other authors suggest an overarching topic like spatiality or apparatus, which would aim for and indicate comparability with other media genres. (e.g. Böhme 2007; Camper 2005 and Galloway 2004) Case Studies are still rare, especially of war-related games, and usually remain rather broad (cf. e.g. Nieborg 2003: 13). An integrative model of programmable media analysis thus needs to introduce a ‘common denominator’ that allows for the correlation of existing approaches and might serve as a corrective measure if needed. The level of technology, beginning with a closereading of program code and its repercussions as a defining property of programmable media, could meet those requirements; I will briefly summarize the key ideas as a starting point for this discussion. This particular use of program code is roughly modeled after Henri Bergson’s analogy between the “cinematographic apparatus“, a fairly intuitive and technical understanding of the then novel media genre film as e.g. “moving images“, and fundamental techniques of cognition, i.e. of “[approaching] reality“ and structuring thinking (for a concise rendering of some of Bergson’s key arguments cf. e.g. http://www.horschamp.qc.ca/new_offscreen/Bergson_film.html). In the following chapter, the model of programmable technologies shall be similarly applied as an epistemological framework. There are already some references to program code in the current game studies discourse. Rolf Behn’s pioneering article from 1984 was based on a number of observations and assertions which, however, were not systematically elaborated at the time. (Behn 1984: 685) According to Behn, the shift from absolute to relative, quantifiable player success in a game (not reaching a goal but attaining the highest score) is necessitated by what he calls ‘computerization’ [Computerisierung], a term which has by now become almost irrelevant and historicized given the current pervasiveness of computer applications. High-level observations like this have become scarce, with most games research focusing on fairly recent phenomena; yet, extrapolating from it with respect to war games demonstrates its utility, e.g. by illustrating the agenda shift from absolute criteria (epitomized by books like Michael Walzer’s Just and Unjust Wars) to infinitesimally small, offsettable [gegeneinander aufrechenbare] subgoals. This shift, as Behn would have it, is made plausible through computers or, as can be argued from a contemporary perspective, through various technologies related to digital games, which have made an impact on society as a whole by this time.

64

A Model of Programmable Media Analysis Behn offers other insightful observations on the phenomenon of digital games imposing their program code logic on the player; for instance, since only orthogonal movement was ‘allowed’ (i.e. algorithmically registered) in early games, the player had to (and did) ‘interpolate’ diagonal movement by repeatedly switching between minute orthogonal movements. In line with the model proposed for this book, this could be considered a ‘kinetic algorithm’, i.e. a semiautomatable complex movement according to the technical understanding of a diagonal line as an interpolation of orthogonal movement on a screen grid consisting of pixels as the smallest representational unit. This form of kinetic automation is encouraged by the game application, which demands quick input that would otherwise be impossible to perform. With digital games still in their infancy, Behn did not link these observations which would only be possible from an ‘ex-post’ perspective. The pedagogical conclusions are thus not sustainable from a contemporary standpoint and reflect the mixture of fascination and fear of computers in ‘home’ environments prevalent at the time. Twenty years later, layers of technological sophistication and ‘interface’ (following Sherry Turkle) almost prevent this level of understanding by marking it as seemingly ‘evident’. Lars Konzack mentions program code as the second “layer” of his game-analytical model, nevertheless conceding that researchers often “may not have access to the source code” of a game or may not “be able to translate the code into anything meaningful” (Konzack 2002: 91/92). The concluding suggestion that program “code may be analyzed indirectly from a functionality perspective” would produce overlap with the third layer (“functionality”) in Konzack’s model, but is a plausible supposition that will be pursued further below by concentrating on middleware and other functionally differentiatable sections of program code rather than on individual instructions. Nieborg takes up this element (Nieborg 2003: 5) but similarly offers no theoretical perspective except for sparse hints e.g. at the focus on lighting computations in the game engine used for the game Tom Clancy’s Splinter Cell. Rolf Nohr explicitly mentions the level of program code in his notes on a theory of transparency [Transparenztheorie] of games (i.e. digital games), describing it as a necessary common denominator to think of digital games as part of the “medium of the computer” and go beyond “image” or “narrative”. (Nohr 2004: 97) Nohr concludes with Hartmut Winkler that the program code is so “evidently abstract” (98) that it does not allow for naturalization; according to Winkler, the code as a formal language is “independent” of actual practices and “applications” or, as he puts it rather normatively, “secured against the intervention of symbolic practices”. (98) Consequently, Nohr leaves the issue at that and refers to it as “the

65

Real Wars on Virtual Battlefields? technical” (e.g. 101) to describe its symbolic dissolution in the course of his argument. A first working hypothesis on the implications of programmable media technologies that will eventually lead up to a conceptual connector shall assume that programmable media, owing to a number of role models like the hypertext movement, but also tabletop-roleplaying games, transform the act of reading away from hermeneutic inquiry towards ‘text processing’. Put differently, the reader/player implicitly accepts the technological framework and its criteria like optimization and discretization (i.e. defining a task as a series of regular patterns that are algorithmically expressible) in their practices of media usage, e.g. gleaning and categorizing as much information as possible from a text. Although his focus is a different one, Claus Pias provides ample evidence for this thesis in his thoughts about text-adventures, the genre of digital games most heavily indebted to textual representation and literary forms, as structural role models. (Pias (2) 2002: 89-) Pias actually represents one of the very few academic authors who actually work with program code (94), documenting how the textually presented environment in a text-adventure game is internally organized as a network of nodes, i.e. quantifiable, and conceptually originates from a technical optimization solution for the precursor of the internet, the ARPANET. (91) Players are expected to specify what they want to do in an English sentence (“Take the apple and put it in the bag”); understanding the way this input is parsed by the computer, experienced players soon begin to specify only the information the ‘parser’ (i.e. the program code section interpreting the input text) actually computed (“take apple, put in bag”). Thus, many readily adopted the principle of optimization inherent in programmable media, reducing both input and parsing time by keeping the text as short as possible; this differs notably from the maxims that apply while reading a book or writing a letter. As will be elaborated below, the pattern cannot only be applied to writing but also reading practices within HCI. Another interesting case of attempting to read program code in academic writing is Nick Montfort’s article “Combat in Context” (Montfort 2006), which attempts a ‘close reading’ of the very old game Combat (1977) according to a set of categories derived from the aforementioned model by Konzack. Describing the ‘archaic’ platform’s technical specifications, i.e. five movable objects, modifiable background color and stationary graphics, (4) effectively demonstrates the impact technological contingencies had on the spectrum of conceivable design solutions from a programmer’s perspective at that time. This finding backs the initial hypothesis that military conflict was one of the very few readily available themes fit for semanticizing using this rudimentary spectrum of symbolic manipula-

66

A Model of Programmable Media Analysis tion; it will be exemplified later on in this book how this pattern, on a much more subtle level, still applies with game engines and other middleware. Suggesting that disassembled machine code is actually fully understandable by “dedicated scholars” (6) might be fairly optimistic, yet Montfort develops a few interesting arguments from this approach. Most notably, he argues that the difficulty of programming the VCS led to the reliance on a simple working ideal type of game mechanism with numerous rule variations (6) rather than graphical variations to maintain motivation. This process is roughly comparable to the musical pattern of ‘variations on a theme’. Thus, departing from technological role models, the musical phenomenon could arguably be explained analogously in technological terms as experimenting with new playing techniques, considering the average player or also contemporary musical instrument technologies as ‘hardware’ to work with. Moving on to the production of war games, this historically comparative edge shall be preserved. Accordingly, Montfort notes how the availability of more hardware memory later provoked a shift from game variants to adding visual details and variation as the most common form of ‘enhancing’ a game ‘skeleton’. (6) All game types in Combat are attributed to different aspects of warfare like tanks, bi-planes and jets; the reference to the contemporary game Pong in some sub-games (e.g. the game mode Tank Pong) indicates that technological feasibility motivated genreblending despite the ‘military’ context and shows how strongly dependent ‘genre’ was on specific reusable algorithms. The difference between tanks and planes lay merely in the speed variable being either tied to the main loop (i.e. continuous movement at a given speed) before player input or to the player input loop (i.e. movement rate depending on key pressed). Montfort does not dwell upon the fact that a generic military theme was chosen for a launch title for the Atari VCS and shipped by default with all machines sold (1). Yet, it appears worth mentioning that the game, to a certain degree, became associated with the respective hardware. While technological restrictions would have made a more specific scenario hard to implement, Combat with its tanks, jets and planes epitomizes an understanding of war that seems far removed from contemporary notions. Yet, in a comparable way, Air Force Delta Storm (2001), which will be revisited in more detail later on, was a launch title for the then-new XBOX console showcasing the resourcefulness of the machine through complex environments and visual effects; Call of Duty 2, a war game set in the Second World War, fulfilled a similar function demonstrating the new capabilities of the XBOX360 as will be elaborated upon in a dedicated paragraph below.

67

Real Wars on Virtual Battlefields? Coming back to the ‘reading as text processing’ analogy, a crucial consequence is the fact that readers increasingly come to accept other forms of contiguity than mere causal links (e.g. quantitative distribution of words) in a text. This practice can be found in established and discretely labeled recent reading techniques like ‘Flächenlesen’ [area reading], ‘scan reading’, ‘photo reading’, ‘spectrum reading’ [Spektrumlesen] and ‘alpha reading’ which are offered as courses and more and more considered as formative cultural techniques. Arguably, as I will try to demonstrate throughout this book, these developments must be related to the pervasive influence of programmable media in every aspect of society, warfare being only one on the list. The ‘photo reading’ method proposed by Paul R. Scheele is even explicitly described as a “game”. Explicit reference to fast reading techniques can frequently be found in military coursework (e.g. http://web.nps.navy.mil/~menissen/mn3309/readguid. htm) and complies with categories of military training; in the fourth chapter, observations like these will be recapitulated and synthesized into a model of military self-conception within the boundaries of its dependence on programmable media (for an introductory perspective on the military origins of speed reading cf. e.g. http:// zenhabits.net/2007/05/overclock-your-reading-speed). The Speed Reader research project even directly correlates reading with a racing-themed videogame to explore the pedagogical potential of “[driving] through a book“; (cf. e.g. http://www.siggraph.org/s2001/con ference/educators/papers.html) the technologically induced implications of this technique, e.g. the formalized “speed-reading protocol“ and the psycho-linguistic segmentation of a text into individual words which “flash to [the reader] rather than appearing in a static line across a page“, clearly exhibit the programmable media rationales that informed their conception and envisage the user as part of the socio-technical setup. At the same time, the interchangeability of semantics (in this case the race track metaphor for a book) will in the following chapter be elaborated into a model of confounding explanatory paradigms based on program code. As indicated above, the new reading styles propose forms of internal text coherence other than hermeneutic ones. While Seymour Chatman held a “legal brief or a sermon” as ‘non-narrative’ texts (Chatman 1980: 122) because of their “logical” instead of temporal order and because they allegedly render discourse time irrelevant, media genres like digital games propagate a broader notion of narrative. For instance, players are regularly confronted with tabular, numerical data and, accordingly, devise techniques to dissect and convert them into meaningful information.

68

A Model of Programmable Media Analysis

Illustration 4: Screenshot from Pacific Theater of Operations II

Lev Manovich already posited numerical representation as one of the key elements of his concept of ‘new media’. (Manovich 2001: 27) Despite its laudable originality, the idea is occasionally flawed by terminological or technological imprecision. Manovich’s claim that a “new media object” (given his background he refers exemplarily to digital images) can be “described formally” (27) or mathematically is only partially true, e.g. in the case of vector graphics, while bitmap graphics can be mathematically modified but not described. For this argument, I will try to apply the idea rather to mathematically defined gameplay patterns like the quantification of victory conditions in the example given above, Pacific Theater of Operations II (KOEI: KOEI 1992), which it fits much more precisely. In ethnographic research, a broader notion of ‘narrative’ has already been tentatively applied to quantitative data, mainly motivated by an interest in bridging the methodological gap between qualitative and quantitative data and arguing for a similar level of richness in both. (Cf. e.g. Elliott, Jane Using Narrative in Social Research: Qualitative and Quantitative Approaches London: SAGE Publications 2005: 60/61) Adopting this perspective and substituting fictional media content like quantitative data in digital games for empirical data in ethnographic research hints at an increasing familiarity with numerical literacy among digital game users which should not be analyzed using narratological means but might rather include e.g. narrative constructs as ‘proxies’ for interpreting this abundant numerical da-

69

Real Wars on Virtual Battlefields? ta. The idea of algorithmic manipulation, which states that “new media” or digital media objects are usually modified according to formulaic patterns like noise reduction of a digitized image, is more widely acceptable; however, it is not clear how this would apply to other types of ‘objects’. Since Manovich considers digitization (28) as the process of numerical encoding, he basically refers to the epistemological shift e.g. of converting text from a sheet of paper into a sequence of binary information; instead, I would argue that the application on nonlinear constellations in a game and a closer qualitative look at the transformation process would yield more valuable insights. Furthermore, Manovich’s catchy conclusion that “media becomes programmable” (27, italics take from the source) obscures more than it explains since it is not the digital image that becomes programmable but the software tool used to execute the algorithm. In modern tools, e.g. Adobe Photoshop since its seventh installment, image manipulation can both be performed using the graphical interface or external scripting languages like Visual Basic or Javascript. (Cf. e.g. http://media.wiley.com/product_data/excerpt/50/ 07645245/0764524550.pdf). Thereby, program code, although interpreted by the computer in an abstracted form from machine code, enters the production process more and more directly and influences routine practices; this topic will also be examined closely in a dedicated paragraph later in this chapter. An interesting assumption for programmable media analysis is the fact that, according to the OOP paradigm, nonlinear constellations and procedures in digital games are stored and used as objects on the same program code level as, for instance, audiovisual objects. The algorithmic handling of interactemes, i.e. algorithmically described rule ecologies producing a spectrum of emergent instantiations, follows similar patterns like the handling of digital images, sound files or text ‘strings’ (text in programmable applications is usually stored as ‚strings’, i.e. variables which are reserved for text instead of numerical data. Strings are usually delimited with their declaration and can contain only a fairly limited amount of text to save computation time); they are stored as modular, i.e. reusable and re-combinable code in so-called ‘libraries’, either triggered by a variable state or a random frequency function. One example would be ‘respawn’ routines that create new enemy objects at pre-specified or random coordinates; Tom Clancy’s Ghost Recon is a strong case of re-used respawn code, creating new enemy objects after the player has passed a given point in the game, which in that context has a probably unintended de-realizing effect because the respawning is insufficiently narratively embedded. America’s Army has also characteristically implemented random respawn points as

70

A Model of Programmable Media Analysis well as other instances of randomization (extraction points, even objectives) since version 2.4. Thus, the romantic notion of reading still popular in the 20th century, i.e. reading as immersion, a ‘flight of fancy’ or a process of moral refinement, (an insightful overview of such 19th century reading habits and ascriptions can be found in chapter 20 from: St. Clair 2004: 394-413) is thus transformed according to criteria such as efficiency and response time, while the player/user increasingly perceives text (or any other on-screen information) according to cognitive and interpretive filters modeled after their continuous experience with programmable media to be able to deal with the otherwise overwhelming influx of data. The very instructive excerpts of interviews conducted by Sherry Turkle over several years provide ample support of this hypothesis. (Cf. e.g. Turkle 2005: 268-270) For instance, the thirteen-year-old Tim playing SimLife is described as routinely ignoring text messages from the game which he does not understand, claiming that “it’s not what’s important”. Turkle explains this behavior by differentiating between surface (or ‘interface’) transparency and internal transparency, elaborating how an understanding of the interfaces or “simulations” (267) of a machine allegedly replaced the desire to understand the machine itself. However, employing (game) programming technologies as metaphors for understanding user behavior, Tim’s playing strategy can also be essentially re-read as a cognitive ‘filter’ (i.e. a ‘reading algorithm’) for categorizing in-game information according to relevance to save ‘processing time’ and allow for fluid gameplay which would hint at a, however implicit, adoption of the properties of underlying program code. Another interviewee, a tenth-grader named Marcia playing Sim City, had adopted the maxim that “raising taxes always leads to riots”, being apparently unable to distinguish between its applicability in the game (or ‘interface’) and in real-world situations. While Turkle’s diagnosis of Marcia’s inability to “‘read’ a simulation” may hold true, focusing on technological contingencies yields a different, less valorizing interpretation. To be able to play successfully, Marcia shifted her playing algorithm (the avoidance of tax raises) from mutable to immutable processing space just as routine algorithms, e.g. graphical display routines like e.g. z-buffer calculations, are usually gradually moved from software to hardware processing in technological development history, improving processing time at the expense of dynamic mutability. Thus, the players’ cognitive and interpretive strategies exhibit striking similarities with the technological framework they originate from, a process explicitly encouraged by digital games or simulations as part of the gameplay system. Claus Pias describes this goal as the “duties of the player”, analyz-

71

Real Wars on Virtual Battlefields? ing programmable technologies, in his particular case the common ping program (designed to check the availability of a host in a network), to support his thesis that the user is actually ‘designed’ by the computer as a “Gestalt of [his] ports” [Gestalt der Anschlüsse], i.e. as a component which must be continually present and alert. (Pias 2004:1) Other vantage points are thus required that allow for understanding digital games first and foremost as programmed applications, the projected usage of which is naturalized by compliance with culturally communicated notions of play. The present focus on technologies allows for a re-reading of established approaches, e.g. narratological tools like theories of characters [Figurentheorie], in the context of programmable media. This evidently does not imply substituting these approaches, but rather emphasizing their dependence on the respective media technologies. In this sense, an important criterion for the taxonomization of characters is their function in naturalizing program structures. The most basic form of character interaction in digital games illustrates this principle by semanticizing one of the simplest program structures, the conditional loop. In games like the paradigmatic Ultima series but also many Japanese role-playing games, NPCs would cycle through a repertoire of phrases when conversation is initiated; occasionally, this repertoire changes depending on whether the player has triggered a specific event in the game, i.e. a status variable. In pseudo-code, this could be expressed as: if (QuestSolved == 0) { switch (NPC_Counter) { case 0 : print (“Hello, my name is XYZ.“); break; case 1 : print (“I’m the local innkeeper.“); break; default : print (“Nice weather today, isn’t it?“); } NPC_Counter++; } else {[…]} // progress with the main game script

Starting with the sixth installment of the Ultima series, digital games temporarily shifted to a different ‘character algorithm’ scheme, the database query. Once the player had entered a word, the character would reply with the matching entry from its dataset; these replies would contain suggestions for further ‘search terms’,

72

A Model of Programmable Media Analysis so that learned conversational routines ideally came to represent the filter by which players ‘scan’ the text snippets for new relevant terms. This system is functionally and algorithmically consistent with the text ‘parser’ used in the aforementioned text-adventure example. The ensuing ‘conversation’ thus closely resembles traversing a hypertext with similar ‘player algorithms’ as in the use of an online search engine; terms like ‘name’ and job’, which were always valid by default, yielded the initial inventory of search terms from which to traverse the dataset. While the technological contingencies thus clearly impact media usage, it can only be speculated as to how far this understanding retroacts on the re-reading of other media genres. Multiple-choice conversation, offering the player one of multiple context-sensitive responses, has emerged as the current standard and represents a polylinear tree structure, i.e. a series of branching choices which lead into different ‘branches’ of the conversation, with the option, however, of going back to the initial set of choices. This structure is also commonplace in games programming and fractally recurs on different scale levels. In a similar vein, temporality in digital games should be tied to the respective media technologies as e.g. Michel Chion interprets temporality in film against the ‘perpetual circular motion’ of the projector as an analogy and structuring metaphor. In this sense, the database as frame of reference re-interprets temporal relations as potentialities and their actuation as a definable variable, generally de-emphasizing it in the process. Consequently, temporality was not found to be a major issue in the close-readings of military-themed games throughout this book. In this sense, relational notions of temporality like ‘access times’ and frame cycles (i.e. the successive processing of commands over several frames) substitute traditional criteria of temporal organization in media texts. Taking a step back, the (often overlooked) omnipresence of programming as daily practice in contemporary modern societies strikes a similar chord. Tasks like programming a DVD recorder to capture a specific channel at a specified time and date are not marked as ‘programming’ since the interface effaces the underlying technologies and provides the necessary layer of abstraction. However, similar e.g. to the re-writing of programmable applications like game MODs, they assume the reference medium (in this case the continuous TV program) as an input stream of ‘material’ and presuppose the semi-automatable manipulation of this stream, i.e. the timed backup. Some recording devices even allow for taking in discrete variables from the input stream using a technique similar to ‘event handlers’ in scripting languages by receiving e.g. markers that allow for cutting out advertisements without needing to physi-

73

Real Wars on Virtual Battlefields? cally watch the program. These technologies, coupled e.g. with downloadable sections of periodical news programs online, fundamentally re-shape the epistemological framework of those news programs which used to be functionally dependent to some degree on the fact that they were being regularly broadcast at fixed intervals, so that users would even take this into account for planning their evening. The American poet Robert Pinsky elaborated on the parallels between poetry and programming as early as 1981, spearheading the hypertext literature movement (cf. Edge (German version) 3/4 2007: 95), which indicates how media technologies are discussed differently and to different extents in the history of a medium. Pinsky is representative of the early enthusiastic embracing (and theorization) of programming as a cultural technique - which was later cast off most notably through its particular structural properties like the incremental perfection and modular development of specific types of algorithms. After the liberalization of access to production tools, program code is now re-entering popular conceptions about programmable media with different accentuations. The feasibility of OOP and other technical concepts in the context of game studies has already been tentatively explored by Mirjam Eladhari and her colleagues, mostly from a design perspective, in order to automate the generation and optimization of nonlinear storytelling constructs. In her abstract, Eladhari admits that the “largest amount of text in a computer game consist of program code” (Eladhari 2003), deploring the fact that “one does seldom get access to the massive amount of program code” for analytical purposes. (13) In this book, I propose to remedy this deficit by examining middleware, i.e. solidified patterns of code that are shared by many games and practices such as code reuse. Eladhari investigates three important computer-science concepts (coupling, dependency and normalization) in the context of digital games, albeit with the purpose of ‘optimizing’ in-game narrative by “eliminating redundancy” (37) and “inconsistent dependencies”. Eladhari thus subscribes to the plausibility of digital game narrative as a database by intuitively observing and proposing to tweak the tools used for its modification, proposing ‘coherence’ in increasingly complex narrative architectures as an implied maxim. It is striking that, with programmable systems becoming the central point of reference in the current media landscape, the utopian notion of being able to model narrative exhaustively appears to be fulfillable. This positivist and rather ahistorically naïve [geschichtsvergessen] belief, implying notions such as ‘optimizing’ a narrative sequence, can be clearly identified as stemming from the increasingly naturalized technological framework or programmable media. It is

74

A Model of Programmable Media Analysis useful, however, to re-read these concepts in terms of their cultural functions, specifically with reference to warfare (normalization is but one topic that, in various contexts, constantly reappears in military discourse), assuming, as Eladhari does, that the steadily increasing complexity of game narratives is also a pattern which verifies and, by virtue of its application, further institutionalizes the effects of program code on cultural production. Eladhari demonstrates how broken causal relations (cf. e.g. the wizard/dragon example, cf. 36/37) in the poly-linear narrative structure of a digital game partially reveal the scriptedness of the game and lead the player to adopt his playing behavior accordingly. The player strategy Eladhari describes of abandoning hermeneutic reasoning and simply visiting all available locations in the game (to find the dragon in the example scenario) coincides with the aforementioned brute-force algorithm which originates from cryptology but, as a pattern of media usage, seems fairly uncommon and can plausibly be understood as a ‘playing algorithm’, i.e. a semiautomatable behavioral script modeled after the experience with the game as program. One result Eladhari proposes is the technique of causal normalization, allowing for digital games to be “analyzed in software engineering terms” (Eladhari/Lindley 2003: 294) in order to minimize story-logic conflicts and to more effectively design complex nonlinear narrative causality systems. The ‘solution’ for storytelling conflicts using software engineering devices like e.g. data flow diagrams (297f.) is at least debatable for design purposes and theoretically assumes that the system of interrelated narrative nodes would be parsable by a script (or, analogously, traversable by an AI entity), i.e. the digital game would effectively ‘play itself’ to obtain an experimental overview of all conceivable instantiations. Another concept Eladhari introduces is ‘data coupling’ and/or ‘control coupling’ which denotes the level of autonomy of a “functional module”. Data coupling describes the fact that changing data in one module modifies the data in the coupled modules; control coupling equally applies for behavior changes in root and coupled module. (298) Generally, minimizing coupling phenomena using data flow diagrams results in code that is easier to modify; however, Eladhari does not indicate precisely how this technique could be translated into a narrative system. The whole process requires low-level segmentation, i.e. the definition of granular narrative elements which opposes print-based narrative segmentation like Vladimir Propp’s ‘functions’ or comparable structuralist categories. The analogy of the database, e.g. evident in the comparison of narrative ruptures and “data coupling and control coupling” (292) in database design, is thereby intuitively implied. From that angle, Eladhari/Lindley’s proposal of formulat-

75

Real Wars on Virtual Battlefields? ing normal forms for causal relationships in narrative, modeled after Edgar F. Codd’s five normal forms for relational databases, represents an original and applicable way of conceptualizing digital game narrative. Another prominent game studies approach that uses technical concepts from computer science as structural devices is the Game Ontology Project (GOP) at Georgia Tech Institute; the eponymous notion of ‘ontology’ is clearly derived from the computer science rather than the philosophical sense. While Eladhari and colleagues only superficially theorize their deliberate use of these concepts, (Zagal et al. 2005: 4ff.) the GOP agenda mentions borrowing the concept of ‘ontology’ from computer science terminology, yet utilizing common digital game terminology like “collectables“ and “side quests“ within the individual ontology elements. The creation of ‘ontology’ is at least partially community-driven with multiple editors and uses the MediaWiki framework (cf. http://www.mediawiki.org/) as ‘middleware’, both in the technical and conceptual sense, to organize its elements. The consensual syntax for describing the ontological elements thereby takes up program code principles; for instance, elements are nested with top-level elements like “rules” and “goals” visible at all times and “children” (a programming term for subordinate nodes) displayed on every element page. Moreover, the internal structure of every element description is consistently segmented into “strong” and “weak” examples, imposing a binary distinction onto the process of collective authoring that implies quantifiability. The nesting of elements is thereby directly congruent with the equivalent pattern of storing pages as nodes in the MediaWiki framework; therefore, this pattern might have been at least partially shaped by the technological contingencies of the authoring environment. The pervasiveness of MediaWiki for scholarly purposes would actually justify a closer investigation of its formative impact on theory formation, read against the backdrop of the model sketched below in this chapter. Before the metaphorical and structural richness of program code for programmable media analysis can be explored, a prototypical code example will serve to illustrate some elementary properties like variable declarations, class definitions and loops. Since isolating a code segment from a full game would complicate the act of retracing the original functionality, the example excerpt at hand implements a paradigmatic particle fountain, i.e. a ‘simulation’ of individual water droplets forming a fountain, and is written in processing, taken from the official documentation. (Cf. http:// processing.org/learning/topics/simpleparticlesystem.html)

76

A Model of Programmable Media Analysis // A class to describe a group of Particles // An ArrayList is used to manage the list of Particles class ParticleSystem { ArrayList particles; Vector3D origin; ParticleSystem(int num, Vector3D v) { particles = new ArrayList(); for (int i = 0; i < num; i++) { particles.add(new Particle(origin)); // Add “num“ amount of particles to the arraylist } } void run() { // Cycle through the ArrayList backwards for (int i = particles.size()-1; i >= 0; i--) { Particle p = (Particle) particles.get(i); p.run(); if (p.dead()) { particles.remove(i); } } } void addParticle() { particles.add(new Particle(origin)); } void addParticle(Particle p) { particles.add(p); } }

The object class ParticleSystem provides a function by the same name to initialize a new particle system which can be assigned two variables: the number of particles to be emitted by the fountain at the same time and a vector specifying the starting position of the fountain; thus, a new fountain with 50 particles can be created at the origin of the internal coordinate system by typing ParticleSystem(50, [0,0,0]). The function run() constitutes the main loop of every particle fountain created, cycling through all particles which

77

Real Wars on Virtual Battlefields? are internally stored in a list to determine when a particle needs to be removed. The code determining the behavior of each particle, e.g. the initial velocity and direction as well as the impact of ‘gravity’, is triggered for each particle through the line p.run();, which is iterated over every item on the list; the respective commands are not shown in the excerpt above for reasons of clarity but basically resemble the structure of the superordinate ParticleSystem function and describe a ‘nested’ object.

Illustration 5: Particle fountain screenshot In the context of digital games, this paradigmatic code structure can be semanticized in different ways rather than just in order to simulate a fountain of water, e.g. as an enemy shooting ‘particles’ at the player character, which would then follow different, slightly more complex routines than simply being pulled down by ‘gravity’ or, using a conditional rather than permanent loop, as an explosion of dust marking the impact of a projectile in a war game. Moving on from this to a more complex application of paradigmatic algorithms, the functions of program code patterns inherently producing game genre differentiations can be re-traced and precisely localized in time; simpler games which are usually not in the center of academic attention are better suited to this purpose since their defining mechanics are less concealed, although there may be similar findings in large-scale commercial games. The game Every Extent and other recent independent games can be read as manifestations of OOP principles (even though object-orientation in terms of program code represents a spectrum rather than a binary criterion and can be more or less intensely implemented) by offering indirect means of interaction. The game takes place on a single screen with objects traversing the screen from all sides, mostly in groups. Collisions are usually harmful, yet players can cause their rather

78

A Model of Programmable Media Analysis amorphous avatar character to detonate and the resulting explosion affects all adjacent opponents. Their explosion, in turn, can again affect nearby opponents, ideally creating a prolonged chain reaction that gradually increases its destructive power. These complex interrelations of objects are handled through simple behavior scripts where both ships and explosions are considered as ‘objects’ which semi-autonomously move and react to collisions by executing their own little script. Thus, aspects of OOP like inheritance and polymorphism or simply semi-independent agents enable new but increasingly popular formal conventions like Every Extent in the first place; the shift from subject-centered games with a protagonist to ‘distributed’ identificational constellations in digital games thus arguably stems from the playful exploration of the inherently ludic applicability of given code structures. The game Tower Defense is another case in point which illustrates the same principle and has indeed spawned a discernible ‘genre’, at least in popular discourse; this genre formation process can itself be conceptualized as a distributed simulation process with a large community of amateur designers creating variations of the core concepts, often as a proxy for learning the implicated programming techniques, and with the sustainable elements emerging over time through experimentation and ‘quasi-natural’ selection through the same community of readerauthors. Following this analytical mode, it could be worthwhile to consider an alternative history of digital games reflecting the respective programming paradigm they embody. The game Tetris, for example, is characteristic of a time where sequential/procedural programming was still the dominant form of expression. Instead of treating every falling piece as an ‘object’, only a fixed number of lines up to the topmost filled line is checked for complete fills which are then statically removed by lines instead of by ‘objects’ from the rectangular grid that represents the playing field; once blocks are placed within the board, they are not treated internally as separate entities any longer (for a more comprehensive overview of a prototypical Tetris re-creation cf. http://cslibrary.stanford.edu/112/Tetris-Archi tecture.html). Gauntlet is a similar key game reflecting a paradigmatic programming concept; the home computer conversions of the dedicated arcade machine used to represent the multitudes of opponents on screen which characterize the game as modified font characters, i.e. literally as text. Thus, only the characters that constituted a contemporary computer screen had to be modified during a frame instead of all enemies in a given environment. (Cf. Retro Gamer 23 2006: 61) In an online forum, users have been attempting to collectively deconstruct and reverse-engineer the programming techniques behind Gauntlet; the fact that they informally refer to the

79

Real Wars on Virtual Battlefields? genre as a “‘lots of enemies’ game“ (cf. http://www.atariage. com/forums/index.php?showtopic=59430) suggests that this kind of programming technique of displaying numerous enemies on screen was indeed re-used, re-created and diversified among contemporary games and, subsequently, crystallized into a discursively operable genre label. Post #6 demonstrates how users intrinsically typologize their suggestions on how the game algorithms might have worked (proposing four techniques to implement multiple moving objects, i.e. “sprite reuse”, “sprite multiplexing”, “draw[ing] the sprites in bitmap mode” and “character mode graphics”) based on more or less extensive technical knowledge; the author of the post even characteristically, in a quasi-program code manner, ‘declares’ his suggestions (like an algorithmic function) e.g. as combinable with other techniques and listing the respective positive and negative effects. This fairly rigid ‘scheme’ of organizing information is geared towards easy, structured ‘usability’ and ideally enables the reader/user to ‘calculate’ the best configuration of techniques for a given problem by balancing all the factors. In a programming history of digital game genres, one of the most recent steps to date is procedural content generation which similarly carries a number of epistemological implications; this technique refers to content-like images or sound effects being described by algorithms (or ‘procedures’) rather than stored as binary information, similar e.g. to fractal shapes found in many natural phenomena. One application of this effect implemented by the company Pro FX is a ‘texture aging’ effect which actually emulates the aging process as a routine rather than using pre-produced texture layers etc. to mimic the visible symptoms of aging on surfaces (as e.g. in the Dosch Design Texture Aging Kit, which has been used for the development of digital games and other programmable applications). Previous genres depending on distinctive programming syntagms have already been tentatively re-written and blended with more recent elements. For instance, in the popular Plasma Pong, (http://www.plasmapong.com/) the paradigmatic game of Pong retained its core principles of simple rebounding physics which were already expressible with much earlier forms of code; yet, the simple ball behavior triggers a complex fluid dynamics engine which encapsulates object-oriented coding and retroactively modifies the behavior of the ball. A ubiquitous but marginalized instance of digital game development as a channeled unfolding of program code potentialities is the case of self-modifying code, i.e. program code that rewrites itself. Chris Crawford describes it as a powerful but hard-tocontain tool that was used e.g. in early copy-protection. With its focus on emergent behavior, it exhibits a few conceptual similarities with OOP, which was developed much later. Crawford also points to

80

A Model of Programmable Media Analysis instances of self-modifying rule-sets in non-digital games like Illuminati or Magic: The Gathering (Crawford 2003: 17), which again confirms the notion that programmability is not necessarily a property of computer-based media product, but has been explored at different times and in different contexts. Complementary to the actual games, established and consolidated playing practices like ‘speed-runs’ illustrate the strong (at least implicit) awareness of the technological framework behind digital games among its practitioners. A speed-run is an attempt to complete a game or part of a game in the minimum amount of time by any means necessary; playing sessions are usually recorded and compared online (a popular and comprehensive repository is the Speed Demos Archive at http://speeddemosarchive.com/). For instance, in games utilizing a 3D environment, speed-runners often keep the ‘camera’ turned towards the floor, navigating the environment by memory; the reason behind this is to minimize the calculations performed by the game software to display the environment and, thus, optimize their performance up to the millisecond. Consequentially, with the standardization of playing strategies, players gradually adopt the same internal measure of precision that the game engine allows for (i.e. the frame rates), with resulting performance discrepancies only discernible by technical means, as in the case of completion time differences of a few milliseconds. The notion of optimization which appears to be inherent in the games as programmed media products thus almost completely normalizes playing strategies, which sharply contrasts with the ‘openness’ offered by the game’s rule system and algorithmic complexity. To conclude, the emergent ‘alternative’ theorization and media usage patterns described in this paragraph, which are arguably tied directly to the consensual reading of games as ‘programmable’, exhibit and further propagate program code literacy analogous to visual and other forms of literacy; this assumption will be pursued and elaborated throughout the whole argument.

2.2.1 LUDICITY IN HUMAN-COMPUTER INTERACTION (HCI) From a technological perspective, instead of assuming the object of a computer game as a given, it appears theoretically more rigid to consider ludicity as an inherent property of programmable media in order to tackle the controversial definition of what can be considered a ‘game’ from another angle. Therefore, the later focus on military-related themes and on corresponding notions of historiography also requires the consideration of phenomena which go beyond the strict definition of a (digital) game.

81

Real Wars on Virtual Battlefields? Inverting the usual scholarly premises and prioritization, digital games shall tentatively be defined as a particular instantiation of programmable media usage in general, canalizing the always inherent ludicity into recognizable rule ecologies and laying out their functionality by tying them in with hermeneutical conceptions of play. In this sense, the ‘game’ is rather considered as a pattern of media reception rather than as a ‘fixed’ media object, which was already exemplified e.g. in the case of file-sharing military documents in the introduction. The link between creating or modifying and actually playing digital games was much more visible in the early days of CVGs as e.g. Düssler already pointed out in 1989. (“The relationship between ‘teleplayers’ and programmers thus is extremely close, despite huge qualitative differences of performance”; Düssler 1989: 24) Currently, the conflation of ‘reading’ and ‘writing’ is becoming relevant again, although the whole process has moved from a niche phenomenon into the spotlight of mainstream culture. While this historio-graphical pattern allows for instructive historical crossreferences, e.g. to the coupling of reading and writing in the 18th century (cf. Kittler 1985: 138), the analogy appears somewhat flawed upon closer inspection, especially since the current process is much more representative of society as a whole than 18th century literary production was and since the practices of referencing and even re-using others’ ‘texts’ are different in a digital medium. The inherently ludic quality of human-computer interaction (HCI) has already been intuitively acknowledged in the on-topic literature. Francis Hunger, a media artist with a strong theoretical perspective, points out the playful elements in the ‘hacker’ archetype, using the movie Wargames (1983), however, as an illustration without explicitly accounting for its fictional status (e.g. Hunger 2005: 12). While the movie adequately renders many aspects (as well as clichés) of HCI imagery in the early 1980s, in this case e.g. the playful and innocent way the protagonist manipulates his school’s computer, it can only provide starting points for further thought, not substantiate any hypotheses. The repeated focus on films and other media artifacts as ‘mirrors’ of contemporary culture (e.g. 14/15) is indicative of the level of ‘naturalization’ this argumentative pattern has attained in media art discourse. Additional references, such as the quoted empirical finding that playful interaction with computers is allegedly rather a male phenomenon (13), sketch a rather ‘essentialist’ portrayal of gender in HCI. Whilst the tendency to tie playfulness in HCI so closely to gender is bound to produce generalizations, some ascriptions from the article are in fact useful, e.g. the ‘trial and error’ component of both hacking and game playing. (15)

82

A Model of Programmable Media Analysis An interesting, yet unrelated idea presented in the same article is the description of the open source movement and its maxims as a “gigantic, continuous game”; (14) particular ludic elements of open source practices are, unfortunately, not elaborated on. However, on looking more closely, the specific forms of collaboratively written program code with an agreed-upon ‘goal’ (i.e. a pre-defined technical specification of the desired application and the elements of the programming language as an inherently binding ‘rule ecology’) do indeed exhibit overlaps with a game setup and might serve as a starting point for further thought. Madey, Freeh and Tynan studied a large number of open source projects and analyzed their distribution using computational tools like cluster analysis and power law relationships, ultimately attempting to model the behavior of the open source community using the software toolkit SWARM, an agent-based authoring environment employing very similar techniques to those used e.g. for artificial intelligence in digital games. I will come back to social modeling and the implications of the swarm paradigm in more detail in the following chapter. For the time being, one of the most important, in this case computationally derived, results holds that the community and project work can be considered a “self-organizing system” (Madey et al. 2004: 204), which is simultaneously one of the most common structural descriptions of a (digital) game. The authors point out that using the SWARM toolkit to “grow an artifical Sourceforge”, i.e. one of the biggest repositories of open source software, (220) allowed for observing “invariant properties of the simulation” (221), which are then compared with observations and statistical data on the ‘real’ Sourceforge project, in order to discern flaws in the simulation and to refine recursively the model which also represents a ‘self-organizing system’. Therefore, this example is modeled after or at least derives its plausibility from the semi-ludic programmable technology, i.e. the software toolkit, used to determine it. Another paper co-authored by Daniel Eckert, Stephan Koch and Johann Mitlöhner attempts to analyze cooperative behavior in open source projects using an iterated version of the prisoner’s dilemma, arguably the most prominent mathematical game model (cf. http://oss2005.case.unibz.it/Papers/5.pdf). Thus, both on the level of social interaction within and the theoretical modeling of the open source movement, ludic patterns can be traced and are intuitively applied. Similarly, elements of play can be found in paradigmatic and widely used computer applications. For instance, Jayakanthan describes the programming language Logo, developed by Seymour Papert for use in schools in the 1960s, as “one of the earliest applications of computer games as a form of education” (Jayakanthan 2002: 99). For instance, the ‘turtlegraphics’ component of Logo vi-

83

Real Wars on Virtual Battlefields? sualizes and semanticizes the abstract writehead used to draw vector graphics in Logo as a turtle which can be controlled to perform various tasks. This facilitates reading the sequential drawing process of even complex images like flowers, using mathematically expressed movements, as an instance of both ‘paidia’ and ‘ludus’ according to Caillois, i.e. free-form play which is consensually constrained by the program code syntax and lexicon. The anthropomorphization mainly stems from the fact that the writehead is directed, i.e. ‘looks’ in one direction and cannot simply move in any direction, but has to ‘turn’ as if it had a ‘body’; this is primarily a purely systemic rather than a visual criterion, i.e. the tip of the triangle representing the turtle symbolically represents its head mostly by virtue of its systemic behavior (cf. the illustration within the official online documentation at http://el.media.mit.edu/logofoundation/logo/turtle.html). The inherent joy in writing LOGO programs as a ludic activity results from the transcoding of abstract formulae into visual patterns and the concurrent asymptotic optimization of the user’s mental projection of the expected outcome, adhering to Csikszentmihalyi’s concept of flow, i.e. the ideally parallel increase of challenge and capacities, which is a key element in games. That is, the user/player repeatedly tries out new formulae, in practice mostly variations of previous expressions, and is only able to project the resulting image approximately; at the same time, the projections become more and more accurate just as, in a game, experienced players combine more and more complex combinations of rules into automatable patterns or behavioral scripts that allow for ever more exhaustive strategies. A related case study is Margaret Daisley’s interpretation of computer-mediated communication (CMC) as an inherently ludic activity. Daisley argues from a pedagogical perspective, attempting to teach instructors in classroom contexts how to utilize the emergent playful usage of online queries and other instances of CMC with children. (Daisley, Margaret “The Game of Literacy. The Meaning of Play in Computer-Mediated Communication” Computers and Composition 11 1994: 107-119) This relatively early approach can be instrumental in understanding digital games first and foremost as computer applications; at the same time it is representative in its attempts to apply this knowledge (for pedagogical purposes) directly, instead of systematically exploring its basic assumptions. This is not to criticize the approach but only to suggest that the momentum of applicability might be a reason for this phenomenon being still comparatively understudied. The pedagogical program implies that the work/play shift, which e.g. Csikszentimihalyi sees as having been re-negotiated e.g. with the emergence of theme parks as confined escapist spaces, is reconciled by the inherent playfulness

84

A Model of Programmable Media Analysis of CMC. While CMC might in fact be positively harnessed in educational contexts, I would argue that in the case of digital war games the claim needs to be inverted: rather than the sphere of work becoming ‘fun’, the sphere of ‘play’ is gradually and barely perceptibly turned into work. Extrapolating from this pattern can provide a number of useful vantage points. Following Sherry Turkle’s binary opposition of programming styles, even theory formation using programmable media like Google Scholar and image databases can be plausibly considered as inherently ludic. Crawford strikes a similar chord from a broader commonsensical perspective, pointing out episodically how playful rumination has historically operated as the origin of fundamental theoretical progress (Crawford (2) 2003: 231). This resonates with earlier cases like research on the open source movement, where the tools of study were found to fundamentally shape the conceptual imagination of the researchers. The inherent ludicity hypothesis can be illustrated from yet another angle, following the traditionally prevalent technological definition of a computer as an “interpreted automatic formal system” or “formal game” (e.g. Haugeland 1989: 48). Haugeland among others draws a structural parallel between the principles underlying automated computation and those found in familiar games like chess, Go, and Tic-Tac-Toe; “formal games” are thereby characterized by the elements of token manipulation, digitality and finite playability. Those systems, moreover, are found to be mediumindependent, i.e. theoretically separatable from the materially of the tokens which could e.g. be either printed cards or wooden board pieces; (58) in this regard, they coincide with Seymour Chatman’s or Werner Wolf’s concept of narrative as a structure (or ‘frame’ in the case of Wolf) which is ideally independent of the ‘carrier’ medium. This property can be used to distinguish formal games from other forms of ludic activity (like soccer) that are not independent of their material medium (e.g. the material of the ball). A second important criterion is formal equivalence¸ necessitating corresponding position constellations, starting positions and validity of moves in both systems. (62) Haugeland presents an interesting analogy, drawing on those two principles, by stating that, although “brain cells and electronic circuits” are different media, both can host formally equivalent, medium-independent systems; thus, a “computer mind”, i.e. a reasoning machine, and a ‘biological’ mind can be considered as equally comparable as computer chess vs. genuine chess. (63) Like the model proposed in this chapter, Haugeland’s argument relies on conceptualizations [Anschauungen] of games rather than actual games on the object level. Throughout the chapter, Haugeland uses examples from board games, mostly Solitaire, as examples of formal systems, taking the structural correspondences with computers as

85

Real Wars on Virtual Battlefields? programmable systems as self-evident. ‘Goals’ (49) are mentioned as one element typical of a game; as shown in the example by Margaret Daisley above, apart from fixed goals, sub-goals can be applied locally and dynamically and modified recursively, as in the case of looking up a topic in a search engine where intermediate findings can alter the search strategies or define new sub-goals within the superordinate task of performing online research on a given topic. Haugeland’s accounts of how game players are able to perform complex finite actions through an ordered combination of simple actions (65) strikingly resemble algorithms used in digital games. For instance, Haugeland describes how a player determines whether a chess king is in check by evaluating the respective position of every one of the opponent’s figures in relation to his king; since this process is performed quickly it appears as one coherent evaluation. This move of ‘cycling’ through all opposing pieces as objects and performing a cognitive simulation of all their possible moves appears very similar to a strategy of solving the problem using program code and supports the notion of ‘playing algorithms’, i.e. semiautomatable playing strategies implicitly modeled after the observed behavior of a programmable system. A computer chess AI, depending on the contingencies of its programming in OOP, would follow a very similar pattern by means of the method of moving the ‘chess king’ object through the potential positions of all remaining opponent figure objects. An illustrative digital game example of the ‘inherent ludicity’ assumption is the game Enter the Matrix (Shiny 2003), which implements a ‘hacking mode’, consistent with the narrative theme of the franchise, as a model or metaphor of the fundamental reality-game relation in parallel with the regular first-person shooter gameplay. The player utilizes a text-based interface to unlock materials and even, within the constraints of the parser, to create and manipulate objects in the 3D environment. Understanding and mastering the in-game syntax, which is significantly consistent with established DOS conventions, is staged as and inherently functions like a game, giving the player an initially very limited repertoire of keywords that need to be applied and combined to access other areas of the conceptual ‘playing field’ semanticized as virtual ‘hard drives’ or ‘folders’ within the game narrative. Many players still remember the DOS commands and practices as already historicized ‘cultural techniques’ and, to modify McLuhan’s hypothesis, increasingly interpreted them as a ‘game’ after they became outmoded. Following Roger Caillois’s typological axes, this macro interacteme would be an instance of paidia (free-form play) with a strong focus on mimicry (or simulation) and arguably a hint of Ilinx, i.e. a sense of desirable immersion [Aufgehen] in the game. (Caillois 2001: 12)

86

A Model of Programmable Media Analysis It might thus be worthwhile trying to map concepts of ludic activity onto a technology-centered framework of analysis, e.g. showing how culturally conditioned and binding ludic patterns ex-post legitimate and pre-structure the interaction with a programmed medium, instead of trying to explain issues related to digital games with an a priori understanding of play, complemented by surface knowledge of the underlying technologies. In this sense, not specifically war-themed games but digital games in general, by virtue of their technological makeup and the social practices implied by their usage, can be argued to generate patterns of thinking ‘compatible’ with contemporary military policy standpoints with regard to training and public consensus. I will elaborate on this very basic claim in Chapter 4a with reference to the impact of programmable media on military self-reflection and planning, particularly after the 2003 U.S. Army Transformation Roadmap (cf. http://www.army.mil/2003Tran sformationRoadmap/).

2.2.2 A VOCABULARY OF ALGORITHMS A suitable first step towards programmable media analysis would be to obtain an inventory of conventional algorithms, just as literary genre theory cannot function without a consensual set of defining narrative devices or topoi. In a traditional understanding shaped by authors like Alastair Fowler, the loss of “generic rules” in a literary form effectively amounts to its “death”. While this extreme reading is certainly criticizable, (cf. e.g. Fishelov 1989: 27f.) the call for reconstructing “conventions” that provide the “key” to a genre by close reading could be extended to digital games. Fishelov concedes that some “allegorical traditions in poetry” or other literary forms may be almost impossible to re-construct without a ‘key’ in the form of related contemporary formal conventions or contextual knowledge; analogously, some developments in ‘historical’ war games, e.g. the representation of battlefields through grids, necessitate background knowledge about the algorithmic conventions, e.g. the inability of contemporary hardware to perform multiplications and divisions, as well as the adequate assessment of contemporary notions of warfare. As will be shown later, both overlap in many cases as e.g. Pias exemplified with reference to the Korean War. Often, repositories of digital game algorithms are categorized by components like e.g. character animation or texture mapping (Sánchez-Crespo Dalmau 2003: 356, 445), which provides some insight into how to usefully subdivide a program code vocabulary; the fundamental distinction between algorithmically describing indoor and outdoor environments is such an apparently counter-intuitive ex-

87

Real Wars on Virtual Battlefields? ample. Compiling an ‘exhaustive’ selection of digital games in general would go beyond the scope of this book; in chapters 4b and 4c, however, I will elaborate on a number of examples pertaining to Gulf War representations and algorithms characteristic of military games respectively. In contrast to literary conventions, programming idioms used in digital games are evidently more deterministic because they need to be ‘machine-readable’, i.e. processable. An analysis of characteristic uses would need to begin at a higher level of abstraction. For instance, the game Desert Strike: Return to the Gulf (1992) popularized the isometric representation of battlefields by coining a framework for representing environments and mathematically describing movement in isometric projection. At that time, other developers had to reverse-engineer the technology instead of licensing the respective program component and would use it in observably different ways; nonetheless, as a precursor of middleware technology, these practices underscored the formative effect of signature algorithms on contemporary digital game production. With this focus, quantitative methods of analysis that would be much less compatible with a literary or film studies perspective on digital games (like measuring the density of ‘object’ interaction distributions and instances of data exchange between algorithms) appear useful; one such example, which will be taken up again in the following chapter, is the exchange of variables between physics and AI components in a game engine. Missions in the abovementioned Desert Strike all start out in the same emergently linear fashion with the helicopter departing from its aircraft carrier, requiring the player to approach the shore in a necessarily onedimensional way. Upon arrival at the coastline, the game abruptly increases nonlinearity with a number of objectives to fulfill in any order and corresponding opponents/reinforcements to distribute, thus implicitly capturing the ‘calm before the storm’ trope as rooted in a now-historical, positivist understanding of war prevalent during the production of the game. These shifts in interaction density can be observed on various scale levels and, as in the case of Desert Strike, need not be quantifiable in every case but can also be useful as an intuitive, approximative approach towards programmable media design. Accordingly, digital games are being discussed in computer science terms, such as the criterion of ‘branching complexity’ used in graph theory (http://interviews.slashdot.org/ comments.pl?sid=165947&threshold=1&commentsort=0&mode=thr ead&pid=13846357#13846685). Chris Crawford has already proposed ‘measuring’ interactivity by degrees instead of as a “boolean” property; although Crawford’s concern (as a designer) is the

88

A Model of Programmable Media Analysis “quality” of interaction, these concepts can be equally useful for critical analysis (Crawford (2) 2003: 6).

2.3 Reading (Digital) Games as Code Using the phenomenon of program code as a conceptual point of origin for programmable media analysis implies the consideration of programming as a cultural technique like reading and writing. As a preliminary working hypothesis that will be pursued throughout the following chapters, digital games (specifically war games) can be said to propagate program code literacy as a widespread phenomenon, occasionally overlaying other (hermeneutic) forms of literacy and meaning-making. For instance, guidance systems built into most current games already gradually reveal the game’s algorithmic framework; Chris Crawford was one of the first designers to actually “[supply] some of the AI routines to the player” within a contextsensitive guidance system for the war game Patton vs. Rommel, thus making its logic observable and re-constructible (Crawford 2003: 425f.). Again, it is worthwhile to point tentatively to continuities from earlier periods, when forms of literacy were considered ‘arcane knowledge’ and a source of empowerment [Herrschaftswissen], countered by developments like slave literature in 18th and 19th century America, where particular (media) texts have been ‘reverseengineered’ to develop a recursively refined understanding of the conventions of textual and media production. It will be indicated later on how MODs of popular war games occasionally perform similar cultural functions, even though many stay within the conceptual frame suggested by the game software as contingent ‘toolkit’. In Patton vs. Rommel, the same AI routines that measured the player’s front line consistency to calculate and control opponent behavior were also used for the in-game guidance system which pointed players to ragged segments in their frontlines to facilitate ‘optimization’. The military ideal type of a ‘straight’ front line is thus firmly established [festgeschrieben] in the game’s rule ecology using algorithms, not visual representational forms. Evidently, this pattern is both characteristic of a now-historicized concept of warfare (and war games) and heavily indebted to late Cold War paradigms. As will be shown later, more recent patterns like ‘information warfare’ strategies are arguably ‘programmed into’ recent war game rule ecologies using similar techniques. With Balance of Power, Crawford attempts to subjectivize this feature by differentiating it into four ‘advisors’ which relay some AI feedback to the player, thus adding semi-random ‘bias’ to the data that the player had to sift. Reverse-engineering the implied ‘perso-

89

Real Wars on Virtual Battlefields? nalities’ of the advisors by understanding their patterns of bias through observation is another inherently ludic element that involves reading even statements connected to an in-game character/advisor as program code in order to play the game most successfully. Patton Strikes Back was the first war game written by Crawford that featured analysis routines created solely for player guidance which were not adapted from enemy AI code. By creating an AI instance that continually evaluates and assesses the player’s actions, making available the results to the player as a ‘corrective’ [Korrektiv], these games serve as a concrete and more complex instantiation of what Claus Pias termed the “user as ‘gestalt’ of [their] ports”, shaped by the technological contingencies of digital (war) games. Applied on a large scale, i.e. in numerous current games being played by a significant number of people, this process alleviates the still present fear of the concept of ‘artificial intelligence’, especially in military applications, as documented in films like Stealth (2005), by suggesting its hermeneutic ‘understandability’.

Illustration 6: Screenshot from Super Metroid All algorithmically defined patterns of interaction provided by a digital game, from simple movement to the complex symbolic manipulation of the game environment, more or less directly encourage reading the game as program code and act as a proxy for players to perceive the environment (and game content) according to those ‘systemified’ algorithmic patterns. A very basic but influential example is the Metroid game series, particularly the 2D installments which, at first glance, almost prototypically follow the conventions of the exploration-based action-adventure genre they co-defined. As an unusual form of interaction, though, the heroine can not only aim horizontally and vertically but also in 45° angles, which requires the player to overlay a cognitive ‘grid’ on the level topography and cognitively simulate displacing it to assess whether moving

90

A Model of Programmable Media Analysis could bring an enemy into the grid. Thus, by producing a particular form of player behavior, this movement algorithm foregrounds the algorithmic setup of the game itself, i.e. the virtual ‘grid’ and the levels being assembled from cubic building blocks facilitate calculation. Iconic but older and not precisely military-themed examples like this serve to demonstrate the general concept more clearly; contemporary interaction patterns are often more refined and, therefore, too subtle to be instantly visible. A more recent example transposed to a 3D environment is the game Shinobi for PlayStation2, in which the ‘dash’ attack also makes a geometric, i.e. algorithmicizable, perception of the environment plausible by allowing the protagonist to ‘teleport’ forward a fixed distance; since these ‘dash’ attacks are often strung together by proficient players (cf. e.g. a speed run at http://www.gametrailers.com/player/usermovies/70672.html), the dash distance constitutes a fixed ‘measure’ for those advanced players to ‘quantify’ the game environments and adapt their playing behavior accordingly. A third example of how digital games inherently lend themselves to ‘reading’ them as program code are hacks of games which actually play out current media users’ increasing algorithmic literacy, made manifest in their capacity for quasi-intuitively ‘reading’ and ‘writing’ algorithmic constellations. In the case of a user authoring ‘automated’ runs of Super Mario World levels (cf. e.g. http://www.metafilter.com/64388/Automated-Mario), this is achieved by re-reading familiar actual game constellations in abstracted algorithmic terms by recombining the separate, systemically relevant level elements like enemies, conveyor belts and platforms to create a reproducible instantiation of emergent game behavior. The popular game Linerider (cf. http://www.official-linerider.com/play. html) similarly requires the player to maximize the emergent complexity of algorithmic behavior in a free-form setup, derived from a physics system based on slopes and gravity. The impact of programmed applications on cultural techniques like reading and writing goes beyond digital games. As an analogy to the previous notion of reading as text-processing, it appears plausible to assume that writing, i.e. a conglomerate of techniques of structuring textual information, is informed and filtered by experiences with media technologies, be it a sheet of paper or a programmable application like Microsoft Word. Sub EnterCurrentDate() ‘ EnterCurrentDate Macro ‘ Macro recorded 15/03/2005 by UserName ‘ Selection.InsertDateTime DateTimeFormat:=“dd-MM-yy“,

91

Real Wars on Virtual Battlefields? InsertAsField:=False, _ DateLanguage:=wdEnglishAUS, Calendar Type:=wdCalendarWestern, _ InsertAsFullWidth:=False End Sub

(Visual Basic for Applications (VBA) code example, recorded within Microsoft Word) Recurring structural aspects of text layout in text-processing tools like nested bulleting, indentions and numbering are internally differentiated as codified patterns, i.e. manifestations of a particular algorithmic setup; similarly, a textually represented date can be inserted into MS Word as a code ‘object’ using the VBA code above. It is especially users intensely exposed to computer-based tools who increasingly adopt these learned patterns of organizing when writing on paper, a transitory illustration of how the tools shape the cognitive organization of information, i.e. by first creating a ‘data structure’ before developing complex thoughts, which mimics standardized practices rooted in computer programming (for examples of computer-based data structures cf. the Dictionary of Algorithms and Data Structures, compiled by the U.S. National Institute of Standards and Technology: http://www.nist.gov/dads/). Thus, digital games, especially those essentially relying on the organization of information as the main game principle like MMOGs, appear as a continuation of software tools, feeding into this modern-day notation system [Aufschreibesystem] in the Kittlerian sense. One recent cultural technique which exemplifies this is the writing of helper programs and most notably macros which are literally designed as ‘extensions of man’ (according to McLuhan) in that they e.g. extend the concept of ‘touch’ to immaterial data. This perceived ‘hapticity’ is achieved e.g. in many game applications on the Nintendo Wii console by the controller physically vibrating when data modification is possible, e.g. hovering over a button or input field. Writing modular code for numerous popular computer applications that extend the core functionality has now become a common phenomenon. The same development can also be observed in digital games, most prominently in the form of macros written in the internal scripting language of MMOGs, where players ‘program’ the projected behaviour of their in-game character during their ‘absence’, i.e. these everyday practices permeate all spheres of life and conceptually bridge the gap between leisure and work. A widely acknowledged distinction of programming styles and cultures in this regard is Sherry Turkle’s dichotomy of engineering vs. bricolage, which she terms “hard” vs. “soft” mastery (cf. Turkle

92

A Model of Programmable Media Analysis 2005(1): 101/102). This distinction allows for a number of associative links; Turkle describes it e.g. mapped onto the distinction between scientific and artistic or between top-down and bottom-up approaches to technology. Although she takes children as an example, she also elaborates upon how far programming styles can be generalized into ways of interpreting the world and problem solving in general. With programming having seeped into mainstream culture in recent years, not least through digital games, the application of these thoughts onto current concepts of warfare will be a recurring theme in the following chapters; with game-like programmed applications being used in every aspect from training to operations planning, I will argue that the logic of these programmable media observably shapes both the execution and interpretation of contemporary warfare. To set the stage for the following chapters, it is possible to differentiate further and extrapolate from this idea and to apply it tentatively to more recent developments, such as ‘extreme programming’, a programming style which is applied with particular eagerness in digital games development (cf. Schofield 2007). Numerous elements of ‘extreme programming’ indicate how far programmers conceptualize their own team-driven cooperative work as a quasi-programmed organization. For instance, a practice called ‘test-driven development’ foresees the writing of “small piece[s] of code that describe[...] how they want a new feature [...] to function”; that is, authors of code are required to write their own automatable test routines, the passing of which signals that a modular piece of the code is now operational and considered ‘autonomous’ in terms of further development; the epistemological implications of this style will be revisited in the following chapter. (1) To introduce another facet of reading digital games as code, both the production and reception of digital games can be considered an optimization process, or are at least driven by the logic of optimization; for that reason, the binary distinction between both is actually misleading, as will be elaborated upon with respect to military-themed games. The production of games involves optimizing graphics routines, e.g. through the strategic deployment of visual detail according to projected player behavior and perceptional patterns, e.g. positioning sophisticated effects where players are likely to notice them and balancing richness of features against usability. The reception of games involves optimizing reaction times, the distribution of attention and the reverse-engineering of the game code, particularly its weaknesses with regard to exploiting it. Using optimization logic as a conceptual connector allows for transferring patterns, e.g. emphasizing the inherent ludicity in the production process as well as the inherent ‘work routines’ often involved in

93

Real Wars on Virtual Battlefields? playing a game. Indeed, especially player discourse focusing on MMOGs like World of Warcraft often describes the iterative refinement of playing strategies and character development in terms similar to the setting off of items against each other in an Excel spreadsheet; some examples even discuss explicitly labeled “[theories]” about in-game algorithmic relations (cf. e.g. http://forums. worldofwarcraft.com/thread.html?topicId=91629683), which clearly exhibits overlaps with structures of academic discourse. Potential indicators of programmable media effects will be further explored in a paragraph on forum analysis in chapter 4b. Analyzing patterns of collective para-media reception like these forum transcripts through the lens of the technological apparatus provides the groundwork for re-reading the reception of older media formats based on the paradigm of programmable media. For instance, the logic of optimization/debugging is tentatively extended into practices of film reception, like spotting and archiving logical or chronological glitches, that has gained immense popularity in recent years (cf. e.g. http://www.moviemistakes.com/). While authors like Jason Mittell trace this phenomenon back to the increasing, individually insurmountable complexity of recent TV formats (as well as other media genres), and e.g. Henry Jenkins describes collective media usage with Pierre Levy as a product of the versatile communication channels available online, the contingencies of programmable media technologies should be considered as a complementary factor. This includes e.g. both the ‘watching as debugging’ patterns as described above, which consciously narrow down reception to a quasi-algorithmic process of comparing new scenes as ‘data’ against the collective factual knowledge within the community, as well as the use of databases and other contingent, programmable tools to archive and massively quantify the findings. Investigating the trickle-down effects on the interpretation of other media genres in greater detail goes beyond the scope of the argument at hand, but certainly warrants a closer look. Encouraging a pervasive reading of topical war films like Jarhead for logical inconsistencies can be understood as a direct extrapolation from the programmable media patterns described in this book. The particular glitches found by the community in Jarhead (e.g. improbable unit logos displayed on buildings in the background and firing behavior of weapons; cf. http://www.moviemistakes.com/film5357) exhibit both an astounding attention to technical detail and deep knowledge of military insignia and weapons properties, i.e. ultimately quantifiable information. This experimental outlook again reaffirms the initial hypothesis that hermeneutic reading is gradually supplanted by other forms of contiguity.

94

A Model of Programmable Media Analysis

2.4 Socio-Cultural Implications of Analyzing Programmable Media The increasingly popular RFID chip technology shall serve as an introductory example to illustrate the potential of socio-cultural transformations implied in the OOP paradigm and made plausible predominantly by digital games. RFID chips, i.e. cheap and small chips that can be implanted into different types of products, can be essentially understood as a way of extending the principles of OOP into the sphere of physical objects. For instance, at an experimental supermarket in 2004, all products were outfitted with an RFID chip to register their position, e.g. being taken away from or put back into the shelf, which produces intricate conglomerates of user data derived from numerous products (as objects or, more precisely, ‘event handlers’; for a practical introduction into the topic using Adobe Actionscript as reference language, cf. e.g. http://www.adobe.com/devnet/action script/articles/event_handling_as3.html) contributing simple data to a central server in a ‘semi-autonomous’ manner (for this actual usage scenario cf. http://www.wired.com/wired/archive/12.07/ shoppers.html). Another such case that has recently exemplified this principle is an application of RFID-type technology used in sports shoes that exchange data with their owner’s IPod to turn it into a pedometer application, measuring the number of steps and total distance (cf. e.g. http://www.wired.com/science/discoveries/ news/2006/11/72202). A team of computer scientists demonstrated how the unique ID the device uses for communicating with the IPod can be easily traced and used e.g. for surveillance purposes (cf. e.g. http://uwnews.washington.edu/ni/relatedcontent/2006/ November/rc_parentID28494_thisID28578.pdf). Even more importantly for this argument, the desire to voluntarily inscribe themselves into a digital architecture, i.e. making themselves a systemic component is apparently pervasive among enough media users to be tapped for economic purposes already. In this regard, I want to focus less on the already well-studied utopian pleasure of becoming part of ‘the machine’ and transcending corporeality and more on the desire for quantifiability and computability, i.e. of making one’s actions accountable by registering every ‘step’ in order to perform often pointless calculations on this emerging dataset such as e.g. averaging the walking distance by day of the week. The structural effects of technologies on the interpretation of everyday processes are of course not new; for instance, Arnold Pacey describes how 17th century societies understood processes in terms of machines as e.g. in the case of mechanical clocks being

95

Real Wars on Virtual Battlefields? read as symbols of “cosmic order” (Pacey 1991: 97ff.). The roughly contemporaneous case of musket reloading techniques drawing on Roman military routines, which were pioneered by Maurice of Nassau during the Eighty Years’ War, is especially relevant for the topic at hand since it exhibits many conceptual links with program code and involves warfare. Since muskets fired inaccurately and took a long time to reload, von Nassau first grouped musketeers together to maximize firepower which, however, posed the problem of controlling larger groups, especially on a scale that exceeded shouting range. To tackle this, von Nassau first broke down the complex reloading process into 42 discrete steps, like assuming priming position and inserting the gun powder. By drilling these steps into the troops, their consecutive execution could be cognitively ‘automated’, which provided an immense advantage in the chaos of battle. Next, the terms of command referencing these steps were standardized to ensure uniformity among units. The units were hierarchically organized in that one row of six to twelve soldiers would fire, then move out of the way to reload; this allowed for continuous firing. Most importantly, to allow for the control of larger units, the soldier performing a subtask, or often the commanding officer of his subunit, would repeat the shouted command; thus, a command would ‘spread’ across an enormous army with a minimum delay or ‘latency’ in engineering terms. These techniques are partially rooted in the logic of mechanization, conceiving of human soldiers as extensions of the mechanical apparatus of their guns, anticipating elements of the conveyor belt and other icons of industrialization. More importantly, however, they are also congruent with principles of OOP, which I assume as a structural imperative for contemporary warfare. The individual soldier could be conceived of as an indivisible object that might be grouped in ‘classes’ (i.e. small units) and would semi-independently react to input by triggering ‘methods’ and outputting ‘status variables’. The notion of compensating for the inaccuracy of contemporary weapons by firing volley shots like Van Nassau’s soldiers (99) furthermore exhibits similarities with programming practices like ‘brute force’ (i.e. determining an optimal strategy not by emulating actual strategic rationales, but by iterating a simplified version of a conflict/problem in every conceivable constellation, storing, assessing and sorting the results for later reference, using a single effectiveness variable) enabled by constantly expanding computational capacities (or army dimensions). This analogy is not intended to suggest conceptual overlap but to provide a framework for putting the effect of digital games in military contexts into historical perspective. As an insightful side note which, however, will not be further pursued in this book, Geoffrey Parker, in his extensive analysis of

96

A Model of Programmable Media Analysis infantry volley fire as a role model for what the Pentagon recently termed “Revolutions in Military Affairs” (or RMAs, cf. Parker 2007: 332), points out the Western specificity of this military strategy, enabling century-long Western military superiority. A culturally comparative perspective would certainly make for valuable future research. Maurice of Nassau was consecutively featured as the AI representative of the Dutch faction in the strategy game Age of Empires III; the Age of Empires and, most prominently, the Empire Earth. Franchises of digital war games also iconically implemented the notion of “revolutions in military affairs” by implementing discrete “epochs” while playing through the game (like e.g. the Dark Age, Digital Age and Nano Age in Empire Earth), each both represented and essentially characterized by a set of key weapon technologies. Moreover, the notion of RMAs will be revisited in chapter 4a by applying the above findings to current warfare practices whilst maintaining the focus on program code. Equally, linking aesthetics as a parameter of warfare to vantage points such as computability and program code allows for some revealing historical cross-referencing which, however, can only be touched upon briefly here. According to Köppen, Heichrich von Kleist differentiated between the formula and the metaphor as “both ends of human capability [Fähigkeit]”, a distinction which Köppen subsumes under the fitting chapter heading “The Beauty of the Parabola“ [Die Schönheit der Parabel]. (Cf. Köppen 2005: 66-71)

Illustration 7: Ballistics in the game Full Spectrum Warrior In Kleist’s time, towards the end of the 18th century, both formulae and metaphors metonymically represented the binary distinction criterion between all human beings. In his application to the concept of war, Köppen only relates mathematical principles to the sphere of ballistics, but provides some interesting historical data.

97

Real Wars on Virtual Battlefields? For instance, from 1934 onwards, ballistics became part of high school education [Gymnasialausbildung] both for practical and, albeit probably implicitly, ideological reasons. (Köppen 2005: 67) In a peculiarly circular argument, issues like ballistics currently again enter ‘civilian’ areas of life, e.g. in playing and, more directly, (re)programming games with military references where ballistics are made visible as a combination of variables handled by the respective game engine. I will attempt tentatively to transfer this concept to media representation of war in contemporary (Western) cultures; the apparently unintuitive concept of weapons as media proposed e.g. by McLuhan has thereby been fortified recently by new technological developments, including weapon systems like drones simultaneously providing ‘media’ imagery, and media technologies like digital games used as weapons within ‘information warfare’ scenarios. One nod towards the socio-cultural transformations triggered by programmable media is the case of authors strongly exposed to the logic of programmable media who already exhibit a distinctly technical understanding of human features. Chris Crawford’s cognitive model of “feature extraction” (Crawford (2) 2003: 294f.) exemplifies this notion, describing how viewers mentally re-construct a highly dithered image of a face on a computer screen. The cognitive process is described in terms very similar to automated image recognition algorithms, i.e. dissecting the face (or a rectangular subsection of the image) into expected shapes and putting on hold uninterpretable data for later reconsideration. Interpretations like these which become increasingly consensual within society need to be taken into account for a model of programmable media usage as a collective activity, spanning different games over a longer period of time. The fact that media developments enter “collective symbolism” [Kollektivsymbolik] and, by inference, textual (as well as other media) production is already consensual and has been investigated e.g. through retracing common metaphors that reflect media-induced cultural transformations as in the case of 19th century ‘universal’ historiography. (Fohrmann 2005: 377) Another important theoretical vantage point for extending the logic of program code to socio-cultural processes is Lewis Mumford’s concept of the “megamachine” (e.g. Mumford 1977: 229), referring to an integral conceptual ‘machine’ assembled from myriads of workers e.g. by pharaohs and kings that transformed the contemporary concept and scope of manual labor. Similar to von Nassau, rulers conceived this technique of social organization for a given purpose (e.g. constructing a monument) within the paradigm spectrum of contemporary technologies. Another similarity is the common assumption of making human operators ‘compatible’ with their tools,

98

A Model of Programmable Media Analysis a goal still present in current warfare maxims. This historically localizable pattern can be found amalgamated into a common gameplay pattern found e.g. in military-themed games like the Ghost Recon series, where players derive pleasure from embedding themselves into a squad or other group with AI-driven non-player character (NPC) instances, often lead by an NPC. The science-fiction game Wing Commander also occasionally subjugates the player to higherranking wingmen, which requires ‘following orders’; all these patterns can be understood as multiform semanticizations of what Pias generically describes as the user being ‘played’ by the computer. To take this idea further, the example of Wing Commander demonstrates that this pattern is at least partially modular, reusable code since it reappeared, within a more clear-cut military scenario, in the ‘sequel’ Strike Commander that took over and elaborated the basic technological frame.

2.4.1 AGENDA SETTING As any other medium, digital games implicitly exert complex agenda-setting effects. Unlike the first inquiries into the topic in the early 1970s which focused exclusively on news media and journalistic coverage (cf. e.g. McCombs, M.E., Shaw, D.L. “The Agenda-Setting Function of Mass Media” The Public Opinion Quarterly 36:2 1972: 176/77), fictional media content will be considered on the same level, especially since, as will be elaborated in the following chapter and has already been exemplified through the News at Seven case, the discursive and epistemological gap between both is being constantly, playfully re-negotiated in programmable media genres. However, the only effects which have received systematic, scholarly attention so far, especially with military-themed games, are related to violence. These effects are implicitly assumed in the case of military-endorsed game production, be it large-scale development, as in the case of America’s Army, or more experimental approaches like browser-based advertising games commissioned e.g. by the Australian Armed Forces (cf.http://www.defence.gov.au/Army/online_ games.htm). The relatively strong impact of interactive (i.e. programmable) vs. ‘static’ or hermetic media genres has recently been backed e.g. by a psychological study on interactivity in advertising.. (Schlosser 2006: 18). A comparison of customers exposed to an advertising webpage using static images as opposed to interactive product presentations yielded the surprising result that users of the interactive advertisement were significantly more susceptible to “false memories”, i.e. ‘errors’ consciously inserted into the post-

99

Real Wars on Virtual Battlefields? exposure memory test, which plays a crucial role in agenda setting policies and community discourse management. One of the most basic effects is probably the quantitative distribution of games focusing on particular military scenarios; conglomerates of heterogeneous scenarios in games spanning multiple conflicts constitute a more complex variety of this pattern. The noticeable accumulation of Vietnam-related digital games surrounding the initiation of the 2003 invasion of Iraq could certainly be explained from multiple angles, many of them ‘mundane’ and technical, such as the archives of digital assets like textures and models (for a peripheral but publicly available example cf. e.g. the materials collection for Marathon: Vietnam at 1000) {

20

float angle = atan2(source.xv, source.yv);

21

//Battles are represented as colour streaks.

22

tstreak(source.x,

source.y,

angle,

l,

source.thecolor); 23

};

24 };

(Source code of the battle module written in processing, cf. http://www.blprnt.com/processing/warpaint/Battle.pde; line numbers added by the author) Even the variable names “dice1” and “dice2” (line 5/6) directly reference the icosahedral dice that are being used for probability distributions in pen&paper role-playing games like Dungeons & Dragons; the internal battle algorithm similarly uses a random number from 1 to 20 to determine the damage dealt by each opponent which then influences the shape of the colored streak. Moreover, the internal nomenclature of variables references digital games, since each particle, labeled Wandered in the ‘battle’ and ‘streak’ function, has a “health” variable that governs its perseverance on the battlefield. In a sense, applications like Warpaint2 can be read as revitalizing previous conceptions of war, such as the aesthetic appreciation of the battlefield as a spectacle of form and movement (as sketched earlier), which are conceptually ‘untainted’ by the traumatic experiences of the 20th century; these are tentatively re-inserted into public dis-

382

Macro Effects of Programmable Media course in the form of simulacra. Within these consensually constrained boundaries, idealized conceptions of war as a ‘natural principle’, conducted within an ordered and stable framework, can still be temporarily upheld or reinstated, given that military conflict is algorithmically expressible in similar ways like other natural phenomena. The epistemological framework is not constituted by divine providence or man-made concepts like nation but technological differences like the shifting between program code and game engine or between high-level programming languages and scripting languages as abstractions. As has been argued throughout this book, programmable media have been influential in catalyzing public conceptions of war and even shaping actual policies accordingly, not least by homogenizing iconic tools and, thereby, practices involved. One prominent case is the 2005 Treaty of Prüm, which provided for the technological merging of data sources between the EU member states and related measures to counteract e.g. terrorist activities and illegal migration; it was incorporated into the legal framework of the EU in June 2007. In this representative document, political rationales and technological feasibility coalesce against the backdrop of an increasingly naturalized everyday use of programmable technologies as media. Most notably, the politico-technological épistémè (in the sense of Foucault) directly affects public discourse and, thus, allows e.g. for implementing inconvenient and unpopular policies in a different guise. For instance, while the project of a joint EU database combining biometric and other data was not feasible both for technological and political reasons, such as the lack of public support, the proposition of ‘interrelated’ data sources combined through a common interface (cf. NI-21: 2) that has been effectuated with the Treaty of Prüm is factually identical but dilutes potential controversy by inserting various ‘buffer’ layers of interdependent technologies into public discourse, similar e.g. to the hardware/software displacement described earlier in various contexts. The same discursive effect has been observed within actual military decisionmaking processes; as Jim Dunnigan reports on the functions of war-gaming in the planning phase of the 2003 invasion of Iraq, iterated war games using a database of historical military stratagems suggested that “a 21st century version of the World War II ‘blitzkrieg’” was the ideal military procedure for victory within the specified conditions. (Dunnigan 2004) Thus, both the (controversial) historical nomenclature and the strongly associatively loaded ‘template’ strategy had been made plausible within military planning through the corresponding tagging and systemic compatibility within the programmable war game environment used. With the explicitly comparative angle chosen for this book, attempting to wrap up the

383

Real Wars on Virtual Battlefields? main results covered in the previous chapters would be methodologically counter-productive; ideally, the arguments presented above raised productive questions and selectively analyzed some interconnections rather than narrowing down the discourse by settling for one aspect. One objective was to point out that not only militarythemed digital games or even specific genres like first-person shooters but programmable media in general, by virtue of their sociotechnological makeup, must be critically analyzed, especially against the backdrop of concurrent developments in military technologies. This does not imply a condemnation of digital games as such or a call for a more restrictive legal framework but, inversely, calls for an in-depth look at the complexities of programmable media at the military-civilian margin; instead of pointing at uncritical patriotic narrative or rhetoric in digital war games, a broader perspective is necessary to contextualize existing debates revolving around issues like violence and potential desensitizing effects of imagery. The case of grassroots media production already indicated how digital war games, like any other media genre, produce their own ‘antidote’ in the form of alternative historiographical concepts and the algorithmic ‘reverse-engineering’ of the systemic relations underlying contemporary warfare. The fear that programmable media ‘abolish’ historical perspectives and usher in sentiments of unrestrained neo-militarism is, therefore, unfounded, although the logic of the changing frame medium compared to radio (World War II) or television (Vietnam) has profound historiographical implications. Among these are the devaluing of ‘experts’ as discourse framing figures and the re-coding of traditional factors like eyewitness accounts and contemporary audiovisual documents within a quasi-algorithmic web of authenticity effects, like e.g. the fragments from a Hitler speech recording, digitized and randomly looped from an in-game radio as a quasiacousmatic audio object in Return to Castle Wolfenstein (Gray Matter Interactive et al.: Activision, 2001). Thus, the argument presented in this book takes in inter-activity and programmability in a re-accentuated sense as patterns of media reception and usage rather than as properties of media products; in this sense, fan site toolkits have been found to be more appropriately categorized as ‘inter-active’ than digital games as such and the concept is seamlessly extensible to older media artifacts, like the initially mentioned Krieg dem Kriege [War against War]. Following up this last example, the historically comparative angle has been a pervasive element of the argument at hand, especially since this type of approach is still rather uncommon within the novelty rhetoric of the digital games studies literature. Currently overused terms like ‘simulation’ can thus be re-investigated by applying them in retrospect to previous media

384

Macro Effects of Programmable Media technological configurations like the military photography of Frank Hurley (1885-1962). Hurley thoroughly understood photographic technologies and exemplifies a particular way of using them to ‘document’ military activities like the battlefields of the two World Wars. Hurley attempted to retouch photographs during the Second World War for added ‘authenticity’. This use of photography can equally be characterized as ‘simulation’, i.e. as a process of understanding the photographic material obtained from a physically experienced situation as ‘material’ to play around with in a quasi-inter-active process aimed at instantiating a subjective, procedural interpretation of realism. Consequently, this practice can be argued to have more in common with authenticity effects in programmable media production than with contemporary photography. (for examples cf. Mules, 2000: 314) According to this logic, no single photograph could actually ‘document’ the impression of war, so that other means were necessary to ‘simulate’ the soldier’s battle experience in the form of a “tableau”. For instance, Hurley juxtaposed representations of a horizon characterized by a particularly dramatic lighting situation with battlefield scenes from other photographs as in his famous ‘documentation’ of the aftermath of the 1917 battle at Paesschendaele in Belgium. Such composite pictures had been a common practice in the 19th century but had never before been systematically applied to military photographs.

6.9.1 A CULTURALLY-COMPARATIVE PERSPECTIVE The argumentation at hand could only touch upon a number of related issues which might be profitably reconsidered using the proposed model, but which need to be left for further inquiry. One such particularly rich approach is the culturally comparative perspective on war and programmable media, which has not yet been systematically explored. Taking into account examples of non-Western military-themed digital games or critically reviewing the theoretical bias within ‘game studies’ methodology will be important steps for further research; as a starting point, though, revisiting the central issue of ‘computability’ as a connector between programmable media and military logic should allow for the tentative inclusion of cultural contexts. A first step could be to look at instances of canon formation both in military usage of digital games and in academic coverage. For instance, Ralf Streibl, a German computer scientist and early scholarly adopter of digital games mentions the games History Line, Victory at Sea, Platoon, Commando Lybia, Desert Storm and Silent Thunder as examples of games with strong historical references; apart from the diverse scenarios, these games form an exqui-

385

Real Wars on Virtual Battlefields? sitely heterogeneous corpus and more than likely a German bias as evidenced e.g. through History Line (Blue Byte Studio: Blue Byte Studio, 1992), a German turn-based strategic simulation set in the unusual First World War context and Commando Lybia (Robert Pfitzner: independent, 1986), which became widely known in Germany through the local copy scene. While this singular canon might indicate the validity of the overall approach, mapping numerous canons from various, also non-Western sources can be suitable for sketching a rough ‘geography’ of conceivable war game syntagmas, which, assumedly, only partially overlaps with political but even more so with technological geographies. (for further elaboration on the notion of ‘technological geography’, cf. e.g. Inkinen, 2000) Analogously, the disposition towards programmable media technologies represents a complementary axis to interpret cultural alignments and refine the technological geography. For instance, a fundamentally critical federal response to the issue of military recruitment through digital games in Germany from 2002 (cf. Deutscher Bundestag, 2002) suggests that along the demarcation line of media technology adoption, Germany must be considered a different intra-Western cultural area; in this respect, countries/militaries like Singapore and Australia appear ‘culturally’ closer to the US than Germany. Despite their methodological problems, e.g. overemphasizing the determinism of technological developments, historiographical models, like the aforementioned categorization of time periods according to weapons technologies by Raudzens (Raudzens, 1990), have already suggested similar criteria for transcultural alignments and should be critically re-read against the paradigmatic functions of programmable media technologies instead of being entirely discarded. Another ‘implied’ aspect in Raudzens’ historical overview is the historiographical impact of nationalist and other interests constraining the interpretation of weapons as ‘keys’ to victory. For instance, the “American Rifle” allegedly invented in Kentucky/Pennsylvania by “non-Europeans” is still widely held responsible for winning the War of Independence, even though, upon closer inspection, this attribution appears disproportionately exaggerated. (414) Nonetheless, the same discursive operations still apply with game engine technologies as e.g. in the case of the SouthKorean War Rock engine Jindo 3D, named after a South-Korean island, or in multiple cases of Korean developers like AcroGames licensing iconic Western game engines like the Unreal Engine 3. (Cf. e.g. http://play.tm/wire/click/1629544) Following up the use of implied canons, one potentially adaptable model of the cultural functions of media, although also originally restricted to textual media, is Winfried Fluck’s ‘cultural spectrum’ [das Kulturelle Imaginäre]. (Cf. e.g. Fluck, 1996) Fluck describes how the American

386

Macro Effects of Programmable Media novel contributed to the social status and stabilization of the medium of literature through its functional differentiation (especially with regard to historical objects). In this sense, literature allows for a diffuse, flexible cultural spectrum which can support ideas and beliefs that are not (yet) culturally ‘thinkable’ [kulturfähig] and thus enables the constant transgression and extension of one’s own selfdescription. In his work, Fluck draws heavily on Wolfgang Iser’s literary anthropology and his assumption of the need for literature/fiction as an anthropological constant, something which hints at methodological bias and should be handled carefully. Nonetheless, the application of the ‘cultural imaginary’ on American literature might prove insightful, especially since the assumed simultaneity of the current state and ‘possible’ contingencies that Fluck posits as a methodological starting point is even more explicit in the case of programmable media, as has been found e.g. with regard to ‘authenticity’ in the third chapter. As to the topic of program code structures translated into economic models and vice versa, some widely adopted recent concepts in economics, tackling aspects like intercultural competence and knowledge management from a corporate perspective, have taken up hardware/software metaphors and rhetoric to make ‘culture’ conceptually compatible and manageable within corporate structures modeled after program code. While e.g. Geert Hofstede explicitly intends his dictum of culture as “software of the mind” to be metaphorical, acknowledging the limits of predicting human organizational behavior, (Hofstede, 2005: 3) the structure of Hofstede’s book directly reflects the need for a consistent taxonomy (or ‘ontology’ in computer science terms) to be translatable into pseudo-code; once a concept has been established (or ‘declared’ in programming terminology) in the book (like e.g. ‘uncertainty avoidance’ or ‘longand short-term orientation’), it is iteratively applied in all applicable contexts and even represented by a fixed headline structure. (Cf. vvi) Culture in this sense is understood as a range of instances, like family and educational background, “programming” individuals and creating ‘scripts’ or “patterns” (2) which are assumed to be default reactions to given stimuli. While the basic assumption behind this technical notion of culture is not at all new, the concrete application of programming terminology and its use to delineate organizational principles for globally operating companies are fairly recent phenomena. At the same time, program code acts as a conceptual proxy to making anthropological notions of culture (as “patterns of thinking, feeling, and acting”, 3) operable in economics and corporate environments; the fact that culture as “mental software” is inherently described as a rule set for “the social game” (4) supports the inherent ludicity of program code posited earlier in this book. Further

387

Real Wars on Virtual Battlefields? differentiations of the model, like the three pyramidal layers of mental programming (i.e. “human nature”, “culture” and “personality”, 4) which members of a given society are subjected to, are implicitly congruent with extrapolations found in program code like the various abstraction levels of programming languages (low-level languages, high-level languages, and scripting languages). While a comprehensive take on the topic would go beyond the scope of this concluding outlook, a substitutive closer look at an Egyptian proposal for terrorist modeling using Complex Adaptive Systems (CAS) and mathematical game theory (Ahmed, 2005) might serve to broaden the perspective and exemplify the unexpectedly homogenizing effects of programmable media technologies on concepts of warfare. The direct experience with ‘terrorist’ activities on Egyptian ground since the 2004 Sinai bombings, mostly stemming from ongoing Israeli-Palestinian frictions, could be expected to produce characteristically different approaches and findings from regional studies on and policies concerned with terrorism than in the US or Europe. Yet, the example study similarly proposes an example of terrorism modeling that appears equally indebted to the logic of digital games and yields surprisingly convergent results. The general underlying structure of the article at hand is a simple syllogism. First, complex systems in an abstract sense like “the brain, the immune system [or] the economy” (2) are presented as a homogeneous field which should not be disassembled but studied as a whole. Then, terrorism is implicitly posited as a ‘complex system’ (3) which appears perfectly plausible according to the tenets of common sense. Finally, terrorism is conceptually linked to other ‘complex systems’ and their particular methodology, as in the aforementioned examples of the economy, “the ecology” (2) and central body functions by virtue of the previously established, discrete category of the CAS label; therefore, the interchangeability of explanatory modes, as identified earlier in digital games, is taken for granted. The authors do acknowledge the inherent unpredictability of a CAS, but argue that continually tweaking and refining the parameters is a sufficient, quasi-quantifiable solution; this premise subscribes to the positivist frame of thinking that has been identified as an inherent consequence of programmable media, encouraged by formal properties like the organic, modular model of writing program code. The most basic recurring argument is, thus, that the inherent unpredictability of terrorism as a CAS can be reduced by “mathematical and computer models”. (4) Moreover, the authors are representative in that they depart from the merely intuitive observation that computational principles, like “agents [using] reinforcement learning” (3), appear to be, at least semantically, congruent with observed and popularly communicated properties of terrorist

388

Macro Effects of Programmable Media activities. For instance, one element derived from intuitive observation that directly feeds into their mathematical model is the truism that the “eradication of (even wrong) ideas is extremely hard, and that it is more practical to contain them”. (3) Thus, through ‘cognitive simulation’, i.e. by testing the provisional equation with a few key values for its compatibility with ‘real-world’ observations, the authors define a framework of suggested actions; this approach appears perfectly congruent with practices of defining simulation models in digital games, both for social systems and other phenomena like the physics of water and other natural phenomena. Another familiar application in the article is its use of mathematical percolation theory, (3/4) a particular form of abstraction which posits the comparability of terrorist influence on “passive sympathizers” (which share the terrorists’ motives but not their disposition to act) with percolation, i.e. the movement and filtering of liquids through porous material, thereby translating terrorism into the conceptual domain of chemistry. Serge Galam, whom Ahmed and colleagues quote on applying percolation theory to social systems, had already suggested policies based on this conceptual paradigm, identifying the “dimension of the social space”, consisting mainly of the number of shared beliefs, as the critical variable which should be reduced in order to curb terrorist potential; by analogy, this dimension would be roughly equivalent to and, thus, imaginable as the thickness of the porous layer. Galam has also already published extensively on the topic with a prescriptive impetus. (Cf. e.g. http://eprintweb.org/S/authors/cond-mat/ga/Galam) The implied analogy between terrorism and “related systems” (4) like natural disasters, which has already informed mainstream discourse, e.g. in the form of topoi like “regional conflagration” (cf. e.g. http://www.globalresearch.ca/index.php?context=va&aid=7176), is thus conceptually stabilized through a quasi-circular argument on the basis of the common denominator of algorithmic expressibility. Observations from these “related systems” are thus tentatively applied also to terrorism, e.g. the hypothetical case that power shortage in one city due to an earthquake can have ripple effects on other cities, although this has not yet been formalized or systematically unfolded. In their conclusion, the authors explicitly mention the “war on terror” (4) as the framework for their political recommendations; the mathematical notation is thereby used directly in the propositions, e.g. cautioning that “excessive force may increase [i.e. the social dimension variable]” (5). The symbolic connotations of this form of rhetoric complement the impression that both the conceptualization of terrorism, and, by inference, society in general, as a computer program is workable within this discourse, as is the point that potential cultural discrepancies are at least partially le-

389

Real Wars on Virtual Battlefields? veled by the common epistemological framework of programmable technologies. To conclude, the convergent development of programmable media may indeed ‘virtualize’ elements of warfare, including even the sensorial experience of the physical battlefield, but it simultaneously represents a very ‘real’ and fiercely contested battlefield itself which extends not only into COTS games but every aspect of programmable media production and consumption. Indeed, as elaborated throughout this book, the conceptual momentum of programmable media in both civilian and military contexts must be understood as inherently ludic and informs media usage in general; characteristically new military strategies mentioned above like open development contests (cf. e.g. the DARPA Urban Challenge or the Northrop Grumman Lunar Lander Challenge) and organizational structures like the Stryker brigade are manifestations of this complex phenomenon which is mirrored in the particularly popular media genre of digital games. At the DARPATech 2007 conference, a military research agenda was sketched which revolves around the concept of “programmable matter”, i.e. amorphous chemically synthesized materials which can take on multiple shapes by reassembling themselves much like a computer program composed from semi-autonomous modules, which temporarily epitomizes in a way the conceptual dynamics of programmable media by planning to reinsert them into material reality. (Cf. Zakin, 2007) The concept itself exhibits the appeal of having the flexibility of software applications with material tools too, a concept nurtured not least through parallel developments, like e.g. the differentiation of macros as multiform, hybrid ‘tools’ in MMORPGs (cf. e.g. Carless, 2004: 125-128), within the sphere of digital games. The project sketch, however, even goes a step further by explicitly stating that “programmable matter” could be “the ultimate way to prevent technological surprise”. (1) That is, the fear of global terrorism as an enemy, which arguably embodies programmable media principles as has been argued above, is not only countered by relying on more sophisticated programmable applications but even by aiming at projecting these principles onto military hardware. Thus, the ‘virtual battlefields’ of digital games and simulation neither represent a unidirectional development nor are they independent of geo-political constellations; in this context, digital games should be seen less as an end in themselves but rather as the currently most visible manifestation of program code in everyday life as well as in contemporary warfare.

390

Bibliography

Abelson, Harold, Sussman, G.J. (1996): The Structure and Interpretation of Computer Programs, Cambridge, MA: MIT Press. Aday, Sean, Cluverius, John, Livingston, Steven (2005): “As Goes the Statue, so Goes the War: The Emergence of the Victory Frame in TV Coverage of the Iraq War”. Journal of Broadcasting & Electronic Media 49:3, p. 314-331. Affron, Matthew, Antliff, Mark (1997): Fascist Visions: Art and Ideology in France and Italy, Princeton, NJ: Princeton UP. Ahmed, E. et al. (2005): “On Complex Adaptive Systems and Terrorism”. Nonlinear Sciences 15 Jan., http://arxiv.org/abs/nl in.AO/0501032. Allen, Graham (2003) Roland Barthes, London: Routledge. Altheide, David L., Grimes, Jennifer N. (2005): „War Programming: The Propaganda Project and the Iraq War“. The Sociological Quarterly 46, p. 617-643. Anderson, Benedict (1991): Imagined Communities. Reflections on the Origin and Spread of Nationalism, London: Verso. Anonymous (2003): “A World at War”. Military Images 25:2, p. 19. Asendorf, Christoph (1991): “Space Writings. Künstliches Licht und Raumerfahrung zwischen den Weltkriegen” In: Martin Stingelin/Wolfgang Scherer (Hg.) HardWar/SoftWar, München: Fink, p. 159-171. Barami, Bahar (2004): “Market Trends in Homeland Security Technologies”. paper presented at IEEE Conference on Technologies for Homeland Security, Cambridge, MA: Volpe Center, http://www.volpe.dot.gov/ourwork/dimensions/05060 4/trends.doc. Bates, Robert H. et al. (1998): Analytic Narratives, Princeton, NJ: Princeton UP. Bauerlein, Mark (2005): “Deconstructing Chomsky”. Reason Magazine 1 Apr. http://www.thefreelibrary.com/_/print/Print Article.aspx?id=130136930. Beaumont, Roger A. (1971): “Images of War: Films as Documentary History”. Military Affairs 35:1, p. 5-7.

391

Real Wars on Virtual Battlefields? Becker, Howard S. (1982): Art Worlds, Los Angeles, CA: U of California P. Behn, Rolf (1984): „Videospiele als Abbild der Wirklichkeit“. In: K.-J. Kreuzer (Hg.), Handbuch der Spielpädagogik, Düsseldorf: Schwann, p. 683-695. Benjamin, Walter (1979): “Theories of German Fascism: On the Collection of Essays War and Warrior”. New German Critique 17, p. 120-128. Benjamin, Walter (2005): “The Work of Art in the Age of Mechanical Reproduction” (on-line reproduction; original text: 1936). http://www.marxists.org/reference/subject/philosophy/works/ ge/benjamin.htm. Bjork, Staffan, Holopainen, Jussi (2005): Patterns in Game Design, Hingham, MA: Charles River Media. Bøgh-Andersen, P., Holmqvist, B., Jensen, J.F. (1994): The Computer as Medium, Cambridge: Cambridge UP. Boczkowski, Pablo (1999): “Mutual Shaping of Users and Technologies in a National Virtual Community”. Journal of Communication 49:2, p. 86-108. Böhme, Stefan (2007): “Normalismus im Computerspiel” (published online). 20 Apr. http://ag-games.de/system/files/Normalismus _B%C3%B6hme.pdf. Bolter, Jay David (1991): Writing Space: Computers, Hypertext and the Remediation of Print, Hillsdale, NJ: Lawrence Erlbaum. Bonk, C.J., Dennen, V.P. (2005) “Massive Multiplayer Online Gaming: A Research Framework for Military Training and Education” (Technical Report 2005-1). U.S. Department of Defense http://www.adlnet.gov/downloads/189.cfm. Bowles, Mark D. (2000): “Liquifying Information: Controlling the Flood in the Cold War and Beyond” In: Miriam R. Levin (Hg.), Cultures of Control, Amsterdam: Harwood Academic Publishers, p. 225-246. Broyles Jr., William (1984): „Why Men Love War“. Esquire 102, p. 55-65. Buckingham, David (2006): “Studying Computer Games”. In: Diane Carr et al. (Hg.), Computer Games: Text, Narrative and Play, Malden, MA: Polity Press, p. 1-14. Bürger, Peter (2005): “Paradigmenwechsel im US-amerikanischen Kriegsfilm? Ein Überblick”. In: S. Machura, R. Voigt (Hg.), Krieg im Film, Münster: LIT Verlag, p. 237-265. Bürger, Peter (2007): „Die Unterhaltungsindustrie als Rüstungssektor“. Telepolis 4 Apr. http://www.heise.de/tp/r4/artikel/25/ 25002/1.html. Büttner, C.. et al. (Hg.) (2004): Der Krieg in den Medien, Frankfurt/Main: Campus.

392

Bibliography Caillois, Roger (2001): Man, Play and Games, Urbana/Chicago, IL: U of Illinois P. Camper, Brett (2005): “Reveling in Restrictions: Technical Mastery and Gameboy Advance Homebrew Software Development”. www.lcc.gatech.edu/~bogost/courses/spring06/lcc2700/. Cardelli, Luca, Wegner, Peter (1985): “On Understanding Types, Data Abstraction and Polymorphism”. ACM Computing Surveys 17:4, p. 471-522. Carless, Simon (2004): Gaming Hacks, Cambridge, MA: O’Reilly. Center of Military History (CMH) (1999): “Army Museums, Historical Artifacts, and Art (AR 870-20)”. http://www.usapa.army.mil/pdf files/r870_20.pdf. Center for Nonproliferation Studies (CNS) (2002): “Literature Review of Existing Terrorist Behavior Modeling”. http://cns.miis.edu/ pubs/reports/terror_lit.pdf. Cherny, Lynn (1999): Conversation and Community: Chat in a Virtual World, Stanford, CA: CSLI. Chion, Michel (1994): Audio-Vision. Sound on Screen, New York, NY: Columbia UP. Church, Doug (1999): “Formal Abstract Design Tools”. Gamasutra 16 Jul., http://gamasutra.com/features/19990716/design_tool s_01.htm. Clauset, Aaron/Young, Maxwell (2005): “Global Terrorism Follows a Power Law”. physicsweb, http://physicsweb.org/articles/news/ 9/2/6/1. Clauset, Aaron/Young, Maxwell (2006): “Scale Invariance in Global Terrorism”. arxiv.org, http://arxiv.org/pdf/physics/0502014v2. Clavell, Richard (2003): McLuhan in Space, Toronto, ON: U of Toronto P. Clemons, E.K., Santamaria, J.A. “Maneuver Warfare. Can Modern Military Strategy Lead You to Victory?”. Harvard Business Review April 2002, p. 3-11. Colton, Timothy (1969): “The ‘New Biology’ and the Causes of War”. Canadian Journal of Political Science 2:4, p. 434-447. Cook, Daniel (2007): “The Chemistry of Game Design” Gamasutra 19 Jul., http://www.gamasutra.com/view/feature/1524/the_ch emistry_of_game_design.php. Coonfield, Gordon (2006): “Thinking Machinically, or, the TechnoAesthetic of Jackie Chan: Toward a Deleuze-Guattarian Media Studies”. Critical Studies in Media Communication 23:4, p. 285301. Correia, Cynthia Cheng (2006): “War Games: Gaining Advantage on the Business Battleground”. Directions Magazine 16 Aug., http://www.directionsmag.com/article.php?article_id=2252&trv =1.

393

Real Wars on Virtual Battlefields? Croteau, D., Hoynes, W. (2002): Media, Society: Industries, Images and Audiences, Thousand Oaks: Pine Forge Press. Crawford, Chris (1982): De Re Atari. A Guide to Effective Programming (Atari home computer manual, scanned and published online), http://www.atariarchives.org/dere/. Crawford, Chris “Play and Mentation” (1994/95): Interactive Entertainment Design 8, http://www.erasmatazz.com/library/J CGD_Volume_8/Play_Mentation.html. Crawford, Chris (2003): On Game Design, Indianapolis, IN: New Riders. Crawford, Chris (2003): The Art of Interactive Design, San Francisco, CA: No Starch Press. Cronin, Ciara (2003): “Democracy and Collective Identity: In Defence of Constitutional Patriotism”. European Journal of Philosophy 11:1, p. 1-28. Cronin, Eric, et al. (2004): “An Efficient Synchronization Mechanism for Mirrored Game Architectures”. Multimedia Tools & Applications 23:1, p. 7-30. Crook, Paul (1994): Darwinism, War and History. The debate over the ‘biology of war’ from the ‘Origin of Species’ to the First World War, Cambridge: Cambridge UP. Cutumisu, M. et al. (2005): “Generating Ambient Behavior in Computer Role-Playing Games”. Lecture Notes in Artificial Intelligence 3814, p. 34-43. Deloura, Mark (2001): Game Programming Gems, Hingham, MA: Charles River Media. Der Derian, James (1997): “Interview With Paul Virilio” http://proxy.arts.uci.edu/%7Enideffer/_SPEED_/1.4/articles/d erderian.html. Der Derian, James (1998): The Virilio Reader, Malden, MA: Blackwell Publishing. Devji, Faisal (2006): “Global War on Terror as De-Militarization”. ISIM Review 18, p. 30-31. Deutscher Bundestag (2002): “Haltung der Bundesregierung zur Rekrutierung von jungen Menschen für den Militärdienst durch den Einsatz von Computerspielen wie z.B. America's Army” (Drucksache 14/9764). http://dip.bundestag.de/btd/14/097/ 1409764.pdf. Dondis, Donis A. (1973): A Primer of Visual Literacy, Cambridge, MA: MIT Press. Dornan, Cain (2006): “Interview with John Carmack”. Gamer Within 1 May, http://www.gamerwithin.com/?view=article&article=13 19&cat=2.

394

Bibliography Drucker, Johanna (2001): “Digital Ontologies: The Ideality of Form in/and Code Storage – or – Can Graphesis challenge Mathesis”. Leonardo 34:2, p. 141-145. Dunnigan, James F. (1997): The Complete Wargames Handbook 2nd Ed. (out of print; published on-line), http://www.hyw.com/ Books/WargamesHandbook/Contents.htm. Dunnigan, James F. (2004): “Wargames and the War in Iraq”. Strategy Page 23 Mar., http://www.strategypage.com/war games/articles/wargame_articles_2004919233.asp. Düssler, Stefan (1989): Computer-Spiel und Narzissmus – Pädagogische Probleme eines neuen Mediums, Eschborn bei Frankfurt: Verlag Dietmar Klotz. Dy, Bernard (2005): „Game Reviews: Call of Duty: United Offensive“. Military History 22: 3, p. 71&79. Egozy, Eran (2005): “Eyetoy: Antigrav - Tales of Woe” (presentation held at Games Developers Conference 2005). http://www.harmonixmusic.com/gdc/woe.ppt. Eladhari, Mirjam (2003): „Object-Oriented Story Construction in Story-Driven Computer Games“. Gamasutra 10 Nov., http://www.gamasutra.com/education/theses/20031110/ eladhari_01.shtml. Eladhari, Mirjam, Lindley, Craig (2002): “Causal Normalization: A Methodology for Coherent Story Logic Design in Computer RolePlaying Games”. In: J. Schaeffer, M. Müller, Y. Björnsson (Hg.), Computers and Games: Third International Conference, CG 2002, Berlin: Springer, p. 292-307. Fishelov, David (1989): Metaphors of Genre: The Roles of Analogies in Genre Theory, University Park, PA: Penn State Press. Fishwick, P.A. (2002): „Aesthetic Programming: Crafting Personalized Software“. LEONARDO 35:4, p. 383-390. Fitzsimons, Bernard (2003): “Developments in Database Design”. Military Training & Simulation News 5:6, p. 36-40. Flake, Gary William (1998): The Computational Beauty of Nature, Cambridge, MA: MIT Press. Fluck, Winfried (1996): "The American Romance and the Changing Functions of the Imaginary“. New Literary History 27:3, p. 415457. Fohrmann, Jürgen (Hg.)(2005): Gelehrte Kommunikation, Wien: Böhlau. Fong, Gwenda (2004): “Adapting COTS Games for Military Simulation”. Proceedings of the 2004 ACM SIGGRAPH International Conference on Virtual Reality Continuum and its Applications in Industry, p. 269-272.

395

Real Wars on Virtual Battlefields? Ford, Mike, Barlow, M., Lewis, E. (2003): „An Initial Analysis of the Military Potential of COTS Games“. Proceedings of SimTect 2003 Adelaide, http://www.siaa.asn.au/get/2391114970.pdf. Friedman, Ted (1999): “The Semiotics of Sim City”. First Monday 4:4, http://www.firstmonday.org/issues/issue4_4/friedman/in dex.html. Friedrich, Ernst (1991): Krieg dem Kriege (23rd Ed.), Frankfurt a.M.: Zweitausendeins. Fritz, Jürgen (Hg.) (1988): Programmiert zum Kriegsspielen. Weltbilder und Bilderwelten im Videospiel, Bonn: Bundeszentrale für Politische Bildung. Fuller, Samuel H./Lesser, Victor R. (1974): “Microprogramming and its Relationship to Emulation and Technology”. State of the Art Report on Microprogramming and Systems Architecture Berkshire: Infotech Ltd., http://research.microsoft.com/~gbell/ CGB%20Files/Microprogramming%20and%20Relationship%20t o%20Emu%20and%20Tech%207405%20c.pdf. Gaddis, John Lewis (2002): The Landscape of History: How Historians Map the Past, New York, NY: Oxford UP. Galloway, Alexander R. (2004): “Social Realism in Gaming”. Game Studies 4:1, http://www.gamestudies.org/0401/galloway/. Galloway, Alexander R. (2004 )(2): PROTOCOL - How Control Exists After Decentralization, Cambridge, MA: MIT Press. Garfinkel, Simson (2000): Database Nation, Sebastopol: O’Reilly. Geertz, Clifford (2002): “Thick Description. Towards an Interpretive Theory of Culture”. In: Lyn Spillman (Hg.), Cultural Sociology, Malden, MA: Blackwell, p. 63-69. Geifman, Anna (1993): Thou Shalt Kill: Revolutionary Terrorism in Russia, 1894-1917, Princeton, NJ: Princeton UP. Genette, Gerard (1997): Paratexts: Thresholds of Interpretation, Cambridge: Cambridge UP. Gieselmann, H. (2002): Der Virtuelle Krieg. Zwischen Schein und Wirklichkeit im Computerspiel, Hannover: Offizin. Gieselmann, H. (2007): “Ordinary Gamers. The Vanishing Violence in War Games and its Influence on Male Gamers”. Eludamos 1:1, http://www.eludamos.org/index.php/eludamos/article/vie w/9/20. Gilleade, K.M., Dix, A., Allanson, J. (2005): “Affective Videogames and Modes of Affective Gaming”. Proceedings of DiGRA 2005: Changing Views – Worlds in Play, http://www.comp.lancs.ac.uk /computing/users/dixa/papers/DIGRA2005/. Giors, John (2004): “The Full Spectrum Warrior Camera System”. Gamasutra 25 Mar., http://www.gamasutra.com/gdc2004/feat ures/20040325/giors_01.shtml.

396

Bibliography Giroux, Henry A. (2004): “War on Terror – The Militarization of Public Spaces and Culture in the United States”. Third Text 18:4, p. 211-221. Goggin, Joyce (2005): “Metaphor and Madness: Stacking the Deck on Iraq”. Bad Subjects 26 May, http://bad.eserver.org/reviews /2005/goggin.html. Goldstrasz, Thomas, Pantle, Henrik (2002): “Informatik & Heeresgerät. Die Entwicklung des Computers während des Zweiten Weltkriegs”. http://waste.informatik.hu-berlin.de/Diplom/WW2 /weltkriegscomputer.html. Grammont, Steve (2007): „Combat Mission: Shock Force Q&A: Simulating Modern War“. Gamespot, http://www.gamespot.com /pc/strategy/combatmissionshockforce/news.html?sid=616587 4&mode=previews&print=1. Gray, Jonathan, Mittell, Jason (2007): “Speculation on Spoilers: Lost Fandom, Narrative Consumption and Rethinking Textuality”. Particip@ations 4:1, http://www.participations.org /Volume%204/Issue%201/4_01_graymittell.htm. Grieveson, L. (2004): Policing Cinema: Movies and Censorship in Early-Twentieth-Century Cinema, Berkley, CA: U of California P. Gunter, G. et al. (2006): “A Case for a Formal Design Paradigm for Serious Games”. Proceedings of Code Human Systems/ Digital Bodies Oxford, OH: iDMAa, http://www.units.muohio. edu/codeconference/papers/papers/Gunter%20Kenny%20Vick %20paper.pdf. Gunzenhäuser, R. (2002): „Raum, Zeit und Körper in Actionspielen“. dichtung-digital.com 4:22, http://www.dichtung-digital.com/ 2002/03-22-Gunzenhaeuser.htm. Hafez, Kai (2000): Islam and the West in the Mass Media. Fragmented Images in a Globalizing World, Cresskill, NJ: Hampton Press. Hafez, Kai (2005): “Islamismus und Medien – eine unheilvolle Symbiose.” (speech transcript). In: Jörgen Klußmann (Hg.), Terrorismus und Medien. Eine komplexe Beziehung, Bonn: Evangelische Akademie im Rheinland, p. 55-60. Haffa, Robert P. Jr., Patton, James A. Jr. (1998): “Gaming the 'System of Systems'”. Parameters. US Army War College Quarterly Spring, p. 110-121. http://www.carlisle.army.mil/usa wc/Parameters/98spring/patton.htm. Hagen, Wolfgang (2007): “Wie ist eigentlich eine so zu nennende ‘Medienwissenschaft’ möglich?” (lecture at the University of Vienna). http://whagen.de/get.php?page=vortraege/2007/2007 0606Wien/MedienWissenschaftMoeglich.pdf. Hall, Justin (2001): “GameStudies.org”. hypertext kitchen, http:// www.hypertextkitchen.com/Features/Hall.html.

397

Real Wars on Virtual Battlefields? Hanson, Victor Davis (2003): “Military Technology and American Culture”. The New Atlantis. A Journal of Technology & Society 1:1, p. 29-36. http://www.thenewatlantis.com/archive/1/TNA 01-Hanson.pdf. Harris, J.LL., Taylor, P.A. (2005): Digital Matters. Theory and Culture of the Matrix, London: Routledge. Haugeland, John (1989): Artificial Intelligence. The Very Idea, Cambridge, MA: MIT Press. Hendry, J., Raven, M. (2002): Japan at Play: The Ludic and Logic of Power, London: Routledge. Henz, Brian J., Shires, Dale R. (2003): “An Object-Oriented Programming Framework for Parallel Finite Element Analysis with Application: Liquid Composite Molding”. ipdps – International Parallel and Distributed Processing Symposium (IPDPS '03), p. 253. Hoffmann, Stefan (2002): Geschichte des Medienbegriffs, Hamburg: Meiner. Hofstede, Geert (2005): Cultures and Organization: Software for the Mind, Columbus, OH: McGraw Hill Professional. Huizinga, Johan (2003): Homo Ludens. The Sociology of Culture (digital print), London: Routledge. Hunger, Francis (2005): “War Games. Hacker Spielen” (lecture transcript). http://www.hgb-leipzig.de/~francis/irmielin/ writings/wargames-hacker_spielen_e.pdf. Inkinen, Tommi (2000): “From the Past to the Future. A Need for New Geographical Knowledge”. Geography Online 1:1, http://w ww.siue.edu/GEOGRAPHY/ONLINE/gov1n1a1.html. Jaiswal, N.K. (1997): Military Operations Research: Quantitative Decision Making, Norwell, MA: Kluwer Academic Publishers. Jameson, Frederick (1972): The Prison-House of Language. A Critical Account of Structuralism and Russian Formalism, Princeton, NJ: Princeton UP. Jayakanthan, R. (2002): “The Application of Electronic Games in the Field of Education”. The Electronic Library 20:2, p. 98-102. Jeffords, S., Rabinovitz, L. (1994): Seeing through the Media: The Persian Gulf War, New Brunswick, NJ: Rutgers UP. Jenkins, Henry (2003): “War Games”. MIT Technology Review online 7 Nov., http://www.technologyreview.com/articles/03/11/wo_j enkins110703.asp?p=1. Jenkins, Henry (2002): “Interactive Audiences. The 'Collective Intelligence' of Media Fans”. In: The New Media Book London: British Film Institute. Jenkins, Henry (2004): “Game Design as Narrative Architecture”. In: Noah Wardrip-Fruin and Pat Harrigan (Hg.), First Person: New

398

Bibliography Media as Story, Performance, Game, Cambridge: MIT Press, p. 118-130. Johnson, N. et al. (2005): “From Old Wars to New Wars and Global Terrorism”. arxiv.org 29 Jun., http://lanl.arxiv.org/abs/ physics/0506213. Joint United States Public Affairs Office (1967): “The Use of Superstitions in Psychological Operations in Vietnam. Policy No. 36”. http://www.psywarrior.com/superstition.html. Juul, Jesper (2002): “Introduction to Game Time”. In: Noah Wardrip-Fruin, P. Harrigan (Hg.), First Person: New Media as Story, Performance, and Game, Cambridge, MA: MIT Press, p. 131-142. Juul, Jesper (2005): half-real. Video Games between Real Rules and Fictional Worlds, Cambridge, MA: MIT Press. Katz, Warren (1994): “Military Networking Technology Applied to Location-Based, Theme Park and Home Entertainment Systems”. Computer Graphics, 28:2, p. 110-112. Kaur, Raminder/Sinha, Ajay J. (2005): Bollywood: Popular Indian Cinema Through a Transnational Lens, New Delhi: Sage Publications India. Kittler, Friedrich (1985): Aufschreibesystem 1800/1900, München: Fink. Kittler, Friedrich (1986): Grammophon, Film, Typewriter, Berlin: Brinkmann & Bose. Kittler, Friedrich (1995): „There is no Software“ ctheory 18 Oct., http://www.ctheory.net/printer.aspx?id=74. Kittler, Friedrich (2006): “Rock Me, Aphrodite”. Telepolis 24 May, http://www.heise.de/tp/r4/artikel/22/22695/1.html. Kirsh, David/Maglio, Paul (1992): “Perceptive Actions in Tetris”. Control of Selective Perception, Proceedings of the AAAI Spring Symposium, http://adrenaline.ucsd.edu/kirsh/articles/ perceptiveactions/perceptive.pdf. Kopp, Carlo, Mills, Bruce (2002): “Information Warfare and Evolution” (lecture transcript). http://www.csse.monash.edu.au /courseware/cse468/Lectures/_JIW-2002-2-CK-BIM.pdf. Kopp, Carlo (2006): “Shannon, Hypergames and Information Warfare”. Monash University Courseware, www.csse.monash.ed u.au/courseware/cse468/2006/Lectures/_JIW-2002-1-CK.pdf. Köppen, Manuel (2005): Das Entsetzen des Beobachters. Krieg und Medien im 19. und 20. Jahrhundert, Heidelberg: Universitätsverlag Winter Heidelberg. Konzack, Lars (2002): “Computer Game Criticism: A Method for Computer Game Analysis”. Proceedings of Computer Games and Digital Cultures Conference, Tampere: Tampere UP, http://www .digra.org/dl/db/05164.32231.

399

Real Wars on Virtual Battlefields? Kreimeier, Bernd (2002): “The Case for Game Design Patterns”. Gamasutra 13 Mar., http://www.gamasutra.com/features/200 20313/kreimeier_pfv.htm Kreimeier, Klaus (2002): Die UFA-Story. Geschichte eines Filmkonzerns, Frankfurt a.M.: Fischer. Kreimeier, Klaus (2005): “Die Konfiguration des Bösen”. In: HeinzPeter Preusser (Hg.), Amsterdamer Beiträge zur Neueren Germanistik, Krieg in den Medien, Amsterdam: Rodopi, p. 281290. Kühme, Dorothea (1997): Bürger und Spiel: Gesellschaftsspiele im Deutschen Bürgertum zwischen 1750 und 1850, Frankfurt a.M.: Campus. Kurita, Isamu (1995): “Japanese Art and the Japanese View of Nature”. http://www.moa-inter.or.jp/english/setsugekka/setsu -kurita.html. Lakoff, George (2006): „Metaphor and War, Again“. http://www.alt ernet.org/story/15414/. Lecky-Thompson, Guy (1999): “Algorithms for an Infinite Universe”. Gamasutra 17 Sep., http://www.gamasutra.com/view/feature/ 3377/algorithms_for_an_infinite_universe.php?print=1. Lederer, Susan E. (1998): „Repellent Subjects: Hollywood Censorship and Surgical Images in the 1930s“. Literature and Medicine 17:1, p. 91-113. Lekan, T.M., Zeller, T. (2005): Germany’s Nature.: Cultural Landscapes and Environmental History, New Brunswick, NJ: Rutgers UP. Leonard, David (2004): “Unsettling the Military Entertainment Complex: Video Games and a Pedagogy of Peace”. Studies in Media & Information Literacy 4:4, http://www.utpjournals.com /jour.ihtml?lp=simile/issue16/leonardfulltext.html Lenoir, Tim, Lowood, Henry (2000): "All But War Is Simulation: The Military-Entertainment Complex". Configurations 8:3, p. 289335. Levy, Pierre (2001): Cyberculture, Minneapolis, MN: University of Minnesota Press. Lewis, Michael, Jacobson, Jeffrey (2002): ”Game Engines in Scientific Research”. Communications of the ACM 45:1, p. 2731. Lintermann, Bernd, Deussen, Oliver (2005): Digital Design of Nature: Computer-Generated Plants and Organics, Berlin: Springer. Locke Siegfried, Susan (1993): “Naked History: The Rhetoric of Military Painting in Postrevolutionary France”. The Art Bulletin 75:2, p. 235-258.

400

Bibliography Lustick, Ian (2003): “Virtual Pakistan and Its Futures: An Exercise in the Deployment of Constructivist Theory”. http://www.psych. upenn.edu/sacsec/abir/_private/publications/VirtualPakistan_ CAEG.doc MacKenzie, J.M. (1995): Orientalism: History, Theory and the Arts, Manchester: Manchester UP. MacMillan, J. et al. (2005): “DARWARS: An Architecture That Supports Effective Experiential Training”. DARWARS Research Papers, http://www.darwars.com/downloads/2005%20IITSEC %20White%20Paper%20v2.pdf. Madey, G., Freeh, V., Tynan, R. (2005): "Modeling the F/OSS Community: A Quantitative Investigation". In: Stephan Koch (Hg.), Free/Open Source Software Development, Hershey, PA: Idea Publishing, p. 203-222. Magy, S.E., Smith, B.K. (2006): „Learning Through Game-Modding“. ACM Computers in Entertainment 4:1, http://portal.acm.org/ citation.cfm?id=1111293.1111301&coll=ACM&dl=ACM Manninen, Tony, Kujanpää, T. (2005): „The Hunt for Collaborative War Gaming – CASE: Battlefield 1942“. Game Studies 5:1, http://www.gamestudies.org/0501/manninen_kujanpaa/. Manovich, Lev (1997): “Cinema as a Cultural Interface”. Essays, http://www.manovich.net/TEXT/cinema-cultural.html. Manovich, Lev (1999): “Database as Symbolic Form”. Convergence: The International Journal of Research into New Media Technologies 5:2, p. 80-99. Manovich, Lev (2001): The Language of New Media, Cambridge, MA: MIT Press. Mayer-Schönberger, Viktor (2007): “Useful Void: The Art of Forgetting in the Age of Ubiquitous Computing” (on-line paper). 9 May, http://ksgnotes1.harvard.edu/Research/wpaper.nsf/rw p/RWP07-022/$File/rwp_07_022_mayer-schoenberger.pdf. McGrath, A. E. (2001): A Scientific Theology Vol. 1: Nature Grand Rapids, MI: Wm. B. Eerdmans. McKenzie, F.D. et al. (2005): “Crowd Modelling for Military Simulations Using Game Technology”. International Journal of Intelligent Games and Simulation 4:1, p. 35-54. McLuhan, M. (2002): Understanding Media, Cambridge, MA: MIT Press. Mezihorak, Pavel (2004): “The State of Game Development in Eastern Europe”. Gamasutra 24 Nov. http://www.gamasutra.co m/features/20041124/mezihorak_01.shtml. Midlarsky, Manus I. (Hg.) (2000): Handbook of War Studies II, Ann Arbor, MI: U of Michigan P. Mishra, V. (2002): Bollywood Cinema: Temples of Desire, New York, NY, Routledge.

401

Real Wars on Virtual Battlefields? Mitcham, Carl (1999): „Why Science, Technology and Society Studies?“. Bulletin of Science, Technology and Society 19:2, p. 128-134. Mittell, Jason (2007): “Speculation on Spoilers: Lost Fandom, Narrative Consumption and Rethinking Textuality”. Particip@tions 4:1, http://www.participations.org/Volume%20 4/Issue%201/4_01_graymittell.htm. Montfort, Nick (2006): “Combat in Context”. Game Studies 6:1, http://gamestudies.org/0601/articles/montfort. Morris, Rosalind (2004): “Images of Untranslatability in the US War on Terror”. interventions 6:3, p. 401-423. Morris, Sue (2002): “First-Person Shooters. A Game Apparatus”. In: Geoff King, Tanja Krzywinska (Hg.), Screenplay. cinema/ videogames/interfaces, London: Wallflower Press, p. 81-97. Moxley, F.I. et al. (2001): “A Foundation for Coalition Interoperability using NATO’s C3 Technical Architecture”. Crosstalk. The Journal of Defense Software Engineering 8, p. 4-8. Mules, Warwick (2000): “Lines, Dots, and Pixels: The Making and Remaking of Printed Images in Visual Culture”. Continuum: Journal of Media & Cultural Studies 14:3, p. 303-316. Müller, Thomas, Spangenberg, P.-M. “Fern-Sehen – Radar – Krieg“ Martin Stingelin, Wolfgang Scherer (Hg.), HardWar/SoftWar, München: Fink, 1991: 275-302. Mumford, Lewis (1977): Mythos der Maschine. Kultur, Technik und Macht, Frankfurt a.M.: Fischer. Natkin, S., Vega, L. (2004): “A Petri Net Model for Computer Games Analysis”. International Journal of Intelligent Games and Simulation 3.1, p. 37-44. Nekrashenko, A.N. (2004): “Computer War Games: How to Estimate Decisions made by C+C Trainees”. Military Thought 13:2, p. 6366. Von Neumann, John/Morgenstern Oskar (1973): Spieltheorie und Wirtschaftliches Verhalten (3rd unchanged Ed.), Würzburg: Physica Verlag. Newman, James (2002): “The Myth of the Ergodic Videogame. Some Thoughts on Player-Character Relationships in Videogames”. Game Studies 2:1, http://www.gamestudies.org/0102/ newman/. Newman, James (2004): Videogames. Routledge Introductions to Media and Communications, Oxford: Routledge. Nicolai, Georg Friedrich (1918): The Biology of War, New York: NY Century Co. Nieborg, D. (2003): “A First Person Analysis: Towards an Academic Method for Analyzing First Person Shooters”. http://www.game

402

Bibliography

space.nl/content/FPA2003.pdf#search=%22nieborg%20first%20 person%20analysis%22. Nieborg, D. (2006): "First Person Paradoxes - The Logic of War in Computer Games". In: K. Oosterhuis, L. Feireiss (Hg.), Game Set and Match II. On Computer Games, Advanced Geometries and Digital Technologies, Rotterdam: Episode Publishers, p. 107115. Nohr, Rolf (2004): “Das Verschwinden der Maschinen. Vorüberlegungen zu einer Transparenztheorie des Games”. In: B. Neitzel, M. Bopp, R. Nohr (Hg.), See? I’m Real… Multidisziplinäre Zugänge zum Computerspiel am Beispiel von Silent Hill, Münster: Lit Verlag, p. 96-125. Novak, Jasminko, Voigt, Benjamin (2007): “Mashups: Strukturelle Eigenschaften und Herausforderungen von End-User Development im Web 2.0“. I-COM, http://www.ifi.unizh.ch/im/fileadm in/user_upload/i-COM07.Mashups.Novak.Preprint.pdf. Oatley, Keith (2000): “Shakespeare's Invention of Theatre as Simulation that Runs on Minds” (conference paper). Artificial Intelligence & Simulation of Behaviour Birmingham, http://ww w.cs.bham.ac.uk/~axs/aisb2000/papers/oatley.pdf. O'Connell, Mary Ellen (2005): “When is a War not a War? The Myth of the Global War on Terror?”. ILSA Journal of International & Comparative Law 12:2, http://papers.ssrn.com/sol3/papers. cfm?abstract_id=893822. Pacey, Arnold (1991): Technology in World Civilization, Cambridge, MA: MIT Press. Parker, Geoffrey (2007): “The Limits to Revolutions in Military Affairs: Maurice of Nassau, the Battle of Nieuwpoort (1600), and the Legacy”. The Journal of Military History 2:71, p. 331-72. Pearce, Celia (2002): “Towards a Theory of Game”. http://www.cp andfriends.com/writing/first-person.html. Pias, Claus (2001): „Synthetic History“. Archiv für Mediengeschichte (Themenheft: Mediale Historiographien) 1, http://www.uni-esse n.de/~bj0063/texte/history.pdf. Pias, Claus (2001/02): „Welt im Raster. Historische Szenen strategischer Interaktivität”. Ästhetik und Kommunikation 115, p. 39-50. Pias, Claus (2002): “Was Uns Entscheidet. Zur Techno-Logik der 'Besten Wahl'”. Neue Rundschau. Themenheft Entscheide Dich! 2, http://www.uni-essen.de/~bj0063/texte/rundschau.pdf. Pias, Claus (2002)(2): Computer Spiel Welten, Berlin: Diaphenes, http://e-pub.uni-weimar.de/volltexte/2004/37/.

403

Real Wars on Virtual Battlefields? Pias, Claus (2004): “Die Pflichten des Spielers. Der User als Gestalt der Anschlüsse.”. In: M. Warnke, G.C. Tholen, W. Coy (Hg.), Hyperkult II, Bielefeld: transcript. Raessens, J., Goldstein, J. (Hg.)(2005): Handbook of Computer Game Studies, Cambridge, MA: MIT Press. Rautzenberg, Markus (2002): Spiegelwelt. Elemente einer Aisthetik des Bildschirmspiels, Berlin: Logos. Regan, Patrick M. (1994): “War Toys, War Movies, and the Militarization of the United States, 1900-85”. Journal of Peace Research, 31:1, p. 45-58. Reichmuth, P./Werning, S. (2006): “Pixel Pashas, Digital Djinns” ISIM Review 18, p. 46-47 http://www.isim.nl/files/Review_18/ Review_18-46.pdf. Rothfeder, Jeffrey (2004): “Terror Games”. Popular Science 264:3, http://www.popsci.com/gear-gadgets/article/2004-02/terrorgames. Roland, Alex (1992): “Theories and Models of Technological Change: Semantics and Substance”. Science, Technology & Human Values 17:1, p. 79-100. Rötzer, Florian (2005): “Aufmerksamkeitsterrorismus und die Medien”. In: Jörgen Klußmann (Hg.), Terrorismus und Medien. Eine komplexe Beziehung, Bonn: Evangelische Akademie im Rheinland, p. 21-40. Rumbke, Leif (2006): „Kinetische Semiotik im Klassischen Computerspiel“. http://www.rumbke.de/data/text/_x++%20% 20leif%20rumbke%202006.pdf. Salen, K., Zimmerman, E. (2003): Rules of Play, Cambridge, MA: MIT Press. Sánchez-Crespo Dalmau, D. (2003): Core Techniques and Algorithms in Game Programming, Indianapolis, IN: New Riders Publishing. Sandkühler, G. (2004): “Der Historiker und Silent Hill. Prospektives Quellenstudium”. In: B. Neitzel, M. Bopp, R. Nohr (Hg.), See? I’m Real… Multidisziplinäre Zugänge zum Computerspiel am Beispiel von Silent Hill, Münster: Lit Verlag, p. 213-227. Schlosser, Ann (2006): “Learning Through Virtual Product Experience. The Role of Imagery on True and False Memories”. Journal of Consumer Research 33, p. 377-383. Schmid, Katja (2006): “Bilderkriege in Glaubenskämpfen”. Telepolis, http://www.heise.de/tp/r4/artikel/21/21940/1.html. Schofield, Bill (2007): “Embracing Fun: Why Extreme Programming is Great for Game Development”. Gamasutra 1 Mar., http://ww w.gamasutra.com/features/20070301/schofield_01.shtml.

404

Bibliography Scott, Karen D., White, Anne M. (2003): “Unnatural History? Deconstructing the Walking With Dinosaurs Phenomenon”. Media, Cultura and Society 25, p. 315-332. Scriba, Christoph J. (1998): "Zwei Beispiele für die Anwendung der Algebra auf technische Fragestellungen im 17. Jhdt.: Überlegungen des Oxforder Mathematikers John Wallis". In: Günter Bayerl/Wolfhard Weber (Hg.), Sozialgeschichte der Technik. Ulrich Troitzsch zum 60. Geburtstag, Münster: Waxmann, p. 95-111. Sims, Karl (1991): “Artificial Evolution for Computer Graphics”. Computer Graphics 25:4, p. 319-328. Singer, Peter W. (2007): “Outsourcing War”. Brookings Institute 10 Nov., http://www.brookings.edu/articles/2005/0301usdepartm entofdefense_singer.aspx. Snow, C.P. (1998): The Two Cultures, Cambridge: Cambridge UP. Sontag, Susan (2002): “Real Battles and Empty Metaphors”. The New York Times 10 Sep., http://select.nytimes.com/gst/abstra ct.html?res=F20B1FFF3D550C738DDDA00894DA404482 Spinuzzi, Clay (2002): “Towards a Hermeneutic Understanding of Programming Languages”. Currents in Electronic Literacy. http://www.cwrl.utexas.edu/currents/spring02/spinuzzi.html. Spreen, Dierk (2001/02): “Schachmatt? Nach Kasparows Niederlage”. Ästhetik & Kommunikation 115, p. 35-37. St. Clair, Robert (2000): “Visual Metaphor, Cultural Knowledge, and the New Rhetoric“ In: Jon Reyhner et al. (Hg.), Learn in Beauty: Indigenous Education for a New Century, Flagstaff, AZ: Northern Arizona U, p. 85-101. St. Clair, William (2004): The Reading Nation in the Romantic Period, Cambridge: Cambridge UP. Stahl, Roger (2006): “Have You Played the War on Terror?”. Critical Studies in Media Communication 23:2, p. 112-130. Stockwell, Stephen, Muir, Adam (2003): “The Military-Entertainment Complex: A New Facet of Information Warfare”. fibreculture 1. http://journal.fibreculture.org/issue1/issue1_stockw ellmuir.html. Straatman, Remco et al. (2005): “Killzone’s AI: Dynamic Procedural Combat Tactics” (conference presentation). GDC, http://www.c gf-ai.com/docs/straatman_remco_killzone_ai.pdf. Streibl, Ralf E. (1997): “Was Ihr Wollt: PowerScene für Krieg und Frieden”. FIFF-Kommunikation 14:1, p. 4-7. Streibl, Ralf E. (1997)(2): “GAME OVER: Die Rüstungsspirale auf Diskette und CD-ROM”. 10th symposium Psychology of Peace, Konstanz, http://fiff.informatik.uni-bremen.de/ruin/kicffp. htm.

405

Real Wars on Virtual Battlefields? Suid, Lawrence H. (2002): Guts and Glory: The Making of the American Military Image in Film, Lexington, KY: UP of Kentucky. Taylor, T.L. (2006): Play between Worlds. Exploring Online Game Culture, Cambridge, MA: MIT Press. Teng, Oyang (2007): “Video Games and the Wars of the Future”. Excutive Intelligence Review 10 Aug., http://www.larouchepub .com/lym/2007/3431videowars_future.html. Thompson, Clive (2002): “Online Video Games are the Newest Form of Social Comment”. Slate 20 Aug., http://www.slate.com/id/2 070197/ Thompson, Clive (2004): “Turn Left at the Zombie”. Slate 10 Sep., http://www.slate.com/id/2106514/. Thompson, Clive (2004)(2): “Game Theories”. The Walrus 18 May, article not archived yet, online copy available on the author’s weblog at: http://www.collisiondetection.net/mt/archives/2004 /05/game_theories_m.html. Thompson, Clive (2007): “Cynical, Subversive BlackSite Bashes Bush's War Machine“. Wired 19 Nov., http://www.wirHg.com/ print/gaming/gamingreviews/commentary/games/2007/11/ga mesfrontiers_1119 Treuille, Adrien et al. (2006): “Continuum Crowds”. ACM Transactions on Graphics 25:3, http://grail.cs.washington.edu /projects/continuum-crowds/. Turkle, Sherry (1995): Life on Screen, New York, NY: Simon & Schuster. Turkle, Sherry (2005): The Second Self. Computers and the Human Spirit, Cambridge, MA: MIT Press. Turkle, Sherry (2005)(2): “Computer Games as Evocative Objects: From Projective Screens to Relational Artifacts”. In: J. Raessens, J. Goldstein (Hg.), Handbook of Computer Game Studies, Cambridge, MA: MIT Press, p. 267-279. Uricchio, William (2005): „Simulation, History, and Computer Games“. In: J. Raessens, J. Goldstein (Hg.), Handbook of Computer Game Studies, Cambridge, MA: MIT Press, p. 327341. Vees-Gulani, Susanne (2003): Trauma and Guilt. Literature of Wartime Bombing in Germany, Berlin: Walter de Gruyter. Virilio, Paul (1989): War & Cinema. The Logistics of Perception, London: Verso. Von Keitz, Ursula/Hoffmann, Kay (Hg.) (2001): Die Einübung des dokumentarischen Blicks. Fiction Film und Non Fiction Film zwischen Wahrheitsanspruch und expressiver Sachlichkeit 1895-1945, Marburg: Schüren.

406

Bibliography Ward, Paul (2002): “Video Games as Remediated Animation” In: Geoff King, Tanja Krzywinska (Hg.), Screenplay. cinema/videogames/interfaces, London: Wallflower Press, p. 122-135. Weber, Samuel (2005): Targets of Opportunity: On the Militarization of Thinking, New York, NY: Fordham UP. Wenner, Lawrence A. (1989): Media, Sports and Society, Thousand Oaks, CA: SAGE Publications. Werber, Niels (2005): “Medien des Kriegs. Zur Semantik des Weltverkehrs.”. In: H.-P. Preusser (Hg.), Krieg in den Medien, Amsterdam, New York: Rodopi: p. 105-126. Werning, Stefan (2007): “Some Ontological Considerations on the Destruction of Cities in Interactive Media and its Discursive Implications” In: Andreas Böhn, Christine Mielke (Hg.), Die Zerstörte Stadt. Mediale Repräsentationen urbaner Räume von Troja bis SimCity, Bielefeld: transcript, p. 311-339. Werning, Stefan (2007)(2): „America’s Army“. In: Friedrich v. Borries et al. (Hg.), Space Time Play, Heidelberg: Birkhäuser, p. 436438. Whannel, Garry (1986): “The Unholy Alliance: Notes on Television and the Remaking of British Sport 1965-1985”. Leisure Studies, 5:2, p. 129-145. Winkler, Hartmut (2004): Diskursökonomie. Versuch Über die Innere Ökonomie der Medien, Frankfurt/Main: Suhrkamp. Wolf, Mark J.P. (2003): The Video Game Theory Reader, Oxford: Routledge. Zagal, José P. et al. (2005): “Towards an Ontological Language for Game Analysis”. Proceedings of DiGRA 2005 Conference, http://www.lcc.gatech.edu/~mateas/publications/OntologyDIG RA2005.pdf. Zakin, Mitchell (2007): “The Next Revolution in Materials”. DARPATech (DARPA 25th Systems and Technology Symposium) 7 Aug., http://www.darpa.mil/dso/newsevents/Zakin_The_Next _Revolution_in_Materials_FINAL.pdf

Designer Diaries DD-NAM67-1 Mathijs de Jonge, Shellshock: NAM’67, 25 May, 2004 http://www.gamespot.com/xbox/action/shellshocknam67/new s.html?sid=6099225 DD-NAM67-2 Mathijs de Jonge, Shellshock: NAM’67, 15 Jun., 2004 http://www.gamespot.com/xbox/action/shellshocknam67/new s.html?sid=6100720

407

Real Wars on Virtual Battlefields? DD-NAM67-3 Doug Walker, Shellshock: NAM’67, 10 Sep., 2004 http://www.gamespot.com/ps2/action/shellshocknam67/news .html?sid=6107181 DD-CNCG-1 Amer Ajami, Command & Conquer: Generals, 11 Mar., 2002 http://www.gamespot.com/pc/strategy/commandconquergener als/news.html?sid=2854173 DD-CGT-1 Jim Bambra, Conflict: Global Terror #1, 3 Aug., 2005 http://www.gamespot.com/pc/action/conflictglobalterror/news .html?sid=6130252 DD-MOHEA-1 Chris Cross/Dale Dye, Medal of Honor: European Assault, 7 Apr., 2005 http://www.gamespot.com/xbox/action/medalofhonor4/news.html ?sid=6121863 DD-SSI-1 Retro Gamer 43: 82-87, Developer Lookback: Strategic Simulations Inc. (part 2 of 2), 2007 DD-LOTR-1 Digital Actors in Rings Can Think” Wired 13 Dec., 2002 http://www.wirHg.com/entertainment/music/news/2002/12/ 56778

News Items NI-1 Electronic Arts, “Original Willis Jeep in Medal of Honor: Airborne”. 23 May, 2006 http://www.electronicarts.de/publish/page204348234638815.php3?messageid=367 NI-2 Slashdot, “Another Robotic Vehicle to Help Soldiers”. 5 June, 2005 http://hardware.slashdot.org/article.pl?sid=06/06/05/01 24210&from=rss NI-3 CVG, “Interview: David Braben”. 11 Jan., 2006 http://www.co mputerandvideogames.com/article.php?id=132379&skip=yes NI-4 The Daily Star, “Liberation War-based 3D video game launched”. 23 Mar., 2004 http://www.thedailystar.net/2004/03 /23/d40323011919.htm NI-5 Stars & Stripes, “DARWARS: Helping the Force Fight Insurgents”. 23 Jan., 2006 http://www.estripes.com/article.asp ?section=104&article=33630&archive=true NI-6 BBC News, “US Military Creates a Second Earth“. 23 Feb., 2004 http://news.bbc.co.uk/2/hi/technology/3507531.stm NI-7 Oman3D, “Interview with Afkar Media“. May, 2005 http://www .oman3d.com/features/interview_afkar/ NI-9 Serious Games Source, “Army Game Project's Frank Blackwell on America's Army” 5 Sep., 2006 http://seriousgamessource.c om/features/feature_090106_aa.php

408

Bibliography NI-10 Heise Online, “Das Unsichere Internet als Folge Miserabler Militärtechnik” 30 Sep., 2006 http://www.heise.de/newsticker/ meldung/78891 NI-11 Washington Post, “Way Radical, Dude” 9 Oct., 2006 http://www.washingtonpost.com/wp-dyn/content/article/2006 /10/08/AR2006100800931_pf.html. NI-12 Telepolis, “Kampfroboter zum Schutz von Grenzen, Flughäfen oder Pipelines“. 15 Nov., 2006 http://www.heise.de/tp/r4/artik el/23/23972/1.html. NI-13 BBC News, “Pentagon boosts 'Media War' Unit” 31 Oct., 2006 http://news.bbc.co.uk/2/hi/americas/6100906.stm NI-14 Telepolis, “Bildbereiningung durch Google Earth“. 21 Jan., 2007 http://www.heise.de/bin/tp/issue/r4/dl-artikel2.cgi?arti kelnr=24483 NI-15 Slate, “The Lameness of World of Warcraft” 14 Nov., 2006 http://www.slate.com/toolbar.aspx?action=print&id=2153757 NI-16 Tagesspiegel, “Sprechtaste am Gewehr” 6 Feb., 2006 http:// archiv.tagesspiegel.de/archiv/06.02.2007/3051577.asp NI-17 Telepolis, “Polizei 2.0“. 16 Feb., 2007 http://www.heise.de/tp /r4/artikel/24/24664/1.html. NI-18 Serious Games Source, “Games For Health 2006: Addressing PTSD, Psychotherapy & Stroke Rehabilitation with Games & Game Technologies ” 2006 http://seriousgamessource.com/feat ures/feature_052306.php NI-19 IGN, “America’s Army invades XBOX 360“. 11 Jun., 2007 http://xbox360.ign.com/articles/795/795776p1.html. NI-20 gamepolitics, (LiVEJOURNAL blog) “Video Game Company Rattles Sabres with Iran” 12 Jun., 2007 http://gamepolitics.live journal.com/305166.html. NI-21 eMediaWire, “Before Oscar Night: Magic Picture Frame Author Michael Class Picks Best Movies of 2006” 17 Feb., 2007 http:// www.emediawire.com/releases/2007/2/emw505647.htm NI-22 Telepolis, “’In Europa entsteht ein sicherheitsindustrieller Komplex!’” 8 Oct., 2007 http://www.heise.de/tp/r4/artikel/26 /26361/1.html NI-23 IBM, „Mercury Computer Systems helps IBM bring gamechanging technology to the market fast“. 25 Jul., 2006 ftp://ftp.software.ibm.com/software/solutions/pdfs/ODB0174-00B.pdf NI-24 Gamasutra, „Nintendo Rethinks the Development Pipeline“. 15 Mar., 2007 http://gamasutra.com/features/20070315/duffy _pfv.htm NI-25 iTnews, „CSIRO electrical shirt to give soldiers a buzz on the battlefield“. 1 Nov., 2007 http://www.itnews.com.au/News/641

409

Real Wars on Virtual Battlefields? 25,csiro-electrical-shirt-to-give-soldiers-a-buzz-on-the-battlefield .aspx NI-26 National Defense Magazine, “Army licenses Game to Entertainment Company“. Dec., 2004 http://www.nationaldefe nsemagazine.org/issues/2004/Dec/ArmyLicenses.htm NI-27 CNN Money, “Microsoft ESP Debuts as a Platform for Visual Simulation” 14 Nov., 2007 http://money.cnn.com/news/newsfe eds/articles/prnewswire/AQW08614112007-1.htm NI-28 Firing Squad, “World War II Online Reloaded” 14 Jul., 2004 http://www.firingsquad.com/print_article.asp?current_section= Games&fs_article_id=1513 NI-29 TAZ, “Die Farbe des Krieges“. 26 Mar., 2003 http://www.taz.de/pt/2003/03/26/a0189.1/text NI-30 Wired “Army Sets Up New Office of Videogames” 12 Dec., 2007 http://games.slashdot.org/article.pl?sid=07/12/12/2314 204&from=rss NI-31 New York Times “More Than Just a Game, but How Close to Reality?” 3 Apr., 2003 http://tech2.nytimes.com/mem/technolo gy/techreview.html?res=9B0CE3D81F39F930A35757C0A9659C 8B63 NI-32 Eurogamer “Saudi refuses to play Israeli“. 4 Jul., 2006 http://www.eurogamer.net/article.php?article_id=65870 NI-33 Telepolis “Science-Fiction für das US-Heimatschutzministerium” 31 May, 2007 http://www.heise.de/tp/r4/artikel/ 25/25404/1.html. NI-34 Tagesspiegel "Sprechtaste am Gewehr“. 7 Feb., 2007 http:// archiv.tagesspiegel.de/archiv/06.02.2007/3051577.asp NI-35 Government Computer News 9 Nov., 1998 http://www.gcn.com/print/17_30/33914-1.html?topic=news NI-36 Heise Online 28 Oct., 2006 http://www.heise.de/newsticker/ meldung/80173/from/rss09 NI-37 1115.org “War as Soap Opera, and Other Rumsfeldian Revelations“. 13 Dec., 2006 http://www.1115.org/2006/12/13 /war-as-soap-opera-and-other-rumsfeldian-revelations/ NI-38 heise online “Electronic Arts startet mit Ingame-Werbung” 31 Aug., 2006 http://www.heise.de/newsticker/meldung/77528 NI-39 develop “Code Warriors” 79, 2007/08: 39-42 NI-40 Associated Press “5 Nation War Games in Bay of Bengal” 6 Sept., 2007 http://usatoday.com/news/world/2007-09-06-162 9053733_x.htm NI-41 Village Voice “War Games“. 19 Nov., 2002: 40

410

Bibliography

Military Publications MP-ND-1 National Defense, „Army Strives for Training that Resembles Combat“ May 2006 http://www.nationaldefensemag azine.org/issues/2006/may/ArmyStrives.htm MP-ND-2 National Defense, „Adapting Video Games for Military Applications“ Feb. 2001 http://www.nationaldefensemagazine. org/issues/2001/Feb/Adapting_Video.htm MP-DTI-1 „Unconventional Warfare“ Defense Technology International Nov./Dec. 2006: 26-28 http://www.nxtbook.com/nxt books/mh/dti1106/ MP-DOD-1 „Blogs Study May Provide Credible Information“ Transformation (Department of Defense) 29 Jun., 2006 http://w ww.defenselink.mil/transformation/articles/200606/ta062906b.html. MP-DOD-2 “U.S. Army's 4th Infantry Division Goes 'Modular'” Transformation (Department of Defense) 16 Dec., 2004 http://www.defenselink.mil/home/features/4th_infantry/1215 2004.html. MP-MIT-1 “Reining in the Blogs of War” Military Information Technology 9:9, 2005 http://www.military-information-technology.com/article.cfm?DocID=1242 MP-AT-1 “Recruiters scour online gaming market” 12 Apr., 2007 http://www.armytimes.com/news/2007/04/gns_army_gamers_ 070412/ MP-DTIC-1 “Network-Centric Operations Case Study: The Stryker Brigade Combat Team” Defense Technical Information Center (DTIC) 2005 http://stinet.dtic.mil/oai/oai?verb=getRecord& metadataPrefix=html&identifier=ADA439066 MP-DTIC-2 “Virtual Environment Technology for MOUT Training” Jul., 1997 http://handle.dtic.mil/100.2/ADA328001

Forum Transcripts FT-AA-1 America’s Army, “Spawning with Random Times; [eliminate] Spam?” 26 Jan., 2006 http://forum.americasarmy.c om/viewtopic.php?p=2135674 FT-AA-2 America’s Army, “Who wants New Guns?” 21 Nov., 2005 http://forum.americasarmy.com/viewtopic.php?t=196141 FT-AA-3 goarmy.de, “Spam” Feb. 1, 2005 http://americas-army.4players.de/thread.php?goto=nextoldest &threadid=551630 FT-AA-4 America's Army, “Gloves in the Game” 18 Oct., 2006 http://forum.americasarmy.com/viewtopic.php?t=247095

411

Real Wars on Virtual Battlefields? FT-AA-5 America's Army, „How Would the Real Army Handle the Bridge Crossing Map?“ 2 Oct., 2006 http://forum.americasarmy .com/viewtopic.php?t=246601 FT-AA-6 America's Army, „Special Forces?“ 23 Aug., 2006 http://forum.americasarmy.com/viewtopic.php?t=247231 FT-AA-7 America’s Army, “Going to Iraq” 13 Mar., 2006 http://forums.ubi.com/groupee/forums/a/tpc/f/5311051192/ m/9381055224 FT-KW-1 Kuma/War, “Kuma Radio Feature” Aug. 8, 2005 http://www.kumawar.com/ubbthreads/showflat.php?Cat=&Nu mber=12180&page=0 FT-D88-1 DNA88.com, “How to Build a Game Engine From Scratch” 16 Jul., 2004 http://www.dna88.com/forum/forum-article272. html.

412

Kultur- und Medientheorie Erika Fischer-Lichte, Kristiane Hasselmann, Alma-Elisa Kittner (Hg.) Kampf der Künste! Kultur im Zeichen von Medienkonkurrenz und Eventstrategien Dezember 2009, ca. 300 Seiten, kart., zahlr. Abb., ca. 28,80 €, ISBN 978-3-89942-873-5

Jürgen Hasse Unbedachtes Wohnen Lebensformen an verdeckten Rändern der Gesellschaft Juni 2009, 254 Seiten, kart., zahlr. Abb., 24,80 €, ISBN 978-3-8376-1005-5

Thomas Hecken Pop Geschichte eines Konzepts 1955-2009 September 2009, 568 Seiten, kart., 35,80 €, ISBN 978-3-89942-982-4

Leseproben, weitere Informationen und Bestellmöglichkeiten finden Sie unter www.transcript-verlag.de

2009-09-21 11-10-07 --- Projekt: transcript.anzeigen / Dokument: FAX ID 02b6221424961686|(S.

1-

3) ANZ1240.p 221424961694

Kultur- und Medientheorie Christian Kassung (Hg.) Die Unordnung der Dinge Eine Wissens- und Mediengeschichte des Unfalls Juni 2009, 476 Seiten, kart., zahlr. z.T. farb. Abb., 29,80 €, ISBN 978-3-89942-721-9

Susanne Regener Visuelle Gewalt Menschenbilder aus der Psychiatrie des 20. Jahrhunderts Dezember 2009, ca. 220 Seiten, kart., zahlr. Abb., ca. 25,80 €, ISBN 978-3-89942-420-1

Karlheinz Wöhler, Andreas Pott, Vera Denzer (Hg.) Tourismusräume Zur soziokulturellen Konstruktion eines globalen Phänomens Februar 2010, ca. 330 Seiten, kart., ca. 29,80 €, ISBN 978-3-8376-1194-6

Leseproben, weitere Informationen und Bestellmöglichkeiten finden Sie unter www.transcript-verlag.de

2009-09-21 11-10-07 --- Projekt: transcript.anzeigen / Dokument: FAX ID 02b6221424961686|(S.

1-

3) ANZ1240.p 221424961694

Kultur- und Medientheorie Cristian Alvarado Leyton, Philipp Erchinger (Hg.) Identität und Unterschied Zur Theorie von Kultur, Differenz und Transdifferenz

Barbara Gronau, Alice Lagaay (Hg.) Ökonomien der Zurückhaltung Kulturelles Handeln zwischen Askese und Restriktion

Dezember 2009, ca. 450 Seiten, kart., ca. 32,80 €, ISBN 978-3-8376-1182-3

Dezember 2009, ca. 350 Seiten, kart., zahlr. Abb., ca. 32,80 €, ISBN 978-3-8376-1260-8

Moritz Csáky, Christoph Leitgeb (Hg.) Kommunikation – Gedächtnis – Raum Kulturwissenschaften nach dem »Spatial Turn«

Insa Härtel Symbolische Ordnungen umschreiben Autorität, Autorschaft und Handlungsmacht

Februar 2009, 176 Seiten, kart., 18,80 €, ISBN 978-3-8376-1120-5

Stephan Ditschke, Katerina Kroucheva, Daniel Stein (Hg.) Comics Zur Geschichte und Theorie eines populärkulturellen Mediums August 2009, 366 Seiten, kart., zahlr. Abb., 29,80 €, ISBN 978-3-8376-1119-9

Gunther Gebhard, Oliver Geisler, Steffen Schröter (Hg.) Von Monstern und Menschen Begegnungen der anderen Art in kulturwissenschaftlicher Perspektive November 2009, ca. 250 Seiten, kart., ca. 26,80 €, ISBN 978-3-8376-1235-6

April 2009, 326 Seiten, kart., 32,80 €, ISBN 978-3-8376-1042-0

Rebekka Ladewig, Annette Vowinckel (Hg.) Am Ball der Zeit Fußball als Ereignis und Faszinosum Juli 2009, 190 Seiten, kart., 20,80 €, ISBN 978-3-8376-1280-6

Christoph Neubert, Gabriele Schabacher (Hg.) Verkehrsgeschichte und Kulturwissenschaft Analysen an der Schnittstelle von Technik, Kultur und Medien November 2009, ca. 250 Seiten, kart., ca. 26,80 €, ISBN 978-3-8376-1092-5

Sacha Szabo (Hg.) Kultur des Vergnügens Kirmes und Freizeitparks – Schausteller und Fahrgeschäfte. Facetten nicht-alltäglicher Orte Oktober 2009, 315 Seiten, kart., zahlr. Abb., 29,80 €, ISBN 978-3-8376-1070-3

Leseproben, weitere Informationen und Bestellmöglichkeiten finden Sie unter www.transcript-verlag.de

2009-09-21 11-10-07 --- Projekt: transcript.anzeigen / Dokument: FAX ID 02b6221424961686|(S.

1-

3) ANZ1240.p 221424961694

2008-05-27 12-26-20 --- Projekt: transcript.titeleien / Dokument: FAX ID 02a8179786122216|(S.

2

) T00_02 seite 2 - 746.p 179786122240