Content uploaded by Rain Liivoja
Author content
All content in this area was uploaded by Rain Liivoja on Oct 27, 2018
Content may be subject to copyright.
Routledge Handbook of the
Law of Armed Conflict
Edited by Rain Liivoja and Tim McCormack
First published 2016
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
711 Third Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2016 Selection and editorial matter, Rain Liivoja and Tim McCormack; individual
chapters, the contributors
The right of the editors to be identified as the authors of the editorial material, and of
the authors for their individual chapters, has been asserted in accordance with sections
77 and 78 of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or utilised 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 publishers.
Trademark notice: Product or corporate names may be trademarks or registered
trademarks, and are used only for identification and explanation without intent to
infringe.
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging in Publication Data
Names: Liivoja, Rain, editor. | McCormack, Timothy L. H., editor.
Title: Routledge handbook of the law of armed conflict / edited by Rain Liivoja and
Tim McCormack.
Other titles: Handbook of the law of armed conflict
Description: Milton Park, Abingdon, Oxon : Routledge, 2016. | Includes bibliographical
references and index.
Identifiers: LCCN 2015037705| ISBN 9780415640374 (hbk) | ISBN 9780203798362 (ebk)
Subjects: LCSH: War (International law)
Classification: LCC KZ6385 .R685 2016 | DDC 341.6–dc23
LC record available at http://lccn.loc.gov/2015037705
ISBN: 978-0-415-64037-4 (hbk)
ISBN: 978-0-203-79836-2 (ebk)
Typeset in Bembo
by Wearset Ltd, Boldon, Tyne and Wear
603
35
Emerging technologies of warfare
Rain Liivoja, Kobi Leins and Tim McCormack*
Advances in science and technology have had a major impact on the conduct of war throughout
history.1 On a popular view, the development of warfare has been punctuated by so- called
‘revolutions in military affairs’, periods of rapid change in military thinking and practice; scient-
ific and technological transformations have played a significant role in such change.2 At the time
of writing, we appear to be in the midst of one such revolution in military affairs, namely an
information revolution.3 This revolution has been facilitated by the development of digital com-
puters and their wide adoption in society at large and in military systems in particular.
Assuming that there is an ongoing revolution in military affairs, the question arises as to whether
it ought to be accompanied by a ‘revolution in military legal affairs’,4 or whether the existing law
of armed conflict (LOAC) is capable of accommodating recent technological developments. This
chapter does not purport to provide a comprehensive answer to this question, nor does it seek to
evaluate the lawfulness or otherwise of specific means or methods of warfare. Rather, it addresses
some of the key legal implications and regulatory challenges arising from the military applications
of certain types of technology.5 We begin with information technology and the notion of ‘cyber
warfare’ (Section 1). We then turn to robotics, specifically remotely controlled, automated and
autonomous military systems (Section 2). Finally we consider nanotechnology (Section 3). We
conclude the chapter with a brief reflection on the formal legal review of new means and methods
of warfare in the context of emerging technologies of warfare (Section 4).
* The research for this chapter was supported under Australian Research Council’s Discovery Projects
funding scheme (project number DP130100432) and by a Society in Science – Branco Weiss Fellowship
(administered by ETH Zurich). We are grateful to Chris Jenks, Natalia Jevglevskaja, Robert J Mathews,
Tim McFarland and Angus Willoughby for comments on an earlier draft. The responsibility for the
present text, however, remains ours.
1 See generally Martin van Creveld, Technology and War: From 2000 BC to the Present (rev edn, Free Press
1991); Max Boot, War Made New: Technology, Warfare, and the Course of History, 1500 to Today (Gotham
Books 2006).
2 See the seminal work by Michael G Vickers and Robert C Martinage, ‘The Revolution in War’ (Center
for Strategic and Budgetary Assessments, December 2004).
3 Boot (n 1) 305 et seq.
4 This term has been used, although not specifically in relation to technology, by Charles J Dunlap Jr, ‘The
Revolution in Military Legal Affairs: Air Force Legal Professionals in 21st Century Conflicts’ (2001) 51
Air Force Law Review 293.
5 For other broad discussions, see eg Special Issue: New Technologies and Warfare (2012) 94(886) IRRC 457;
Dan Saxon (ed), International Humanitarian Law and the Changing Technology of War (Martinus Nijhoff 2013);
Hitoshi Nasu and Robert McLaughlin (eds), New Technologies and the Law of Armed Conflict (Asser 2014).
R Liivoja et al
604
1 Information technology
1.1 Technology
The start of the ongoing information revolution in military affairs can be reasonably precisely
dated to the late 1930s and early 1940s. This was the time when the first digital computers were
built, many of which were preoccupied by military tasks such as ballistics calculations and code
breaking.6 Of course, computers have developed quite spectacularly from what in hindsight
appears to be a rather humble beginning as advanced calculators, into systems that process and
store exponentially increasing amounts of data, which is often strategically significant and/or
economically valuable.
Also, computers are widely used to exert control over the physical world. Computerised
industrial control systems (ICS) are common in many different sectors of industry and infra-
structure.7 A comparatively simple programmable logic controller (PLC) can run a specific
device – such as an elevator, a set of traffic lights or an industrial appliance – by ‘monitor[ing]
the state of input devices and make[ing] decisions based upon a custom program to control
the state of output devices’.8 A distributed control system (DCS) is more complex, coord-
inating between multiple sub- systems involved in carrying out an entire industrial process, for
example, at an oil refinery, water and wastewater treatment plant, power plant or chemical
manufacturing plant.9 Another type of ICS, a supervisory control and data acquisition (SCADA)
system, is
designed to collect field information, transfer it to a central computer facility, and display
the information to the operator graphically or textually, thereby allowing the operator to
monitor or control an entire system from a central location in near real time.10
SCADA systems are widely used in distribution systems, such as water distribution and waste-
water collection systems, oil and natural gas pipelines, electrical grids, as well as rail, air traffic
control and other public transportation systems.11
Computers are routinely connected to each other to form networks. These networks can, in
turn, be connected to other networks, ultimately making up the Internet. This allows for the
remote access and manipulation of data, such as accessing a website stored on a server or saving
documents in the ‘cloud’. ICS are also often networked. Indeed, the architecture of DCS and
SCADA presumes a degree of interconnection between computers. Also, many ICS are con-
nected to the Internet to facilitate remote access. The same is true for various computerised
6 See Martin Campbell-Kelly, Willian Aspray, Nathan Ensmenger and Jeffrey R Yost, Computer: A
History of the Information Machine (3rd edn, Westview 2013).
7 Keith Stouffer, Suzanne Lightman, Victoria Pilletteri, Marshall Abrams and Adam Hahn, ‘NIST Special
Publication 800-82: Guide to Industrial Control Systems (ICS) Security’ Revision 2 – Final Public
Draft (National Institute of Standards and Technology, US Department of Commerce, February
2015).
8 Advance Micro Controllers Inc, ‘What is a Programmable Logic Controller (PLC)?’ (undated), www.
amci.com/tutorials/tutorials-what-is-programmable-logic-controller.asp.
9 Stouffer et al (n 7) 2–10.
10 Ibid 2–5.
11 Ibid.
Emerging technologies of warfare
605
devices, ranging from mobile phones to refrigerators – giving rise to the phenomenon referred
to as the ‘Internet of Things’.12
Computers and networks have vulnerabilities, that is to say flaws or weaknesses in their
design or operation, which can be exploited to violate information systems’ key security
attributes – confidentiality, integrity or availability.13 A breach of confidentiality entails access to
data by an unauthorised person, for instance for the purposes of cyber espionage. This was the
case with the alleged copying by Chinese hackers of sensitive design data of the US’s F- 35 Joint
Strike Fighter14 and the blueprints of the new headquarters of the Australian Secret Intelligence
Organisation.15 A breach of integrity involves the modification of data on the system or the
system’s configuration. This would include the defacement of websites, as well as the loading by
hackers of pro- Daesh videos and messages to the US military’s YouTube and Twitter accounts.16
A violation of availability denies access to data and of services of the system, to those authorised
to use them. An example would be the launch of such a number of queries to a system as to
overwhelm it and to shut it down or to render it inaccessible, referred to as a denial- of-service
attack. Prominent examples include the disabling of governmental websites of Estonia (in April
2007), Georgia (in July/August 2008) and Germany (in January 2015) by cyber activities ostens-
ibly emanating from Russia.17
Perhaps more seriously, given the wide use of computerised ICS, cyber operations altering
data can provoke real- world physical events. The most prominent example was the use of the
Stuxnet virus to infect the ICS of an Iranian uranium enrichment facility; the virus manipu-
lated the rotational speeds of centrifuges, causing them significant damage.18 It does not require
much imagination to envision the potentially catastrophic destruction that could be caused by
cyber operations against air traffic control systems, emergency services, dams or nuclear power
plants.
1.2 Legal challenges
Since the late 1990s, significant effort has gone into clarifying the international legal framework
applicable to hostile cyber operations. Numerous papers and books have appeared on the law of
12 See eg Dave Evans, ‘The Internet of Things: How the Next Evolution of the Internet is Changing
Everything’ White Paper (Cisco Internet Business Solutions Group, April 2011); Jacob Morgan, ‘A
Simple Explanation of “The Internet Of Things” ’ Forbes (13 May 2014), www.forbes.com/sites/
jacobmorgan/2014/05/13/simple-explanation-internet-things-that-anyone-can-understand.
13 See eg George Loukas, Cyber-Physical Attacks: A Growing Invisible Threat (Butterworth-Heinemann
2015) 2–3.
14 See Siobhan Gorman, August Cole and Yochi Dreazen, ‘Computer Spies Breach Fighter-Jet Project’
Wall Street Journal (21 April 2009).
15 See Ben Grubb, ‘Blueprints for New ASIO Headquarters “Stolen” ’ The Age (Melbourne, 27 May
2013).
16 See Mana Raibee, ‘US Central Command’s YouTube, Twitter Accounts Suspended after Hacking by
IS Supporters’ New York Times (12 January 2015), www.nytimes.com/video/multimedia/1000
00003445205.
17 See Michelle Martin and Erik Kirschbaum, ‘Pro-Russian Group Claims Cyber Attack on German
Government Websites’ Reuters (7 January 2015), www.reuters.com/article/2015/01/07/us-germany-
cyberattack-idUSKBN0KG15320150107.
18 See eg Holger Stark, ‘Mossad’s Miracle Weapon: Stuxnet Virus Opens New Era of Cyber War’ Der
Spiegel International (8 August 2011), www.spiegel.de/international/world/mossad-s-miracle-weapon-
stuxnet-virus-opens-new-era-of-cyber-war-a-778912.html.
R Liivoja et al
606
cyber warfare,19 with the most prominent scholarly publication being the 2013 Tallinn Manual
on the International Law Applicable to Cyber Warfare,20 a statement of 95 rules considered to be lex
lata by an international group of experts meeting at the invitation of the NATO Cooperative
Cyber Defence Centre of Excellence.21
There appears to be broad agreement on a few important issues. Most fundamentally, hostile
cyber operations undertaken by belligerents in the context of an ongoing armed conflict are,
like other military operations, governed by LOAC.22 In particular, where such cyber operations
amount to ‘attacks’ within the meaning of LOAC, they are subject to the key principles of dis-
tinction and proportionality, and necessitate the taking of certain precautionary measures.23
These general propositions, however well accepted, inexorably lead to further questions.
First, which cyber operations amount to ‘attacks’? Second, are cyber operations that do not
amount to attacks subject to restrictions of LOAC? Third, are cyber operations capable of trig-
gering the application of LOAC in the absence of ‘ordinary’ hostilities? While we readily
acknowledge that a number of further LOAC issues arise from cyber operations, we limit our
attention here to these three key questions as they are in some ways foundational to any discus-
sion about the law of cyber hostilities.
1.2.1 Cyber operations as ‘attacks’
Additional Protocol I defines ‘attacks’ as ‘acts of violence against the adversary, whether in
offence or defence’.24 According to the prevailing view, it is the anticipated violent effect or
consequence of an act that allow it to be characterised as an attack.25 Commentators point out
that the wilful releases of pathogens or toxic chemicals, entailing no overtly violent action,
undoubtedly constitute attacks for the purpose of LOAC because of their harmful or even lethal
consequences.26 Thus, the lack of violent action cannot exclude cyber operations from consti-
tuting attacks as long as the consequences are violent.
19 See eg Symposia in (2012) 17 JCSL 183, (2013) 84 International Law Studies 1; Heather Harrison
Dinniss, Cyber Warfare and the Laws of War (CUP 2012); Marco Roscini, Cyber Operations and the Use
of Force in International Law (OUP 2014); Jens David Ohlin, Claire Finkelstein and Kevin Govern (eds),
Cyber War: Law and Ethics for Virtual Conflicts (OUP 2015).
20 Tallinn Manual on the International Law Applicable to Cyber Warfare (CUP 2013), https://ccdcoe.org/
tallinn-manual.html.
21 For critical comments, see eg Rain Liivoja and Tim McCormack, ‘Law in the Virtual Battlespace: The
Tallinn Manual and the Jus in Bello’ (2012) 15 YBIHL 45; Dieter Fleck, ‘Searching for International
Rules Applicable to Cyber Warfare: A Critical First Assessment of the New Tallinn Manual’ (2013) 18
JCSL 331; Oliver Kessler and Wouter Werner, ‘Expertise, Uncertainty, and International Law: A
Study of the Tallinn Manual on Cyberwarfare’ (2013) 26 Leiden JIL 793.
22 See ICRC, ‘International Humanitarian Law and the Challenges of Contemporary Armed Conflict’
31st International Conference of the Red Cross and Red Crescent, 28 November to 1 December
2011, Doc 31IC/11/5.1.2 (October 2011) 38; Tallinn Manual (n 20) r 20. For further references, see
David Turns, ‘Cyber War and the Concept of “Attack” in International Humanitarian Law’ in Saxon
(ed) (n 5) 209, 221 (citing academic works); Cordula Droege, ‘Get Off my Cloud: Cyber Warfare,
International Humanitarian Law, and the Protection of Civilians’ (2012) 94(886) IRRC 533, 567
(citing views of states); Roscini, Cyber Operations (n 19) 21–22 (citing views of states).
23 See ICRC (n 22) 37; Tallinn Manual (n 20) rr 31, 51, 53–59.
24 API art 49(1).
25 Michael N Schmitt, ‘Cyber Operations and the Jus in Bello: Key Issues’ in Raul A Pedrozo and Daria
P Wollschlaeger (eds), International Law and the Changing Character of War (US Naval War College 2011)
89, 93–94; Roscini, Cyber Operations (n 19) 179; for a detailed discussion, see Turns (n 22) 221–224.
26 See eg Schmitt, ‘Key Issues’ (n 25) 94; Bill Boothby, ‘Where do Cyber Hostilities Fit in the Inter-
national Law Maze?’ in Nasu and McLaughlin (eds) (n 5) 59, 60.
Emerging technologies of warfare
607
There is broad support for the proposition that injury or death to persons, or damage or
destruction to objects, amounts to violence, and that cyber operations that are reasonably
expected to, or which do, have such consequences amount to attacks.27 On one view, physical
damage of this nature is not merely a sufficient but a necessary condition for a cyber operation
to constitute an attack.28 This reading relies on the literal meaning of the word ‘violence’ and
draws support from the formulation of the principle of proportionality in Additional Protocol I,
which speaks of ‘loss of civilian life, injury to civilians, damage to civilian objects, or a combina-
tion thereof ’.29 According to an alternative view, the disabling of an object, even without
causing any physical damage or destruction, also constitutes an attack.30 The proponents of this
approach rely in particular on the definition of military objects in Additional Protocol I, which
implies that the capture or neutralisation of an object is a possible aim of an attack.31
One construction attempts to reconcile these two approaches. This moves the focus from
physical destruction to ‘the fact that it [the object] is no longer completely suitable for its
intended purpose’.32 It does so by interpreting the notion of ‘damage’ to mean ‘the loss of func-
tionality that permanently renders the object inoperable or that necessitates some form of
repair’.33 However, even amongst those who adopt this approach to ‘damage’, there is no unan-
imity as to the nature or extent of the repair necessary. Circumstances where ‘restoration of
functionality requires replacement of physical components’ seems to be more readily accepted
as damage.34 Going further, some would argue, for example, is that the reloading of ‘the operat-
ing system or any software essential to operation’ would qualify as damage; however, replacing
data ‘merely stored on the system’ would not.35
This discussion is closely associated with the problem as to whether data could qualify as an
‘object’. If that were the case, the destruction of data could be considered an attack in itself –
that is to say, even in the absence of physical damage or loss of functionality of a device. There
does not appear to be much support for the data- as-object position.36 This is unsurprising given
that the ICRC Commentary to the 1977 Additional Protocols indicates that the word ‘object’, as
used in Additional Protocol I, means ‘something that is visible and tangible’.37 But, as we have
argued elsewhere, this corporeal conception of objects under LOAC may be outdated.38 The
economic value of data can be enormous and its permanent destruction may have far more
serious consequences than the destruction of some tangible objects. As a result, it may prove
27 See Tallinn Manual (n 20) r 30.
28 Schmitt, ‘Key Issues’ (n 25) 94–95; Yoram Dinstein, ‘The Principle of Distinction and Cyber War in
International Armed Conflicts’ (2012) 17 JCSL 261, 264.
29 Michael N Schmitt, ‘Wired Warfare: Computer Network Attack and Jus in Bello’ (2002) 84(846)
IRRC 365, 376–377.
30 Knut Dörmann, ‘The Applicability of the Additional Protocols to Computer Network Attacks: An
ICRC Approach’ in Karin Byström (ed), International Expert Conference on Computer Network Attacks and
the Applicability of International Humanitarian Law, 17–19 November 2004, Stockholm, Sweden: Proceedings
of the Conference (Swedish National Defence College 2005) 139, 142–143; ICRC (n 22) 37.
31 API art 52(2).
32 Michael N Schmitt, ‘Rewired Warfare: Rethinking the Law of Cyber Attack’ (2014) 96(893) IRRC
189, 203.
33 Ibid.
34 Tallinn Manual (n 20) commentary to r 30, ¶ 10.
35 Schmitt, ‘Rewired Warfare’ (n 32) 15.
36 Tallinn Manual (n 20) commentary to r 30, ¶ 6. But, see Nils Melzer, ‘Cyberwarfare and International
Law’ UNIDIR Resources (2011) 31.
37 AP Commentary ¶ 2008.
38 See Liivoja and McCormack (n 21) 53–54.
R Liivoja et al
608
difficult in the long run to ‘maintain a normative distinction between harm caused to physical
objects and that caused to data’.39
1.2.2 Cyber operations not amounting to ‘attacks’
The vast majority of cyber operations would not meet the fairly restrictive definition of attacks
outlined previously. This raises the question whether such operations are subject to any restric-
tions under LOAC. Notably, Article 48 of Additional Protocol I requires parties to a conflict to
‘direct their operations only against military objectives’ and Article 57(1) stipulates that, ‘[i]n the
conduct of military operations, constant care shall be taken to spare the civilian population,
civilians and civilian objects’. There are, however, differences in opinion as to the significance
of the use of the ostensibly broader term ‘operations’ instead of ‘attack’ in these provisions.
Some commentators do not hold the choice of terminology to be determinative: they argue
that the prohibition contained in Article 48 ‘is not so much on targeting non- military objectives
as it is on attacking them, specifically through the use of violence’.40 This view draws contextual
support from the position of Article 48 within the Protocol: the provision introduces a series of
more specific articles, all dealing with attacks.41 Also, according to the ICRC Commentary on
Article 48, ‘the word “operations” should be understood in the context of the whole of the
Section [on methods and means of warfare]; it refers to military operations during which viol-
ence is used, and not to ideological, political or religious campaigns’.42 If this line of reasoning
is accepted, cyber operations not amounting to attacks are not restricted by the principle of
distinction and thus ‘are permissible against non- military objectives, such as the population’.43
Alternative views note that the drafters’ choice to use the term operations instead of attack
must have some significance. Those views rely on the ICRC Commentary on Article 48 which
accepts a dictionary definition whereby ‘ “military operations” refers to all movements and acts
related to hostilities that are undertaken by armed forces’.44 This has led to the argument that the
applicability of LOAC restraints on cyber operations depends on whether they constitute part
of ‘hostilities’.45 Accordingly, given that ‘hostilities’ is a broader concept than ‘attacks’, cyber
operations ‘directly adversely affecting military operations or military capacity’ of the adversary
must also be subject to restrictions imposed by LOAC:46
[C]yber operations aiming to disrupt or incapacitate an adversary’s computer- controlled
radar or weapon systems, logistic supply or communication networks may not directly
cause any physical damage, but would certainly qualify as part of the hostilities and, there-
fore, would have to comply with the rules and principles of [international humanitarian
law] governing the conduct of hostilities. The same would apply to cyber operations intrud-
ing into the adversary’s computer network to delete targeting data, manipulate military
orders or change, encrypt, exploit or render useless any other sensitive data with a direct
(adverse) impact on the belligerent party’s capacity to conduct hostilities.47
39 Schmitt, ‘Rewired Warfare’ (n 32) 16.
40 Schmitt, ‘Wired Warfare’ (n 29) 376 (original emphasis); see also Turns (n 22) 217; Roscini, Cyber
Operations (n 19) 178.
41 Schmitt, ‘Wired Warfare’ (n 29) 376.
42 AP Commentary ¶ 1875.
43 Schmitt, ‘Wired Warfare’ (n 29) 376.
44 AP Commentary ¶ 1875.
45 Melzer (n 36) 27.
46 Ibid 28.
47 Ibid 28 (footnotes omitted).
Emerging technologies of warfare
609
A variation of this approach relies on the commentary to Article 57, which explains that the
term ‘ “military operations” should be understood to mean any movements, manœuvres and
other activities whatsoever carried out by the armed forces with a view to combat’.48 Con-
sequently, in order to be subject to the principles of distinction, proportionality and precaution,
a cyber operation must be combat related, in other words it ‘must be associated with the use of
physical force, but it does not have to result in violent consequences itself ’.49
On either of the broader conceptions of ‘operations’, however, propaganda, espionage and
psychological operations – sometimes compared with sub- attack cyber operations – would not
be governed by the principles of distinction, proportionality and precaution because they are
neither part of hostilities nor undertaken with a view to combat.50
To conclude the discussion about the difference between attacks and operations, it is worth-
while to note that certain persons and objects enjoy special protection under LOAC. In par-
ticular, medical personnel, units and transports must be ‘respected and protected’.51 This
prohibition ‘includes, but is not limited to the prohibition of attacks’.52 Accordingly, irrespec-
tive of what the correct interpretation of the notions of attack and operation may be, any cyber
operation that impedes, prevents or otherwise adversely affects the carrying out of humanitarian
functions of medical services – for example, altering the GPS data of a medical aircraft or the
digital personal records of patients – is prohibited.53
1.2.3 Cyber operations as a trigger for armed conflict
The preceding discussion presupposed that an armed conflict was already in existence. The
question remains, however, whether cyber operations, in the absence of hostilities involving
conventional military means, could constitute an armed conflict and thus trigger the application
of LOAC.
Even though the Geneva Conventions and the Additional Protocols apply, according to
their own terms, in ‘armed conflict’, these treaties do not establish what constitutes such a con-
flict.54 The now- standard threshold test, reflecting the relevant treaty provisions and customary
law, was formulated by the ICTY Appeals Chamber in Tadic´ as follows: ‘An armed conflict
exists whenever there is a resort to armed force between States or protracted violence between
governmental authorities and organized armed groups or between such groups within a
State.’55
The application of this test to establish the existence of an armed conflict involves a signi-
ficant preliminary question of attribution. The usual rules of state responsibility apply when
determining whether a cyber operation is attributable to a state but demonstrating the requisite
link between the operation and a state or a non- state armed group can be difficult.56 It may well
be that cyber operations are undertaken by independent ‘hacktivists’ whose conduct cannot be
48 AP Commentary ¶ 2191.
49 Harrison Dinniss (n 19) 201.
50 See Droege (n 22) 556.
51 See in particular API arts 12(1), 15(1), 21, 23(1), 24; cf Tallinn Manual (n 20) rr 70–71.
52 See Tallinn Manual (n 20) commentary to r 70, ¶ 4.
53 See ibid commentary to r 70, ¶ 4; commentary to r 71, ¶ 3.
54 See GCI–IV common arts 2 and 3; API art 1; APII art 1.
55 Prosecutor v Tadic´ (Decision on the Defence Motion for Interlocutory Appeal on Jurisdiction) (Case no IT-94-1,
ICTY Appeals Chamber, 2 October 1995) ¶ 70.
56 See eg Marco Roscini, ‘Evidentiary Issues in International Disputes Related to State Responsibility for
Cyber Operations’ (2015) 50 Texas International Law Journal 233.
R Liivoja et al
610
attributed to any state or a non- state armed group, in which case a situation of armed conflict
would not arise, irrespective of the extent of damage caused.
If the attribution problem can be overcome, the question is whether the requirements of the
Tadic´ test can be met.
According to the prevailing view, no particular degree of violence or intention is required for
the existence of an international armed conflict – all that is necessary is ‘a resort to armed force
between States’.57 On this view, where a single cyber operation, launched by one state against
another, amounts to an ‘attack’ as discussed previously, an international armed conflict would be
triggered.58 Admittedly, the uncertainty as to what constitutes an attack in the cyber context would
obviously create some difficulties in this context. But were one to accept the competing view that
the applicability of LOAC requires a greater extent, duration or intensity of hostilities, even the
2010 Stuxnet operation, which succeeded in causing physical damage to centrifuges in a nuclear
fuel processing plant, might not be sufficient to trigger an international armed conflict.59
The threshold for a non- international armed conflict is higher: the armed force used must
have a certain degree of intensity – in the words of the ICTY, ‘protracted violence’ – and each
non- state armed group involved must have a certain level of organisation.60 To amount to pro-
tracted violence, ‘cyber attacks have to be frequent enough to be considered related, [but] they
clearly do not have to be continuous’.61 The requirement of organisation means, for example,
that a number of individuals sharing a common purpose, but engaging in cyber operations
against a state or a non- state armed group independently of each other, would not constitute an
organised armed group.62 However, a group that operates as a unit in a coordinated fashion – for
example, where members of the group have assigned roles and they act upon orders issued vir-
tually from a recognised leader – could be seen as organised, even though the members never
meet face to face and may never even discover each other’s true identity.63 An additional diffi-
culty is created, however, by Additional Protocol II: Article 1(1) of the Protocol presumes that
an organised armed group is ‘under responsible command’. This requirement has been inter-
preted to mean that the organised armed group ‘must be in a position to implement the Proto-
col’64 and that this entails ‘the possibility to impose discipline’.65 There is obvious doubt whether
a group organised virtually would have the tools to impose the kind of discipline that is typical
of military organisations. However, there is room for the argument that the ‘responsible
command’ requirement does not limit the applicability of Common Article 3,66 and that, at any
rate, ‘being in a position to implement’ the law merely requires ‘the organisational ability to
comply with the obligations of international humanitarian law’.67
57 See further Caitlin Dwyer and Tim McCormack, ‘Conflict characterisation’ ch 3 in this volume, s 2.2.
58 See eg Michael Schmitt, ‘Classification of Cyber Conflict’ (2012) 17 JCSL 245, 251–252; see also
Roscini, Cyber Operations (n 19) 134–135.
59 See Tallinn Manual (n 20) commentary to r 22, ¶¶ 12, 14.
60 See further Dwyer and McCormack (n 57) s 2.3.
61 Schmitt, ‘Classification’ (n 58) 258; see also Roscini, Cyber Operations (n 19) 152–153.
62 Schmitt, ‘Classification’ (n 58) 256; Tallinn Manual (n 20) commentary to r 23, ¶ 15.
63 See Schmitt, ‘Classification’ (n 58) 256; Tallinn Manual (n 20) commentary to r 23, ¶¶ 13, 15; for an
analysis of different scenarios pertaining to the level of organisation, see Roscini, Cyber Operations
(n 19) 155–157.
64 AP Commentary ¶ 4470.
65 Prosecutor v Bemba (Decision on the Confirmation of Charges) (Case no ICC-01/05-01/08, ICC Pre-Trial
Chamber III, 15 June 2009) ¶ 234.
66 Tallinn Manual (n 20) commentary to r 23, ¶ 14, n 202.
67 See Prosecutor v Boškoski and Tarc´ulovski.
(ICTY-04-82-T, ICTY Judgment, 10 July 2008) ¶ 205.
Emerging technologies of warfare
611
Whatever the correct interpretation, the intensity and organisation requirements imposed by
the LOAC of non- international armed conflicts are significant. Thus the plausible view has been
expressed that ‘cyber operations alone can trigger a non- international armed conflict in only rare
cases’.68
2 Robotics
2.1 Technology
The term ‘robotics’ refers to ‘the science and technology of robots’.69 There is a modicum of
agreement that a robot is ‘an artificial device that can sense its environment and purposefully act
on or in that environment’.70 In other words, a robot has sensors to monitor the environment,
processors to make decisions as to how to react to the environment (thus incorporating some
degree of programmability or artificial intelligence) and actuators to effect change in the environ-
ment.71 What remains controversial, however, is whether a device must have some meaningful
degree of autonomy in order to be deemed a robot.72 Autonomy, in this context, refers to the
capacity of the device to undertake some action without intervention from a human
operator.73
While it is important to appreciate that different devices have different levels of autonomy
with respect to each of their functions – being autonomous is a matter of degree74 – it does not
appear to be easy, or particularly helpful in this context, to try to establish a requisite minimal
degree of autonomy for something to be called a ‘robot’. The field of ‘robotics’ and the notion
of ‘robotic technology’ are broad enough to cover sophisticated electromechanical systems that
rely heavily on a sensing–processing–acting loop but have very limited autonomy. For example,
the area of telerobotics deals with devices where high- level cognitive decisions are made by a
human operator while the device is responsible for the mechanical implementation of these
decisions some distance away.75 There is, however, very little autonomy involved. For the pur-
poses of this discussion, we take the widest possible view of robotics: we believe that utilising
narrow definitions has the potential of ignoring how devices with a greater degree of autonomy
are essentially incremental developments of less- sophisticated devices.
Especially in view of a broad definition of the term, military applications of robotics are
numerous. One commonplace example is the use of fly- by-wire (FBW) technology on military
aircraft. With a FBW system, the connection between the cockpit controls (such as the yoke or
joystick, and rudder pedals) on the one hand, and the flight control surfaces and engines on the
68 See Tallinn Manual (n 20) commentary to r 23, ¶ 7.
69 Bruno Siciliano and Oussama Khatib, ‘Introduction’ in Bruno Siciliano and Oussama Khatib (eds),
Springer Handbook of Robotics (Springer 2008) 1, 1.
70 Alan Winfield, Robotics: A Very Short Introduction (OUP 2012) 8–9.
71 Cf P W Singer, Wired for War: The Robotics Revolution and Conflict in the Twenty-First Century (Penguin
2009) 67; Armin Krishnan, Killer Robots: Legality and Ethicality of Autonomous Weapons (Ashgate 2009) 9.
72 Cf US Department of Defense, Joint Robotics Program Master Plan FY2005 (2005) 1–2 (suggesting that a
robot ‘works automatically or operates by remote control’); Krishnan (n 71) 9 (‘A robot must exhibit
some degree of autonomy, even if it is only very limited autonomy.’).
73 See Winfield (n 70) 10.
74 See, notably, Thomas B Sheridan and William L Verplank, ‘Human and Computer Control of Under-
sea Teleoperators’ Technical Report (Massachusetts Institute of Technology, 14 July 1978).
75 See eg Günter Niemeyer, Carsten Preusche and Gerd Hirzinger, ‘Telerobotics’ in Bruno Siciliano and
Oussama Khatib (eds), Springer Handbook of Robotics (Springer 2008) 741, 741.
R Liivoja et al
612
other hand, is electronic rather than mechanical.76 Essentially, the same concept is used in many
remotely controlled weapon stations (RCWS) – systems controlled by operators who are not in
direct (physical) contact with the weapons. For example, a gun turret may be mounted on a
vehicle and be controlled by an operator inside. Such systems are widely used both on land, for
example on armoured fighting vehicles,77 as well as on naval vessels.78
The rapidly increasing ability to combine robotics with sophisticated sensing systems (radars,
electro- optical and infrared sensors) and a communication link has paved the way for advanced
remotely controlled military systems. Remotely piloted vehicles (RPVs) are now available in all
traditional domains of warfare – land, sea and air.79 Remotely piloted aircraft (RPA) – also called
unmanned aerial vehicles (UAVs) or, popularly, ‘drones’ – are used widely for intelligence,
surveillance and reconnaissance purposes. According to a 2013 report, ‘[b]oth military and civil-
ian UAVs are in use by almost every country, including nearly 60 with their own manufacturing
capability’.80 RPA carrying weapons, on the other hand, have remained the purview of a
considerably smaller number of states: a 2014 report identified 23 states as potentially developing
some type of armed RPA.81 Thus, much of the prominence of RPVs has to do with their exten-
sive use by the US: between 2002 and early 2015, it had carried out more than 500 strikes using
RPA, with the greatest number in Pakistan.82
Military systems are obtaining an increasing degree of autonomy in relation to some of their
functions.83 For example, some RPA currently in operation, such as the RQ- 4 Global Hawk,
have a considerable ability to navigate on a pre- defined flight plan without any human inter-
vention. The Northrop Grumman X- 47B, a strike fighter- sized RPV prototype, has demon-
strated the feasibility of unsupervised aerial refuelling.84 Certain weapon systems are also capable
of operation without continuous control by an operator. Notable examples include ‘close- in
weapon systems’ (CIWS) on naval vessels, and their land- based counterparts, C- RAMs (short
for ‘counter rocket, artillery and mortar’). In a nutshell, CIWS and C- RAM are rapid- fire,
computer- controlled, radar- guided guns designed to engage automatically and defeat incoming
missiles and other close- in threats.85
76 See eg Nick Lee-Frampton, ‘Anything but Simple: The F-16’s FBW Flight Control System’ (January/
February 2000) 54(9) New Zealand Engineering 13.
77 See eg Ezio Bonsignore and David Eshel, ‘Remotely Controlled Weapon Stations: Technologies and
Markets’ (2009) 33(7) Military Technology 52.
78 See eg Luca Peruzzi, ‘Remote Control Cannon Proliferation at Sea’ (2013) 37(4) Armada International 26.
79 For an overview of the technology, see eg US Department of Defense, ‘Unmanned Systems Integrated
Roadmap FY2013–2038’ Ref No 14-S-0553 (2013).
80 ‘UAV Roundup 2013’ (July–August 2013) Aerospace America 26, 26; see also Lynn E Davis, Michael J
McNerney, James S Chow, Thomas Hamilton, Sarah Harting and Daniel Byma, Armed and Dangerous?
UAVs and U.S. Security (RAND 2014).
81 Davis et al (n 80) 7.
82 The Bureau of Investigative Journalism, ‘Get the Data: Drone Wars’, www.thebureauinvestigates.
com/category/projects/drones/drones-graphs.
83 As the US Defense Science Board has noted,
[a]utonomy is better understood as a capability (or a set of capabilities) that enables the larger
human-machine system to accomplish a given mission, rather than as a ‘black box’ that can be dis-
cussed separately from the vehicle and the mission.
US Defense Science Board, ‘Task Force Report: The Role of Autonomy in DoD Systems’
(July 2012) 21, http://fas.org/irp/agency/dod/dsb/autonomy.pdf
84 See ‘Fueled in Flight: X-47B First to Complete Autonomous Aerial Refueling’ NAVAIR News (22
April 2015), www.navair.navy.mil/index.cfm?fuseaction=home.NAVAIRNewsStory&id=5880.
85 See eg Raytheon, ‘Phalanx Close-In Weapon System: Last Line of Defense for Air, Land and Sea’,
www.raytheon.com/capabilities/products/phalanx.
Emerging technologies of warfare
613
2.2 Legal challenges
Philip Alston, the UN Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions,
noted in a 2010 report that
[a]lthough robotic or unmanned weapons technology has developed at astonishing rates,
the public debate over the legal, ethical and moral issues arising from its use is at a very early
stage, and very little consideration has been given to the international legal framework
necessary for dealing with the resulting issues.86
The widely reported and often contentious use of RPA for ‘targeted killings’ has, however,
fuelled an intense debate about the legality and morality of remotely controlled weapons.87
The single most controversial international law issue in this regard has been the determina-
tion of the appropriate regulatory framework. The use of violence in the course of an armed
conflict is subject to the restrictions of LOAC. This framework is relatively permissive: in par-
ticular, it accepts the lethal targeting of persons on the basis of their status (combatants) or their
conduct (direct participation in hostilities), and tolerates civilian deaths insofar as they are incid-
ental to an attack against a legitimate military objective and proportionate to the military
advantage expected from the attack.88 Use of armed force outside an armed conflict, on the
other hand, is governed by the fairly restrictive human rights law. Under human rights stand-
ards, a state may use lethal force only where ‘it is required to protect life . . . and there is no other
means, such as capture or non- lethal incapacitation, of preventing that threat to life’.89
Given the disparity of these standards, the lawfulness of a particular RPV attack would
heavily depend on whether LOAC is applicable. The initial issue in this respect is the existence
of an armed conflict that would trigger the application of LOAC.90 But even when there is an
armed conflict, LOAC would apply to hostilities only in the context of that armed conflict.
Hence, to apply LOAC to a specific attack, there must be a connection between that attack and
the armed conflict. This issue has become particularly problematic in relation to US RPA strikes
against al- Qaeda in Pakistan: it is unclear whether these can be seen as occurring in the context
of an armed conflict, such as the non- international armed conflict against al- Qaeda in
Afghanistan.91
86 Philip Alston (Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions), Interim Report
to the General Assembly, UN Doc A/65/32 (23 August 2010) ¶ 29.
87 See eg P W Singer, Wired For War: The Robotics Revolution and Conflict in the 21st Century (Penguin
2009); Claire Finkelstein, Jens David Ohlin and Andrew Altman (eds), Targeted Killings: Law and
Morality in an Asymmetrical World (OUP 2012); Bradley Jay Strawser (ed), Killing by Remote Control: The
Ethics of an Unmanned Military (OUP 2013); Christian Enemark, Armed Drones and the Ethics of War:
Military Virtue in a Post-Heroic Age (Routledge 2014); John Kaag and Sarah Kreps, Drone Warfare (Polity
2014); Special Issue: Unmanned Vehicles, Legal, Social and Ethical Issues (2012) 21(2) JLIS; Special
Issue: Legal and Ethical Implications of Drone Warfare (2015) 19(2) International Journal of Human
Rights 105.
88 See generally Part II of this volume.
89 Philip Alston (Special Rapporteur on Extrajudicial, Summary or Arbitrary Executions), Report to the
Human Rights Council, Addendum: Study on Targeted Killings, UN Doc A/HRC/14/24/Add.6 (28
May 2010) ¶ 32.
90 See generally Dwyer and McCormack (n 57).
91 See eg Chris Jenks, ‘Law from Above: Unmanned Aerial Systems, Use of Force, and the Law of Armed
Conflict’ (2010) 85 North Dakota LR 649; Ian Henderson and Bryan Cavanagh, ‘Unmanned Aerial
Vehicles: Do They Pose Legal Challenges?’ in Nasu and McLaughlin (eds) (n 5) 193, 200–201.
R Liivoja et al
614
Where LOAC is applicable, it is doubtful whether remotely controlled weapon systems pose,
in and of themselves, any novel issues in terms of rules governing the conduct of hostilities.
Weapon- specific rules place no restriction on the development, acquisition or use of weapons
that are controlled remotely.92 Of course, no weapon can be placed lawfully on an RPV that
could not be placed lawfully on a manned vehicle. Thus, for example, a state party to the 2008
Convention on Cluster Munitions93 would be precluded from arming its RPVs with cluster
munitions.
Complying with the principles of discrimination and proportionality, and taking precaution-
ary measures, does not appear to be more difficult when launching an attack by means of a
weapon carried by an RPV as compared to a comparable manned vehicle. To the contrary, in
some circumstances RPVs may enhance compliance with LOAC. Removing the operator from
immediate danger and making use of the advanced sensing capabilities of RPVs may allow for
more thoroughly considered and accurate targeting decisions.94 Paradoxically, LOAC may thus
encourage the use of RPVs.95
An oft- expressed concern in relation to RPVs has to do with the status of their operators
under LOAC. While difficulties in classifying persons for the purposes of LOAC are hardly a
problem specific to this operational context,96 it is certainly true that civilian operators of
remotely controlled systems can easily become direct participants in hostilities, with all the
attendant consequences.97 This would be the case not only where a remote operator, say,
launches an attack from an RPV, but also where the operator of an unarmed RPV passes
information about the location of adversary military forces to ground troops for the purposes of
launching a land- based attack.
Additional legal and ethical issues arise from the increased autonomy of military systems,
which has already generated quite a voluminous body of literature.98 Broad questions as to the
safety of a system will arise from any kind of operation where a human is not in control, such as
the unsupervised flight of an RPA along a predetermined flight path – notwithstanding that
human- controlled operation might not, on the evidence, be any safer.99 But a weapon system
that can identify targets and launch attacks upon them without attack- specific human oversight
generates significant issues under LOAC.
International law does not specifically restrict the degree of autonomy that can be imple-
mented in weapon systems. Thus the key problem is whether the weapon system is capable of
92 See eg Meredith Hagger and Tim McCormack, ‘Regulating the Use of Unmanned Combat Vehicles:
Are General Principles of International Humanitarian Law Sufficient?’ (2011) 21 JLIS 74, 84–85.
93 (3 December 2008) 2688 UNTS 39.
94 Cf Jenks (n 91) 668–669; Henderson and Cavanagh (n 91) 205–206.
95 See Michael W Lewis and Emily Crawford, ‘Drones and Distinction: How IHL Encouraged the Rise
of Drones’ (2013) 44 Georgetown JIL 1127.
96 See eg Nelleke van Amstel and Rain Liivoja, ‘Private military and security companies’ ch 36 of this
volume, s 2 (in relation to employees of private companies).
97 See API arts 51(3); APII art 13(3); Michelle Lesh, ‘Direct participation in hostilities’ ch 10 in this
volume.
98 See eg Patrick Lin, George Bekey and Keith Abney, ‘Autonomous Military Robotics: Risk, Ethics,
and Design’ (Ethics + Emerging Sciences Group, California Polytechnic State University, 20 December
2008); Ronald C Arkin, Governing Lethal Behaviour in Autonomous Robots (CRC Press 2009); Krishnan
(n 71); Jai Galliott, Military Robots: Mapping the Moral Landscape (Ashgate 2015); ‘Autonomous Weapon
Systems: Technical, Military, Legal and Humanitarian Aspects’ Report of the ICRC Expert Meeting,
Geneva, 26–28 March 2014 (9 May 2014), www.icrc.org/eng/assets/files/2014/expert-meeting-
autonomous-weapons-icrc-report-2014-05-09.pdf.
99 See Chris Jenks, ‘False Rubicons, Human Placebos and Increasingly Autonomous Weapon Systems’
(forthcoming).
Emerging technologies of warfare
615
operating in compliance with LOAC. In particular: Can the system adequately distinguish
between combatants and civilians, and military objectives and civilian objects? Can the system
assess collateral damage? Can the system recognise surrender?
Arguably these are primarily technical challenges, potentially capable of a technical solution,
rather than legal problems requiring a regulatory remedy.100 In any event, one should not assume
that any autonomous systems would have to be given the same amount of discretion as a soldier,
only to find that the system is incapable of exercising that discretion in conformity with LOAC
(especially, assessing the maximum permissible collateral damage). The LOAC compliance of a
particular system may be ensured by programming it to identify a very narrow range of objects
that by their clearly identifiable nature or behaviour are military objectives (say, submarines or
incoming missiles) and to deploy them only where no civilians or civilian objects can be identi-
fied in the vicinity. This may well be easier to achieve in some operational settings, such as
remote deserts, open oceans and depopulated areas.101
Significant legal issues do arise in terms of accountability. In autonomous systems, certain
decisions normally made by humans are ‘delegated’ to the system such that the need for input
from a human operator is reduced and the operation of the system is determined by computer
software.102 Yet it would be fanciful to think that the system thereby becomes an accountable
agent and human beings are removed from the decision- making process; rather, the nature of
the decisions made by humans changes: such decisions will be more general, made ahead of
deployment and involve greater degree of forecasting.103 To put it differently, an autonomous
system will have been designed, built, programmed, approved and deployed by people, who
will ultimately remain responsible for the ‘conduct’ of the system. Admittedly, allocating that
responsibility between those human actors may be a difficult enterprise. Part of this difficulty
results from the assumption underlying LOAC and international criminal law that the operator
of a weapon system has ultimate control over it. With autonomous – or, for that matter, other
highly advanced – systems that would not necessarily be the case: the designer and programmer
could have more control over a system than the person switching it on. Current modes of indi-
vidual criminal responsibility might not be adequate to reflect this new reality.104
Finally, RPVs or autonomous systems can potentially affect other areas of LOAC, for
example, the law of occupation. Under LOAC, a territory is considered occupied when it falls
under the actual control of the adversary.105 Traditionally, this has meant ‘boots on the ground’
– the presence of military personnel exercising authority in the territory. Technology can
undoubtedly allow certain kinds of military operations to be carried out without such presence
of personnel. This raises the question whether the control potentially exercisable through a
combination of RPVs and autonomous systems can be such as to warrant a reinterpretation of
certain basic tenets of the law of occupation.
100 On the distinction, see Tim McFarland, ‘How Should Lawyers Think about Weapon Autonomy?’
(2016) IRRC (forthcoming).
101 See William H Boothby, Conflict Law: The Influence of New Weapons Technology, Human Rights and
Emerging Actors (Asser 2014) 111.
102 Cf ibid 105 (‘The significant element of autonomy is the ability of the system to decide a course of
action from a number of alternatives without depending on human oversight and control.’).
103 See McFarland (n 100).
104 See Tim McCormack and Tim McFarland, ‘Mind the Gap: Can Developers of Autonomous Weapons
Systems be Liable for War Crimes?’ (2014) 90 International Law Studies 361.
105 HR art 42.
R Liivoja et al
616
3 Nanotechnology
3.1 Technology
Nanotechnology – conceptualised in the late 1950s, and introduced to a wider audience in the
1970s and 1980s106 – refers to ‘technology at the nanoscale’.107 The broad term ‘technology’
includes applications from just about every field of science. Nanoscale, for its part, refers to the
size of these applications, and covers a range of one to 100 nanometres.108 (A nanometre is one-
billionth of a metre – a factor of 10−9. To put this in perspective, a nanometre is to a metre what
a marble is to the size of the Earth.)109 Thus, nanotechnology involves the deliberate manipula-
tion of material on the atomic or molecular level to produce novel materials, devices and
systems.110
Much of nanotechnology is concerned with nanomaterials, that is to say materials that have
one or more dimensions on the nanoscale.111 The most common type of these materials, a nano-
particle, has all three dimensions on the nanoscale.112 Nanoparticles occur naturally in the
environment, such as in volcanic ash, sea spray, clay and milk, and in some man- made sub-
stances, such as depleted uranium. In the field of nanotechnology, however, scientists and engi-
neers have an increasingly sophisticated ability to manipulate, design and develop novel
nanoparticles.
Chemicals at the nanoparticle size behave differently from chemicals at ‘normal size’. They
may be more conductive, have stronger bonds or be more chemically reactive than larger par-
ticles of the same substance. For example, gold, chemically inert at normal size, can serve as a
chemical catalyst at the nanoscale.113 Furthermore, surface charge – the potential electrical
energy between the particle surface and the medium in which it moves114 – appears to play a
greater role at the nanoscale.
As a consequence, substances made of or incorporating nanoparticles may have markedly
different, and often enhanced, mechanical, catalytic and thermal properties. Nanoparticles can
be utilised, for example, to create more resistant surfaces and to produce active filters to protect
against toxic chemical and biological threats. Also, nanotechnology ‘helps to enhance specific
biological reactions as drug delivery systems help to pass through biological barriers’,115 leading
to, inter alia, more efficient (targeted) drug delivery, and creating new opportunities for cell
imaging.
106 See especially Richard Feynman, ‘Plenty of Room at the Bottom’ (Talk given to the American Phys-
ical Society, Pasadena, December 1959), www.pa.msu.edu/~yang/RFeynman_plentySpace.pdf;
K Eric Drexler, Engines of Creation: The Coming Era of Nanotechnology (Anchor 1986).
107 Jeremy Ramsden, Nanotechnology: An Introduction (Elsevier 2011) 2.
108 Ibid.
109 Jennifer Kahn, ‘Nanotechnology’ National Geographic (June 2006) 98.
110 For definitions along these lines, see eg Interagency Working Group on Nanoscience, Engineering
and Technology, ‘National Nanotechnology Initiative: Leading to the Next Industrial Revolution’
(Committee on Technology, National Science and Technology Council, February 2000) 15; The
Foresight Institute, ‘About Nanotechnology’, www.foresight.org/nano.
111 Ramsden (n 107) 101–103.
112 Ibid.
113 See eg Naomi Lubick and Kellyn Betts, ‘Silver Socks have Cloudy Lining: Court Bans Widely Used
Flame Retardant’ (2008) 42 Environmental Science and Technology 3910.
114 See International Union of Pure and Applied Chemistry, Compendium of Chemical Terminology (2nd
edn, Blackwell Scientific Publications 1997).
115 Johann Ach and Beate Luttenberg (eds), Nanobiotechnology, Nanomedicine and Human Enhancement (Lit
2008).
Emerging technologies of warfare
617
What are often referred to as ‘nanorobots’ are effectively assemblies of atoms that have the
ability to trap, transport and alter other atoms. They are ‘programmable, potentially self-
replicating, molecular machines made of specifically arranged atoms’.116 Scientists working on
this cutting- edge research describe their beneficent vision of a future in which ‘nanorobots’
would ‘cruise around inside the body, communicating with each other and performing various
kinds of diagnoses and therapy’.117 The possibility of using ‘nanorobots’ to seek out specific
organs, and even specific cells, opens up new approaches to much less invasive, highly targeted
and hence more effective administration of drugs or of micro- surgical techniques.
As this brief discussion indicates, the applications for nanotechnology are many and they
span, inter alia, biotechnology, medicine, microelectronics, energy conversion and storage, coat-
ings, textiles, pharmaceuticals, cosmetics, food, manufacturing and security. Many of these
applications are already exploited in a wide range of commercial products.118 For example, gold
nanoparticles, mentioned earlier, are used in biosensors, cancer cell imaging and drug delivery
to cell membranes.119
In the same way that other areas of science and technology can be dual- purpose, there is little
doubt that nanotechnology may also be repurposed in a military context: the distinct properties
of materials at the nanoscale offer new potential applications for armed conflict. The potential
military uses of nanotechnology – both offensive and defensive – have been explored in the
literature, although not exhaustively.120 A notable difficulty in this respect is the futuristic, and
sometimes classified, nature of research in this area, which means that potential military nano-
technology applications and estimates of development time frames have to be extrapolated from
present science and technology.121
While many potential military applications of nanotechnology remain speculative, it is pos-
sible to give a few illustrations. The unique properties of matter observed at the nanoscale have
been utilised to improve sensors,122 to enhance protective vests worn by military and para-
military personnel123 as well as to develop new types of armour and communication equip-
ment.124 Nano- enhanced energy storage is being developed with implications for fuel cells,
soldier systems, small robots, RPVs and outer space technology.125 Nanotechnology may
116 Jun Wang and Peter J Dortmans, ‘A Review of Selected Nanotechnology Topics and their Potential
Military Applications’ Report no DSTO-TN-0537 (DSTO Systems Sciences Laboratory 2004) 4.
117 Kristna Weidner,’Nanomotors are Controlled for the First Time, within Living Cells’ Penn State News
(11 February 2014), http://news.psu.edu/story/303296/2014/02/10/research/nanomotors-are-con-
trolled-first-time-inside-living-cells.
118 Maxine McCall, ‘Nanoparticles and Nanosafety: The Big Picture’ Conversation (6 February 2014),
http://theconversation.com/nanoparticles-and-nanosafety-the-big-picture-22061.
119 Madhusudhan R Papasani, Guankui Wang and Rodney A Hill, ‘Gold Nanoparticles: The Importance
of Physiological Principles to Devise Strategies for Targeted Drug Delivery’ (2012) 8(6) Nanomedicine:
Nanotechnology, Biology and Medicine 804.
120 See in particular Wang and Dortmans (n 116); Jürgen Altmann, Military Nanotechnology (Routledge
2006); Margaret Kosal, Nanotechnology for Chemical and Biological Defense (Springer 2009); Wilson
Wong, Emerging Military Technologies: A Guide to the Issues (Praeger 2013) ch 5.
121 Altmann (n 120) 71.
122 Margaret Kosal, ‘The Security Implications of Nanotechnology’ (2013) 66 Bulletin of the Atomic Scien-
tists 58.
123 See eg Kanesalingam Sinnppoo, Lyndon Arnold and Rajiv Padhye, ‘Application of Wool in High-
Velocity Ballistic Protective Fabrics’ (2010) 80 Textile Research Journal 1083.
124 Wang and Dortmans (n 116).
125 Altmann (n 120) 78–79.
R Liivoja et al
618
improve the penetration capability and accuracy of projectiles, and enhance camouflaging.126
Furthermore, given its therapeutic potential, nanotechnology could also ‘facilitate weapons
based on enhanced delivery mechanisms for toxic substances, on tailored compounds capable of
targeting specific physiological functions, and on complex multilayered stealth designs’.127
3.2 Legal challenges
Since nanotechnology includes manipulation of chemical and biological functions, specific rules
which govern chemical and biological weapons may have relevance where nanotechnology is
used in the military context, even though no nanotechnology- derived weapons appear to be in
production as yet. In the chemical and biological warfare context, there are at least four broad
areas where nanotechnology has regulatory implications: (1) nano- enabled delivery methods,
(2) novel nanotechnology- based biochemical weapons, (3) nanoparticles and nanomaterials with
toxicological or deleterious health properties and (4) nanotechnology- enabled evasion of medical
countermeasures.128
The 1925 Geneva Protocol prohibits the use of ‘asphyxiating, poisonous or other gases, and
of all analogous liquids, materials or devices’ and ‘bacteriological methods of warfare’.129 This
prohibition is extended by the 1972 Biological Weapons Convention (BWC) and the 1993
Chemical Weapons Convention (CWC) to cover the development, production, acquisition,
stockpiling and transfer of biological and chemical weapons respectively.130 The creation of
synthetic DNA has effectively increased the ‘continuous biochemical threat spectrum’ with the
BWC and CWC prohibitions ‘overlapping in their coverage of mid- spectrum agents such as
toxins and bioregulators’.131
To determine their scope of application, the BWC and CWC use broad terms such as
‘microbial or other biological agents’, ‘toxins’ and ‘toxic chemicals’,132 making the conventions
applicable to any new technologies that involve chemical or biological agents used for hostile
purposes. The conventions, and the corresponding rules of customary international law, remain
applicable regardless of the size of the matter in question.
That said, the BWC is concerned with living organisms and substances produced by such
organisms. It has been suggested that artificially created nanoparticles or ‘nanorobots’ fall within
the scope of the BWC prohibition if they replicate the behaviour of known biological agents.133
On the other hand, nanoparticles that interact with their host through their chemical action on
126 Thomas Faunce and Hitoshi Nasu, ‘Nanotechnology and the International Law of Weaponry:
Towards International Regulation of Nano-Weapons’ (2010) 20 JLIS 21; Hitoshi Nasu, ‘Nano-
technology and Challenges to International Humanitarian Law: A Preliminary Legal Assessment’
(2012) 94(886) IRRC 653.
127 Juan Pardo-Guerra and Francisco Aguayo Ayala, ‘Nanotechnology and the International Regime on
Chemical and Biological Weapons’ (2005) 2 Law and Business 1.
128 Kosal, Chemical and Biological Defense (n 120) ch 4.
129 Protocol for the Prohibition of the Use of Asphyxiating, Poisonous or Other Gases, and of Bacterio-
logical Methods of Warfare (17 June 1925) 94 LNTS 65.
130 Convention on the Prohibition of the Development, Production, and Stockpiling of Bacteriological
(Biological) and Toxin Weapons and on their Destruction (10 April 1972) 1015 UNTS 163, arts 1
and 3; Convention on the Prohibition of the Development, Production, Stockpiling and Use of
Chemical Weapons and on their Destruction (13 January 1993) 1974 UNTS 45, art 1.
131 Mark Wheelis and Malcolm Dando, ‘Neurobiology: A Case Study for the Imminent Militarization of
Biology’ (2005) 87(859) IRRC 560.
132 BWC art 1(1); CWC art 2(2).
133 Robert Pinson, ‘Is Nanotechnology Prohibited by the Biological and Chemical Weapons Conven-
tions?’ (2004) 22 Berkeley Journal of International Law 279, 298–300.
Emerging technologies of warfare
619
life processes fall squarely under the prohibition contained in the CWC, regardless of their
size.134
Significantly, however, the prohibition of biological weapons extends to ‘equipment or
means of delivery designed to use such [biological] agents or toxins for hostile purposes or in
armed conflict’ and the definition of chemical weapons expressly includes ‘munitions and
devices’ for the release of toxic chemicals.135 This would also capture any nanotechnological
means of delivering biological agents or toxic chemicals into the human body.
Nanoscale objects can, however, affect human physiology by way of their physical properties
rather than by their chemical composition. For example, the inhalation of silver nanofibres has
been shown to cause an acute inflammatory reaction in the lungs, with the severity of reaction
dependent on the length of the fibre.136 Also, an accumulation of nanoparticles in the human
body – for example the build- up of nanogold, used as an anti- inflammatory agent for conditions
such as arthritis137 – could be manipulated externally (that is, heated) to cause damage to the
human body. Finally, ‘nanorobots’ could potentially be programmed to cause mechanical injury
at the cellular level without any biochemical action. The absence of a toxic chemical or disease-
causing organism in these instances suggests that the prohibitions contained in the BWC and
CWC could be inapplicable. Arguably, however, the broad customary law prohibition of pois-
oning, codified in the Hague Regulations,138 would prohibit such uses of nanotechnology.139
Also, causing tissue damage by means of nanoparticles or ‘nanorobots’ that due to their size or
other properties cannot be detected by industry- standard medical imaging devices may run afoul
of the ban on ‘any weapon the primary effect of which is to injure by fragments which in the
human body escape detection by X- rays’.140
Rules and principles on the protection of the environment may also play a role in the use and
application of nanotechnology for military purposes. The ENMOD Convention prohibits the
‘military or any other hostile use of environmental modification techniques having widespread,
long- lasting or severe effects as the means of destruction, damage or injury to any other State
Party’141 and Additional Protocol I bans methods or means of warfare ‘which are intended, or
may be expected, to cause widespread, long- term and severe damage to the natural
environment’.142 While the precise environmental impact of engineered nanomaterials and nan-
oparticles remains unclear, there is evidence that at least some of them pose significant risks to
the natural environment.143 This raises the question whether the precautionary principle recog-
nised in environmental law applies equally during times of armed conflict.144
134 Ibid 300–304.
135 BWC art 1(2); CWC art 2(1)(b)–(c).
136 Anja Schinwald, Fiona A Murphy, Adriele Prina-Mello, Craig A Poland, Fiona Byrne, Dania Movia,
James R Glass, Janet C Dickerson, David A Schultz, Chris E Jeffree, William MacNee and Ken Don-
aldson, ‘The Threshold Length for Fibre-Induced Acute Pleural Inflammation: Shedding Light on the
Early Events in Asbestos-Induced Mesothelioma’ (2012) 128(2) Toxicological Sciences 461.
137 Christine T N Pham, ‘Nanotherapeutic Approaches for the Treatment of Rheumatoid Arthritis’
(2011) 3 Wiley Interdisciplinary Reviews: Nanomedicine and Nanobiotechnology 607.
138 HR art 23(a).
139 On the scope of the prohibition on the use of poison and poisoned weapons, see Robert J Mathews,
‘Chemical and biological weapons’ ch 12 in this volume, in particular s 4.3.
140 Protocol (I) on Non-Detectable Fragments (10 October 1980) 1342 UNTS 168.
141 Convention on the Prohibition of Military or any Other Hostile Use of Environmental Modification
Techniques (10 December 1976) 1108 UNTS 151 art I.
142 API art 35(3).
143 See Nasu (n 126) and the sources cited therein.
144 See Laurent Hourcle, ‘Environmental Law of War’ (2001) 25 Vermont LR 653.
R Liivoja et al
620
4 Weapons review
Article 36 of Additional Protocol I provides an obligation to conduct due diligence in relation
to some military applications of technology:
In the study, development, acquisition or adoption of a new weapon, means or method of
warfare, a High Contracting Party is under an obligation to determine whether its employ-
ment would, in some or all circumstances, be prohibited by this Protocol or by any other
rule of international law applicable to the High Contracting Party.
Even though this provision is contained in Additional Protocol I that is applicable in armed
conflict, the effect of Article 36 is such that legal review needs to be undertaken irrespective of
whether the state in question is involved in an armed conflict.
Article 36 is broad in scope. In terms of the object of review, it refers to any ‘new weapon,
means and method of warfare’. This clearly covers, for example, novel projectiles as well as
innovative devices for propelling them. The reference to methods of warfare also subjects to
legal review military technology that, despite not being directly used to harm the adversary,
nonetheless changes the way in which hostilities are undertaken. Thus, for instance, armed
RPVs, even if they carry previously known and reviewed weapons, may become new means or
methods of warfare and thus subject to a legal review.
Article 36 is also extensive in terms of the relevant legal framework. The review obligation
extends well beyond the rules contained in Additional Protocol I itself. The phrase ‘any other
rule of international law applicable to the High Contracting Party’ clearly refers to other LOAC
treaties as well as customary international law.145 Moreover, the ICRC Commentary makes it
plain that the phrase also ‘refers to any agreement on disarmament concluded by the Party con-
cerned, or any other agreement related to the prohibition, limitation or restriction on the use of
a weapon or a particular type of weapon, concluded by this Party’.146 The broad language of
Article 36 suggests that the compatibility of weapons, means and methods of warfare with
applicable human rights law or environmental law obligations must also be assessed, to the
extent that those branches of international law apply in armed conflict.
The broad scope of Article 36 and the complexity of new military platforms can make the
legal review of new weapons, means and methods a difficult task. The reviewers of weapons
must not only have a thorough knowledge of the applicable law, they must also have an under-
standing of the ‘engineering design, production and testing (or validation) methods, and the way
in which the weapon might be employed on the battlefield’.147 Thus, the obligation under
Article 36 necessitates close collaboration between scientists, engineers and lawyers so as to
ensure compliance with the requirement to review weapons. Also, given the immediate military
applications of some advances in technology, Article 36 reviews may need to be conducted
earlier in the development process than previously.
Unfortunately, Article 36 suffers from a striking lack of national implementation. Only a
handful of states parties to Additional Protocol I, and two non- parties (the US and Israel), have
in place procedures for systematically conducting reviews of new weapons, means and methods
of warfare. Even when such reviews are conducted, they do not necessarily reveal all of the
145 AP Commentary ¶ 1472.
146 Ibid.
147 Alan Backstrom and Ian Henderson, ‘New Capabilities in Warfare: An Overview of Contemporary
Technological Developments and the Associated Legal and Engineering Issues in Article 36 Weapons
Reviews’ (2012) 94(886) IRRC 483, 484.
Emerging technologies of warfare
621
relevant legal issues: as pointed out earlier, some emerging technologies raise legal difficulties
not in relation to specific prohibitions or restrictions arising under LOAC, arms control law or
even human rights law, but because they challenge some of the assumptions that underlie these
legal regimes (for example, what constitutes ‘attack’ or how accountability with respect to
systems with a degree of autonomy is implemented). Thus, while a more diligent national
application of Article 36 should certainly be welcomed, even broader legal reviews may be
necessary for some types of emerging technology.
5 Concluding remarks
Rapid technological developments in the defence sector have justifiably caused concern, espe-
cially within civil society and among scholars. This has led to calls for additional regulation. It
has been suggested that ‘[t]he world needs a Geneva Convention for cybercombat’.148 It has
been argued that states should ‘control and regulate the proliferation of drones through the judi-
cious use of international law’149 and that they ‘should prohibit the creation of weapons that
have full autonomy to decide when to apply lethal force’.150 It has been claimed that ‘there is an
urgent need for regulating nano- weapons under the international law of weaponry’151 and that
‘[p]ast methods for other technologies are not adequate to deal with nanotechnology’152 such
that reducing the risk of misuse requires both ‘[t]raditional and innovative new approaches to
non- proliferation and counterproliferation’.153 In short, there appears to be a growing sentiment
in some quarters that the existing law is inadequate. Many LOAC specialists, on the other
hand, tend to be more optimistic. One commentator concludes that, at least with respect to tar-
geting rules, ‘the existing body of law is capable of being applied to novel weapons
technologies’.154
We take an intermediate position. We believe that much of LOAC is flexible enough to be
applied to any weapon technology that might be fielded. That said, some of the current scient-
ific and technological developments do point to weaknesses or uncertainties in the law. Some
of these problems could, however, be overcome by more up- to-date interpretations of the
current law, rather than the adoption of new, ever- more intricate rules. For instance, in the
context of cyber operations, revisiting the meaning of the term ‘damage’ would help clarify
some of the uncertainties. A major difficulty, however, is that the use of emerging technologies
– including specifically cyber capabilities and RPVs – constitutes an area where states have been
exceedingly reluctant to express their legal views clearly.155 We share the concern that this
148 Karl Rauscher, ‘It’s Time to Write the Rules of Cyberwar’ IEEE Spectrum (27 November 2013),
http://spectrum.ieee.org/telecom/security/its-time-to-write-the-rules-of-cyberwar.
149 Kaag and Kreps (n 87) 151; Sarah Kreps and Micah Zenko, ‘The Next Drone Wars: Preparing for
Proliferation’ (2014) 93(2) Foreign Affairs 68.
150 Human Rights Watch, Losing Humanity: The Case against Killer Robots (2012) 46.
151 Faunce and Nasu (n 126) 59; see also Altmann (n 120).
152 Kosal, ‘Security Implications’ (n 122) 59.
153 Kosal, Chemical and Biological Defense (n 120) 8.
154 William H Boothby, ‘Legal Challenges of New Technologies: An Overview’ in Nasu and McLaughlin
(eds) (n 5) 21, 25. Boothby concedes, though that
if novel technology should emerge which raises humanitarian concerns that cannot easily be
addressed by the application of existing law, it would be for the international community of states
to decide whether new, specific treaty regulation is required to address such concerns.
155 See Michael N Schmitt and Sean Watts, ‘The Decline of International Humanitarian Law: Opinio Juris
and the Law of Cyber Warfare’ (2015) 50 Texas International Law Journal 189.
R Liivoja et al
622
unwillingness to express opinio juris will have detrimental effects on the development of
LOAC.156
Finally, it is worth keeping in mind that efforts to ban outright military applications of par-
ticular kinds of technology may have unintended consequences: unqualified restrictions,
however well intentioned, can prove to be counter- humanitarian. As regards autonomous
systems, for example, it is worth recalling the 1988 incident where the US missile cruiser USS
Vincennes shot down an Iranian civilian airliner, mistaking it for an attacking military fighter.
The disaster could have been avoided, however, had the commander of the Vincennes relied on
the data supplied by the Aegis Combat System, which was consistent with a civilian aircraft.
This suggests that, at least in some situations, an autonomous system would react more appro-
priately to potential threats than a combatant. RPVs and nanotechnology serve as illustrations of
dual- use technology, the regulation of which has its inherent challenges. Any restriction of the
use of such technology would need to accommodate the research and development of beneficial
(civilian) applications.
156 Ibid.
